Self-Anti-Disturbance Control of a Hydraulic System Subjected to Variable Static Loads
Round 1
Reviewer 1 Report
“VSL Follow-up Analysis Based on SAD Control Strategy Op-timization of Hydraulic System”
In this paper, authors proposed an improved Self-Turbulent Flow (STF) algorithm, based on the real-time acquisition and monitoring of Lubricating Oil Static Pressure (LOSP) in hydraulic system to simulate VSL, is proposed. In this topic, a mathematical model of the Electro-Hydraulic Servo (EHS). It is an interesting topic. In addition, the authors are suggested to address the following comments in order to meet the requirements of the journal and to be suitable for publications.
1) There are some typos in the manuscript that need to be fixed.
2) Avoid abbreviations in abstract.
3) The main findings are not clear in abstract.
4) Validation of your mathematical model with experimental needs to be cleared.
Comments for author File: 
Comments.pdf
Author Response
July 7, 2022
Dear Editor
Editor-in-Chief
Applied Sciences
We are very grateful for your help and suggestions in publishing the manuscript to the Applied Sciences, titled “VSL Follow-up Analysis Based on SAD Control Strategy Optimization of Hydraulic System.” The paper was coauthored by Xigui Wang, Jian Zhang, Yongmei Wang, and Chen Li. We have checked the manuscript and revised it according to the comments. Overall, the comments have been fair, encouraging and constructive. We have learned much from it. We submit here the revised manuscript to meet the evaluation conditions and requirements of the reviewers.
Response to Reviewers' Comments:
Reviewer1:
On behalf of all members of our team, I would like to thank the Reviewer 1 for approving and accepting this manuscript. The authors have once again fully reviewed and refined this revised manuscript, and reflected the refined content in this article. In the long-term in the future, our research team will submit more excellent and high-quality articles to your journal.
Q1. There are some typos in the manuscript that need to be fixed.
Sincerely respond to Q1 raised by Reviewer 1:
The authors thank the Reviewer 1 for their reasonable suggestions and valuable comments on this manuscript. According to the questions raised by the Reviewer 1, the authors have carried out comprehensive and detailed inspections and corrections in the preface and the full text of the manuscript, especially the grammatical errors and poor language polish in this article are inadequate. Revision 3 examples are listed to illustrate the demo.
Example 1:
Unrevised original:
- A hydraulic system typically transfers a fluidic medium from one hydraulic element to another, capable of delivering serialized variable hydraulic loads to a contact lubrication area [1, 2].
Revised manuscript:
- A hydraulic system typically transfers a fluidic medium from one hydraulic element to another, capable of delivering serialized variable hydraulic loads to contact lubrication areas [1, 2].
Example 2:
Unrevised original:
- This stated novel idea is preferable in JC power rear transmission systems to generate stable output power and torque.
Revised manuscript:
- This stated novel idea is preferred in the warship power rear transmission systems for consistent output power and torque.
Example 3:
Unrevised original:
In addition, considering the practical application reasons such as volume/cost/weight, the fly in the ointment is that the full state feedback cannot be realized in this study, which will prompt further discussions on the output feedback control of the hydraulic system.
Revised manuscript:
Apart from that, considering the practical application reasons such as volume/cost/weight, the fly in the ointment is that the full state feedback cannot be perfectly realized in this study, which will prompt further in-depth discussions on the output feedback control of the hydraulic system.
Q2. Avoid abbreviations in abstract.
Sincerely respond to Q2 raised by Reviewer 1:
The authors thank the reviewer 1 for reasonable suggestions and valuable comments on this manuscript. Based on the questions raised by reviewer 1, the author has conducted a comprehensive and detailed inspection and correction of the abstract section as suggested by reviewer 1.
Unrevised original:
Abstract: Hydraulic system lubricating oil is subject to serialized Variable Static Loads (VSL) following-up performance. An improved Self-Turbulent Flow (STF) algorithm, based on the real-time acquisition and monitoring of Lubricating Oil Static Pressure (LOSP) in hydraulic system to simulate VSL, is proposed. In this topic, a mathematical model of the Electro-Hydraulic Servo (EHS) control system for LOSP acquisition is presented, and the STF controller is designed for numerical analysis. The Self-Anti-Disturbance (SAD) control strategy for LOSP of EHS system is discussed, which is used for quadratic optimization, pole placement, PID and STF control, and the LOSP simulation model of STF control is constructed by SIMULINK module. Numerical simulation results indicate that the overshoot is significantly reduced. The proposed SAD control algorithm is verified by experiments, and the LOSP acquisition followability and monitoring accuracy are greatly improved. Toward this goal, a variable hydraulic LOSP acquisition and monitoring can be effectively and stability adjusted by pre-designed EHS control system in the field of power hydraulic fluid lubrication.
Revised manuscript:
Abstract: Hydraulic system lubricating oil is subject to serialized variable static loads following-up performance. An improved self-turbulent flow algorithm, based on the real-time acquisition and monitoring of lubricating oil static pressure in hydraulic system to simulate variable static loads, is proposed. In this topic, a mathematical model of the electro-hydraulic servo control system for lubricating oil static pressure acquisition is presented, and the self-turbulent flow controller is designed for numerical analysis. The self-anti-disturbance control strategy for lubricating oil static pressure of electro-hydraulic servo system is discussed, which is used for quadratic optimization, pole placement, PID and self-turbulent flow control, and the lubricating oil static pressure simulation model of self-turbulent flow control is constructed by SIMULINK module. Numerical simulation results indicate that the overshoot is significantly reduced. The proposed self-anti-disturbance control algorithm is verified by experiments, and the lubricating oil static pressure acquisition followability and monitoring accuracy are greatly improved. Toward this goal, a variable hydraulic lubricating oil static pressure acquisition and monitoring can be effectively and stability adjusted by pre-designed electro-hydraulic servo control system in the field of power hydraulic fluid lubrication.
Q3. The main findings are not clear in abstract.
Sincerely respond to Q3 raised by Reviewer 1:
The authors thank Reviewer 1 for sound suggestions and valuable comments on this manuscript. Based on the questions raised by Reviewer 1, the authors refine and summarize the bottleneck issues in the Abstract section as suggested by Reviewer 1.
In this subject, a mathematical model of electro-hydraulic servo control system for lubricating oil static pressure acquisition is proposed, and a self-turbulence controller is designed for numerical analysis. The control strategy of lubricating oil static pressure active disturbance rejection of electro-hydraulic servo system is discussed, which is used for secondary optimization, pole placement, PID and self-turbulent flow control, and a simulation model of lubricating oil static pressure active disturbance rejection is established. The proposed active disturbance rejection control algorithm is verified by experiments, which greatly improves the followability and monitoring accuracy of lubricating oil static pressure acquisition.
Q4. Validation of your mathematical model with experimental needs to be cleared.
Sincerely respond to Q4 raised by Reviewer 1:
The authors have revised the experimental validation part of the presentation in this manuscript as suggested by Reviewer 1. The proposed mathematical model needs to be explicitly validated experimentally.
To verify the effectiveness of the proposed four control strategies, experimental following-up error performance is shown through quantitative acquisition and monitoring. The following-up performance of the Case 4 controller outperformed the other controllers, which is evident. The following performance of the Case 2 controller is relatively weak, which indicates that the parameter uncertainty has a great influence on the high frequency band of the controller. Also, the following-up performance of the conventional PID controller (Case 1) is the least ideal compared to other controllers. From these experimental results, it can be seen that the following-up ability of the controller of case 3 is greatly improved compared with that of case 1 and case 2. An improved STF algorithm - optimized for SAD control strategy - is introduced in the EHS system with LOSP op, which is a novel move. In addition, the following-up performance of the case 4 controller is better than that of the case 3. Therefore, it is verified that the variable hydraulic LOSP acquisition and monitoring proposed in this paper can be controlled by a pre-designed EHS system.
