*Review* **Review of Research and Development of Hydraulic Synchronous Control System**

**Ruichuan Li 1, Wentao Yuan 1,\*, Xinkai Ding 1,\*, Jikang Xu 2, Qiyou Sun <sup>1</sup> and Yisheng Zhang <sup>1</sup>**


**Abstract:** Hydraulic synchronous control systems are widely used in various industrial fields. This paper deeply analyzes the research status and development trend of the hydraulic synchronous control system. Firstly, it gives a brief introduction of the research significance control theory and control methods of the hydraulic synchronous control system. Secondly, the hydraulic synchronization control system is classified, the synchronization error is analyzed, and some solutions to synchronization error are given. Then, according to the classification of the hydraulic synchronous control system, relevant research is carried out. In this paper, three control modes (equivalent, master–slave and cross-coupling) and related control algorithms (fuzzy PID control, sliding mode control, robust control, machine learning control, neural network control, etc.) of closed-loop hydraulic synchronous control systems are studied in detail. Finally, the development trend of the hydraulic synchronization control system is predicted and prospected, which can provide some reference for promoting the research and application of hydraulic synchronization technology in the future industrial field.

**Keywords:** hydraulic synchronization; closed loop control; synchronization error; control mode; control algorithm

#### **1. Introduction**

Hydraulic transmission has the advantages of large output power, stability and reliability, fast response speed, compact and flexible installation, and can achieve a wide range of stepless speed regulation. It is often widely used in large-scale mechanical equipment, metal processing equipment, large-scale metallurgical equipment, agricultural machinery equipment, aerospace, and other related industrial fields [1–4] and has become an important technical force to promote the development of mechanical equipment. Additionally, the hydraulic synchronous control system is an important component in the field of hydraulic transmission, which has a great influence on the research and development of this field.

The development of the hydraulic synchronous control system has gone through a long process. With the unremitting efforts of many scholars and experts, the hydraulic synchronous control system is also developing and progressing [3,5–8]. Joseph Bramah [9], an Englishman, invented the world's first hydraulic press in 1795, and the hydraulic field began to develop gradually. Around the 1870s, the steam engine-driven water pump was gradually applied to hydraulic equipment such as extruders and shears. In the 1930s, the integrated hydraulic system composed of a pump, valve, and actuator was gradually applied to the hydraulic synchronization equipment, and the hydraulic synchronization control system developed rapidly. In the late 1970s, the feedback element began to be applied in the hydraulic closed-loop synchronous control system. Compared with the previous open-loop control, the synchronization accuracy and response speed have been significantly improved. At the same time, the first fuzzy controller developed by the University of London in the United Kingdom was gradually applied to the hydraulic

**Citation:** Li, R.; Yuan, W.; Ding, X.; Xu, J.; Sun, Q.; Zhang, Y. Review of Research and Development of Hydraulic Synchronous Control System. *Processes* **2023**, *11*, 981. https://doi.org/10.3390/ pr11040981

Academic Editors: Francisco Ronay López-Estrada and Guillermo Valencia-Palomo

Received: 4 March 2023 Revised: 17 March 2023 Accepted: 21 March 2023 Published: 23 March 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

synchronous control system, which effectively solved the control problem of the nonlinear system. However, the fuzzy rules and system design methods are completely based on experience, and it is difficult to accurately and stably control the complex system. In 1981, Canadian scholar G. Zames [10] first proposed H∞ optimal control, and then robust control was widely used in electro-hydraulic servo systems and multi-cylinder hydraulic synchronous systems. Since the 20th century, with the development of electronic components and artificial intelligence, adaptive fuzzy PID, robust control, neural network and other algorithms have been gradually applied to hydraulic synchronization systems. Because of its adjustable control parameters and strong adaptability, it can achieve highprecision synchronization accuracy and good robustness in the control of complex nonlinear hydraulic synchronization systems.

The hydraulic synchronous control system usually requires the actuators in the hydraulic transmission mechanical equipment to operate synchronously with the same displacement or speed at the same time [11]. Generally, due to the existence of nonlinear factors such as leakage, impurities, machining error, and component wear in the hydraulic synchronization control system, it is difficult to ensure the accuracy of the hydraulic synchronization system. Although the existing conventional and simple hydraulic synchronization system is low in cost, it cannot meet the requirements of high-precision synchronization equipment. In order to make the hydraulic synchronization equipment have good operation performance, some relevant scholars have optimized the hydraulic components and hydraulic circuit, but the synchronization accuracy improved by this method is limited. Other scholars have used closed-loop control to apply relevant intelligent control algorithms to the hydraulic synchronization control system, and the control accuracy and reliability and stability have been greatly improved.

Gu et al. [12] focused on the phenomenon that the synchronous open-loop hydraulic synchronous control system of two hydraulic cylinders has no feedback regulation, and there is a constant accumulation of errors and low control accuracy. Through comparing several improvement schemes, the hydraulic synchronous circuit was optimized, and finally the scheme of reasonable configuration of the one-way valve by the synchronous motor for flow compensation was adopted. Through repeated optimization of the opening of the one-way valve, the synchronous error of the two cylinders in the full stroke of 500 mm was not more than 3 mm. Chen et al. [13] proposed a fuzzy control method for piston motion trajectory synchronization aiming at the synchronization error caused by the load imbalance of dual hydraulic cylinders. Based on the fuzzy control, two fuzzy controllers were introduced to eliminate the track tracking error of hydraulic cylinders and the synchronization error between hydraulic cylinders. The experiment shows that the composite control method can control the synchronization error within ±10 mm. Yang et al. [14] proposed a single neuron PID cross-coupling control strategy for the synchronization error caused by the uneven stress of the two hydraulic cylinders of the hydraulic bending machine during the working process, and compared it with the PID control effect under master–slave and cross-coupling. The simulation results showed that the maximum synchronization error of the master–slave PID control was 0.9 mm, and the maximum synchronization error of the cross-coupling PID control was 0.79 mm. The two-cylinder system controlled by neuron cross-coupling PID had a faster response, and the maximum synchronization error was 0.27 mm.

Traditional PID control needs to rely on accurate mathematical models because its parameters are not adjustable. It can achieve a good control effect in a linear system. For complex nonlinear hydraulic systems, some scholars combine PID control with fuzzy control and adaptive control to achieve high synchronization control accuracy. Zhang et al. [15] proposed a synchronization control strategy based on fuzzy PID control, which uses fuzzy rules to realize the real-time adjustment of PID control parameters. Taking the piston displacement of a hydraulic cylinder as an indicator, the joint simulation of AMESim and Simulink shows that the synchronization accuracy of double hydraulic cylinders under this control strategy is between 0 and 0.66 mm, which meets the operation requirements of the shear-type synchronous lifting mechanism. Liu et al. [16] investigated the hydraulic pin type lifting system of the jack up wind power installation ship, a two-stage hydraulic synchronous control system. The speed tracking and displacement coupling synchronization control strategy was adopted for single pile legs to improve the synchronization accuracy, and the fuzzy PID control algorithm was adopted to improve the robustness of the system. The joint simulation of AMESim and Simulink software showed that the maximum synchronization error of the sensor was still less than 0.3 mm when it was disturbed, the system had a higher synchronization accuracy, and had the advantages of stable operation, no overshoot and higher reliability. Li et al. [17] designed a fuzzy adaptive PID controller to be used in the steel plate hydraulic synchronization technology. The joint simulation of AMEsim and Simulink software showed that the displacement synchronization error of the two hydraulic cylinders was always less than 0.1 mm, and the proposed control strategy had a faster response speed, stronger stability, and a higher control accuracy. Zhao et al. [18] proposed a fuzzy PID controller based on particle swarm optimization to solve the problem of synchronization error caused by friction and uneven stress of a new type of hydraulic fan barring drive cylinder group. The particle swarm optimization algorithm was used to iteratively optimize the quantization factors Ke, Kec, and scale factor Ku of the fuzzy controller. Simulation analysis by Simulink software showed that the hydraulic synchronization system under particle swarm composite PID control had a higher synchronization accuracy than traditional PID control, and that fuzzy PID control had the advantages of smaller overshoot, shorter adjustment time, and stronger adaptive ability. Wang et al. [19] proposed a control method combining adaptive robust error sign integration control and extended state observation (ESO) control and combined them with cross-coupling control. The simulation results showed that the control strategy can achieve good synchronous tracking performance in the double hydraulic cylinder actuator system. Fang et al. [20] introduced fuzzy PID control into the hydraulic servo system of a C32 friction welding machine, which improved the closed-loop control characteristics of the welding process. However, due to the small design tonnage, the difficulty of multi-cylinder synchronous control was not considered. In order to solve the difficulty of multi-cylinder synchronization control accuracy, Wang et al. [21] proposed a synchronization controller based on a neural network to solve the synchronization error caused by the difficulty of the coordinated operation of the hydraulic multi-actuator synchronization system. Through the backstepping technology, the synchronization accuracy of multiple hydraulic cylinders was greatly improved. Finally, the effectiveness of the control strategy was verified by simulation and experimentation.

According to the above research, closed-loop control is often used in the hydraulic synchronization control system to improve the accuracy of the hydraulic synchronization control system. This is because there is no feedback signal regulation in the hydraulic synchronous control system under open-loop control, and its control accuracy depends entirely on the accuracy of the hydraulic components to control the actuator. In addition, the synchronization error of the hydraulic system will continue to accumulate under long-term operation, so its synchronization accuracy and anti-interference are poor. The hydraulic synchronous control system under closed-loop control adds online detection and feedback of the actuator displacement and speed between the controller and the controlled object, and the feedback value is compared with the preset control parameters. The controller outputs the signal adjustment amount to compensate the synchronization error of the hydraulic system, which can significantly improve the synchronization accuracy and stability of the hydraulic synchronous control system [22–25].

This paper analyzes the development process and research significance of the hydraulic synchronous control system, studies three control modes and related intelligent control algorithms of the closed-loop hydraulic synchronous control system of the hydraulic synchronous control system, summarizes the causes of synchronization errors, and puts forward some corresponding solutions to improve the control accuracy of the hydraulic synchronous system, Finally, based on the above research, the development trend

of hydraulic synchronous control system is predicted and prospects are proposed. With guaranteed control accuracy of the hydraulic synchronization system, the research in this paper can provide some measures to reduce the synchronization error, and has certain reference significance for the selection of control methods and control algorithms.

#### **2. Research Significance of the Hydraulic Synchronous Control System and Development History of Control Method**

#### *2.1. Research Significance of the Hydraulic Synchronous Control System*

Hydraulic synchronous control systems are a branch of hydraulic systems, and their research direction mainly includes hydraulic synchronous control methods and control algorithms. The hydraulic synchronous system has the advantages of high control accuracy, fast response speed, and easy application in high-power equipment, so it has a pivotal position in synchronous control systems [26], and its research significance has the following aspects.


#### *2.2. Development History of Hydraulic Synchronous Control Method*

With the development of modern control technology, all kinds of hydraulic equipment require a higher and higher accuracy of the hydraulic synchronous system. According to the development process of the synchronization control method, it can be divided into the first, second, and third generations of the synchronization control method [27].

The first generation of the hydraulic synchronization control method mainly adopts open-loop control to regulate the output of the actuator, using hydraulic components to change the pressure change in the hydraulic system to control the output of the hydraulic source and change the flow and pressure of the fluid in the hydraulic line, so as to control the synchronization movement of the actuator. The first-generation control method is low cost, simple, and practical, but the synchronization accuracy of the control system is not high and is generally used for industrial equipment with low control accuracy requirements.

With the social demand for high-precision machining and the improvement of sensor and control technology, the requirements for hydraulic synchronous equipment have gradually increased. The second-generation hydraulic synchronization control method widely uses the electro-hydraulic control method, introducing a signal (speed, load, displacement, etc.) feedback link, detecting feedback on the output quantity of the controlled object, comparing the feedback signal value with the ideal target value, and outputting the regulation electric signal to the electro-hydraulic valve by the controller, thus regulating the actuating element to reduce the synchronization error. The second generation of hydraulic synchronization control method adopts closed-loop control, mostly used for hydraulic system control using an electronic component system-hydraulic pilot control-hydraulic actuator, which regulates the synchronization error and increases the synchronization accuracy of hydraulic actuator.

