Machines, Volume 11, Issue 1 (January 2023) – 124 articles

Surface defect inspection is vital for the quality control of products and the fault diagnosis of equipment. Defect inspection remains challenging due to the low level of automation in some manufacturing plants and the difficulty in identifying defects. To improve the automation and intelligence levels of defect inspection, a CNN model is proposed for the high-precision defect inspection of USB components in the actual demands of factories. First, the defect inspection system was built, and a dataset named USB-SG, which contained five types of defects—dents, scratches, spots, stains, and normal—was established. The pixel-level defect ground-truth annotations were manually marked. This paper puts forward a CNN model for solving the problem of defect inspection tasks, and three strategies are proposed to improve the model’s performance. The proposed model is built based on the lightweight SqueezeNet network, and a rich feature extraction block is designed to capture semantic and detailed information. Residual-based progressive feature integration is proposed to fuse the extracted features, which can reduce the difficulty of model fine-tuning and improve the generalization ability. Finally, a multi-step deep supervision scheme is proposed to supervise the feature integration process. The experiments on the USB-SG dataset prove that the model proposed in this paper has better performance than that of other methods, and the running speed can meet the real-time demand, which has broad application prospects in the industrial inspection scene. Full article

The opposed-power reciprocating pump has the characteristics of high pressure, large flow, and high efficiency and energy saving. However, due to the special structure of the opposed-power reciprocating pump, existing theoretical methods cannot analyze its dynamic performance. Therefore, this paper proposes a method of analyzing the power end of the opposed-power reciprocating pump. Firstly, according to the working principle and structural characteristics of the traditional plunger pump, the novel and complex structure of the opposed-power reciprocating pump is analyzed by analogy, and the force analysis model of the crankshaft is established. The dynamic analysis model of the Matlab program is used to solve the dynamic load and section stress in the working process, and the variation law of crankshaft load is obtained. The 25 most critical working conditions are selected for analysis, and the most critical station and section of the crankshaft are obtained. With the connection between ANSYS Workbench and Solidworks, the model is imported into ANSYS Workbench, the load on the crank pin is loaded by APDL command flow, and the static analysis of the crankshaft is carried out to obtain the stress and strain of the crankshaft. Finally, the static and fatigue strength of the dangerous section is checked, and it is proven that the strength and stiffness of the crankshaft meet the design requirements. The results show that the dynamic analysis results of the crankshaft under critical working conditions are consistent with the finite element analysis, verifying the rationality of the method and providing a reference for the improvement and optimized design of the crankshaft of the opposed-power reciprocating pump. Full article

Edge computing can avoid the long-distance transmission of massive data and problems with large-scale centralized processing. Hence, defect identification for insulators with object detection models based on deep learning is gradually shifting from cloud servers to edge computing devices. Therefore, we propose a detection model for insulators and defects designed to deploy on edge computing devices. The proposed model is improved on the basis of YOLOv4-tiny, which is suitable for edge computing devices, and the detection accuracy of the model is improved on the premise of maintaining a high detection speed. First, in the neck network, the inverted residual module is introduced to perform feature fusion to improve the positioning ability of the insulators. Then, a high-resolution detection output head is added to the original model to enhance its ability to detect defects. Finally, the prediction boxes are post-processed to incorporate split object boxes for large-scale insulators. In an experimental evaluation, the proposed model achieved an mAP of 96.22% with a detection speed of 10.398 frames per second (FPS) on an edge computing device, which basically meets the requirements of insulator and defect detection scenarios in edge computing devices. Full article

Exploring the transformation of spatial mechanisms from their unfolded to controlled folding states to meet the requirements of various application scenarios has long been a hot topic in mechanical structure research. Although conventional spatial mechanisms can be designed to meet almost any application scenario, the design’s complex and excessive combinations of structural components, kinematic pairs, and drive units are unavoidable. It introduces many problems, such as poor reliability, drive complexity, and control difficulties. Based on 4D printing technology, the design of self-folding spatial mechanisms that use pre-stressed response properties under predetermined thermal excitation to achieve different shrinkage ratios integrates the control and drive system and the structural components and kinematic pairs. It brings novel features of self-folding while effectively avoiding many problems associated with conventional mechanical design. Further, the pre-stressed response model introduces the self-folding spatial mechanisms’ excitation, morphing, and driving investigation. Self-folding spatial mechanisms with different shrinkage ratios were prepared via fused deposition modeling, which verified the theoretical analysis and pre-stress response model and the design’s correctness and feasibility by experiments. The existing 4D printing technology lacks a paradigmatic design method in the application field. Contrarily, this work organically combined the conventional mechanical structure design with materials and fabrication via fused deposition modeling. A systematic study of self-folding spatial mechanisms from structural design to morphing control was carried out. This design is expected to introduce a novel paradigm of 4D printing technology in conventional mechanical design and has considerable application prospects in spherical radar calibration mechanisms. Full article