We are very hoped to publish this article in your journal, and I thank you on behalf of our group. We apologize for what we have not done well. We hope we will continue to submit better articles to you.
Thank you in advance for considering this revised submission.
Sincerely Yours
Corresponding author:
Xigui Wang Professor/Doctoral Supervisor
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Jian Zhang
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Yongmei Wang Professor/Master's Supervisor
School of Motorcar Engineering, Heilongjiang Institute of Technology, No. 999, Hongqidajie Road, Daowai District, Harbin, 150036, PRC
Chen Li
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Author Response File: 
Author Response.docx
Reviewer 2 Report
There are great inaccuracies in the terminology used and the symbolic representation of hydraulic components. The hydraulic symbol "Pilot Operated Relief Valve" (Fig. 1) corresponds to the element "Pressure relief valve Screw-in Direct operated". There is no pilot line and connection point, which has a major impact on the function and the compiled mathematical model (Fig. 2). What is the real hydraulic component of the "Adjustable Flow Throttle Valve (AFTV)" (Fig. 1)? 2-way flow control or 3-way flow control valve? It is necessary to state the correct detailed symbol of the component in the circuit diagram! The description of the system function (lines 142, 143) states "The surplus flow is returned to the lubricating oil tank in the Secondary Lubricating Oil Circuit (SLOC) of the hydraulic system and stored in the hydraulic accumulator". The location of the accumulator in the hydraulic circuit (Fig. 1) is missing. The authors use the term (lines 198, 199) Self-Turbulent Flow (STF). If the authors have in mind turbulence in the hydraulic sense, it is necessary to state the relationship e.g. to the Reynolds number. In Chapter 4. The authors present an experimental verification of the "following-up error". When it comes to accuracy, it is probably appropriate to state the metrological parameters of the measurement system used.
The images have a relatively low quality, I give an adjustment with respect to the standardized display.
Author Response
July 7, 2022
Dear Editor
Editor-in-Chief
Applied Sciences
We are very grateful for your help and suggestions in publishing the manuscript to the Applied Sciences, titled “VSL Follow-up Analysis Based on SAD Control Strategy Optimization of Hydraulic System.” The paper was coauthored by Xigui Wang, Jian Zhang, Yongmei Wang, and Chen Li. We have checked the manuscript and revised it according to the comments. Overall, the comments have been fair, encouraging and constructive. We have learned much from it. We submit here the revised manuscript to meet the evaluation conditions and requirements of the reviewers.
Response to Reviewers' Comments:
Reviewer: 2
The authors thank Reviewer 2 for sound suggestions and valuable comments on this manuscript. Based on the questions raised by Reviewer 2, the authors conducted a comprehensive and detailed inspection and correction of the full text as suggested by Reviewer 2, and explained the problems raised by reviewer 2 one by one.
Q1. There are great inaccuracies in the terminology used and the symbolic representation of hydraulic components. The hydraulic symbol "Pilot Operated Relief Valve" (Fig. 1) corresponds to the element "Pressure relief valve Screw-in Direct operated". There is no pilot line and connection point, which has a major impact on the function and the compiled mathematical model (Fig. 2). What is the real hydraulic component of the "Adjustable Flow Throttle Valve (AFTV)" (Fig. 1)? 2-way flow control or 3-way flow control valve? It is necessary to state the correct detailed symbol of the component in the circuit diagram! The description of the system function (lines 142, 143) states "The surplus flow is returned to the lubricating oil tank in the Secondary Lubricating Oil Circuit (SLOC) of the hydraulic system and stored in the hydraulic accumulator". The location of the accumulator in the hydraulic circuit (Fig. 1) is missing. The authors use the term (lines 198, 199) Self-Turbulent Flow (STF). If the authors have in mind turbulence in the hydraulic sense, it is necessary to state the relationship e.g. to the Reynolds number. In Chapter 4. The authors present an experimental verification of the "following-up error". When it comes to accuracy, it is probably appropriate to state the metrological parameters of the measurement system used.
Sincerely respond to Q1 raised by Reviewer 2:
The authors have carefully checked the terms and symbols used in this manuscript for hydraulic components (including pilot lines and connection points). The detailed necessary relevant explanations and instructions are given here.
Pilot Operated Relief Valve (PORF) is a kind of indirect load safety valve. It is composed of a main valve and a pilot valve (also known as "auxiliary valve"). The main valve is driven by the medium discharged from the pilot valve. Because the pressure in the system is in the form of pulses, it is also called "pulse safety valve". On its own, the pilot valve itself is a direct-loaded (Pressure relief valve Screw-in Direct operated) safety valve. When the medium force reaches the opening pressure of the pilot valve, the pilot valve opens first, and the discharged medium enters the main valve from the bypass pipe. When the protected system is in normal operating condition. The slide valve in the pilot valve cavity is in the open state, and the system pressure is transmitted from the main valve inlet to the air chamber above the main valve disc (piston) through the lower conduit, the pilot valve cavity and the upper conduit. Currently, the gas pressure in the upper air chamber is equal to system pressure at the main valve inlet. The sealing surface of the pilot valve is in a sealed state to ensure that the gas in the air chamber above the main valve cannot be discharged from the discharge port of the pilot valve. Since the area of the piston is larger than the sealing surface area of the valve disc, the system pressure produces a downward resultant force on the main valve disc, so that the main valve is in a closed and sealed state. When the system pressure rises and reaches the set pressure, the gas pushes up the valve core of the pilot valve at the sealing surface of the pilot valve, so that the sealing surface is in an open state, and at the same time, the slide valve in the valve cavity of the pilot valve moves upward, closing the valve cavity of the pilot valve. air passage. The medium in the air chamber above the main valve disc is discharged through the open pilot valve sealing surface, which reduces the pressure above the main valve disc. The main valve disc is pushed open by the inlet pressure to depressurize the system. When the system pressure is reduced to a certain value, the pilot valve core will return to its seat downward under the action of the spring, the sealing surface of the pilot valve will be closed, the spool valve of the pilot valve cavity will be opened, the valve cavity will be unblocked, and the system pressure will pass through the pilot valve cavity again. Into the air chamber above the main valve disc, and push the main valve disc to close. In this manuscript, we have successfully applied for a patent for an invention of this type of valve: A pressure regulating valve for marine lubricating oil system (Authorization Number: ZL201310499155.X).
Adjustable Flow Throttle Valve (AFTV) is a valve component that controls fluid flow by changing the throttling section or throttling length. In the hydraulic system of the quantitative pump, the throttle valve and the relief valve cooperate to form three throttle speed control systems, namely the oil inlet throttle speed control system, the oil return circuit throttle speed control system, and the bypass throttle speed control system. The pressure oil flows in from the oil inlet and flows out from the oil outlet through the channel, the throttle groove on the right end of the valve core and the channel. Rotate the handle to move the valve core axially through the push rod, which can change the flow cross-sectional area of the orifice and realize the adjustment of the flow. The function of the spring is to press the valve core to the left against the push rod. The throttle valve can play a role in controlling the fluid flow. Generally speaking, when the pressure difference of the throttle valve is constant, the size of the opening affects the change of the liquid flow. In a nutshell, the throttle valve has three main functions:
- The function of shut-off speed regulation
The shut-off valve changes the flow by controlling the throttling section or length, which is the main function of the throttling valve and its main function.