With the further improvement of automation technology and control technology, higher requirements for hydraulic synchronization accuracy have been put forward. The third-generation hydraulic synchronization control method mainly uses a microprocessor combined with different control algorithms, introduces sensors (force sensors, displacement sensors, angle sensors, speed sensors, etc.) to compare the actual control accuracy and ideal control accuracy of the system, uses a microprocessor as the control unit, all kinds of sensors to detect and feedback the hydraulic synchronization control system, and combines different pairs of control algorithms to reduce synchronization interference and error. The third generation hydraulic synchronization control method has higher automatic intelligence, but its regulating actuators have a certain lag. In the future, the hydraulic synchronization control system can be digitized and intelligently controlled through the construction of a hydraulic characteristic information network, the collection of flow and pressure at the important nodes of the hydraulic circuit, and the active real-time regulation of the hydraulic synchronization control system by making full use of intelligent algorithms and information-variable technology.

#### **3. Analysis of Synchronization Error of Hydraulic Synchronous Control System**

To achieve synchronous control of the hydraulic system, ideally it is only necessary to allow equal pressure and equal amount of hydraulic oil to flow through the actuator with identical structural parameters. Generally, the hydraulic synchronization system has synchronization errors due to oil pollution and leakage in the hydraulic synchronization system, uneven friction and load of the actuators, asymmetrical arrangement, back pressure of the actuators, oil pressure fluctuation, manufacturing and installation errors of components, and other factors. Even if closed-loop control is adopted, the hydraulic synchronization system under actual working conditions is complex and nonlinear due to its control hysteresis. It is difficult to establish its accurate mathematical model to achieve high-precision control, so the generation of synchronization error is inevitable [28–33].

At present, the hydraulic synchronous control system mainly uses the error feedback correction method to reduce the error accumulation of the hydraulic actuator, so as to improve its synchronization accuracy, but the method has a certain lag. In the future, with the continuous development of artificial intelligence and big data, active feedback adjustment is expected to solve this problem. There are many reasons for synchronization error of the hydraulic synchronization control system, which are mainly summarized in the following 10 aspects.


#### **4. Research on Hydraulic Synchronous Control System**

The hydraulic synchronization control system is widely used because of its advantages such as easy control, simple structure, and being able to adapt to high power and complex environments. At the same time, the continuous improvement and optimization of the hydraulic synchronization system has always been a research hotspot in the hydraulic field [34]. According to the different application fields, control modes and tasks, the hydraulic synchronization system can be classified into several categories [35–37].

According to whether there is feedback in the hydraulic synchronization control system, it can be divided into "open-loop synchronization" or "closed-loop synchronization". Open-loop synchronization is on a simple scale, has low control accuracy, and poor robustness. Closed-loop synchronization can be widely used in various hydraulic equipment; because of the feedback regulation of its intelligent controller, its synchronization accuracy is significantly improved, its complexity and processing scale become larger, and the robustness of the system is better. In addition, different intelligent control algorithms can be combined in closed-loop control to select the best control scheme, but the closed-loop control system under complex machine learning and a neural network often has the phenomenon of over-learning, which increases the control cost of the system. In addition, the closed-loop hydraulic synchronization control system can be subdivided into "equivalent synchronization", "master–slave synchronization" and "cross-coupling synchronization", according to the different control modes.

According to whether the hydraulic synchronous control system has compensation, it can be divided into "uncompensated synchronization" and "compensated synchronization". The uncompensated hydraulic synchronous system is simple, has low control accuracy and poor robustness, while the compensated hydraulic synchronous control system timely compensates the synchronization error, so its control accuracy is high, and its robustness is strong. In addition, the compensated hydraulic synchronization system can be divided into "pre-valve compensation hydraulic synchronization" and "post-valve compensation hydraulic synchronization" according to the different compensation positions.

According to the different control elements, the hydraulic synchronization control system can be divided into "pump-controlled synchronization" and "valve-controlled synchronization". The pump-controlled synchronization efficiency is high, the response speed is slow, the control accuracy under constant load condition is high, and the valvecontrolled synchronization efficiency is low, but the robustness is good, the response speed is fast, and the synchronization accuracy under dynamic load condition is high. In addition, according to the different types of control valves, the valve-controlled hydraulic synchronization control system is divided into "flow synchronization", "electro-hydraulic proportional synchronization", and "electro-hydraulic servo synchronization".

In addition, other hydraulic synchronization systems are divided into "hydraulic cylinder synchronization" and "hydraulic motor synchronization" according to the different actuating elements. According to different control tasks, it can be divided into "position synchronization" and "speed synchronization". According to the number of actuators, it can be divided into "double actuator synchronization" and "multiple actuator synchronization". According to different installation methods, it can be divided into "horizontal synchronization" and "vertical synchronization". The classification of hydraulic synchronization control systems is shown in the following Figure 1.

This paper will carry out a detailed study on the hydraulic synchronization system according to the relevant classification in Figure 1 above.

**Figure 1.** Classification diagram of hydraulic synchronous control system.

#### *4.1. Open-Loop and Closed-Loop Hydraulic Synchronization Control System*

The hydraulic synchronous control system can be divided into open-loop hydraulic synchronous control systems and closed-loop hydraulic synchronous control systems according to whether there is feedback signal adjustment [38].

#### 4.1.1. Open-Loop Hydraulic Synchronous Control System

There is no feedback signal regulation in the hydraulic synchronization system of openloop control, and its synchronization control accuracy depends entirely on the accuracy of the hydraulic components (such as step valve, throttle valve, and speed regulating valve) to control the actuator. The output signal of the actuator is not detected and fed back in the open-loop control hydraulic synchronization system, so the accuracy and anti-interference of the open-loop control are poor. Common open-loop synchronous control circuits include a mechanical rigid synchronous circuit, flow control valve synchronous circuit, series cylinder synchronous circuit, synchronous cylinder synchronous circuit, parallel motor, or parallel pump synchronous circuit. Generally speaking, because of its low cost and simple structure, the open-loop control hydraulic synchronization control system is widely used in situations with small load, low synchronization accuracy requirements, simple control loop, and close synchronization distance. The open-loop control of the hydraulic synchronization system, due to the absence of feedback compensation, will make the synchronization error gradually accumulate and increase, and cannot eliminate the errors caused by hydraulic oil leakage, load angle bias, and the manufacture and installation of hydraulic components.

To make the hydraulic synchronization control system under open-loop control realize accurate synchronization, Ding et al. [39] proposed a load sensing synchronization control method consisting of a load sensing unit and a synchronization valve. The load sensing unit consists of a load sensing pump and a load sensing valve. The load sensing pump provides the pressure and flow required by the system through the pressure-closed-loop control. The load sensing valve can improve the ability to resist offset load through pressure compensation. The synchronizing valve is located between the load sensing pump and the load sensing valve to achieve an even distribution of the flow supplied by the pump. The experimental results show that compared with the traditional synchronous valve control, this control mode has stronger control accuracy, efficiency, and synchronous speed regulation capability.

#### 4.1.2. Closed-Loop Hydraulic Synchronous Control System

With the development of modern control theory and control technology, various closed-loop hydraulic synchronous control systems are widely used in high-precision hydraulic synchronous drive equipment [22,40,41].

The hydraulic synchronous control system under closed-loop control has a high synchronization accuracy because of the feedback regulation. The controller can conduct online feedback comparison between the output value and the ideal setting value and reduce the control deviation after regulation. The control modes of the hydraulic synchronous control system can be divided into master–slave control, equivalent control, and cross-coupling control according to the different feedback control modes of output values [42,43].

Lorenz et al. [44] proposed two synchronous control modes: equivalent control and master–slave control. Equivalent control [45–47] is a parallel structure, in which multiple hydraulic actuators form a closed loop independently, tracking the target output value signal set by the control system, so that each actuator can achieve synchronous control. Since each actuator in the equivalent control synchronizes the feedback signal independently and tracks the motion error, the synchronous tracking error of each actuator determines the error of the equivalent control. The equivalent control block diagram of the hydraulic synchronization control system is shown in Figure 2.

**Figure 2.** Equivalent control block diagram of hydraulic synchronous control system.

Equivalent control requires that each hydraulic component and electrical component of the hydraulic synchronization system must have a high matching consistency, and if the consistency of each component is good, the hydraulic system with equivalent control has a high synchronization accuracy.

The master–slave control [48–50] is a series structure. First, the output of an actuator is selected as the ideal output signal in the master–slave controlled hydraulic synchronization system; that is, one actuator is the main hydraulic oil circuit, and the other actuator is the slave hydraulic circuit. Then, the error feedback signal in the main circuit is input to the other slave hydraulic circuits through the feedback detection elements in the main hydraulic circuit. The master–slave control has a delay feedback that means the slave hydraulic circuit follows the signal of the main hydraulic circuit; especially in the start stop phase of the hydraulic synchronous control system, this delay is more significant. The slave actuating element has a large tracking lag to the master actuating element, which will cause the master–slave control error. The master–slave control block diagram of the hydraulic synchronization control system is shown in Figure 3.

**Figure 3.** Master–slave control block diagram of hydraulic synchronous control system.

There is a certain delay error in the master–slave-controlled hydraulic synchronization system, and to reduce the tracking delay error of the master–slave-controlled hydraulic circuit, a front feedback device needs to be added to feed back the synchronization error signal of the system to the control link [51].

There is no coupling relationship between the master–slave control and the hydraulic branches of the same control, so the controller is relatively simple, and the synchronization accuracy is relatively general. Therefore, the cross-coupling control is gradually studied and applied in the hydraulic synchronization control system by scholars. Cross-coupling control [52,53] establishes a coupling relationship between hydraulic branches, detects the output deviation values of multiple hydraulic branches, and feeds them back to the controller. When the synchronous deviation exceeds a certain range, the synchronous controller immediately compensates for the coupling of two hydraulic actuators, thereby effectively reducing the synchronous error. The cross-coupling control combines the advantages of master–slave control and equivalent control, and the synchronization accuracy is further improved. The cross-coupling control block diagram of hydraulic synchronization control system is shown in Figure 4.

**Figure 4.** Cross-coupling control block diagram of hydraulic synchronous control system.

However, the cross-coupling control has certain limitations for any hydraulic system with more than two hydraulic branches. When there are many controlled hydraulic actuators, the control mode will become too complex, and the robustness, stability, and synchronization of the hydraulic synchronization control system will become worse. To solve the synchronization accuracy of multiple actuators, adjacent cross-coupling is gradually being studied by more and more scholars. Adjacent cross-coupling hydraulic synchronization control [54,55] is to locally synchronize the actuating elements. When the controller synchronizes each hydraulic branch, it outputs synchronization error feedback for its own hydraulic branch and outputs synchronization error feedback to its adjacent hydraulic branch, to control the output error compensation of adjacent hydraulic branches and achieve synchronization control of each hydraulic branch. The adjacent cross-coupling control block diagram of the hydraulic synchronization control system is shown in Figure 5.

**Figure 5.** Adjacent cross-coupling control block diagram of the hydraulic synchronization control system.

Wu et al. [56], based on the adjacent cross-coupling synchronization control, designed the adjacent cross-coupling fuzzy self-tuning integral separation PID synchronization controller by combining fuzzy control with integral separation PID control, combined with three hydraulic synchronous closed-loop control modes (master–slave, equivalent, and adjacent cross-coupling). Through the joint simulation with AMESim and Simulink software, the three closed-loop synchronous control methods were compared and analyzed. The results show that in the four cylinders synchronous control system, the overshoot under the master–slave synchronous control mode is 6%, and the maximum synchronous error is 0.26 mm; the overshoot under the same synchronous control mode is 4.2%, and the maximum synchronous error is 0.15 mm; the overshoot of adjacent cross-coupling synchronization control mode is 2.8%, and the maximum synchronization error is 0.12 mm. The reason for the above phenomenon is that the master–slave synchronous control first sets a main hydraulic branch, and the output of the other slave hydraulic branches should follow the set main hydraulic branch, so there will be some hysteresis in the control process. All controllers of equivalent synchronous control follow the same set ideal output value, and the actuating elements of each hydraulic branch are not regulated by synchronous controller. The adjacent cross-coupling synchronous control can not only track and control the given ideal output value, but also compensate the synchronous error of the actuating elements of different adjacent hydraulic branches. Therefore, the control accuracy of adjacent cross-coupling synchronous control under closed-loop control is higher than that of master–slave control and equivalent control.