As the main engineering power plant, diesel engines are irreplaceable in the future. However, the stringent emission regulations impose many tough requirements to their developments. Recently, flexible fuel injection strategy has been recognized as an effective technology in creating an advanced spray and mixture formation and improving combustion efficiency indirectly. However, the detailed combustion and emission behaviors under flexible fuel injection are still unknown. Therefore, this paper aims to investigate the combustion and emission characteristics under flexible fuel injection and explore an optimal injection strategy for high-efficiency combustion. A numerical simulation method is conducted by coupling the large-eddy simulation (LES) model and the SAGE combustion model. Then, the spray mixing, combustion flame propagation and emissions formation under various multiple-injection strategies are investigated. Results reveal that initial an ultrahigh injection pressure has a significant influence on the spray’s axial penetration while dwell time mainly affects the spray’s radial expansion. Under an initial ultrahigh injection pressure, the turbulence kinetic energy (TKE) becomes larger, and the vortex motions are stronger, contributing to a better spray turbulent mixing. Meanwhile, a snatchier flame structure with a favorable level of equivalence ratio and a homogeneous temperature distribution is obtained. In this way, the peak heat release rate (HRR) could increase by 46.7% with a 16.7% reduction in soot formation and a 31.4% reduction in NO_x formation. Full article

Considering the problem that the early fault signal of rolling bearing is easily interfered with by background information, such as noise, and it is difficult to extract fault features, a method of rolling bearing early fault diagnosis based on the threshold acquisition U-Net (TA-UNet) is proposed. First, to improve the feature extraction ability of U-Net, the channel spatial threshold acquisition network (CS-TAN) and the dilated convolution module (DCM) based on different dilated rate combinations are introduced into the U-Net to construct the TA-UNet. Among them, the CS-TAN can adaptively learn the threshold, reduce the interference of noise in the signal, and the DCM can improve the multi-scale feature extraction ability of the network. Then, the TA-UNet is used for early fault diagnosis, and the method is divided into two steps: The model training phase and the vibration signal fault feature extraction phase. In the first step, additive gaussian white noise is added to the vibration signal to obtain the noise-added vibration signal, and the TA-UNet is trained to learn how to denoise the noise-added vibration signal. In the second step, the trained TA-UNet is used to extract the fault features of vibration signals and diagnose the early fault types of rolling bearing. The two-step method solves the problem that U-Net, as a supervised neural network, needs corresponding labeled data to be trained, as it realizes the fault diagnosis of unlabeled data. The feature extraction capability of the TA-UNet is evaluated by denoising the simulated signal of rolling bearing. The effectiveness of the proposed diagnostic method is demonstrated by the early fault diagnosis of open-source datasets. Full article

One-dimensional (1D) aerodynamic performance predictions are very often conducted by researchers and designers during the preliminary design of centrifugal compressors. This paper focuses on a 1D prediction method for centrifugal compressors with splitter blades, which is rarely seen in the open literature. One-dimensional prediction of aerodynamic overall performance is made for centrifugal compressors with different technical design specifications. However, the aerodynamic overall prediction accuracy relies on the accuracy of the 1D-loss-models used. Therefore, an optimum combination of loss models is proposed by summarizing a variety of loss models presented in the public literature. In addition, an optimization method is utilized to optimize some coefficients involved in loss models in order to improve the generality of the combined model. The modified models obtained in this study are proved to have good predictive accuracy. Full article