- Play the role of load resistance
The throttle valve can also play a load role to a certain extent, which is only a part of the function of the throttle valve. The main function of the throttle valve is still throttle speed regulation.
- The role of pressure buffer
The throttle valve can play a pressure buffering effect on the fluid. When the fluid flows in through the throttle valve, the throttle valve can hinder the operation of the fluid to a certain extent and reduce the impact force.
The circuit of differential connection of hydraulic cylinder is realized by using two-position three-way electromagnetic reversing valve. In the auxiliary description of Figure 1, the sequential action loop controlled by the pressure relay is explained, the 1YA electromagnet of the three-position four-way reversing valve 1 is energized, the left position is connected to the system, the pressure oil enters the left chamber of the hydraulic cylinder A, pushes the piston to move to the right, and the return oil flows back to the fuel tank through the reversing valve 1 to complete the action. 1. When the piston hits the positioning stopper (not shown in the figure), the system pressure rises, so that the pressure relay 1KP installed near the oil inlet chamber of hydraulic cylinder A operates, and an electrical signal is sent to make the two-position four-way reversing valve 2 The electromagnet is energized, the left position is connected to the system, the pressure oil enters the left chamber of the hydraulic cylinder B, pushes the piston to move to the right, and completes action 2. In this way, the A and B hydraulic cylinders act in sequence.
Supplementary Figure 1. Sequential action circuit controlled by three-position four-way pressure relay
A hydraulic accumulator is an energy storage device in a hydropneumatic system. At the appropriate time, the energy in the system is converted into compression energy or potential energy and stored, and when the system needs it, the compression energy or potential energy is converted into hydraulic or pneumatic energy and released to replenish the system. When the instantaneous pressure of the system increases, this part of the energy can be absorbed to ensure the normal pressure of the entire system. Hydraulic oil is an incompressible liquid, and hydraulic oil cannot accumulate pressure energy, and then rely on other media to convert and accumulate pressure energy. In this paper, based on Boyle's law (PVn=K=constant), the energy conversion is completed by compressing the gas, and the accumulator is first charged with a predetermined pressure gas during use. When the system pressure exceeds the internal pressure of the accumulator, the oil compresses the gas, converting the pressure in the oil into gas internal energy. When the system pressure is lower than the internal pressure of the accumulator, the oil in the hydraulic accumulator flows to the external system under the action of high-pressure gas, releasing energy. Selecting the proper charge pressure is the key to this type of accumulator. In supplementary Figure 2, the position of the accumulator in the hydraulic circuit is indicated.
Supplementary Figure 2. Sequential action circuit controlled by three-position four-way pressure relay
The accumulator in this paper is installed before the reversing valve or the oil cylinder, which can absorb or alleviate the shock pressure caused by the sudden reversal of the reversing valve and the sudden stop of the movement of the oil cylinder. When the reversing valve suddenly changes direction, the accumulator absorbs the hydraulic shock, so that the pressure will not increase sharply, the time constant is large, and the frequency characteristics are improved.
In the current study, we focus on the efficiency of gradient-based optimization algorithms to optimize Self-Turbulent Flow (STF) control and investigate the number of forward and adjoint simulations required per unit cost function improvement. Different algorithms are considered for application to an unconstrained distributed control problem involving the optimization of distributed forces in the direct numerical simulation of incompressible turbulent channel flow. Self-Turbulent Flow (STF) control algorithm is formulated as: An implicit Navier-Stokes solution algorithm is proposed for computing turbulent unstructured meshes. The inviscid flux is calculated using the headwind algorithm, and the solution is proposed in time using the backward Euler time-stepping scheme. At each time step, the linear system of equations is approximately solved using a point-implicit relaxation scheme. This approach provides a feasible and powerful algorithm for computing turbulence on unstructured grids. In general, the accuracy of the calculated results is generally consistent with the experimental results.
The authors have made normalized display adjustments for the relatively low quality of images in this manuscript.
We are very hoped to publish this article in your journal, and I thank you on behalf of our group. We apologize for what we have not done well. We hope we will continue to submit better articles to you.
Thank you in advance for considering this revised submission.
Sincerely Yours
Corresponding author:
Xigui Wang Professor/Doctoral Supervisor
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Jian Zhang
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Yongmei Wang Professor/Master's Supervisor
School of Motorcar Engineering, Heilongjiang Institute of Technology, No. 999, Hongqidajie Road, Daowai District, Harbin, 150036, PRC
Chen Li
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Reviewer 3 Report
Dear authors!
The paper presents some control strategies for the Hydraulic System. Classical linear controllers have been synthesized, and results show good transients and overall applicability of the method.
However, some questions appears.
First, it is not clear what "Follow-up Analysis" in the title means. Abbreviations are not common, but explained only in abstract.
Optimization does not present in the paper, instead, authors use classical "Optimal control" which does not directry refer to optimization.
Second, a number of abbreviations make reading the paper difficult. The content of the paper is concerned with application of standard Matlab procedures and does not contain any novel ideas.
It is not clear what is the difference between SAD and classical approached.
My overall opinion is that the paper should be rejected due to lack of novelty.
Author Response
July 7, 2022
Dear Editor
Editor-in-Chief
Applied Sciences
We are very grateful for your help and suggestions in publishing the manuscript to the Applied Sciences, titled “VSL Follow-up Analysis Based on SAD Control Strategy Optimization of Hydraulic System.” The paper was coauthored by Xigui Wang, Jian Zhang, Yongmei Wang, and Chen Li. We have checked the manuscript and revised it according to the comments. Overall, the comments have been fair, encouraging and constructive. We have learned much from it. We submit here the revised manuscript to meet the evaluation conditions and requirements of the reviewers.
Response to Reviewers' Comments:
Reviewer: 3
The authors thank the reviewer 3 for reasonable suggestions and valuable comments on this manuscript. According to the questions raised by the reviewer 3, the author conducted a comprehensive and detailed inspection and correction of the full text according to the suggestions of the reviewer 3, and made necessary explanations to the questions raised by the reviewer 3 in order to expect the reviewer 3's approval and acceptance.
Q1. First, it is not clear what "Follow-up Analysis" in the title means. Abbreviations are not common, but explained only in abstract.
Sincerely respond to Q1 raised by Reviewer 3:
The algorithm can follow-up the path a conflict may follow as it evolves from a chosen starting state (usually the status quo) to a state of interest (such as an attractive solution). The value of the logical method is that it can explain how evolution happened, and the method facilitates computation. To illustrate this algorithm, the mention model has been applied to several real case studies. Since the control accuracy is one of the most important indicators, this research focuses on the high-precision following-up control of the hydraulic system. We propose an adaptive disturbance observer-based asymptotic following-up control strategy for hydraulic systems to achieve our motivation. It is worth noting that parameter uncertainty and disturbance are challenges in the field of hydraulic system control. In this study, time-varying disturbances are compensated by model-based feedforward control terms designed by the symbols of the disturbance observer based on the error function. To avoid the high learning burden of disturbances due to parameter uncertainty, online adaptive control is combined with the disturbance observer described above. In order to ensure the theoretical feasibility of the proposed controller, stability analysis is done by introducing stability theory. The proposed adaptive disturbance observer-based controller can achieve global asymptotic stability when the considered system is subject to mismatch or matching disturbances. The comparative experimental results show that the method can obtain the state-of-the-art hydraulic system following-up performance under the influence of parameter uncertainty and time-varying disturbance. To verify the effectiveness of the proposed controller, a hydraulic system platform, as well as measurement and control systems, have been established. The measurement and control system consists of monitoring software, real-time control software, A/D card (Advantech PCI-1716), D/A card (Advantech PCI-1723), counter card (Heidenhain IK-220); all these cards are 32-bit. The monitoring software is programmed with NI LabWindows/CVI, and the real-time control software is compiled with Microsoft Visual Studio 2005 plus Ardence RTX7.0. Ardence RTX 7.0 is used to provide a real-time working environment for real-time control software under Windows XP operating system. To implement the proposed controller and other comparison controllers defined below, the discretized C++ code was programmed. The control sampling time for the control is 0.4 ms. In this manuscript, some cases are provided to verify the effectiveness of the proposed controller. Furthermore, the effectiveness of parameter adaptation and perturbation observers will be illustrated in these cases.
