Guang-Zhong Cao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Guang-Zhong Cao has a total of 240 co-authors. Here are the top co-authors by number of shared publications:
- Su-Dan Huang
Shenzhen University
31 shared publications - Yeping Peng
Shenzhen University
11 shared publications - Yue-Peng Zhang
Shenzhen University of Information Technology
11 shared publications - Hong-Jin Hu
Shenzhen University
9 shared publications - Jiangbiao He
Department of Electrical Engineering and Computer Science, University of Tennessee, Knoxville, TN, USA
9 shared publications - Junqi Xu
Tongji University
9 shared publications
Latest Publications
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Magnetics
- Vol. 62 (10) , 7002106
A Data-Driven Magnetic Field Model for the Halbach Arrays of the Magnetically Levitated Planar Motor With Variable Harmonic Coefficients
Halbach arrays are widely employed in magnetically levitated planar motors (MLPMs) because of their magnetic field concentration properties. The end effect of a Halbach array causes magnetic field distortion, making it difficult to establish an accurate magnetic field model and thus tending to degrade the achieved control performance. In this paper, a novel data-driven magnetic field model with variable harmonic coefficients, which can accurately characterize the distorted magnetic field, is proposed. A linear relationship between the harmonic coefficients and the air gap length is derived to reduce the computational complexity of the model. The harmonic coefficients are optimized through Bayesian regularization to enhance the generalization ability of the model and minimize the induced model errors. The experimental results show that the proposed model agrees well with the measured magnetic field and exhibits low computational time consumption, demonstrating its potential for real-time MLPM control.
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Magnetics
- Vol. 62 (10) , 7002106
A Data-Driven Magnetic Field Model for the Halbach Arrays of the Magnetically Levitated Planar Motor With Variable Harmonic Coefficients
Halbach arrays are widely employed in magnetically levitated planar motors (MLPMs) because of their magnetic field concentration properties. The end effect of a Halbach array causes magnetic field distortion, making it difficult to establish an accurate magnetic field model and thus tending to degrade the achieved control performance. In this paper, a novel data-driven magnetic field model with variable harmonic coefficients, which can accurately characterize the distorted magnetic field, is proposed. A linear relationship between the harmonic coefficients and the air gap length is derived to reduce the computational complexity of the model. The harmonic coefficients are optimized through Bayesian regularization to enhance the generalization ability of the model and minimize the induced model errors. The experimental results show that the proposed model agrees well with the measured magnetic field and exhibits low computational time consumption, demonstrating its potential for real-time MLPM control.
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Magnetics
- Vol. 62 (10) , 6200606
An Intelligent Inversion Method to Identify Railway Ballast Gradation Parameters Using Electromagnetic Wave Reflection
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Magnetics
- Vol. 62 (10) , 6200606
An Intelligent Inversion Method to Identify Railway Ballast Gradation Parameters Using Electromagnetic Wave Reflection
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 73 (10) , 14978-14992
Intelligent Explicit Model Predictive Control of Planar Motors for High-Precision Position Tracking Applications
To address the challenges of achieving high-precision motion control under physical constraints in motion systems, an intelligent explicit model predictive control (IEMPC) method for planar motors is proposed in this article. Unlike conventional explicit model predictive control (EMPC) approaches that suffer from excessive memory requirements, limited model flexibility, and dependence on commercial solvers, the proposed method integrates intelligent algorithms with an error state feedback solution space to construct a novel EMPC framework. First, a prediction model of error state is established using an augmented state-space model. Based on this model, a multi-objective optimal position control problem under control input constraints is formulated by incorporating a defined cost function. An offline solving strategy is then developed, in which the error state space is partitioned into multiple convex elliptical regions and explicit error state feedback control laws are derived for each subregion. To solve the constrained multi-objective optimal position control problem, an improved nondominated sorting genetic algorithm II (NSGA-II) is designed, generating a set of Pareto-optimal parameters for the control laws. These parameters are subsequently mapped to corresponding elliptical subregions to determine the control law in each subregion. Finally, the effectiveness and superiority of the proposed method are verified through systematic comparative experiments.