The rollover evaluation index provides an important threshold basis for the anti-rollover control system of vehicle. Regarding the rollover risk of independently driven heavy-duty vehicles, a new rollover evaluation index is proposed, and the feasibility of the improved index was verified through hierarchical control and HIL (hardware-in-the-loop) experiments. Based on an 18-DOF spatial dynamics model of a heavy-duty vehicle, the improved LTR (load transfer rate) index was obtained to describe the dynamic change in the tire’s vertical load. It replaces the suspension force and the vertical inertia force of the unsprung load mass. It avoids the problem of directly measuring or estimating the vertical load in the LTR index. Under the conditions of fishhooking and angle stepping, three types of rollover indicators were compared, and the proposed index can more sensitively identify the likelihood of rollover. In order to apply the improved rollover index to a rollover control well, a hierarchical controller based on the identification of the slip rate of the road surface, ABS control with sliding mode, variable structure and differential braking was designed. Simulations and HIL tests proved that the designed controller can accurately predict the rollover risk and avoid the rollover in time. Under the condition of J-turning, the yaw rate, slip angle and maximum lateral acceleration are reduced by 9%, 16% and 3%, respectively; under the condition of fishhooking, the maximum yaw rate, slip angle and lateral acceleration are reduced by 12%, 18% and 3%, respectively. Full article

Three-dimensional (3D) object detection has a vital effect on the environmental awareness task of autonomous driving scenarios. At present, the accuracy of 3D object detection has significant improvement potential. In addition, a 3D point cloud is not uniformly distributed on a regular grid because of its disorder, dispersion, and sparseness. The strategy of the convolution neural networks (CNNs) for 3D point cloud feature extraction has the limitations of potential information loss and empty operation. Therefore, we propose a graph neural network (GNN) detector based on neighbor feature alignment mechanism for 3D object detection in LiDAR point clouds. This method exploits the structural information of graphs, and it aggregates the neighbor and edge features to update the state of vertices during the iteration process. This method enables the reduction of the offset error of the vertices, and ensures the invariance of the point cloud in the spatial domain. For experiments performed on the KITTI public benchmark, the results demonstrate that the proposed method achieves competitive experimental results. Full article

Manufacturing processes have become highly accurate and precise in recent years, particularly in the chemical, aerospace, and electronics industries. This has attracted researchers to investigate improved procedures for monitoring and detection of small process variations to remain in line with such advances. Among these techniques, statistical process controls (SPC), in particular the control chart pattern (CCP), have become a popular choice for monitoring process variance, being utilized in numerous industrial and manufacturing applications. This study provides an improved control chart pattern recognition (CCPR) method focusing on X-bar chart patterns of small process variations using an ensemble classifier comprised of five complementing algorithms: decision tree, artificial neural network, linear support vector machine, Gaussian support vector machine, and k-nearest neighbours. Before advancing to the classification step, Nelson’s Rus Rules were utilized as a monitoring rule to distinguish between stable and unstable processes. The study’s findings indicate that the proposed method improves classification performance for patterns with mean changes of less than 1.5 sigma, and confirm that the performance of the ensemble classifier is superior to that of the individual classifier. The ensemble classifier can distinguish unstable pattern types with a classification accuracy of 99.55% and an ARL1 of 11.94. Full article

On-board train control equipment is an important component of the Train Control System (TCS) of railway trains. In order to guarantee the safe and efficient operation of the railway system, Predictive Maintenance (PdM) is significantly required. The operation data of the on-board equipment allow us to build fault prediction models using a data-driven approach. However, the problem of unbalanced fault samples makes it difficult to achieve the expected modeling performance. In this paper, a Conditional Generative Adversarial Network (CGAN) is adopted to solve the unbalancing problem by generating synthetic samples corresponding to specific fault labels that belong to the minority classes. With this basis, a CGAN-enhanced eXtreme Gradient Boosting (XGBoost) solution is presented for training the fault prediction models. From the pre-processing to the field data, artificial fault samples are generated and integrated into the training sample sets, and the XGBoost models can be derived with multiple decision trees. Both the feature importance sequence list and the knowledge graph are derived to describe the characteristics obtained by the models. Filed data sets from practical operation are utilized to validate the proposed solution. By comparison with conventional machine learning algorithms, it can be found that higher accuracy, precision, recall, and F1 scores, which are up to 99.76%, can be achieved by the proposed solution. By involving the CGAN strategy, the maximum enhancement to the F1 score with the XGBoost approach reaches 6.13%. The advantages of the proposed solution show great potential in implementing equipment health management and intelligent condition-based maintenance. Full article