Q2. Optimization does not present in the paper, instead, authors use classical "Optimal control" which does not directly refer to optimization.
Sincerely respond to Q2 raised by Reviewer 3:
This paper proposes a Self-Anti-Disturbance (SAD) control strategy for Lubricating Oil Static Pressure (LOSP) acquisition to estimate the state and disturbances acting on an Electro-Hydraulic Servo (EHS) system that are designed and tested. The proposed optimal control strategy is based on an observer-provided system and its estimates that evaluate the absolute value of the error calculated between the available actual measurements. Experimental results are presented to verify the effectiveness of the proposed optimal control strategy in a real environment. Comparisons were made with the case of manual tuning of the gain and classical gain adaptation techniques reported in the literature, where the gain adaptation is based on a fixed rate of change. The proposed optimal control method provides comparable performance or outperforms manual and classical adaptive gain selection. Furthermore, compared to classical techniques, the proposed optimal control technique has the advantage of improving convergence in the case of rate-noise measurements with large estimation errors and limiting observer gain growth.
Q3. Second, a number of abbreviations make reading the paper difficult. The content of the paper is concerned with application of standard Matlab procedures and does not contain any novel ideas.
Sincerely respond to Q2 raised by Reviewer 3:
The authors have revised many abbreviations to make reading the paper easier. The content of the paper involves the application of standard Matlab programs. The measurement and control system consists of monitoring software, real-time control software, A/D card (Advantech PCI-1716), D/A card (Advantech PCI-1723), counter card (Heidenhain IK-220); all these cards are 32-bit . The monitoring software is programmed with NI LabWindows/CVI, and the real-time control software is compiled with Microsoft Visual Studio 2005 plus Ardence RTX7.0. Ardence RTX 7.0 is used to provide a real-time working environment for real-time control software under Windows XP operating system. To implement the proposed controller and the other comparison controllers defined below, discretized C++ code was programmed. The control sampling time for the control is 0.4 ms. In this manuscript, some cases are provided to verify the effectiveness of the proposed controller. Furthermore, the effectiveness of parameter adaptation and perturbation observers will be illustrated in these cases, thereby adding novelty to the research content of this topic.
Q4. It is not clear what is the difference between SAD and classical approached.
Sincerely respond to Q4 raised by Reviewer 3:
Classical approached (take Classic PID control as an example) is still the most widely used control algorithm until now, and this is still used in most control systems. The main advantage of classical PID control is that it does not require a model of the controlled object. The disadvantages of classical PID control can be expressed as follows:
1) Error extraction method (calculate errors directly from given instruction)
2) A way to extract the error differential from the error (using a traditional linear differentiator)
3) The weighted sum strategy is not necessarily the best (proportional, integral, differential terms are multiplied by the amplification factor and then added to calculate the control amount)
4) Integral feedback has many side effects (the error is amplified by the integral and then fed back to the system)
The Self-Anti-Disturbance (SAD) control strategy retains the advantages of classic PID control while making up for its shortcomings.
(1) Error Extraction Method - Arranging the Transition Process
Calculating the error directly according to the given command may lead to poor control effect. For example, some commands contain undesired high-frequency signals. Examples of such signals are: step command, square wave command. In order to solve the high frequency signal, the Self-Anti-Disturbance (SAD) control strategy proposes a method of arranging the "transition process", which is similar to low-pass filtering the given instruction to obtain a more easily implemented instruction, thereby sacrificing A little quickness while greatly reducing overshoot.
(2) The method of extracting the error differential from the error - following-up differentiator
A classic linear differentiator outputs a differential component while following-up a given signal. The Self-Anti-Disturbance (SAD) control strategy realizes the fastest control of discrete systems and eliminates the steady-state chatter of the following-up-differentiator. Extracting the error differential from the error consists of two parts: the differential of the instruction and the differential of the output (because the error is equal to the instruction minus the output). The differentiation of the instruction is done by the following-up-differentiator, and the differentiation of the output is done by the subsequent expansion state observer.
(3) The weighted sum strategy is not necessarily the best - nonlinear feedback
The traditional linear feedback method (that is, the error is directly multiplied by a gain) has insufficient convergence speed and anti-disturbance ability.
The solution of Self-Anti-Disturbance (SAD) control strategy is to replace traditional gain with nonlinear function (with nonlinear feedback instead of linear feedback).
(4) Side Effects of Integral Feedback - Expanded State Observer
One of the main functions of the integral is to remove the disturbance (it can be considered as a simple disturbance observer), but the integral works slowly and can cause overshoot.
SAD directly discards the integral, uses the extended state observer to observe the total disturbance, and compensates the system into a pure integral chain. In this way, it is much easier to control. It can be said that the expansion state observer is the soul and essence of SAD. Using the extended state observer, SAD can theoretically compensate an arbitrary order system into an arbitrary order integral chain, and then a simple linear control method can be used to achieve control, and a better control effect can also be obtained. Therefore, the expanded state observer is the essence of SAD. It is no problem to replace the previous following-up differentiator with another linear filter, and it is no problem to replace the nonlinear feedback with linear feedback, but the idea of observing disturbance and compensating cannot leave.
In summary, the differences between the two can be summarized as follows:
- Different in terms of mathematical models
Classical control theory mainly uses ordinary differential equations, transfer functions and dynamic structure diagrams. It only describes the relationship between the input and output of the system, but cannot describe the internal structure of the system and the changes in the system, and ignores the initial conditions. Information about the internal state of the system cannot be fully described.
2.The foundations are different.
Classical control theory is a branch of automatic control theory based on frequency response method and root locus method. Modern control theory is a control theory based on the state-space method, which is a major component of automatic control theory.
- Different systems
The research object of classical control theory is the automatic control system with single input and single output, especially the linear steady system. The characteristic of classical control theory is that the input and output characteristics (mainly transfer function) are used as the mathematical model of the system, and the graphical analysis methods such as the frequency response method and the root locus method are used to analyze the system performance and design the control device.
In modern control theory, the analysis and design of the control system are mainly carried out by describing the state variables of the system, and the basic method is the time domain method. Modern control theory can deal with a much wider range of control problems than classical control theory, including linear and nonlinear systems, stationery and time-varying systems, single-variable systems, and multi-variable systems.
- Different methods
The mathematical basis of classical control theory is the Laplace transform, and the dominant analysis and synthesis method is the frequency domain method. The methods and algorithms used in modern control theory are also more suitable for digital computers. Modern control theory also offers the possibility to design and construct optimal control systems with specified performance indicators.
We are very hoped to publish this article in your journal, and I thank you on behalf of our group. We apologize for what we have not done well. We hope we will continue to submit better articles to you.
Thank you in advance for considering this revised submission.