- 1 October 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 73 (10) , 14978-14992
Intelligent Explicit Model Predictive Control of Planar Motors for High-Precision Position Tracking Applications
To address the challenges of achieving high-precision motion control under physical constraints in motion systems, an intelligent explicit model predictive control (IEMPC) method for planar motors is proposed in this article. Unlike conventional explicit model predictive control (EMPC) approaches that suffer from excessive memory requirements, limited model flexibility, and dependence on commercial solvers, the proposed method integrates intelligent algorithms with an error state feedback solution space to construct a novel EMPC framework. First, a prediction model of error state is established using an augmented state-space model. Based on this model, a multi-objective optimal position control problem under control input constraints is formulated by incorporating a defined cost function. An offline solving strategy is then developed, in which the error state space is partitioned into multiple convex elliptical regions and explicit error state feedback control laws are derived for each subregion. To solve the constrained multi-objective optimal position control problem, an improved nondominated sorting genetic algorithm II (NSGA-II) is designed, generating a set of Pareto-optimal parameters for the control laws. These parameters are subsequently mapped to corresponding elliptical subregions to determine the control law in each subregion. Finally, the effectiveness and superiority of the proposed method are verified through systematic comparative experiments.
- 5 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. PP (02780046) , 1-13
Composite Control With Prescribed Performance for High-Speed Maglev Train Levitation Systems
This article presents an improved composite control framework for high-performance levitation gap regulation in electromagnetic suspension (EMS) type high-speed maglev trains. The control architecture incorporates three strategically coordinated components: a prescribed performance function (PPF) to enforce transient performance by confining the tracking error within prescribed bounds, a cascaded finite-time extended state observer (CFTESO) for rapid and accurate estimation of lumped disturbances, and a global fast nonsingular terminal sliding mode controller (GFNTSMC) to achieve high-precision reference tracking with strong disturbance rejection. The stability of the closed-loop system and finite-time convergence are rigorously established via Lyapunov analysis. Real-time hardware-in-the-loop (HiL) validation under high-speed operation with real-world disturbances verifies the superiority of the proposed control framework. Compared with conventional control strategies, it performs coordinated optimization of the transient response and steady-state performance, along with higher control accuracy and more effective disturbance rejection.
- 5 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. PP (02780046) , 1-13
Composite Control With Prescribed Performance for High-Speed Maglev Train Levitation Systems
This article presents an improved composite control framework for high-performance levitation gap regulation in electromagnetic suspension (EMS) type high-speed maglev trains. The control architecture incorporates three strategically coordinated components: a prescribed performance function (PPF) to enforce transient performance by confining the tracking error within prescribed bounds, a cascaded finite-time extended state observer (CFTESO) for rapid and accurate estimation of lumped disturbances, and a global fast nonsingular terminal sliding mode controller (GFNTSMC) to achieve high-precision reference tracking with strong disturbance rejection. The stability of the closed-loop system and finite-time convergence are rigorously established via Lyapunov analysis. Real-time hardware-in-the-loop (HiL) validation under high-speed operation with real-world disturbances verifies the superiority of the proposed control framework. Compared with conventional control strategies, it performs coordinated optimization of the transient response and steady-state performance, along with higher control accuracy and more effective disturbance rejection.