In several MIMO system applications, the deviations of some output performance variables from their nominal values are required to be controlled independently, while the other performance variables are required to remain at their nominal value. This problem, named noninteracting control with simultaneous partial output zeroing, is important in the case of the common design of multi-model systems. To this end, the problem of a common noninteracting control with simultaneous common partial output zeroing is formulated. The present paper aims to develop a solution to the problem of multi-model normal linear time-invariant systems via regular and static measurement output feedback. The present approach follows the method developed for the solution of the common I/O decoupling problem. The main results of the paper are the introduction and the formulation of the problem at hand, the establishment of the necessary and sufficient conditions for its solvability, and the derivation of the respective general solution of the controller matrices. For the resulting closed-loop system, the additional design requirement of approximate command following a simultaneous I/O stabilizability is studied using a composite norm 2 type cost function and a metaheuristic algorithm for the derivation of the free parameters of the controller. The present results are illustrated through a numerical example of a nonlinear process with two operating points. Moreover, all the above results are successfully applied to the two-model description of a robot-tracked UGV, using a common controller feeding back measurements of the motor currents and the orientation of the vehicle. Full article

Nanofluids (NFs) synthesized via the suspension of diverse nanoparticles into conventional thermal fluids are known to exhibit better thermal, optical, tribological, and convective properties, photothermal conversion, and heat transfer performance in comparison with traditional thermal fluids. Stability is pivotal to NF preparation, properties, performance, and application. NF preparation is not as easy as it appears, but complex in that obtaining a stable NF comes with the harnessing of different preparation parameters. These parameters include stirring duration and speed, volume, density, base fluid type, weight/volume concentration, density, nano-size, type of mono or hybrid nanoparticles used, type and quantity of surfactant used, and sonication time, temperature, mode, frequency, and amplitude. The effect of these preparation parameters on the stability of mono and hybrid NFs consequently affects the thermal, optical, rheological, and convective properties, and photothermal conversion and heat transfer performances of NFs in various applications. A comprehensive overview of the influence of these preparation characteristics on the thermal, optical, rheological, and properties, photothermal conversion, and heat transfer performance is presented in this paper. This is imperative due to the extensive study on mono and hybrid NFs and their acceptance as advanced thermal fluids along with the critical importance of stability to their properties and performance. The various preparation, characterization, and stability methods deployed in NF studies have been compiled and discussed herein. In addition, the effect of the various preparation characteristics on the properties (thermal, optical, rheological, and convective), photothermal conversion, and heat transfer performances of mono and hybrid NFs have been reviewed. The need to achieve optimum stability of NFs by optimizing the preparation characteristics is observed to be critical to the obtained results for the properties, photothermal conversion, and heat transfer performance studies. As noticed that the preparation characteristics data are not detailed in most of the published works and thus making it mostly impossible to reproduce NF experimental studies, stability, and results; future research is expected to address this gap. In addition, the research community should be concerned about the aging and reusability of NFs (mono and hybrid) in the nearest future. Full article

A collaborative robot, or cobot, enables users to work closely with it through direct communication without the use of traditional barricades. Cobots eliminate the gap that has historically existed between industrial robots and humans while they work within fences. Cobots can be used for a variety of tasks, from communication robots in public areas and logistic or supply chain robots that move materials inside a building, to articulated or industrial robots that assist in automating tasks which are not ergonomically sound, such as assisting individuals in carrying large parts, or assembly lines. Human faith in collaboration has increased through human–robot collaboration applications built with dependability and safety in mind, which also enhances employee performance and working circumstances. Artificial intelligence and cobots are becoming more accessible due to advanced technology and new processor generations. Cobots are now being changed from science fiction to science through machine learning. They can quickly respond to change, decrease expenses, and enhance user experience. In order to identify the existing and potential expanding role of artificial intelligence in cobots for industrial applications, this paper provides a systematic literature review of the latest research publications between 2018 and 2022. It concludes by discussing various difficulties in current industrial collaborative robots and provides direction for future research. Full article