Sincerely Yours
Corresponding author:
Xigui Wang Professor/Doctoral Supervisor
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Jian Zhang
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Yongmei Wang Professor/Master's Supervisor
School of Motorcar Engineering, Heilongjiang Institute of Technology, No. 999, Hongqidajie Road, Daowai District, Harbin, 150036, PRC
Chen Li
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Reviewer 4 Report
The paper is about improved Self-Turbulent Flow algorithm, based on the real-time acquisition and monitoring of Lubricating Oil Static Pressure (LOSP) in hydraulic system.
Scientific soundness of the paper is good.
Practical application of the science is clear.
Paper lies in the scope of the journal.
I have some specific minor comments:
- The introduction does not bring sufficient background for the study. Authors too quickly describe the specific topic. It should be described from the general to the specific. At the beginning you should describe the potential practical application, indicate the technologies and industries where such hydraulic systems are used, for what, why, if it has competition with other systems etc. Please put more information about the background to highlight practical application and range of applications and then you can go to the description of the specific problems with hydraulic system lubricating oil.
- Fig. 6 and 7 could be mixed and then the difference between optimized control strategy and typical PID would be clearly visible. Please think about splitting sections 3.2 and 3.3.
- Please divide conclusions section in two parts: discussion and conclusions. In discussion, please provide more information about simplifications and their possible effect on the results and on the conclusions (separately). Please make a discussion about the possible practical application of the work, especially if it is possible, please do an additional simple analysis of the benefits reached in the exemplary practical application in short term and long-term application. Please discuss future work.
- In conclusions, please add only short statements that contains QUANTITATIVE findings of the work and please highlight the great meaning of the work and new / sufficient contribution to the field. It is very important to present here only main findings in short sentences and with the numbers not only words and description. Words and description should be placed in the discussion section.
Author Response
July 7, 2022
Dear Editor
Editor-in-Chief
Applied Sciences
We are very grateful for your help and suggestions in publishing the manuscript to the Applied Sciences, titled “VSL Follow-up Analysis Based on SAD Control Strategy Optimization of Hydraulic System.” The paper was coauthored by Xigui Wang, Jian Zhang, Yongmei Wang, and Chen Li. We have checked the manuscript and revised it according to the comments. Overall, the comments have been fair, encouraging and constructive. We have learned much from it. We submit here the revised manuscript to meet the evaluation conditions and requirements of the reviewers.
Response to Reviewers' Comments:
Reviewer: 4
Q1. The introduction does not bring sufficient background for the study. Authors too quickly describe the specific topic. It should be described from the general to the specific. At the beginning you should describe the potential practical application, indicate the technologies and industries where such hydraulic systems are used, for what, why, if it has competition with other systems etc. Please put more information about the background to highlight practical application and range of applications and then you can go to the description of the specific problems with hydraulic system lubricating oil.
Sincerely respond to Q1 raised by Reviewer 4:
The authors thank reviewer 4 for their valuable suggestions, which have been revised and reflected in this revised manuscript in accordance with reviewer 4's comments.
In the introduction chapter, the authors describe the research background from general to specific, and describe the potential practical applications in detail, as well as the technologies and industries that use such hydraulic systems, uses, etc. Additional background information is incorporated to highlight practical needs and scope of application.
Supplementary text has been added to the Introduction section and has been embodied in the revised manuscript:
Consider high-precision following-up of hydraulic systems widely used in heavy industry. The outstanding advantage of the hydraulic system is that it can provide a large anti-load stiffness and a large power-to-weight ratio, which is very necessary for the above scenarios. The challenges of high-accuracy following-up control of hydraulic systems are nonlinear characteristics and modeling uncertainty. Specifically, the parameter uncertainty and disturbance are the main stumbling blocks. Therefore, an advanced control strategies that can handle both parameter uncertainty and disturbance are required. At present, there are many research results on this problem, and most of the existing controllers focus on disturbance rejection, such as adaptive robust control, robust adaptive control, continuous robust integration of error feedback control symbols, etc. These control methods have been applied to hydraulic systems to ensure transient performance and specified steady-state performance. However, another concern was raised. It is well known that traditional robust control suffers from the problem of high feedback gain, which reduces the stability margin and even excites high frequency dynamics, which in turn induces instability in practice.
Q2. Fig. 6 and 7 could be mixed and then the difference between optimized control strategy and typical PID would be clearly visible. Please think about splitting sections 3.2 and 3.3.
Sincerely respond to Q2 raised by Reviewer 4:
Classical approached (take Classic PID control as an example) is still the most widely used control algorithm until now, and this is still used in most control systems. The main advantage of classical PID control is that it does not require a model of the controlled object. The disadvantages of classical PID control can be expressed as follows:
1) Error extraction method (calculate errors directly from given instruction)
2) A way to extract the error differential from the error (using a traditional linear differentiator)
3) The weighted sum strategy is not necessarily the best (proportional, integral, differential terms are multiplied by the amplification factor and then added to calculate the control amount)
4) Integral feedback has many side effects (the error is amplified by the integral and then fed back to the system)
The Self-Anti-Disturbance (SAD) control strategy retains the advantages of classic PID control while making up for its shortcomings.
(1) Error Extraction Method - Arranging the Transition Process
Calculating the error directly according to the given command may lead to poor control effect. For example, some commands contain undesired high-frequency signals. Examples of such signals are: step command, square wave command. In order to solve the high frequency signal, the Self-Anti-Disturbance (SAD) control strategy proposes a method of arranging the "transition process", which is similar to low-pass filtering the given instruction to obtain a more easily implemented instruction, thereby sacrificing A little quickness while greatly reducing overshoot.
(2) The method of extracting the error differential from the error - following-up differentiator
A classic linear differentiator outputs a differential component while following-up a given signal. The Self-Anti-Disturbance (SAD) control strategy realizes the fastest control of discrete systems and eliminates the steady-state chatter of the following-up-differentiator. Extracting the error differential from the error consists of two parts: the differential of the instruction and the differential of the output (because the error is equal to the instruction minus the output). The differentiation of the instruction is done by the following-up-differentiator, and the differentiation of the output is done by the subsequent expansion state observer.
(3) The weighted sum strategy is not necessarily the best - nonlinear feedback
The traditional linear feedback method (that is, the error is directly multiplied by a gain) has insufficient convergence speed and anti-disturbance ability.
The solution of Self-Anti-Disturbance (SAD) control strategy is to replace traditional gain with nonlinear function (with nonlinear feedback instead of linear feedback).
(4) Side Effects of Integral Feedback - Expanded State Observer
One of the main functions of the integral is to remove the disturbance (it can be considered as a simple disturbance observer), but the integral works slowly and can cause overshoot.
SAD directly discards the integral, uses the extended state observer to observe the total disturbance, and compensates the system into a pure integral chain. In this way, it is much easier to control. It can be said that the expansion state observer is the soul and essence of SAD. Using the extended state observer, SAD can theoretically compensate an arbitrary order system into an arbitrary order integral chain, and then a simple linear control method can be used to achieve control, and a better control effect can also be obtained. Therefore, the expanded state observer is the essence of SAD. It is no problem to replace the previous following-up differentiator with another linear filter, and it is no problem to replace the nonlinear feedback with linear feedback, but the idea of observing disturbance and compensating cannot leave.
In summary, the differences between the two can be summarized as follows:
- Different in terms of mathematical models
Classical control theory mainly uses ordinary differential equations, transfer functions and dynamic structure diagrams. It only describes the relationship between the input and output of the system, but cannot describe the internal structure of the system and the changes in the system, and ignores the initial conditions. Information about the internal state of the system cannot be fully described.
2.The foundations are different.
Classical control theory is a branch of automatic control theory based on frequency response method and root locus method. Modern control theory is a control theory based on the state-space method, which is a major component of automatic control theory.