Compared with master–slave control and equivalent control, cross-coupling control can obtain a higher accuracy in the hydraulic synchronous system. Therefore, more and more scholars are combining cross-coupling control with other control methods to further develop the field of cross-coupling control. Sun et al. [57] proposed an adaptive robust cross-coupling control strategy by combining the cross-coupling control mode with the adaptive robust control, and its maximum synchronization error is not more than ±0.1 mm, which further improves the synchronization accuracy of the hydraulic press slide leveling electro-hydraulic control system. Liu et al. [58] proposed a control method based on fuzzy adjacent coupling for synchronous control of multiple hydraulic cylinders to be leveled, designed an adjacent coupling controller, fed back the position synchronization error of two adjacent hydraulic cylinders to the synchronous controller, and used the fuzzy controller to obtain the parameters that are difficult to obtain. The simulation and experimental results show that the fuzzy adjacent coupling control can quickly reduce the synchronization error of the hydraulic cylinder in the acceleration and deceleration phase.

Through the above research on the equivalent, master–slave, cross-coupling, and adjacent cross-coupling control of closed-loop hydraulic synchronous control systems, it can be concluded that their control effects and characteristics are shown in Table 1 below:

**Table 1.** Comparison of the effects of four control methods for closed-loop hydraulic synchronization systems.


*4.2. Hydraulic Synchronous Control System without Compensation and with Compensation*

The hydraulic synchronization control system can be divided into the "compensated hydraulic synchronization control system" and "non-compensated hydraulic synchronization control system", according to whether there is compensation in the hydraulic circuit.

#### 4.2.1. Hydraulic Synchronous Control System without Compensation

In the hydraulic synchronous control system without compensation, the actuator is not compensated, and the accuracy of the hydraulic synchronous control system is determined by the manufacturing and installation errors of the hydraulic components. Generally speaking, the control accuracy of the hydraulic synchronous control system without compensation is relatively low, and it is mostly used in situations where the load is not large, and the synchronous accuracy is not required to be too high.

#### 4.2.2. Compensated Hydraulic Synchronous Control System

In the hydraulic synchronous control system with compensation, when the load changes, the hydraulic oil flow from the hydraulic pump to the actuator via the hydraulic valve will change, affecting the control accuracy of the hydraulic synchronous system. The compensated hydraulic synchronous system introduces the compensation valve to regulate the flow, or through the sensor to feedback the error value to the control signal. Through the corresponding control algorithm, the signal to be compensated and adjusted is input to the hydraulic pump/hydraulic motor and control valve to compensate and regulate the flow and pressure in the hydraulic circuit. The control system belongs to closed-loop control. Because it can adjust the flow and pressure of the synchronous control system, the control accuracy of the hydraulic synchronous control system is significantly improved. According to different positions of the compensation valve and control valve, it can be divided into either a "pre-valve" compensation or "post-valve" compensation hydraulic synchronous control system [59,60].

The pre-valve compensation hydraulic synchronization control system is to set the pressure compensation valve at the front of the control valve port. The spring chamber of the pressure compensation valve receives the outlet pressure of the control valve port, and the non-spring chamber is connected to the inlet pressure of the control valve port. The differential pressure before and after the two control valves for valve front compensation is equal and both are constant. The flow that can be achieved is only related to the opening of the control valve and is not affected by the load change.

The post-valve compensation hydraulic synchronization control system is to arrange the pressure compensation valve behind the throttle valve. The spring chamber of the pressure compensation valve receives the maximum load pressure selected by the shuttle valve, and the non-spring chamber is connected to control the pressure behind the controllable throttle valve. The differential pressure before and after the two control valves in the post-valve compensation is equal but not constant; when the flow provided by the pump cannot meet the requirements, the flow of the two branches will be reduced in equal proportion. The actuator with the large load will not stop working, and the function of anti-flow saturation can be realized. The schematic diagram of the hydraulic synchronous control system for "pre-valve" compensation and "post-valve" compensation is shown in Figure 6 [59].

With the development of electro-hydraulic control and hydraulic energy-saving technology, more and more scholars have studied the compensated hydraulic synchronous control system and made a series of achievements. Li [61] established an integrated platform for the synchronous test of sunken ships, analyzed the synchronous hydraulic system of active compensation and passive compensation of sunken ships, further designed the overall hydraulic system of the hydraulic heave compensation test platform, and verified the accuracy of platform synchronous motion and the effectiveness of active compensation through tests. Zheng [62] studied the hydraulic synchronization system for lifting and fishing operations, designed a semi-active wave hydraulic cylinder synchronization compensation system with hydraulic cylinders in series–controlled by using double closedloop PID and compared and analyzed through Matlab software simulation: master–slave PID synchronization control, parallel PID control and cross-coupling PID synchronization control. The results show that the error decay rate of master–slave PID synchronization control is 47.26%, the error decay rate of parallel PID synchronization control is 16.81%,

and the error decay rate of cross-coupling PID control is 69.56%. It further proves that the cross-coupling PID synchronization control can quickly reduce the synchronization error and achieve stable conditions, and its control effect is the best.

**Figure 6.** Schematic diagram of compensated hydraulic synchronous control system. (**a**) Schematic diagram of pre-valve compensation hydraulic synchronization control system; (**b**) Schematic diagram of post-valve compensation hydraulic synchronization control system.

Zhou et al. [63] studied the valve-controlled asymmetric hydraulic cylinder synchronization system, proposed a two-layer fuzzy controller to improve the response speed of the hydraulic cylinder, and designed a feedforward compensation control mode under cross-coupling to reduce the synchronization error. The simulation results show that the tracking accuracy and synchronization accuracy of the hydraulic cylinder with dual compensation under uneven load have been greatly improved.

#### *4.3. Pump-Controlled and Valve-Controlled Hydraulic Synchronous Control System*

Hydraulic synchronization control systems can be divided into "pump-controlled hydraulic synchronization control systems" and "valve-controlled hydraulic synchronization control systems", according to different control modes and control valve components [64].

#### 4.3.1. Pump-Controlled Hydraulic Synchronous Control System

Pump controlled is also called volume controlled [65]. The pump-controlled hydraulic synchronous control system controls the synchronous movement of the actuating elements by changing the displacement of the hydraulic pump to make each actuator input the same amount of oil. Compared with the valve-controlled hydraulic synchronization system, the pump-controlled hydraulic synchronization system adopts the way of pump controlling the actuating elements, which reduces the number of hydraulic valves and the layout of hydraulic pipelines, significantly reduces the overflow throttling loss of the hydraulic system, the leakage of hydraulic oil, the friction heat of the system and other energy loss, and improves the efficiency of the system. However, under the same conditions, the natural frequency of the pump control system is much lower than that of the valve control system, which makes the response speed of the pump control system not high. The volumetric efficiency of the pump-controlled hydraulic synchronization system is high, and a low synchronization error can be obtained under the condition of constant load or small load change.

Peng et al. [66] designed a closed-loop digital PID control algorithm to control the speed of the hydraulic pump-controlled motor due to the synchronization error problem during its operation, and verified the good stability and rapidity of the system by optimizing the PID control parameters through simulation. Zhang et al. [67] put forward a loader lifting device with closed pump-controlled three chamber hydraulic cylinder, which adds a potential energy recovery chamber connected with the accumulator, and changed the original asymmetric two chamber hydraulic cylinder of the boom into a symmetric hydraulic cylinder. The simulation proved that the valve control system can not only significantly reduce energy consumption, but also significantly improve the response speed and control accuracy of the boom.

#### 4.3.2. Valve-Controlled Hydraulic Synchronous Control System

The valve-controlled hydraulic synchronization control system realizes synchronization by controlling hydraulic valves (proportional valves, servo valves, pressure compensation valves, synchronization valves, diverter valves, etc.) and specially designed valve blocks. The valve control system usually selects proportional and servo valves as its control components, which can be known from the basic formulas of pump control system and valve control system [68]: in the case of the same controlled object, the natural frequency of the valve control system is greater than that of the pump control system. The valve control system has faster response and better dynamic characteristics. Therefore, the valve control system is mostly used under the working condition of dynamic load. The components of the valve-controlled hydraulic synchronous control system are arranged separately, which does not affect the performance of the system and is more convenient for maintenance. Due to the throttling speed regulation of the throttle port and overflow valve and the centralized oil supply of the system, the valve-controlled hydraulic synchronous system has the disadvantages of large throttling loss, large pressure loss, large hydraulic oil leakage, and low system efficiency, resulting in large system power loss.

The valve-controlled hydraulic synchronization control system can be divided into "flow hydraulic synchronization control systems", "electro-hydraulic proportional hydraulic synchronization control systems" and "electro-hydraulic servo hydraulic synchronization control systems" according to different types of control valves.

The flow hydraulic synchronization control system [69,70] regulates the flow in the hydraulic circuit through the flow control valve to make the flow into and out of the hydraulic cylinder/hydraulic motor equal, thus realizing the synchronization of the hydraulic system. The flow hydraulic synchronous system often uses a speed-regulating valve, diversion and collection valve, and throttle valve to control the hydraulic circuit. The synchronization accuracy of the hydraulic synchronization system of the speed-regulating valve is generally less than 4–5%, which is often used in situations where the flow is not very large, the speed change is small, and the synchronization accuracy is required to be high. The synchronization accuracy of the hydraulic synchronization system of the diverter and collector valve is generally 1–3%, its deviation correction capability is large, its use is simple, and its cost is low. It is often used in situations where the hydraulic circuit is simple and the flow and load change range are not large. The hydraulic synchronizing system of the throttle valve has low synchronization accuracy, but its cost is low, adjustment is simple, and multi cylinder synchronization can be achieved. It is often used in situations where the synchronization accuracy is not high, the flow is small, the load is not large, and the load is stable.

The electro-hydraulic proportional hydraulic synchronization control system [71,72] is controlled by the electro-hydraulic proportional valve. The electro-hydraulic proportional valve is a new type of electro-hydraulic control element, which is mainly composed of a proportional electromagnet and a working valve core, which can change the displacement of the working valve core according to the size of the electrical signal of the system, to adjust the flow and pressure at both ends of the electro-hydraulic proportional valve. Compared with the traditional hydraulic synchronization control system, the electro-hydraulic

proportional hydraulic synchronization system has greatly improved the response speed, low cost, strong anti pollution ability, and high control accuracy. The electro-hydraulic proportional hydraulic synchronous control system is generally composed of a controller, electro-hydraulic proportional unit, actuator, controlled object, etc. Its control block diagram is shown in the following Figure 7.

**Figure 7.** Block diagram of electro-hydraulic proportional hydraulic synchronization control system.

The electro-hydraulic servo hydraulic synchronization control system [73,74] is controlled by the electro-hydraulic servo valve. The electro-hydraulic servo valve has the advantages of small dead zone, zero covering, fast dynamic response, and high oil cleanliness. The electro-hydraulic servo hydraulic synchronous control system can adjust the flow into the hydraulic actuator at any time, with good dynamic characteristics and high control accuracy [75]. The electro-hydraulic servo hydraulic system has the characteristics of intrinsic nonlinearity and parameter nonlinearity. Intrinsic nonlinearity mainly refers to the dead zone characteristics of control elements and the coupling characteristics of flow in the system, while parameter nonlinearity mainly refers to the uncertainty of temperature change, load change, and oil leakage. The synchronization error caused by the above nonlinear factors can be reduced by selecting an effective control algorithm. The electro-hydraulic servo hydraulic synchronization control system is generally composed of a servo amplifier, controller, electro-hydraulic servo valve, and actuating element. Its control block diagram is shown in Figure 8.

**Figure 8.** Block diagram of electro-hydraulic servo hydraulic synchronization control system.

With the development of modern control, more and more scholars have carried out research related to hydraulic synchronous control systems under valve control and pump control. Prabhakar et al. [76] compared the dynamic performance of pump control and valve control for a hydraulic motor, and the results show that the valve-controlled hydraulic motor system is more sensitive. Zhao et al. [77] established the control model for the four-motor hydraulic synchronous control system controlled by an electro-hydraulic proportional valve based on a fuzzy self-tuning PID control strategy. AMESim/Simulink software simulation analysis showed that the electro-hydraulic proportional valve-controlled four-motor hydraulic synchronous control system has a fast response and high synchronization accuracy.