Recent works have already demonstrated that placing a crescent-shaped block upstream of a cylindrical hole could enhance the cooling performance of flat-plate films. The flow and cooling performance of the crescent-shaped block applied over the pressure and suction sides of the blade is investigated in this article. The Reynolds-averaged Navier-Stokes equations are solved with the Shear Stress Transport model for turbulence closure. Two optimized blocks are obtained from the flat-plate film cooling in our previous work, and two positions on the pressure and suction sides are tested. The blowing ratio varies from 0.5 to 2.0. The results show that when the block is applied on the blade surface, it yields a different cooling performance compared with the flat plate due to different geometry curvature and pressure gradient. The cooling performance on the suction side is slightly higher than on that on the pressure side, while the aerodynamic loss on the suction side is much higher. For the different blocks, the qualitative change of cooling performance vs. blowing ratios held on turbine blades is quite close to that of flat plates. The optimized smaller block in the flat plate provides better cooling performance at lower blowing ratios, while the larger block is superior when the blowing ratios are higher. Full article

Fault location is one of the main challenges in the distribution network due to its expanse and complexity. Today, with the advent of phasor measurement units (PMU), various techniques for fault location using these devices have been proposed. In this research, distribution network fault location is defined as an optimization problem, and the network fault location is determined by solving it. This is done by combining PMU data before and after the fault with the power system status estimation (PSSE) problem. Two new objective functions are designed to identify the faulty section and fault location based on calculating the voltage difference between the two ends of the grid lines. In the proposed algorithm, the purpose of combining the PMU in the PSSE problem is to estimate the voltage and current quantities at the branch point and the total network nodes after the fault occurs. Branch point quantities are calculated using the PMU and the governing equations of the π line model for each network section, and the faulty section is identified based on a comparison of the resulting values. The advantages of the proposed algorithm include simplicity, step-by-step implementation, efficiency in conditions of different branch specifications, application for various types of faults including short-circuit and series, and its optimal accuracy compared to other methods. Finally, the proposed algorithm has been implemented on the IEEE 123-node distribution feeder and its performance has been evaluated for changes in various factors including fault resistance, type of fault, angle of occurrence of a fault, uncertainty in loading states, and PMU measurement error. The results show the appropriate accuracy of the proposed algorithm showing that it was able to determine the location of the fault with a maximum error of 1.21% at a maximum time of 23.87 s. Full article

This paper investigates the path planning problem of an unmanned aerial vehicle (UAV) for completing a raid mission through ultra-low altitude flight in complex environments. The UAV needs to avoid radar detection areas, low-altitude static obstacles, and low-altitude dynamic obstacles during the flight process. Due to the uncertainty of low-altitude dynamic obstacle movement, this can slow down the convergence of existing algorithm models and also reduce the mission success rate of UAVs. In order to solve this problem, this paper designs a state detection method to encode the environmental state of the UAV’s direction of travel and compress the environmental state space. In considering the continuity of the state space and action space, the SD-TD3 algorithm is proposed in combination with the double-delayed deep deterministic policy gradient algorithm (TD3), which can accelerate the training convergence speed and improve the obstacle avoidance capability of the algorithm model. Further, to address the sparse reward problem of traditional reinforcement learning, a heuristic dynamic reward function is designed to give real-time rewards and guide the UAV to complete the task. The simulation results show that the training results of the SD-TD3 algorithm converge faster than the TD3 algorithm, and the actual results of the converged model are better. Full article

This paper develops an inner stator current controller based on an enhanced reaching-law-based discrete-time terminal sliding mode. The problem of tracking stator currents with high accuracy while ensuring the robustness of a six-phase induction motor in the presence of uncertain electrical parameters and unmeasurable states is tackled. The unknown dynamics are approximated by using a time delay estimation method. Then, an enhanced power-reaching law is used to make each stage of the convergence faster. A stability analysis and the system controller’s finite-time convergence are demonstrated in detail. Practical work was conducted on an asymmetrical six-phase induction machine to illustrate the developed discrete approach’s robustness and effectiveness. Full article

This paper presents a new investigation to detect various faults within the three-phase star and delta induction motors (IMs) using a frequency response analysis (FRA). In this regard, experimental measurements using FRA are performed on three IMs of ratings 1 HP, 3 HP and 5.5 HP in normal conditions, short-circuit fault (SC) and open-circuit fault (OC) conditions. The SC and OC faults are applied artificially between the turns (Turn-to-Turn), between the coils (Coil-to-Coil) and between the phases (Phase-to-Phase). The obtained measurements show that the star and delta IMs result in dissimilar FRA signatures for the normal and faulty windings. Various statistical indicators are used to quantify the deviations between the normal and faulty FRA signatures. The calculation is performed in three frequency ranges: low, middle and high ones, as the winding parameters including resistive, inductive and capacitive components dominate the frequency characteristics at different frequency ranges. Consequently, it is proposed that the boundaries for the used indicators facilitate fault identification and quantification. Full article