- Different systems
The research object of classical control theory is the automatic control system with single input and single output, especially the linear steady system. The characteristic of classical control theory is that the input and output characteristics (mainly transfer function) are used as the mathematical model of the system, and the graphical analysis methods such as the frequency response method and the root locus method are used to analyze the system performance and design the control device.
In modern control theory, the analysis and design of the control system are mainly carried out by describing the state variables of the system, and the basic method is the time domain method. Modern control theory can deal with a much wider range of control problems than classical control theory, including linear and nonlinear systems, stationery and time-varying systems, single-variable systems, and multi-variable systems.
- Different methods
The mathematical basis of classical control theory is the Laplace transform, and the dominant analysis and synthesis method is the frequency domain method. The methods and algorithms used in modern control theory are also more suitable for digital computers. Modern control theory also offers the possibility to design and construct optimal control systems with specified performance indicators.
The authors have combed and refined Sections 3.2 and 3.3 proposed by Reviewer 4.
Q3. Please divide conclusions section in two parts: discussion and conclusions. In discussion, please provide more information about simplifications and their possible effect on the results and on the conclusions (separately). Please make a discussion about the possible practical application of the work, especially if it is possible, please do an additional simple analysis of the benefits reached in the exemplary practical application in short term and long-term application. Please discuss future work.
Sincerely respond to Q3 raised by Reviewer 4:
As suggested by reviewer 4, the authors have divided the conclusion part of the paper into two parts: one is the discussion part; which describes more information about the simplification and its possible impact on the results of the numerical analysis and the summary of the conclusions. The second is the conclusion part; it is expected that a further brief analysis of the effects achieved by typical practical applications for short-term and long-term applications will be carried out in conjunction with the ongoing research work and practical applications in the future. The specific revised content and items have been reflected in this manuscript.
Q4. In conclusions, please add only short statements that contains QUANTITATIVE findings of the work and please highlight the great meaning of the work and new / sufficient contribution to the field. It is very important to present here only main findings in short sentences and with the numbers not only words and description. Words and description should be placed in the discussion section.
Sincerely respond to Q4 raised by Reviewer 4:
The authors have accepted the comments of Reviewer 4 and have added a brief statement containing the quantitative findings of the work to the Discussion Section of this revised manuscript, both in short sentences and numerically, and highlighting the importance of the research work and new and sufficient contributions to the field. The specific revised content and items have been reflected in this manuscript.
We are very hoped to publish this article in your journal, and I thank you on behalf of our group. We apologize for what we have not done well. We hope we will continue to submit better articles to you.
Thank you in advance for considering this revised submission.
Sincerely Yours
Corresponding author:
Xigui Wang Professor/Doctoral Supervisor
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Jian Zhang
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Yongmei Wang Professor/Master's Supervisor
School of Motorcar Engineering, Heilongjiang Institute of Technology, No. 999, Hongqidajie Road, Daowai District, Harbin, 150036, PRC
Chen Li
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Dear authors!
Thanks for answering my questions. The questions were not directed in such a way that the function of the individual components of the circuit is not obvious. My objections were directed to a non-standardly drawn schematic where the overall function is not clear. It is clear to me that the benefit is not the hydraulic circuit itself, yet I think the schematic should be drawn in accordance with standards, for example ISO 1219-1:2012 (see also attached file).
Comments for author File: Comments.pdf
Author Response
July 13, 2022
Dear Editor
Editor-in-Chief
Applied Sciences
We are very grateful for your help and suggestions in publishing the manuscript to the Applied Sciences, titled “Self- Anti-Disturbance Control of Hydraulic System Subjected to Variable Static Loads.” The paper was coauthored by Xigui Wang, Jian Zhang, Yongmei Wang, and Chen Li. We have checked the manuscript and revised it according to the comments. Overall, the comments have been fair, encouraging, and constructive. We have learned much from it. We submit here the revised manuscript to meet the evaluation conditions and requirements of the reviewers.
Response to Reviewers' Comments:
Reviewer: 2
Thanks for answering my questions. The questions were not directed in such a way that the function of the individual components of the circuit is not obvious. My objections were directed to a non- standardly drawn schematic where the overall function is not clear. It is clear to me that the benefit is not the hydraulic circuit itself, yet I think the schematic should be drawn in accordance with standards, for example ISO 1219-1:2012.
Sincere response to the questions raised by Reviewer 2:
The authors thank this Reviewer 2 for their sound advice, the authors have drawn schematics in accordance with the standard (ISO 1219-1:2012), and specific modifications have been reflected in the submitted revised manuscript.
Take three standard symbols of hydraulic components as an example to illustrate.
1) The symbols for the hydraulic components of the pilot operated pressure relief valve are as follows:
2) The symbols for the hydraulic components of the 3-way flow control valve are as follows:
The simplified form hydraulic symbols as shown
The detailed form hydraulic symbols as shown
3) The symbols for the hydraulic components of the hydraulic accumulator are as follows:
The authors thank Reviewer 2 for sound suggestions and valuable comments on this manuscript. Based on the questions raised by Reviewer 2, the authors conducted a comprehensive and detailed inspection and correction of the full text as suggested by Reviewer 2, and explained the problems raised by reviewer 2 one by one.
We are very hoped to publish this article in your journal, and I thank you on behalf of our group. We apologize for what we have not done well. We hope we will continue to submit better articles to you.
Thank you in advance for considering this revised submission.
Sincerely Yours
Corresponding author:
Xigui Wang Professor/Doctoral Supervisor
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Jian Zhang
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Yongmei Wang Professor/Master's Supervisor
School of Motorcar Engineering, Heilongjiang Institute of Technology, No. 999, Hongqidajie Road, Daowai District, Harbin, 150036, PRC
Chen Li
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Author Response File: Author Response.pdf
Reviewer 3 Report
Please find a review attached.
Comments for author File: Comments.pdf
Author Response
July 13, 2022
Dear Editor
Editor-in-Chief
Applied Sciences
We are very grateful for your help and suggestions in publishing the manuscript to the Applied Sciences, titled “Self- Anti-Disturbance Control of Hydraulic System Subjected to Variable Static Loads.” The paper was coauthored by Xigui Wang, Jian Zhang, Yongmei Wang, and Chen Li. We have checked the manuscript and revised it according to the comments. Overall, the comments have been fair, encouraging, and constructive. We have learned much from it. We submit here the revised manuscript to meet the evaluation conditions and requirements of the reviewers.
Response to Reviewers' Comments:
Reviewer: 3
The authors thank the reviewer 3 for reasonable suggestions and valuable comments on this manuscript. According to the questions raised by the reviewer 3, the author conducted a comprehensive and detailed inspection and correction of the full text according to the suggestions of the reviewer 3, and made necessary explanations to the questions raised by the reviewer 3 in order to expect the reviewer 3's approval and acceptance.
Q1. The title of the paper is irrelevant due to incorrect use of the term Optimization, and hard to understand because SAD and VSL are not common abbreviations. I propose changing it to "Self- Anti-Disturbance Control of Hydraulic System Subjected to Variable Static Loads". Please consider this variant or choose your one which would reflect the findings more accurate but would not confuse the reader.
Sincerely respond to Q1 raised by Reviewer 3:
The revised submission is titled
Self- Anti-Disturbance Control of Hydraulic System Subjected to Variable Static Loads
Q2. Please re-draw figures presenting the control system or supply them with detailed captions. The impression is that Fig. 3 does not contain any block for the controller presented in Fig. 4. Also, the controller in Fig. 8 is also presented without any interconnection with the system in Fig. 3.