In addition, some scholars combine valve control with pump control. The new synchronous control system with pump control as the main and valve control as the auxiliary can integrate the respective advantages of valve control and pump control, making the system have the characteristics of both control accuracy and response speed. Dong et al. [78] applied valve control and pump control to the synchronous system of a hydraulic excavator to improve the response speed and control accuracy of the system, designed the flow pressure to match the independent control of the inlet and outlet, applied it to the hydraulic excavator system to reduce the pressure fluctuation of the hydraulic excavator boom during operation from 6.9 Mpa to 1.7 Mpa, and at the same time, the energy consumption of one working cycle of the boom was reduced by 15%. The control strategy significantly

reduces the working pressure difference at the valve port and improves the system stability and energy utilization.

#### *4.4. Other Hydraulic Synchronization Control Systems*

According to the different actuators [79], the hydraulic synchronization control system can be divided into the "hydraulic cylinder synchronization" and "hydraulic motor synchronization" systems. The hydraulic motor synchronization control system can be divided into "gear motor synchronization system" and "plunger motor hydraulic synchronization system" according to the type of hydraulic motor. Ideally, by comparison, the synchronization accuracy of the plunger hydraulic synchronous motor can reach more than 99%, and the synchronization accuracy of the gear hydraulic synchronous motor is about 95%, and the external load has a great impact on it.

According to different motion states [80], the hydraulic synchronization control system can be divided into "position hydraulic synchronization control system" and "speed hydraulic synchronization control system". Position synchronization requires that different hydraulic actuators have the same position at the same time during movement or when static. Speed synchronization requires that different hydraulic actuators have the same speed.

According to the number of actuators [81], the hydraulic synchronous control system can be divided into "double actuator" and "multi-actuator" hydraulic synchronous control systems. The hydraulic synchronous system with double actuators has less synchronous error accumulation due to fewer actuating elements. The synchronous system of multiple actuators has many components in the circuit, and there are manufacturing, installation, and leakage errors among the hydraulic components, which makes the accumulation of synchronous errors among its mechanisms relatively large. To improve the synchronization accuracy of multiple actuators, closed-loop control and flow compensation are often used.

According to different installation position [82], the hydraulic synchronization control system can be divided into "horizontal" and "vertical" hydraulic synchronization control system. The actuating element of the horizontal hydraulic synchronous system is fixed horizontally, as the hydraulic cylinder synchronous control system is not affected by its own gravity, and the change in the hydraulic cylinder's own gravity will not affect the synchronization accuracy. In comparison, the vertical hydraulic cylinder synchronization system has its own load under the vertical installation of the hydraulic cylinder, and the uneven change of the oil quality in the hydraulic cylinder also has a certain impact on the synchronization accuracy of the system.

#### **5. Research Status of Control Algorithms of Hydraulic Synchronous Control System**

With the development of modern control methods, hydraulic synchronous systems are more widely used in the engineering field, and at the same time, people also put forward higher requirements for hydraulic synchronization accuracy, anti-interference, stability, and response speed. The hydraulic synchronous system mostly adopts closed-loop control, and the simple closed-loop control can no longer meet the requirements of high-precision synchronous control due to the hysteresis of error feedback and the influence of nonlinear factors. Therefore, scholars at home and abroad combine different control algorithms with the hydraulic synchronous system to further improve the accuracy, stability, and anti-interference ability of the hydraulic synchronous system. The control algorithms of the hydraulic synchronous control system mainly include fuzzy control, PID control, fuzzy PID control, auto disturbance rejection control, sliding mode control, robust control, adaptive robust control, machine learning control, neural network control, and networked control, etc.

#### *5.1. Fuzzy Control*

Fuzzy control makes the control variables of the hydraulic synchronous system fuzzy, uses fuzzy rules, and finally defuzzles to simulate human reasoning and decision-making, to realize the control of the hydraulic synchronous system. The fuzzy control block diagram of the hydraulic synchronization control system is shown in Figure 9.

**Figure 9.** Fuzzy control block diagram of hydraulic synchronization control system.

Fuzzy control is mostly applied to hydraulic synchronous systems that are difficult to establish accurate mathematical models. However, this control algorithm will have a certain steady-state error, which requires researchers to summarize a lot of experience. The accuracy of the control system needs to be guaranteed by establishing complete fuzzy control rules. Liu et al. [83] established a fuzzy feedforward controller for symmetric double hydraulic cylinders, and simulation results show that the synchronization error between the two cylinders is within 15 mm, which verifies the feasibility of the dual hydraulic cylinder synchronization control system under fuzzy control. Chen et al. [84] designed a dual-integrated fuzzy controller based on fuzzy control, in which the fuzzy coordinator is used to process the synchronization error signal of the dual hydraulic cylinders, and the fuzzy tracking controller is used to detect the tracking error of the tracking dual hydraulic cylinders. The test results show that the maximum synchronization error of the two hydraulic cylinders is stable within ±4 mm under the working conditions of unbalanced load, system uncertainty, and large change of system interference.

The synchronization error of the hydraulic synchronous system under the fuzzy control is relatively large, and to make it have a higher synchronization accuracy, it needs to combine other algorithms and control methods to optimize. Li et al. [85] proposed a cross-coupled hydraulic fuzzy synchronization control strategy based on the fuzzy control principle and applied it to the two plug-in main valves of the plug-in electro-hydraulic servo valve, and the simulation results show that the control strategy can effectively compensate the synchronization error. Jia [86] designed a fuzzy-PID controller for the multi-stage hydraulic cylinder synchronization control system by combining fuzzy control and PID control in Simulink. The joint simulation of AMESim and Simulink shows that the synchronization accuracy of the multi-stage hydraulic cylinder synchronization system under fuzzy PID control is more stable and higher than that under fuzzy control.

#### *5.2. PID Control*

The PID algorithm [45,46,87,88] is widely used in hydraulic synchronous systems in recent years due to its simplicity, easy implementation, simple algorithm, and good robustness. Compared with fuzzy control, PID control has a better control effect in linear systems. PID control is composed of a proportional factor P, integral factor I, and differential factor D, and when these three parameters are adjusted properly, the hydraulic synchronous system can obtain a good control effect. The block diagram of PID control of the hydraulic synchronization control system is shown in Figure 10.

**Figure 10.** PID control block diagram of hydraulic synchronization control system.

Wei et al. [89] studied the step response of the four-cylinder synchronous platform under different parameters based on PLC control and the PID algorithm, and the results showed that the top mold hydraulic system under the PID control algorithm had a faster response speed, stable operation, and no overshoot. Li et al. [90] compared and analyzed the dynamic control effects of conventional PID and fractional PID on the motion of hydraulic cylinders based on the synchronous control strategy and method of double-cylinder fourcolumn hydraulic press, and the simulation results show that the fractional PID controller has better anti-interference ability and robustness.

PID control has certain limitations in the control of hydraulic synchronous system because its parameters are not adjustable, and it depends on accurate mathematical models. Therefore, some scholars combine PID control with fuzzy control, adaptive control, neural network control, and other algorithms to further improve the control accuracy of the hydraulic synchronous system. Huang et al. [91], based on the control mode of deviation coupling, used an algorithm to optimize the PID control parameters, and studied the synchronization performance of multi-channel hydraulic cylinders with different parameters. The simulation results show that the control strategy has the characteristics of rapid system response, no overshoot, and a high steady-state accuracy. Yu et al. [92] combined the fuzzy adaptive control and PID control to realize the real-time adjustment of PID control parameters, and adopted the master–slave control mode for the hydraulic lifting synchronization system. The simulation shows that the improved control strategy can make the system respond faster and become more stable when the synchronization error is within the allowable range.

#### *5.3. Fuzzy PID Control*

Some scholars have combined fuzzy control and PID control to apply them to the hydraulic synchronous control system. First, the fuzzy controller performs fuzzy inference on the input control quantity, and then adjusts the parameters of the PID controller according to the working conditions of the hydraulic synchronous system. Then, the PID controller regulates the hydraulic synchronization system based on the deviation signal, thereby controlling the hydraulic actuator to synchronize the movement. The fuzzy PID control block diagram of the hydraulic synchronization control system is shown in Figure 11.

**Figure 11.** Fuzzy PID control block diagram of hydraulic synchronization control system.

Compared with traditional PID control, fuzzy PID control [93] combines the advantages of fuzzy control and PID control. It can not only realize the real-time adjustment of PID control parameters, but also apply to nonlinear systems. Moreover, the hydraulic synchronous system under this control strategy has a higher control accuracy and better control effect. Liu et al. [94] used the fuzzy PID control strategy to control the two-way electro-hydraulic dual-cylinder synchronous drive system. The PID control parameters can be automatically adjusted according to the change in the working condition error during the control process; the simulation results show that there is a small following error between the slave cylinder and the active cylinder under the fuzzy PID control, and the synchronization accuracy is high.

To make the synchronization control effect better, some scholars have proposed multiple fuzzy PID control strategies. Based on the dual fuzzy PID control with compensation factors, Mu et al. [95] proposed an adaptive fuzzy controller based on a model reference to control the hydraulic circuit of three hydraulic cylinders synchronization control, which reduced the impact of the hydraulic synchronization system due to the changes in the nonlinear friction, load, and operating parameters.

The combination of fuzzy PID control, adaptive control, neural network control, and other algorithms can further improve the control effect of the hydraulic synchronous system [96]. Mu et al. [97] designed a fuzzy parameter adaptive PID synchronization control strategy based on master–slave synchronization control for the hydraulic synchronization system of the inclined gating machine and added an accumulator pressure maintaining circuit at the outlet of the hydraulic pump of the hydraulic synchronization system. The simulation results of Simulink and AMESim software show that the control strategy has better synchronization accuracy and was more energy saving than the traditional PID hydraulic synchronization control strategy. Wu et al. [56] compared the synchronization control accuracy of the fuzzy PID controller and the fuzzy self tuning integral separation PID controller. The test showed that the overshoot amount of the fuzzy PID control algorithm was 4.8%, and the maximum synchronization error was 0.19 mm; the overshoot amount of fuzzy self-tuning integral separation PID control algorithm was 2.8%, and the maximum synchronization error was 0.12 mm. The reason for this phenomenon is that the fuzzy self-tuning integral separation PID controller integrated with the integral separation strategy can close the integral link when the synchronization error is large to prevent the control flow from supersaturation, so its control accuracy is higher than the fuzzy PID control.

#### *5.4. Auto Disturbance Rejection Control*

Active disturbance rejection control (ADRC) [98,99] can effectively solve the problem of system disturbance; the algorithm can estimate and compensate the disturbance to the hydraulic synchronous control system in real time, and the system response is more rapid and accurate. Compared with PID control, ADRC is more convenient to adjust parameters and has better control accuracy.

Ren et al. [100] designed an auto disturbance rejection controller to counteract the internal and external disturbances to the speed system, and added PID control in the feedback loop, which enhanced the dynamic adjustment ability and anti-interference ability of the system under disturbances. This significantly improved the synchronization accuracy of the dual electro-hydraulic position servo hydraulic synchronization control system. Lu et al. [101] introduced the auto-disturbance rejection control method into the hydraulic suspension multi-cylinder synchronization system based on the adjacent cross-coupling control method, and compared it with the PID control strategy based on the adjacent cross-coupling. The simulation shows that when the adjacent cross-coupling and auto-disturbance rejection control method is used, the synchronization error of the suspension system group is within ±0.2 mm. When the adjacent cross-coupling and PID control method is used, the synchronization error range of the suspension system group is −2~4 mm, Thus, it is verified that the designed control strategy has strong robustness and adaptability, and has high synchronization accuracy.

#### *5.5. Sliding Mode Control*

In the hydraulic synchronous control system, the sliding mode control sets the corresponding sliding plane according to the dynamic characteristics of the system, and then makes the system gradually converge to the sliding plane through the control function. When the state of the controlled object reaches the sliding plane, the system will move along the plane under the control of the equivalent function. Sliding mode control has the advantages of simple structure, insensitivity to parameter disturbances and changes, and is a nonlinear control strategy based on switching characteristics. Synovial membrane control has higher accuracy and robustness than PID control [102,103].

Guo et al. [104] designed an adaptive sliding film synchronous position control strategy with a feedforward compensator based on the improved sliding film approach law, and through the joint simulation of the dual hydraulic cylinders of the hydraulic support used in mining by Matlab and AMESim, and the comparative analysis with PI control and fuzzy PID control, the superiority of the control strategy was further verified. Zhang et al. [105] studied the performance of master–slave synchronous control of hydraulic support and proposed an adaptive sliding film control method based on sliding film control and fuzzy control. The simulation results show that compared with the fuzzy PID control, the adaptive sliding film control has better stability, synchronization, and higher synchronization accuracy, and can better overcome the problem of large fuzzy PID steady-state error.