The navigation of mobile robots throughout the surrounding environment without collisions is one of the mandatory behaviors in the field of mobile robotics. The movement of the robot through its surrounding environment is achieved using sensors and a control system. The application of artificial intelligence could potentially predict the possible movement of a mobile robot if a robot encounters potential obstacles. The data used in this paper is obtained from a wall-following robot that navigates through the room following the wall in a clockwise direction with the use of 24 ultrasound sensors. The idea of this paper is to apply genetic programming symbolic classifier (GPSC) with random hyperparameter search and 5-fold cross-validation to investigate if these methods could classify the movement in the correct category (move forward, slight right turn, sharp right turn, and slight left turn) with high accuracy. Since the original dataset is imbalanced, oversampling methods (ADASYN, SMOTE, and BorderlineSMOTE) were applied to achieve the balance between class samples. These over-sampled dataset variations were used to train the GPSC algorithm with a random hyperparameter search and 5-fold cross-validation. The mean and standard deviation of accuracy ($ACC$), the area under the receiver operating characteristic ($AUC$), precision, recall, and $F1-score$ values were used to measure the classification performance of the obtained symbolic expressions. The investigation showed that the best symbolic expressions were obtained on a dataset balanced with the BorderlineSMOTE method with $\overline{ACC}\pm SD\left(ACC\right)$, ${\overline{AUC}}_{macro}\pm SD\left(AUC\right)$, ${\overline{Precision}}_{macro}\pm SD\left(Precision\right)$, ${\overline{Recall}}_{macro}\pm SD\left(Recall\right)$, and ${\overline{F1-score}}_{macro}\pm SD(F1-score)$ equal to $0.975\times 1.81\times {10}^{-3}$, $0.997\pm 6.37\times {10}^{-4}$, $0.975\pm 1.82\times {10}^{-3}$, $0.976\pm 1.59\times {10}^{-3}$, and $0.9785\pm 1.74\times {10}^{-3}$, respectively. The final test was to use the set of best symbolic expressions and apply them to the original dataset. In this case the $\overline{ACC}\pm SD\left(ACC\right)$, $\overline{AUC}\pm SD\left(AUC\right)$, $\overline{Precision}\pm SD\left(Precision\right)$, $\overline{Recall}\pm SD\left(Recall\right)$, and $\overline{F1-score}\pm SD(F1-Score)$ are equal to $0.956\pm 0.05$, $0.9536\pm 0.057$, $0.9507\pm 0.0275$, $0.9809\pm 0.01$, $0.9698\pm 0.00725$, respectively. The results of the investigation showed that this simple, non-linearly separable classification task could be solved using the GPSC algorithm with high accuracy. Full article

Continuum robots are good candidates for shape morphing. However, due to the coupled problem between kinematics and statics, the inverse kinematics of continuum robots is highly nonlinear, posing a challenging problem for real-time applications. This paper presents a simplified approach to solving the inverse kinematics of a flexible panel continuum robot efficiently. Through an experiment, two approximate relationships are discovered. First, the arc length of the middle backbone can be estimated from the arc lengths of the two panels; second, the length difference between the two panels can be related to the tip angle of the end-effector. Based on these two discovered relationships, a simplified inverse kinematics method is proposed based on a constant curvature model. This method has been validated by the experimental data with high accuracy of less than 2% error, thereby demonstrating the effectiveness of the proposed method for real-time applications. Full article

In 4D printing, structures with gradients in physical properties are 3D printed in order to dramatically increase deformation. For example, printing bilayer structures with passive and active layers has been proposed, however, these methods have the disadvantages that the material of each layer is mixed, and the modeling process is complicated. Herein, we present a method of creating gradient gels with different degrees of polymerization on the UV-exposed side and the other side using a single material by simply increasing the amount of initiator. This gel is the first example in which the differential swelling ratio between two sides causes the gradient to curl inward toward the UV-exposed side. The mechanical properties (swelling ratio and Young’s modulus) were measured at different material concentrations and structures, and the effects of each on deformation were analyzed and simulated. The results show that adding an initiator concentration of 0.2 (mol/L) or more causes deformation, that increasing the crosslinker concentration by a factor of three or more increases deformation, and that adding a hinge structure limits the gradient gel to deformation up to 90°. Thus, it was found that the maximum deformation can be predicted to some extent by simulation. In the future, we will be able to create complex structures while utilizing simulation. Full article