Sincerely respond to Q2 raised by Reviewer 3:
The authors thank and accept reviewer 3 for his reasonable suggestion, which has been described in detail by the authors.
This paper proposes a Self-Anti-Disturbance (SAD) control strategy for Lubricating Oil Static Pressure (LOSP) acquisition to estimate the state and disturbances acting on an Electro-Hydraulic Servo (EHS) system that are designed and tested. The proposed optimal control strategy is based on an observer-provided system and its estimates that evaluate the absolute value of the error calculated between the available actual measurements. Experimental results are presented to verify the effectiveness of the proposed optimal control strategy in a real environment. Comparisons were made with the case of manual tuning of the gain and classical gain adaptation techniques reported in the literature, where the gain adaptation is based on a fixed rate of change. The proposed optimal control method provides comparable performance or outperforms manual and classical adaptive gain selection. Furthermore, compared to classical techniques, the proposed optimal control technique has the advantage of improving convergence in the case of rate-noise measurements with large estimation errors and limiting observer gain growth.
In related Figures 3.4 and 8, the classical approached (take Classic PID control as an example) is still the most widely used control algorithm until now, and this is still used in most control systems. The main advantage of classical PID control is that it does not require a model of the controlled object. The disadvantages of classical PID control can be expressed as follows:
1) Error extraction method (calculate errors directly from given instruction)
2) A way to extract the error differential from the error (using a traditional linear differentiator)
3) The weighted sum strategy is not necessarily the best (proportional, integral, differential terms are multiplied by the amplification factor and then added to calculate the control amount)
4) Integral feedback has many side effects (the error is amplified by the integral and then fed back to the system)
The Self-Anti-Disturbance (SAD) control strategy retains the advantages of classic PID control while making up for its shortcomings.
(1) Error Extraction Method - Arranging the Transition Process
Calculating the error directly according to the given command may lead to poor control effect. For example, some commands contain undesired high-frequency signals. Examples of such signals are: step command, square wave command. In order to solve the high frequency signal, the Self-Anti-Disturbance (SAD) control strategy proposes a method of arranging the "transition process", which is similar to low-pass filtering the given instruction to obtain a more easily implemented instruction, thereby sacrificing A little quickness while greatly reducing overshoot.
(2) The method of extracting the error differential from the error - following-up differentiator
A classic linear differentiator outputs a differential component while following-up a given signal. The Self-Anti-Disturbance (SAD) control strategy realizes the fastest control of discrete systems and eliminates the steady-state chatter of the following-up-differentiator. Extracting the error differential from the error consists of two parts: the differential of the instruction and the differential of the output (because the error is equal to the instruction minus the output). The differentiation of the instruction is done by the following-up-differentiator, and the differentiation of the output is done by the subsequent expansion state observer.
(3) The weighted sum strategy is not necessarily the best - nonlinear feedback
The traditional linear feedback method (that is, the error is directly multiplied by a gain) has insufficient convergence speed and anti-disturbance ability.
The solution of Self-Anti-Disturbance (SAD) control strategy is to replace traditional gain with nonlinear function (with nonlinear feedback instead of linear feedback).
(4) Side Effects of Integral Feedback - Expanded State Observer
One of the main functions of the integral is to remove the disturbance (it can be considered as a simple disturbance observer), but the integral works slowly and can cause overshoot.
SAD directly discards the integral, uses the extended state observer to observe the total disturbance, and compensates the system into a pure integral chain. In this way, it is much easier to control. It can be said that the expansion state observer is the soul and essence of SAD. Using the extended state observer, SAD can theoretically compensate an arbitrary order system into an arbitrary order integral chain, and then a simple linear control method can be used to achieve control, and a better control effect can also be obtained. Therefore, the expanded state observer is the essence of SAD. It is no problem to replace the previous following-up differentiator with another linear filter, and it is no problem to replace the nonlinear feedback with linear feedback, but the idea of observing disturbance and compensating cannot leave.
Q3. Fig. 6 contains a very strange process which could not occur in practice, see figure below. At time t = 0.3 sec three simultaneous states of amplitude are explicitly shown by crosses, dashed line is strictly vertical at time 0.3 sec. I see no explanation of this phenomenon. Please redraw the plot.
Sincerely respond to Q3 raised by Reviewer 3:
The authors have made revisions as reasonably suggested by reviewer 3 and the revised figures are reflected in this submitted manuscript.
Revised Figure 6
Figure 6. Step response curve of the EHS system after pole optimal configured control
Q4. Please exclude Figure 9. It is not necessary since any reader familiar with Matlab can plot similar picture. But, instead, exact poles and zeros selected for your experiment could be better given numerically.
Sincerely respond to Q4 raised by Reviewer 3:
The authors thank Reviewer 3 for the sound advice and necessary explanation, and the authors sincerely hope that Reviewer 3 will consider this.
The following-up differentiator follows the given signal and obtains its differential equation. Through the following-up differentiator, the transition process is arranged to prevent overshoot due to sudden change of the given signal, so that the controlled object approaches the target smoothly, thus improving the stability of the control system. By eliminating the poles of the system with the configured zeros, the system can be viewed as a control system approximating to 1.
The virtual control volume U(t) is determined by applying a self-turbulent controller design approach to the control subsystem y=U(t)/P1(s), which thus allows for a certain range of uncertainties to exist.
Let the virtual control volume U(t)=q(s)u, then the system becomes y=U(t)/P(s), then the actual control volume becomes u=U(t)/q(s).
If the controlled object conforms to the minimum phase system, then q(s) belongs to a stable polynomial and is known. At this point, a self-anti-disturbance controller can be used to determine the virtual control quantity U(t), and then the problem can be solved by considering U(t) as the input and solving the system u = U(t)/q(s) to obtain the actual control quantity u(t).
Based on the above analysis, the zero-pole diagram of the controlled object is derived using Matlab as shown in Figure 9.
Q5. Please rewrite Formula (26) without “where”. Also, each equation should be better given in a separate line.
Sincerely respond to Q5 raised by Reviewer 3:
The authors thank Reviewer 3 for his pertinent suggestion and have revised Equation (26) based on Reviewer 3's comments.
The revised equation (26) is as follows
(26)
Q6. When the formula is placed within a sentence, please, do not use indent as for a new paragraph. Examples of incorrect style can be found in lines 165, 178, 183, 186, 220 etc. Typically, incorrect lines start with “where” starting with a small letter.
Sincerely respond to Q6 raised by Reviewer 3:
The authors would like to thank Reviewer 3 for his reasonable suggestion, which has been revised in this submitted manuscript in light of the above comments.
Q7. Fig. 12 and 13 refer to one type of the study, but different colors of lines confuse the reader. In Fig. 13, the black line has no corresponding legend. Please unify the color legend in both plots and change legend “Case 1”, “Case 2”... to “PID”, “SAD” ... Also, please give the averaged values of control error in a table to make possible to estimate the overall performance of your approach in comparison with the other ones.
Sincerely respond to Q7 raised by Reviewer 3:
The authors would like to thank Reviewer 3 for his valuable suggestions, based on the above comments the authors have provided the necessary explanations and clarifications.
Comparative cases
In this paper, four cases are provided to verify the effectiveness of the proposed controller. Furthermore, the effectiveness of parameter adaptation and perturbation observers will be illustrated in these cases.