#### *5.6. Robust Control*

Robust control [106] is based on the system state space model, which can solve nonlinear control problems without an accurate mathematical model. Robust control is applied to hydraulic synchronization equipment because it can ensure stability after the model and parameters of hydraulic synchronization system change, and has high robustness.

Zhou et al. [107] designed a robust controller composed of a stabilizing compensator and a servo compensator to solve the problem of small suppression and poor robustness of the multi-cylinder synchronous drive control system of a giant die-forging hydraulic press to system parameter changes. The simulation results show that the multi-cylinder hydraulic synchronous system still has good synchronization accuracy and response speed even in the case of large parameter changes. Liu et al. [108] established the mathematical model of the synchronization system of the giant die-forging hydraulic press, and designed a robust controller for the synchronization control of the giant die-forging hydraulic press. The simulation results show that the system synchronization error increases with the increase of the eccentric load, but the steady state error is basically close to 0, which proves that the robust controller has a good inhibition effect on the nonlinear fluctuation of the synchronization system, and can effectively solve the problem of multi-cylinder synchronization control.

#### *5.7. Adaptive Robust Control*

Robust control is applicable to hydraulic synchronous control systems with a large variation range of uncertain factors and a small stability margin. Because it generally does not work in the optimal state, the steady-state control accuracy of the system is poor. Therefore, in order to make the system work in the optimal state, some scholars have proposed adaptive robust control. Adaptive robust control [109,110] combines the advantages of robust control and adaptive control. Based on robust control, parameter adaptive control is added to the feedforward link of the hydraulic synchronous control system to enable the system to work in the optimal state and ensure that the hydraulic actuator has high-precision motion tracking performance and high synchronization accuracy when moving.

Dou [111] studied the synchronization of unsymmetrical hydraulic cylinders, designed a robust adaptive tracking controller based on the nonlinear coupling model, selected the adaptive backstepping design method, and combined it with the parameter adaptive law to solve the uncertainty of some parameters of the system. The simulation in Amesim software shows that the maximum synchronization error of the two hydraulic cylinders is about 1.4 mm, and this control method has good synchronization. Liu et al. [112] designed an adaptive robust controller based on the new integrated actuator electro-hydraulic composite cylinder to compensate the continuous friction it suffered, introduced the robust integral of the error signal into the controller to compensate the approximate error caused by the external disturbance, and the simulation verified the effectiveness of the control method. Li et al. [113] designed an adaptive robust H∞ control and proposed an adaptive controller to update the parameters to control the influence of uncertain parameters and external disturbances during the movement of double hydraulic cylinders. The good control ability of the control system was verified through simulation, which provided a reference for the synchronous control of multiple hydraulic cylinders.

#### *5.8. Machine Learning Control*

Machine learning [114,115], as an artificial intelligence algorithm with multiple fields, is dedicated to studying how computers simulate and realize human learning behavior and gradually improve and perfect their own performance through autonomous learning. The process of machine learning modeling is to divide the data into training sets and test sets. The test set is the independent data of the training set. It does not participate in the training at all, but is only used for the final evaluation of the model. With the continuous development of artificial intelligence, machine learning is now gradually being applied to the hydraulic synchronous control system.

Compared with traditional control, machine learning control is more stable and reliable, with higher control accuracy. Fu et al. [116] proposed a control method combining fuzzy PID control and feedforward control based on machine learning to solve nonlinear problems such as a sluggish and dead zone in the electro-hydraulic system of the boom of a mainstream pump truck, and fitted the machine learning and fuzzy PID to obtain the control algorithm that can be applied in PLC. The test results show that the control algorithm can greatly improve the tracking control synchronization accuracy of the boom. Wei [117] studied the control method based on the fusion of machine vision, machine learning and the fuzzy model and established the prediction model of slab bending based on machine learning. Subsequently, the author introduced the differential evolution algorithm and Bayesian optimization algorithm to select the optimal parameters of the machine learning algorithm in order to realize the high-precision detection and prediction of slab bending, and the synchronous and accurate compensation of the roll gap tilt compensation value for the hot strip rolling process. The experiment shows that the accuracy of the stochastic forest prediction model based on differential optimization is 96.3% in the range of ±3 mm, and the accuracy of the prediction model based on Bayesian optimization support vector is 99.25% in the range of ±2 mm. The machine learning model can meet the accuracy requirements.

#### *5.9. Neural Network Control*

Neural network is based on simulating the structure and function of the biological brain, and quickly approximates the complex model of nonlinear multi-input and multi-output. Its control parameters can be adjusted continuously with the changes in environment and load conditions, so it can achieve good control accuracy. However, neural networks are prone to over-learning, and their control cost is high. With the development of modern control theory, more and more scholars have applied neural network control to hydraulic synchronous systems and achieved good control results. The relevant neural network control algorithm can automatically adjust the control parameters online, which gives the hydraulic synchronization system a strong anti-interference ability. The neural networks commonly used in the hydraulic synchronization system include the BP neural network and immune neural network.

The BP neural network [118,119] (error return neural network) is widely used in hydraulic neural network synchronization control. The BP neural network uses a nonlinear continuous transformation function to make neurons in the hidden layer have a learning ability. Its core idea is to use error to reverse correct the control system, so that the system has a strong learning ability, memory ability, and nonlinear mapping ability. Liu et al. [120] proposed the master–slave synchronization control strategy of the BP neural network for the four-cylinder pod thruster installation platform that needs synchronization control, and designed a PID controller based on the BP neural network in combination with the PID controller. The simulation verified that the synchronization error of the smooth operation of the four hydraulic cylinders was within 0.1 mm, and the PID control principle under the BP neural network was shown in Figure 12.

**Figure 12.** PID control principle under BP neural network.

The immune neural network [121] is a control algorithm formed by studying the immune algorithm based on the working principle of the biological immune feedback mechanism and applying it to a neural network. Its key is the approximation of antibody suppression function. Based on the property that a neural network can approach any nonlinear function, Liu [122] used the BP neural network to approach the antibody suppression regulation function, and combined it with a PID controller to design an immune neural network control method. This uses the output of the actuator to approach the antibody suppression regulation function in real time, thus realizing the self-adjustment of proportional integral differential parameters, and further improving the control effect of the dual hydraulic cylinders. The simulation results show that when the system damping is 0.1, the maximum synchronization error of two hydraulic cylinders is 0.014 mm, which further verifies that the control algorithm can meet the control requirements of a high precision double-cylinder forging machine.

Neural network control is often combined with fuzzy control, PID control, and adaptive control because of its strong robustness, high control accuracy, and the ability to solve the control of complex, uncertain, nonlinear, and time-varying systems, which further significantly improves the control accuracy of hydraulic synchronous system. Xu et al. [123] combined a neural network and fuzzy control to form a fuzzy neural network synchronization control algorithm, applied it to a double-cylinder forging hydraulic press, and established a mathematical model of the double-cylinder synchronization control. The experiment shows that the fuzzy neural network control algorithm has good synchronous control accuracy.

#### *5.10. Networked Control*

With the continuous development of Internet technology, the hydraulic synchronization control system is gradually developing to wireless network control. Some scholars have used sensors and a wireless network signal transceiver to achieve hydraulic synchronization control, so that it has a faster response speed and higher synchronization accuracy.

To solve the problem of a long-distance hydraulic pipeline and signal line in hoisting a hydraulic lifting system, Dong et al. [124] proposed an intelligent hydraulic synchronous lifting scheme based on a wireless network communication module, and designed a distributed iterative learning controller, which can effectively overcome the impact of communication time delay and saturation. AMESim simulation verified the effectiveness of the control strategy, and the six hydraulic-cylinder system can achieve accurate synchronization control in a short time. Lu et al. [125], based on the robust control theory, set the sensor and actuator nodes as time-driven, set the controller as event-driven, used the wireless networked synchronization system to connect the distributed subsystems that needed synchronous output, and established the error model of the hydraulic synchronization system that could reduce the network delay function. The simulation shows that the displacement of the active cylinder and the slave cylinder can be gradually synchronized, and the experiment shows that the displacement synchronization error of the driving cylinder and the slave cylinder is within ±1 mm.

After the above research on hydraulic synchronization control algorithms, their control effects and characteristics can be summarized as shown in Table 2.

**Table 2.** Comparison of Different Control Algorithms for Hydraulic Synchronization System.



#### **Table 2.** *Cont.*

#### **6. Discussion**

The hydraulic synchronization control system has the characteristic of nonlinearity and is time-varying, and the adoption of an appropriate control algorithm is an important research direction of the hydraulic synchronization system. Common advanced control algorithms include fuzzy control, fuzzy PID control, adaptive robustness control, machine learning control, and neural network control, etc., but they have their own advantages and disadvantages. For a nonlinear, easily disturbed, time-varying hydraulic synchronous control system, a single control algorithm often cannot achieve a good control effect, so some scholars have organically combined multiple control algorithms and thus achieved a good control effect.

With the development of hydraulic synchronization control systems towards automation, low energy consumption, low leakage, low noise, high response, anti-interference, intelligence, and high accuracy, higher requirements are put forward for the control strategy and hydraulic components. The designed hydraulic synchronous system first must meet the requirements of high precision, needs to have fast response, overshooting, strong anti-interference ability and other characteristics, and the selected control strategy can ensure that the system has a good control effect when it is subjected to external disturbance. The prediction and prospect of the development trend of the hydraulic synchronous control system [22,126–133] can be summarized as follows.


#### **7. Conclusions**

The hydraulic synchronous control system is the core of hydraulic synchronous motion machinery and equipment, which determines the function and technical performance of these machines and equipment. The main problems of the hydraulic synchronization system under the current closed-loop control are: there is a feedback adjustment lag, it is difficult to combine between different control algorithms, it is difficult to achieve accurate synchronization under eccentric load and other problems, and with the improvement of control accuracy, the requirements and costs of hydraulic components are getting higher and higher. It is believed that with the continuous development of artificial intelligence and control algorithms in the future, the above problems can be gradually solved. With

the continuous development of control theory, the improvement of computer level and the continuous improvement of new intelligent manufacturing methods and processes, the cost of the hydraulic synchronization system will be gradually reduced, and its combination with intelligent control algorithms will also be closer. Digitalization, integration, intelligence, neural networking, innovation and networking are gradually becoming the future development trend of the hydraulic synchronous control system, among which the more promising development is neural network and networking, which can significantly improve the control accuracy of the hydraulic synchronous system and reduce the time of the feedback adjustment. At the same time, more scholars are also required to carry out relevant research on the hydraulic synchronous control system, so as to continuously promote the progress and development of the hydraulic field.

**Author Contributions:** Conceptualization, R.L. and W.Y.; methodology, X.D.; software, J.X.; validation, W.Y., Q.S. and Y.Z.; formal analysis, R.L.; investigation, X.D.; resources, W.Y.; data curation, X.D.; writing—original draft preparation, W.Y.; writing—review and editing, R.L.; visualization, J.X.; supervision, X.D.; project administration, Q.S.; funding acquisition, R.L. All authors have read and agreed to the published version of the manuscript.

**Funding:** 1. Key R&D plan of Shandong Province, China, grant number 2021CXGC010207; 2. Creation Project of Key Components Of High Performance Sowing and Harvesting and Intelligent Work Tools, grant number 2021CXGC010813; 3. General project of Shandong Natural Science Foundation, grant number ZR2021ME116; 4. Key R&D plan of Shandong Province, China, grant number 2020CXGC011005.