Since wind turbines are exposed to harsh working environments and variable weather conditions, wind turbine blade condition monitoring is critical to prevent unscheduled downtime and loss. Realizing that common convolutional neural networks are difficult to use in embedded devices, a lightweight convolutional neural network for wind turbine blades (WTBMobileNet) based on spectrograms is proposed, reducing computation and size with a high accuracy. Compared to baseline models, WTBMobileNet without data augmentation has an accuracy of 97.05%, a parameter of 0.315 million, and a computation of 0.423 giga floating point operations (GFLOPs), which is 9.4 times smaller and 2.7 times less computation than the best-performing model with only a 1.68% decrease in accuracy. Then, the impact of difference data augmentation is analyzed. The WTBMobileNet with augmentation has an accuracy of 98.1%, and the accuracy of each category is above 95%. Furthermore, the interpretability and transparency of WTBMobileNet are demonstrated through class activation mapping for reliable deployment. Finally, WTBMobileNet is explored in drones image classification and spectrogram object detection, whose accuracy and mAP@[0.5, 0.95] are 89.55% and 70.7%, respectively. This proves that WTBMobileNet not only has a good performance in spectrogram classification, but also has good application potential in drone image classification and spectrogram object detection. Full article

The synchronization control performance of the Fieldbus control system (FCS) is an important guarantee for the completion of multi-axis collaborative machining tasks, and its synchronization control accuracy is one of the decisive factors for the machining quality. To improve the synchronization control accuracy of FCS, this paper first makes a comprehensive analysis of the factors affecting synchronization in FCS. Secondly, by analyzing the communication model of linear Ethernet, a distributed clock compensation method based on timestamps is proposed to solve the asynchronous problem of communication data transmission in the linear ethernet bus topology. Then, based on the CANopen application layer protocol, the FCS communication and device control task collaboration method is proposed to ensure the synchronous control of multiple devices by FCS. Finally, an experimental platform is built for functional verification and performance testing of the proposed synchronization method. The results show that the proposed synchronization method can achieve a communication synchronization accuracy of 50 ns and a device control synchronization accuracy of 150 ns. Full article

Theoretical investigations into the capabilities of a coaxial inertial drive with various operating modes for vibratory conveyors and screens are conducted in the paper. The coaxial inertial exciter is designed with one asynchronous electric motor and the kinematically synchronized rotation of two unbalanced masses. Three variants of angular speeds ratios, namely ω₂/ω₁ = 1, ω₂/ω₁ = –1, and ω₂/ω₁ = 2, are considered. Based on these relations, the circular, elliptical, and complex motion trajectories of the working members are implemented. In the first two cases, single-frequency harmonic oscillations take place. In the latter case, the double-frequency periodic oscillations are excited. The dynamic behavior of the motor’s shaft during its running-up and running-out is considered. The influence of the inertial parameters of the unbalanced rotors and the relative phase shift angle between them on the elliptical trajectories of the vibratory system’s mass center motion is investigated. The use of forced kinematic synchronization provides the motion stability of the vibratory system for all considered working regimes. Full article

Macro–micro dual-drive technology uses a micro-drive system to compensate for motion errors of a macro-drive system, solving the contradiction between large travel and high-precision motion. Additionally, it has a wide range of applications in the ultra-precision field. Therefore, it is necessary to analyze and research the ultra-precision macro–micro dual-drive system. Firstly, this paper analyzes the history of ultra-precision technology development and summarizes the research status of ultra-precision technology processing and application. Secondly, the micro-drive mechanism design and macro–micro-drive mode of macro–micro dual-drive technology, which can solve the contradiction of large stroke and high precision, are reviewed, and the application of macro–micro dual-drive technology in an ultra-precision system is summarized. Finally, the challenges and development trends of the ultra-precision macro–micro dual-drive system are analyzed. The research in this paper will play an important role in promoting the development of the ultra-precision system and macro–micro dual-drive technology. Full article