Auxiliary Fig. 1. Following-up errors of the four controllers for 10 mm - 0.5 Hz sinusoidal motion
First test the normal horizontal trajectory of 0.8965 mm for the four controllers. The following-up errors of the four controllers are shown in Auxiliary Fig. 1. In Figs 12 and 13, the following-up errors of the four controllers are collected in the last 4 seconds, with model-based adjustable compensation, where parameter adaptation and disturbance estimation are assigned to handle parameter uncertainty and time-varying disturbances. This verifies the effectiveness of the adaptive law in the controller, which can reduce parameter uncertainty through parameter adaptation. That is, this demonstrates the effectiveness of the disturbance compensation technique. The following-up performance of the proposed controller is more accurate than other controllers, and the maximum following-up error of the proposed controller is reduced to 40% of the other controllers.
The motivation for this study is the tracking challenge presented in hydraulic systems to develop a new controller that is efficient at the performance level and asymptotically stable at the theoretical level. To achieve this goal, we introduce an adaptive disturbance observer-based asymptotic tracking control algorithm that takes advantage of both adaptive control and disturbance observers. This approach can handle both parameter uncertainty and time-varying perturbations. Based on the proposed algorithm, we demonstrate stability analysis based on Lyapunov stability theory and achieve asymptotic stability. Our experimental tests show that the proposed controller can achieve improved tracking error in various situations. Considering that valve dynamics is neglected and in the controller design, one of our future studies is to investigate valve dynamics modeling techniques to achieve high dynamic tracking performance. In addition, in practical applications, due to cost/weight/volume and other reasons, full state feedback cannot be achieved, so the output feedback control of the hydraulic system will be further considered.
Q8. Literature is given in a strange manner with doubled numbers. Please fix this bug.
Sincerely respond to Q8 raised by Reviewer 3:
The authors would like to thank Reviewer 3 for his reasonable suggestion, which has been revised in the submitted manuscript based on the above comments.
Q9. All controllers are given in s-domain, but, obviously, practical implementation needs discretization. Please give a brief explanation which discrete operator did you use and why: classical shift operator, delta operator or any other. If you are not familiar with delta operator (to my expertise, it is not among standard operators in Matlab), I recommend getting acquainted with works “Delta operator realizations of direct-form IIR filters”, “The choice between delta and shift operators for low-precision data representation”, “Analysis and synthesis of delta operator systems” and other. If you find these works useful for your study, it would be kind of you to quote them.
Sincerely respond to Q9 raised by Reviewer 3:
The authors thank Reviewer 3 for his reasonable suggestion, who have given the necessary explanations based on the above comments.
This paper proposes an output feedback adaptive controller based on an extended state observer with continuous compensation for hydraulic servo control systems. It retains the original model of the main physical properties without increasing the complexity of the system stability analysis. With the development of new parameter adaptation laws, the parameter uncertainty problem is overcome, enabling more precise compensation. On the other hand, well-developed adaptation laws are driven by both following-up error and observation error. Therefore, the burden of extending the state observer to solve the remaining uncertainty is greatly alleviated and high-gain feedback is avoided, which means better following-up performance and robustness are achieved. The designed controller can handle not only matching uncertainty, but also mismatching dynamics, requiring very little system information, and more importantly, it is an output feedback-based approach, i.e., the synthesized controller relies only on the system the input signal and position output signal, greatly reducing the effects caused by signal contamination, measurement noise and other unexpected dynamics. Based on Lyapunov-based analysis it is demonstrated that this strategy provides the specified following-up transient performance and the final following-up accuracy only obtains asymptotic following-up performance in the presence of parametric uncertainty. Finally, comparative experiments are carried out on a hydraulic servo platform to verify the performance of the proposed control strategy for high following-up.
Q10. A more complex description of the plant and hardware should be added. What is the target platform intended to execute the control algorithm in real application? Does it use fixed-point or floating-point arithmetic?
Sincerely respond to Q10 raised by Reviewer 3:
The authors thank Reviewer 3 for his reasonable suggestion and have made the necessary explanations based on the above comments.
The algorithm can follow-up the path a conflict may follow as it evolves from a chosen starting state (usually the status quo) to a state of interest (such as an attractive solution). The value of the logical method is that it can explain how evolution happened, and the method facilitates computation. To illustrate this algorithm, the mention model has been applied to several real case studies. Since the control accuracy is one of the most important indicators, this research focuses on the high-precision following-up control of the hydraulic system. We propose an adaptive disturbance observer-based asymptotic following-up control strategy for hydraulic systems to achieve our motivation. It is worth noting that parameter uncertainty and disturbance are challenges in the field of hydraulic system control. In this study, time-varying disturbances are compensated by model-based feedforward control terms designed by the symbols of the disturbance observer based on the error function. To avoid the high learning burden of disturbances due to parameter uncertainty, online adaptive control is combined with the disturbance observer described above. In order to ensure the theoretical feasibility of the proposed controller, stability analysis is done by introducing stability theory. The proposed adaptive disturbance observer-based controller can achieve global asymptotic stability when the considered system is subject to mismatch or matching disturbances. The comparative experimental results show that the method can obtain the state-of-the-art hydraulic system following-up performance under the influence of parameter uncertainty and time-varying disturbance. To verify the effectiveness of the proposed controller, a hydraulic system platform, as well as measurement and control systems, have been established. The measurement and control system consists of monitoring software, real-time control software, A/D card (Advantech PCI-1716), D/A card (Advantech PCI-1723), counter card (Heidenhain IK-220); all these cards are 32-bit. The monitoring software is programmed with NI LabWindows/CVI, and the real-time control software is compiled with Microsoft Visual Studio 2005 plus Ardence RTX7.0. Ardence RTX 7.0 is used to provide a real-time working environment for real-time control software under Windows XP operating system. To implement the proposed controller and other comparison controllers defined below, the discretized C++ code was programmed. The control sampling time for the control is 0.4 ms. In this manuscript, some cases are provided to verify the effectiveness of the proposed controller. Furthermore, the effectiveness of parameter adaptation and perturbation observers will be illustrated in these cases.
In this paper, an extended state observer-based output feedback adaptive controller with continuous compensation is proposed for hydraulic servo control systems. An extended state observer with a new parameter adaptation law is constructed, and then the matched uncertainty and mismatched uncertainty are resolved through feedforward compensation. In addition, the output feedback is realized since the proposed extended state observer adaptive controller, which can reduce the influence of signal pollution, measurement noise, etc. The synthetic controller not only guarantees the specified following-up transient performance and final following-up accuracy, but also achieves asymptotic following-up performance in the presence of only parametric uncertainty, which is demonstrated by Lyapunov-based analysis. Finally, comparative experiments are carried out on a hydraulic servo platform to verify the high following-up performance of the proposed control strategy. Although this paper considers the nonlinearity of the hydraulic system and significantly improves the following-up performance, there are many other nonlinear factors in the hydraulic system. Future work should focus on dead zone and latency issues. Furthermore, the main idea of the proposed control strategy can be combined with other control objectives such as hydraulic robots, machine tools, etc. We believe that this practical control strategy will help achieve better control performance in the relevant implementations.
We are very hoped to publish this article in your journal, and I thank you on behalf of our group. We apologize for what we have not done well. We hope we will continue to submit better articles to you.
Thank you in advance for considering this revised submission.
Sincerely Yours
Corresponding author:
Xigui Wang Professor/Doctoral Supervisor
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Jian Zhang
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Yongmei Wang Professor/Master's Supervisor
School of Motorcar Engineering, Heilongjiang Institute of Technology, No. 999, Hongqidajie Road, Daowai District, Harbin, 150036, PRC
Chen Li
School of Engineering Technology, Northeast Forestry University, No. 26, Hexing Road, Xiangfang District, Harbin, 150040, PRC
Author Response File: Author Response.pdf