**Acknowledgments:** We would like to thank our tutor, Ruichuan Li, for all his support and guidance. I would like to thank my colleagues for their care and help in my daily work.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

### *Review* **A Review of Automobile Brake-by-Wire Control Technology**

**Xuehui Hua 1, Jinbin Zeng 2, Haoxin Li 2, Jingkai Huang 2, Maolin Luo 2, Xiaoming Feng 3, Huiyuan Xiong <sup>4</sup> and Weibin Wu 2,\***


**Abstract:** Brake-by-wire (BBW) technology is crucial in driverless cars. The BBW technology, which has a faster reaction time and greater stability, can improve passenger safety in driverless cars. BBW technology refers to the removal of some complicated mechanical and hydraulic components from the traditional braking system in favor of using wires to transmit braking signals, which improves braking performance. Firstly, this paper summarized BBW's development history as well as its structure, classification, and operating principles. Subsequently, various control strategies of the BBW system were analyzed, and the development trend and research status of the motor brake-control strategy and wheel-cylinder pressure-control strategy in the braking force-distribution strategy were analyzed respectively, and the brake fault-tolerance technology and regenerative-braking technology were also analyzed and summarized. Finally, this paper summarized the various technologies of BBW, taking the electromechanical brake (EMB) in the braking system as an example to discuss the current challenges and the way forward.

**Keywords:** brake-by-wire technology; motor brake-control strategy; wheel cylinder pressure-control strategy; brake fault-tolerance technology; regenerative-braking technology

#### **1. Introduction**

More effective and energy-saving drive-by-wire technology has evolved with the advent of intelligent and networked automobiles [1,2]. Drive-by-wire technology was initially utilized in the aerospace industry as a crucial component of the braking system. The quick development of intelligent networked vehicles has drawn a lot of attention. For intelligent networked vehicles, the quick advancement of wire-control technology offers not only a reliable control basis but also guidance for developing unmanned vehicles [3].

The most significant and technically complex aspect of chassis technology is the brakeby-wire (BBW) system [4]. The engine, in the case of an electric vehicle, was eliminated, therefore it is unable to supply a vacuum source to the vacuum booster for engine-based braking assistance [5]. As a result, an improvement to the braking system is inevitable. Moreover, more cooperative, intelligent, and responsive motion actuators are needed for the autonomous driving of automobiles. To facilitate braking and effectively recover braking energy to extend the range of electric vehicles, brake-by-wire technology regulates the motor. At the same time, it can execute intelligent driving commands with pinpoint accuracy. The brake-by-wire system will undoubtedly be the best option for braking system improvement.

The brake-by-wire system, as an important safeguard mechanism in automobile active safety, has the advantages of environmental protection, accurate pressure regulation, and fast response time. The BBW systems can readily combine the anti-lock braking system (ABS), electronic stability controller system (ESC), and regenerative braking system (RBS)

**Citation:** Hua, X.; Zeng, J.; Li, H.; Huang, J.; Luo, M.; Feng, X.; Xiong, H.; Wu, W. A Review of Automobile Brake-by-Wire Control Technology. *Processes* **2023**, *11*, 994. https:// doi.org/10.3390/pr11040994

Academic Editors: Francisco Ronay López-Estrada and Guillermo Valencia-Palomo

Received: 22 February 2023 Revised: 19 March 2023 Accepted: 21 March 2023 Published: 24 March 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

through the implementation of synergistic control techniques. Some benefits of these pairings include enhanced brake stability and energy recovery [6,7].

At the time when the BBW system had more mature technology and loading cases, some famous auto parts manufacturers in developed countries, such as Bosch, Siemens, Continental Teves in Germany, Delphi, TRW in the United States, Hyundai, Mobis, Mando in South Korea, Haldex and SKF in Sweden, had carried out research on EMB and developed their own products. Mobis and Mando in Korea, and Haldex and SKF in Sweden, had all conducted EMB research and developed their own products. A number of Chinese universities and automobile companies, including Jilin University, Tongji University, Tsinghua University, Beijing University of Technology, and Geely Automobile, developed preliminary programs [8–11].

#### **2. Development History**

Several types of braking systems have evolved since Wilhelm Maybach invented the drum brake in 1900. Their evolution can be loosely divided into the four stages listed below.

The first stage is ABS, anti-lock braking system. To achieve the best braking effect, the system may automatically alter the wheel brake force while braking. The wheels can prevent skidding during braking by regulating and managing the brake line pressure. To optimize braking performance, keep the wheels at a 15–25% slip rate of motion while rolling [12]. Proportion Integral Derivative (PID) control [13], neural network control [14], fuzzy control [15], fuzzy PID control [16], and logic threshold control [17–20] are the primary control systems utilized for automobile ABS. ABS is standard equipment in almost every car in the globe. The majority of them are based on Bosch's logic threshold ABS control system. The logical threshold control method is straightforward, easy to apply, and extremely adaptable. It maintains the slip rate and angular acceleration in the best range and is mature for the application. Today, improving control techniques is the main focus of ABS optimization. Li made improvements to the automotive ABS system and put out a plan to use the peak wheel speed linkage to address the vehicle speed and standard slip rate. To solve the control problem of an anti-lock braking system (ABS) in an automobile under many challenging driving circumstances. He created a hierarchical intelligent control system that includes organizational coordination, parameter correction, and human control [21,22].

The second stage is ESC/ESP. The electronic stability controller system (ESC) is a new type of vehicle active safety system. It is an extension of the vehicle's anti-lock braking system (ABS) and traction control system (TCS). On this basis, a transverse swing rate (yaw rate) sensor, a lateral acceleration sensor, and a steering wheel angle sensor were added to the vehicle steering when driving. Through ECU control, ensure the vehicle's stability and safety while driving, braking, and steering. The ECU electronic control ensures the vehicle's stability and safety while driving, braking, and steering. The vehicle dynamic stability control system, as shown in Figure 1, provides internal feedback control via ESP. The performance of the wheel will be unstable and non-linear at the limit of adhesion between the tire and the road. In this case, the vehicle dynamics control system will assist the car in maintaining control [23–25]. The sensors provide the ECU with information on the vehicle's current status [26]. Two new pressure limiting valves and two suction valves were added to the ABS's eight solenoid valves, totaling 12 solenoid valves capable of switching the system's active boost circuit and return circuit. To ensure the stability of the vehicle driving, solenoid valves, plunger pumps, and other key components have different parameters of the series of products, which can be diversified with the trial of different models.

The third stage is IPB+RBU. For L3 and L4 of autopilot, Booster proposed a brake-bywire technology, which is IPB+RBU. Integrated Power Brake (IPB) integrates brake vacuum pump, vacuum brake booster, ESP and other important actuators. In the IPB, the highly integrated power brake is complemented by a redundant brake unit. This has ensured that even in the event of an integrated power brake failure, the vehicle can stop safely and reliably without driver intervention with the following advantages:


By analyzing the requirements of the braking system, it is necessary to consider the design of redundant braking. In addition to the primary brake control unit, a secondary brake control unit is required, and the system should also be equipped with extended functions such as system status detection, redundancy control, and back-up vehicle stability control. Booster's solution is to use the IPB as the primary braking system to perform the braking request in most cases. In the case of IPB failure, RBU (redundant brake unit) can be used as the redundant brake.

The fourth stage is EMB. It is electromechanical and can be easily integrated with other electronic control systems of the vehicle to realize more functions, such as braking, ABS, EBS, ESP, automatic driving, and optimized energy recovery. Based on all these advantages, EMB technology is bound to develop vigorously and present serious challenges to the hydraulic braking system in the future.

EMB is a new concept of braking systems. At the stage of EMB, load, cost, reliability and other constraints have blocked the industrialization. However, as a parking brake it is used in industry. However, thanks to the advantages of fast response, energy saving and environmental protection, it has become one of the development directions of the future vehicle braking system.

#### **3. Structure of Brake-by-Wire System**

Electronic components had largely taken the role of mechanical and hydraulic components in the brake-by-wire system. The brake pedal is no longer mechanically connected to the braking wheel cylinders. Electronic brake pedal sensors monitor the driver's braking

activity and translate it into electrical signals for the electronic control unit (ECU) to analyze. Then the ECU sends the proper instructions to the electric drive units, including the high-pressure accumulator and the motor. In order to produce braking power, the motor or high-pressure brake wheel cylinder eventually presses the push rod.

The brake-by-wire system can be classified into two categories: hydraulic and mechanical, depending on the various actuators. The hydraulically controlled actuation system, which includes the electro-hydraulic brake system (EHB), is structurally similar to the conventional hydraulic brake system. It only kept the original hydraulic brake system and swapped out the vacuum assist for the motor assist [27,28]. But, as an intermediate of the brake system with wire control. Major automakers continue to favor EHB due to its outstanding compatibility and simplicity of retrofitting to the original brake system for adoption. The electromechanical brake is a braking system that employs an electric signal to regulate the mechanical structure, eliminating the transmission mechanism such as hydraulic pressure, and using a motor as the power source. Being a mechanically driven system, it is more environmentally friendly and sensitive, and eliminates the brake fluid delay effect. It can also actively implement the longitudinal dynamics integration of EMB and other control systems via algorithmic advancements to achieve a better braking effect. As intelligent car technology evolves, EHB will be replaced by EMB, which is more in line with the needs of future braking systems Because of its distinct features, the EMB system is currently attracting considerable interest from Chinese and global automotive manufacturers. It will become a research center for future automobile braking.

#### *3.1. Electro-Hydraulic Brake System (EHB)*

As a transitional product between the traditional brake system and the brake-bywire system, the operating mechanism of the EHB has replaced the traditional hydraulic brake pedal with an electronic brake pedal and used a hydraulic device instead of a bulky vacuum booster. The hydraulic control unit in the EHB automatically adjusts the braking pressure of the wheels according to different driving conditions to provide sufficient braking force. EHB has eliminated the coupling between the brake pedal and the driver without a vacuum booster, with advantages of compact structure, convenient and reliable control, low braking noise, and good braking comfort. The composition of EHB system can be seen in Figure 2 [29,30]. EHB can also control the braking force of each wheel individually to easily maximize the braking energy recovery. Therefore, EHB has special application value for electric vehicles. Dependent on the hydraulic drive unit, EHB can be classified as a high-pressure accumulator type or an electric pump type. Table 1 displays the hydraulic control system for the electro-hydraulic braking system.

**Figure 2.** The composition of EHB system.


**Table 1.** Hydraulic pressure-control scheme of master cylinder of electro-hydraulic braking system.

#### 3.1.1. High Pressure Accumulator Type

Among them, the hydraulic brake system of high-pressure accumulator type provides the master cylinder hydraulic or wheel cylinder braking force through the high-pressure accumulator high-pressure energy. Thus, the dynamic regulation of braking force is achieved. After obtaining the driver's intention through the brake analog pedal, ECU sends a command to the vehicle controller. The high pressure accumulator, solenoid valve and pump are controlled to produce appropriate hydraulic pressure. When pressure in the highpressure accumulator is insufficient, the hydraulic pump will pressurize the high-pressure accumulator.

In the 1990s, when braking systems were still in their infancy, a high-pressure accumulator kind of hydraulic braking system was devised to overcome the problem of the brake pump's delayed response and low flow rate. In which high pressure is generated by an electric motor plunger pump and stored in advance in a high-pressure accumulator. The high-pressure accumulator received braking fluid at a faster pace, which shortened the brake's response time. When the brake was activated, the high pressure was swiftly released [49]. In 1994, the first high-pressure accumulator brake system was created. It's created using analogy and Saber modeling. Toyota then introduced the first direct-drive wheel cylinder actuator-equipped EHB into production. It has two linear solenoid valves installed in the hydraulic adjustment unit, allowing for accurate control of the pressure in each wheel cylinder. In the system failure backup braking circuit, the front and rear chambers of the brake master cylinder were connected to each of the 4-wheel cylinders via switching solenoid valves. Also, the hydraulic adjustment unit's high-pressure accumulator, hydraulic pump, and motor were all detached separately. To implement the hydraulic power-assist function, the high-pressure accumulator outlet was connected to the brake master cylinder [50]. Bosch improved the design of the Sensotronic Brake Control (SBC), which was based on Toyota and shared many structural and functional similarities with the EHB. With the evident exception that the hydraulic adjustment unit of SBC was outfitted with two separate pistons for the front axle brake circuit. In the event of a system failure, the SBC system can only brake on the front axle wheels. The high-pressure nitrogen leakage from the high-pressure accumulator can be efficiently eliminated by the separating pistons on the front axle brake circuit [8]. The Mercedes R230 SL received the design for the first time in 2001. The high-pressure accumulator received high pressure from the electric motor, which was used to immediately supply braking pressure to the wheel cylinders [51]. They were made by Bosch, Siemens, Continental Teves, and Toyota [52–54]. These pioneering producers and researchers gave EHB a direction for future study. For the earlier concepts, Li proposed an electro-hydraulic braking system and its braking control method using a high-pressure accumulator. The high-pressure accumulator enabled the brake system to be realized with quick and precise pressure control and provided the driver with good pedal feedback.