This work presents the multi-objective optimization results of three experimental cases involving the laser sintering/melting operation and obtained by a virus evolutionary genetic algorithm. From these three experimental cases, the first one is formulated as a single-objective optimization problem aimed at maximizing the density of Ti6Al4V specimens, with layer thickness, linear energy density, hatching space and scanning strategy as the independent process parameters. The second one refers to the formulation of a two-objective optimization problem aimed at maximizing both the hardness and tensile strength of Ti6Al4V samples, with laser power, scanning speed, hatch spacing, scan pattern angle and heat treatment temperature as the independent process parameters. Finally, the third case deals with the formulation of a three-objective optimization problem aimed at minimizing mean surface roughness, while maximizing the density and hardness of laser-melted L316 stainless steel powder. The results obtained by the proposed algorithm are statistically compared to those obtained by the Greywolf (GWO), Multi-verse (MVO), Antlion (ALO), and dragonfly (DA) algorithms. Algorithm-specific parameters for all the algorithms including those of the virus-evolutionary genetic algorithm were examined by performing systematic response surface experiments to find the beneficial settings and perform comparisons under equal terms. The results have shown that the virus-evolutionary genetic algorithm is superior to the heuristics that were tested, at least on the basis of evaluating regression models as fitness functions. Full article

Herein, to accurately predict tool wear, we proposed a new deep learning network—that is, the IE-Bi-LSTM—based on an informer encoder and bi-directional long short-term memory. The IE-Bi-LSTM uses the encoder part of the informer model to capture connections globally and to extract long feature sequences with rich information from multichannel sensors. In contrast to methods using CNN and RNN, this model could achieve remote feature extraction and the parallel computation of long-sequence-dependent features. The informer encoder adopts the attention distillation layer to increase computational efficiency, thereby lowering the attention computational overhead in comparison to that of a transformer encoder. To better collect location information while maintaining serialization properties, a bi-directional long short-term memory (Bi-LSTM) network was employed. After the fully connected layer, the tool-wear prediction value was generated. After data augmentation, the PHM2010 basic dataset was used to check the effectiveness of the model. A comparison test revealed that the model could learn more full features and had a strong prediction accuracy after hyperparameter tweaking. An ablation experiment was also carried out to demonstrate the efficacy of the improved model module. Full article

This study presents a new flexible dynamic model for drive systems comprising lubricated bearings operating under conditions representative of electrified vehicle powertrains. The multi-physics approach importantly accounts for the tribological phenomena at the roller–race conjunction and models their effect on shaft-bearing system dynamics. This is achieved by embedding a non-linear lubricated bearing model within a flexible system level model; this is something which has not, to the authors’ knowledge, been reported on hitherto. The elastohydrodynamic (EHL) film is shown to increase contact deflection, leading to increased contact forces and total bearing stiffness as rotational speeds increase. Results show that for a 68 Nm hub motor operating up to 21,000 rpm, the input bearing EHL film reaches a thickness of 4.15 $\mathsf{\mu }$m. The lubricant entrainment increases the roller–race contact deflection, causing the contact stiffness to increase non-linearly with speed. The contribution of the lubricant film leads to a 16.6% greater bearing stiffness at 21,000 rpm when compared to conventional dry-bearing modelling methods used in current multi-body dynamic software. This new methodology leads to more accurate dynamic response of high-speed systems necessary for the next generation of electrified vehicles. Full article

At present, deep learning technology shows great market potential in broaching tool wear state recognition based on vibration signals. However, traditional single neural network structure is difficult to extract a variety of different features simultaneously and has low robustness, so the accuracy of wear status recognition is not high. In view of the above problems, a broaching tool wear recognition model based on ShuffleNet v2.3-StackedBiLSTM is proposed in this paper. The model integrates ShuffleNet v2.3, which has been channel shuffling, and StackedBiLSTM, a long and short-term memory network, to effectively extract spatial and temporal features for tool wear state recognition. Based on the innovative recognition model, the turbine disc fir-tree slot broaching experiment is designed, and the performance index system based on confusion matrix is adopted. The experimental research and results show that the model has outstanding accuracy, precision, recall, and F1 value, and the accuracy rate reaches 99.37%, which is significantly better than ShuffleNet v2.3 and StackedBiLSTM models. The recognition speed of a single sample was improved to 8.67 ms, which is 90.32% less than that of the StackedBiLSTM model. Full article