#### 3.1.2. Electric Pump Type

In the 1990s, Bosch designed and built the first EHB system based on the more mature ESP system and conducted real-world tests with good results [55]. The electric pump hydraulic braking system is driven by an actuator that directly drives the master cylinder for braking. The design retains the traditional brake anti-lock system ABS (Anti-lock Brake System), and also has the advantages of high control reliability and small changes to the traditional structure, which is now one of the mainstream development trends of the brakeby-wire system. Sun proposed a pump-controlled direct-drive the brake-by-wire unit, which uses pump-controlled direct-drive volumetric servo technology, and directly drives the bidirectional gear pump, to achieve control and rapid adjustment of the brake wheel cylinder pressure, eliminate the throttling loss of the valve control system, and effectively improve the system efficiency. The rotary motor output shaft directly drives the hydraulic pump scheme, and the hydraulic pump forward outlet port is connected to the brake wheel cylinder, that is, the hydraulic pump directly drives the wheel cylinder piston, and the forward inlet port is connected to the low-pressure accumulator to realize the function of controlling the wheel cylinder pressure [56].

Gu presented a passenger automobile brake-by-wire technology. A primary pressure supply unit was directly powered by an electric motor, and an auxiliary pressure supply unit with a high-pressure accumulator was also part of the system. The synergistic action of the two units results in accurate pressure regulation [57]. Li suggested a direct-actuated valvebased quick reaction braking system. It reduced the number of solenoid valves that needed to be configured by driving the valve spool directly with a solenoid linear actuator. And the sliding film variable structure control and adaptive robust control [58] swiftly modify the oil pressure of the brake wheel cylinder; Gong created a novel electromagnetic linear actuator-based braking unit for the all-electric driving characteristics of electric cars. By driving the electro-hydraulic brake unit directly to improve braking performance [59]. As a result, some manufacturers created electronic wedge brake (EWB). This is a self-excitation function linear actuation system [60]. Jo researched EWB modeling and control. A new single-motor EWB has been created. Controlling electronic actuators and self-exciting wedge mechanisms provided braking power. Because EWB did not require vacuum boosters or master cylinders, they were straightforward to configure in the cabin [61]. Streli and Balogh created a mechanical model that takes frictional factors into account and an EWB controller. A sliding mode controller was developed based on a dynamic equation model of the electronic wedge brake. Based on a simplified electronic wedge brake model, Kwangjin Han accomplished clamping force prediction and provided a contact point identification technique. Simulation and experimentation with a prototype EHB were used to confirm this controller's effectiveness [60]. The findings demonstrate that, in comparison to the pre-optimized EHB system, the optimized EHB system can provide the same vehicle stabilization with less boost torque during braking [62].

#### *3.2. Electromechanical Braking System (EMB)*

EMB stands for electronically managed mechanical braking. It entirely gets rid of the hydraulic system. It offers features like an efficient structure and environmental protection, among others. Due to the absence of the braking fluid's delay effect, it is also incredibly sensitive. The management of braking pressure is more exact with active braking function. EMB can simultaneously disconnect the brake pedal force and guarantee pedal feeling in the vehicle. To increase the rate of energy recovery when used in conjunction with the higher controller without altering the system's organizational structure. Simple adjustments to the control algorithm level can integrate EMB with the longitudinal dynamics of other systems. Decoupled composite braking systems can be implemented with wire-controlled actuation. EHB and EMB are two types of linear control technology currently in use. On the basis of Conventional Hydraulic Brake (CHB), EHB adds brake pedal sensor, pedal stroke simulator, master cylinder pressure sensor, pressure regulator, and pressure controller, and each wheel's braking force can be independently controlled. The EHB is based on the hydraulic circuit of the conventional automotive braking system and has a number of inherent flaws, including numerous hydraulic lines, a sizable vacuum booster, a challenging layout and assembly, a lengthy response time to braking, a high level of pedal vibration, and high manufacturing and maintenance costs. Instead of a hydraulic circuit, EMB is an electromechanical system composed primarily of pedal simulators, EMB actuators, and controllers that control the clamping force of the actuator to achieve independent control of each wheel braking force. EMB has numerous advantages over EHB, as shown in Table 2 [6,63].


The EMB is made up of a brake actuator on the brake wheel (which includes a torque motor, gear reducer, differential mechanism, motion converter, and brake caliper), and a central controller, the structure of which is represented in Figure 3. The motor serves as the primary driving source, with the reducer and torsion mechanism amplifying the spinning torque. The rotating motion of the motor is converted into the linear motion of the brake caliper by the motion conversion mechanism (ball screw structure or bevel gear structure), which then acts on the brake disc in the blocking condition.

**Figure 3.** Electromechanical brake system structure composition.

As shown in Figure 4, The EMB actuator is designed using the vehicle's braking performance requirements and installation space limitations as input, the actuator's size limitation, the target maximum braking clamping force and its response time, the braking gap and its gap elimination time, and other performance objectives [64]. The performance criteria for electrical components, mechanical transmission components, and sensors are then obtained individually. Finally, the performance parameters of the specific motor type, operating range, rotational inertia, mass, operating voltage, and operating current, etc., are added to the design of the motor and its drive controller, the design of transmission parts, and the selection of sensors, and the detailed design of the actuator is realized.

**Figure 4.** The flowchart of EMB actuator design.

Electromechanical brake systems were first applied in aircraft, such as the U.S. F-15 fighter jets. A new generation of aircraft will use brake-by-wire technology, according to a 2004 statement from Boeing. Numerous automotive companies and research institutions have become interested in EMB as a result of its successful use in the aviation industry. And it gradually moved on to the field of automotive braking [65,66].

For example, Bosch, Siemens, and Continental Teves in Germany, Delphi and TRW in the United States, Hyundai Mobis and Mando in Korea, Inc., have all demonstrated a great interest in researching EMB actuators since the 1990s. In Sweden, Haldex, SKF had investigated EMB, created their own EMB actuators, and conducted a number of studies on them [67,68].

The EMB designed by Bosch adopts the structure of external motor rotor [67,68]. The motor drove the internal planetary wheel system, and then the rotational motion exported from the planetary wheel system was converted into linear motion through the bevel gear structure. The gears would push the friction blocks to compress the brake discs to achieve the deceleration effect. This structure is more compact and more complex. The planetary gear reducer with a ball screw mechanism was chosen as the answer by the majority of the businesses represented by Continental Teves. The parking brake gear construction was likewise fixedly attached to the rotor. The solenoid's magnetism was dissipated while braking. To fix the rotor, the pawl made contact with the gear. The braking force already applied before halting was maintained in order to achieve the parking brake function, locking the entire transmission mechanism.

Nowadays, EMB research has been conducted in China, with universities serving as the primary research subjects. Early Chinese research on the EMB system dates back to a 2005 patent application by Tsinghua University. Tsinghua University's Zhao examined and compiled data on the structural designs of EMB actuators used both domestically and overseas. A structural plan with a torque motor, planetary gear reduction, and ball screw was his suggestion. The EMB controller without a pressure sensor was created by Han from Jilin University, who also finished the off-line simulation and debugging. Li from Chongqing University finished designing the braking control algorithm for the EMB technology after analyzing the response characteristics of the actuator. The stiffness of the brake caliper and ball screw, which impact the stability of the EMB, were looked into modeling studies by Yu of Shenyang University of Technology and colleagues, who also adjusted the EMB's structural properties. In the design of the EMB actuator, Yun of Zhejiang University converted the rotor to a hollow type and positioned the motion translation mechanism in this area [27,28,69–72]. Based on the existing commercial vehicle disc brake, Song proposed a commercial vehicle driving and parking brake electromechanical actuator based on a dual motor and self-locking mechanism. Sun proposed a dual-planetary EMB system based on the front-drive electric vehicle. Compared with the ordinary EMB structure, the structure of the dual-planetary EMB is optimized, which effectively reduces the motor power and improves the actuator performance. Aiming at the problems of large volume and low braking performance of commercial bus brakes, Shen proposed an auxiliary braking actuator-double stator eddy current brake and main braking actuator-electro-mechanical brake. Hye-Yeon Ryu proposed an EMB system using a motor controller without a large braking force. Zhou proposed an EMB system based on slip rate control. It is concluded that the system can improve the braking effect. Giseo proposed a new EMB clamping force controller, which can control the additional cost of existing sensor installation and response delay. It can be seen that with continuous research and development, electromechanical braking is expected to be popularized in the braking system of future automobiles due to its many advantages.

Research on EMB started late in China. So far, research on EMB systems in domestic universities and research institutes is still based on prototypes and has not yet been commercially applied. In general, there is still a large gap compared with foreign technologies, and the unremitting efforts of relevant scientific and technical workers are needed to narrow the technology gap and catch up with imported advanced levels [67]. Hence, as intelligent technologies and electrification continue to advance. The EMB will almost certainly be more up to date. Because of its distinct advantages, the EMB system is currently attracting considerable interest from domestic and foreign automobile manufacturers. It has become a research priority in the field of automotive braking. However, due to the high safety risk

of motor power supply failure, the EMB system still has a long way to go before it can enter mass production without an additional backup mechanism for brake failure [6,64].

#### **4. The Control Strategy of BBW System**

#### *4.1. Motor Braking Force Distribution Strategy*

In the process of braking force distribution control of the brake-by-wire system, there are mainly electric machine power control and wheel cylinder pressure control. Reasonable distribution of braking force of motor and hydraulic system can improve the braking stability of BBW system, shorten the braking distance and increase the braking comfort. The traditional braking force distribution strategy is usually set up in the form of fixed ratio function of axle load or vehicle deceleration. As the fixed ratio function of the brake force distribution system has a simple and poor adaptability, it cannot better perform the performance of the braking system. Therefore, it is necessary to optimize the design of the braking force distribution control strategy to continuously improve the vehicle braking performance.

The braking force distribution method based on the slip rate is a more reliable distribution method to realize the electric mechanism power distribution. Li at Shanghai Jiaotong University used the control distribution algorithm to achieve the optimal distribution of the tire force based on the symbol-holding quadratic programming method [73]. Peng proposed an optimal braking force distribution strategy along with the constrained optimization problem based on the slip rate as a real-time solution method. The proposed optimal braking force distribution strategy can always ensure that the front wheel slip rate is higher than the rear wheel slip rate under different braking conditions. As a result, it not only improves the stability of vehicle braking, but also shortens the braking distance [74]. For the braking force distribution, H Fennel proposed that the slip rate of the rear wheels should track that of the front wheels in a fixed ratio [75]. In contrast, the ratio is divided into four segments for separate control in the literature [76]. Zhang proposed a hierarchical control system to solve the problems of poor anti-interference performance, poor electromechanical coordination performance and large target tracking errors when autonomous vehicles perform braking. In the upper-level controller, a feedback controller is built based on the desired speed sequence of the unmanned system. The desired braking deceleration is used as the front-end input to compensate for the target braking torque, and the speed error used as the feedback input to correct the target torque difference. In the lower controller, a coordinated braking force distribution algorithm based on fuzzy control is established by considering the characteristics of mechanical braking and motor braking [77].

#### *4.2. Wheel Cylinder Pressure-Control strategy*

The introduction of the brake master cylinder has caused a response delay in the linear hydraulic brake system. Frictional nonlinearity and initial pressure difference have different effects on pressure control. A digitally controlled proportional valve is the most effective and straightforward method in achieving efficient brake pressure modulation. It can achieve the continuous control of hydraulic fluid flow, and thus go on to achieve linear control of hydraulic pressure. To improve the control accuracy of wheel cylinder hydraulics and the performance of the hydraulic braking system, researchers at home and abroad had conducted a comprehensive study on the design and control methods of hydraulic actuators. Li proposed a pressure-control strategy. Based on the open-loop test analysis of the electric master cylinder, the integrated in-line hydraulic brake (IEHB) system was designed. Combined with the pressure segmentation control architecture, a feed forward and feedback PID method based on the auxiliary boost coefficient compensation was used to regulate the electric master cylinder. And a logic valving method was used to control the booster valve, pressure reducing valve and electric pump. Thus, the design of a pressure controller based on this IEHB system was realized [78]. Chen Lv proposed a novel sliding mode control method based on high precision hydraulic feedback modulation. A hydraulic brake system dynamics model including valve dynamics was established. An open-loop load pressure control method based on the linear relationship between pressure drop and coil current in the critical open-loop equilibrium state of the valve was proposed, and the effectiveness and superior performance of the proposed closed-loop modulation method verified [79].

#### *4.3. Stability Control Strategy Technology*

For conventional cars, the electronic stability controller (ESC) system is controlled by direct transverse moment. As the wire-controlled actuation technology increasingly develops, electrical signal transmission has replaced the conventional automotive drive line. It can make the most of the advantages such as the rapid electrical signal transmission to improve the handling stability of the vehicle [80,81]. Therefore, it is urgent to research the dynamics stability control system of brake-by-wire.

Vehicle stability control parameters include the lateral stability control of the vehicle. Factors affecting the lateral stability of the vehicle mainly include the yaw rate and the sideslip angle of the mass center. For the vehicle stability control, there are four main control methods directly transverse moment control, active front wheel steering, active suspension and active anti-roll stabilizer [82,83].

The transverse moment stabilization control system mainly adopts a hierarchical control structure. The computation time is thus reduced by establishing a parallel controller. Its good real-time performance can meet the requirements of the braking system. In response to the above issues, Qin proposed a structure of the new stable hierarchical control system. Based on the feedforward joint control, the upper layer is an additional transverse moment controller, which is able to resist the influence of changing road adhesion conditions. The lower-level torque distribution controller adopts an optimal distribution control algorithm to achieve independent and optimal distribution of braking force to all four wheels of the vehicle [84].

The structure divides the whole control system into three layers as follows:

(1) The layer of vehicle state observation: Real-time monitoring of vehicle parameters;

(2) The layer of motion control target tracking decision: When the vehicle has a tendency to destabilize, the magnitude of the transverse swing torque to prevent transverse swing instability will be calculated according to the driver's intention and the actual state of the vehicle;

(3) The layer of torque distribution control: Optimally distributes the additional transverse sway torque to the brake actuators of the four wheels when the vehicle dynamics constraints are satisfied.

Currently, the dynamics stability control system is mainly through tracking control of the yaw rate and the centroid sideslip angle. Tracking control of yaw rate is commonly used in the regular vehicle driving, which can ensure the vehicle handling stability during normal driving, but it may fail under extreme working conditions. Tracking control of centroid sideslip angle is a good complement to this defect, which can quickly adjust the vehicle state when it loses stability control and ensure the normal vehicle driving. Therefore, most of the research on vehicle stability control focus on coordinating tracking control of the yaw rate and the centroid sideslip angle, and the two cooperate to ensure the handling stability and improve driving comfort and safety.

Mirzaeinejad H et al. proposed a vehicle stability control system as shown in Figure 5. The designed control system is divided into two layers. The upper layer calculates the front outer yaw moment and the corrected lateral force referring to the yaw rate and proposes an algorithm for the distribution of tire braking force. In the lower layer, a new nonlinear multivariable controller is optimally designed. The weighted combination of the front outer yaw moment, corrected lateral force and the tracking error are used to define the vehicle stability performance index. The defined performance index is minimized to obtain a closed form multivariate control law. Through the proposed fuzzy scheduling method, the designed integrated controller is automatically tuned so that the weighting factors can

change softly under different driving conditions, thus realizing the adaptive regulation of vehicle stability control [85].

**Figure 5.** Overall structure of the proposed integrated control system.

Besides, the research on vehicle stability control focuses on the tracking control of yaw rate and center point sideslip angle under extreme operating conditions. Various control algorithms have been used, such as sliding mode control [86,87], fuzzy logic control [88], control allocation methods [89], optimal control [90,91], model predictive control strategies [92], and robust control [93]. In the literature, an extra direct yaw moment control was obtained in real time by tracking and controlling the angular velocity of the transverse swing and the lateral deflection angle of mass center of a robust controller [94]. The literature proposed a vehicle stability control algorithm based on brake-by-wire, which adds a brake redundancy to the vehicle stability control to improve the braking safety [95]. It also proposed a cooperative control of vehicle lateral stability using active front-wheel steering and differential braking, which softly constrains the yaw rate velocity by a model prediction algorithm and controls the centroid sideslip angle to always stay within a certain range so as to maintain the stability and comfort during driving [96].

#### *4.4. Brake Fault-Tolerant Technology*

The brake-by-wire system is expected to completely replace conventional braking systems because of advantages of fast response and high accuracy. However, all mechanical connections between the brake pedal and the brake actuator have been eliminated in the brake-by-wire system. While providing better braking performance, this also poses significant safety challenges to the in-line braking systems. A fault-tolerant and failsafe system architecture is therefore needed. Given such issues, experts have put forth some solutions. Based on the analysis of transient fault propagation characteristics of BBW systems, Shuang Huang proposed a hierarchical transient fault-tolerance scheme with embedded intelligence and resilient coordination. In this scheme, most transient faults can be quickly handled at the node level by a feature-code-based detection method. The remaining are those that cannot be detected directly at the node level. Transient faults that degrade the system performance through fault propagation and evolution can be detected and recovered at the system level by functional and structural models. A sliding mode control algorithm and task reassignment strategy were designed [97]. A comprehensive review of fault-tolerant brake-by-wire system was presented in the literature [98], which discussed fault detection methods for low-cost components, analyzes fault-tolerant design principles between sensors, actuators, communications and provided comprehensive considerations. Liu at Tsinghua University conducted a redundant braking design for the ABS module in the brake-by-wire and designed a redundant ABS control system architecture. By recognizing the driver's driving signals such as braking, steering

and driving, estimating and monitoring each state parameter during vehicle braking, the redundant ABS control system can determine the intervention and withdrawal timing of the redundant braking function through driving intention and vehicle parameters. Among them, pressure controls used variable to gain PID control. When the redundant ABS control system works, the SMC (sliding mode control) algorithm will be used to control the slip rate of the wheels, thus realizing the redundant ABS control function of the vehicle and improving driving safety [99].

#### *4.5. Regenerative Braking Technology*

The concept of energy conservation and environmental protection takes root. People pursue green and renewable environmental energy such as automobile fuel and pay more and more attention to the energy utilization of automobiles [100]. Therefore, the research on automotive energy recovery system is particularly important. Automotive braking energy recovery [101], also known as regenerative braking technology, is a process of vehicle deceleration and braking. Through the energy conversion device, it can convert part of the braking kinetic energy into other forms of energy. While maintaining a steady deceleration of the vehicle, the energy will be saved to a device such as a battery, capacitor or high-speed flywheel for reuse in the vehicle [102,103]. The schematic of the comprehensive vehicle energy recovery system can be seen in Figure 6. Fuel economy can be effectively improved by efficient utilization of regenerative braking and improvement of regenerative braking energy. Current research on regenerative braking technology focuses on the synergistic control of regenerative braking and hydraulic braking. It is necessary to ensure that the total braking force matches the braking force required by the driver [104,105]. This paper analyzes the structure and characteristics of various researchers on braking regeneration strategies in a comparative manner. The following Table 3 describes the relevant researches:

**Figure 6.** Schematic of the comprehensive vehicle energy recovery system.


**Table 3.** List of relevant studies on braking.

#### **5. Challenges and Future Directions of EMB**

The EMB system provides a way for the development of intelligent driving. However, up to now, the brake-by-wire system still cannot be put into use on a large scale, indicating that there are still many issues to be solved.

(1) Safety issues. In the process of automobile development, automobile safety is the first factor to be considered in the design of automobile structure. Automobile safety can be divided into active safety and passive safety. Among them, the braking system belongs to the category of active safety. In order to ensure the safety of drivers and passengers, the design scheme of braking redundancy should be fully considered when designing the electromechanical braking system. The biggest application scenario of EMB system is commercial vehicles. For commercial vehicles, they need to face many complex road conditions, such as high temperature, plateau cold road, muddy road and so on. As the installation form of EMB is similar to that of hub motor, a large number of vibration frequencies will be transmitted to the system, causing the socket to become loose and the actuator to break. In addition, the severe environment will lead to EMB system sensor detection failure due to high temperature or cold. Therefore, the brake redundancy design of EMB will become a significant obstacle to further growth;

(2) Stable response of EMB. EMB is transmitted by electrical signals. The sensor collects the braking signal of the brake simulation pedal and transmits it to the actuator, and the braking time is very short. Therefore, the sensitivity of the sensor will greatly affect the response time of EMB. However, at this point, the sensor's delayed reaction is an issue;

(3) Energy demand of braking force. The traditional vehicle power supply is generally 12 V, and the driving motor of the electromechanical braking system adopts a voltage of 42 V, because it is beneficial to improve the performance of the actuator. However, the high voltage brought by the 42 V power supply system will bring issues such as line insulation, withstand voltage and electromagnetic interference, which also poses a threat to the longevity of the vehicle's complete circuit system;

(4) Accurate control of cars with different parameters. Due to the uncertainty of the vehicle's driving conditions, cargo quality and the nonlinearity of braking, it is difficult to achieve precise control of braking force. Therefore, the next stage of EMB research will be to achieve accurate estimation of vehicle parameters and accurate control of braking force. This requires the use of a variety of control methods to achieve the most suitable control of vehicle braking, and to continuously improve it to improve control accuracy.

EMB will be more in line with the characteristics of automobile integration, intelligence and automation, and more suitable for the future development direction of automobiles. In view of the problems existing in EMB, the development trend is as follows:


In a word, the improvement, optimization, research and development of these technologies of EMB system have important value and role in improving the safety performance of automobile driving, and are conducive to promoting the development of vehicle intelligence and automatic driving technology.

#### **6. Conclusions**

As an emerging product of the era, the brake-by-wire system can provide more stable and faster braking performance for automotive braking. It also puts strict requirements on the control accuracy and robustness of the wire-controlled actuation system. The EHB based on the traditional valve technology has the advantages of high control reliability, low energy consumption, low price and easy modification on the traditional braking system. At present, the improvement of the brake-by-wire system mainly focuses on reducing energy consumption. The main strategies include:

(1) Use Two box or One box architecture solution;

(2) Improve braking accuracy by improving the control module of hydraulic adjustment unit or brake motor;

(3) Design the control method of high pressure accumulator to reduce the braking time and energy consumption;

(4) Design brake regeneration system to improve brake energy utilization.

At present, what is more widely deployed in the brake-by-wire system is EHB. EHB technology is more mature, and more adaptable to the market demand. EHB will be the direction of wireless control braking development at this stage. EMB, due to superior technical conditions, high cost, redundancy backup and thermal reliability technology, still needs to be improved. At present, there is still some distance from full commercialization. However, as the most advanced technology in braking system, EMB system has the advantages of high response speed, energy saving and environmental protection to which the traditional braking systems including EHB cannot be compared. It can be applied in L4 and L5 of automatic driving. For the research of EMB, we can only take this opportunity to rise to the challenges and carry out continuous research in the development of EMB system, and gradually deepen and refine our work. We believe that in the future, when various difficulties that constrain the development of EMB systems are overcome, the conditions for EMB industrialization will slowly mature. We also believe that the electromechanical brake technology has a bright future, and is bound to gain more and more people's attention and favor. Finally, this paper compared the advantages and disadvantages of the braking systems described in the previous paper and reviewed the efficiency of the discussed solutions used for the braking systems. The following Table 4 can describe the comparison of braking system solutions.


**Table 4.** Braking system solution comparison.

**Author Contributions:** Conceptualization, J.Z. and X.H.; methodology, J.Z.; software, H.L.; validation, J.H., M.L. and X.F.; formal analysis, H.X.; investigation, J.Z.; resources, W.W.; data curation, J.Z.; writing—original draft preparation, J.Z.; writing—review and editing, X.H.; visualization, X.H.; supervision, H.L.; project administration, J.Z.; funding acquisition, X.H. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the Guangdong Province Key Field R&D Program—Electric Vehicle Powertrain Design and Optimization Industrial Software (2021B0101220003) and the R&D Program in Key Fields of Guangdong Province—R&D and Application of Key Technologies for New Energy Vehicle Gearboxes with High Efficiency, High Precision, Long Life and Low Noise (2020B090926004).

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** This study did not report any data.

**Acknowledgments:** The authors acknowledge the editors and reviewers for their constructive comments and all the supports on this work.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Abbreviations**

The following abbreviations are used in this manuscript:



#### **References**


**Disclaimer/Publisher's Note:** The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