- Open Access
- 1 February 2026
- Published by Tsinghua University Press in Friction
Wear evolution analysis of artificial joint based on wear particle recognition: A study under insufficient and imbalanced sample scenarios
Revealing the evolution law of wear conditions and mechanisms is critical for assessing the health status of artificial joints and thereby extending their service life. Notably, high-precision wear particle recognition plays a pivotal role in the intelligent wear analysis of joint implants. However, insufficient and imbalanced particle samples often lead to inadequate recognition accuracy, which fails to meet the requirements of reliable wear particle analysis. Such limitations highlight the urgent need for a robust solution to address sample scarcity and imbalance. To this end, an intelligent wear particle recognition method based on a generative adversarial network is proposed to enhance the accuracy of artificial joint wear analysis. Specifically, a multi-scale coordinated attention-based generative adversarial network is designed to produce high-quality wear particle samples, which are used to supplement scarce sample categories and balance the particle distribution. Subsequently, a wear particle classifier is con-structed using a neural network, and the training strategy for this classification model is further optimized to ensure high-precision wear particle recognition. Ablation experiments demonstrate that the proposed method achieves an accuracy, precision, recall, and F1-score of 0.90 for wear particle recognition when using the generated samples. All metrics represent a significant improvement compared to the results obtained from the original sample dataset. Meanwhile, the recognition performance of the proposed method outperforms that of other state-of-the-art algorithms. Based on the recognition results of wear particles from joint implants, the particle size and type distributions are derived to evaluate the wear evolution of artificial joint. These findings hold great value for advancing artificial joint wear analysis and enabling reliable failure prediction.
- 1 February 2026
- Published by Tsinghua University Press in Friction
Open AccessWear evolution analysis of artificial joint based on wear particle recognition: A study under insufficient and imbalanced sample scenarios
Revealing the evolution law of wear conditions and mechanisms is critical for assessing the health status of artificial joints and thereby extending their service life. Notably, high-precision wear particle recognition plays a pivotal role in the intelligent wear analysis of joint implants. However, insufficient and imbalanced particle samples often lead to inadequate recognition accuracy, which fails to meet the requirements of reliable wear particle analysis. Such limitations highlight the urgent need for a robust solution to address sample scarcity and imbalance. To this end, an intelligent wear particle recognition method based on a generative adversarial network is proposed to enhance the accuracy of artificial joint wear analysis. Specifically, a multi-scale coordinated attention-based generative adversarial network is designed to produce high-quality wear particle samples, which are used to supplement scarce sample categories and balance the particle distribution. Subsequently, a wear particle classifier is con-structed using a neural network, and the training strategy for this classification model is further optimized to ensure high-precision wear particle recognition. Ablation experiments demonstrate that the proposed method achieves an accuracy, precision, recall, and F1-score of 0.90 for wear particle recognition when using the generated samples. All metrics represent a significant improvement compared to the results obtained from the original sample dataset. Meanwhile, the recognition performance of the proposed method outperforms that of other state-of-the-art algorithms. Based on the recognition results of wear particles from joint implants, the particle size and type distributions are derived to evaluate the wear evolution of artificial joint. These findings hold great value for advancing artificial joint wear analysis and enabling reliable failure prediction.
- 14 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 73 (3) , 3675-3685
Piecewise Nonlinear Force–Current Model of Long-Stroke Double-Sided Linear Switched Reluctance Motors Using WOA-BPNN Algorithm
To improve the motion performance of long-stroke double-sided linear switched reluctance motors, a piecewise nonlinear force–current model is proposed by integrating the backpropagation neural network (BPNN) with the improved whale optimization algorithm (WOA). WOA features a few parameters and strong global search capability, overcoming the limitations of BPNNs in optimizing their structure and parameters, and improving the accuracy of the BPNN-based model. In long-stroke applications, an uneven double-sided air gap is inevitable, which intensifies the nonlinear relationship between the electromagnetic force and current. To address this issue, a segmentation strategy is employed to divide the long stroke into multiple short-stroke regions, based on the thrust force characteristic obtained from finite element analysis and the actual air gap distribution. This strategy enables the development of an accurate, highly nonlinear piecewise force–current model for long-stroke applications. In each short-stroke region, a nonlinear force–current model represented by a BPNN is developed. A WOA is then designed to optimize both the structure and parameters of the BPNN-based model. Moreover, a refined convergence factor incorporating nonlinear elements is introduced to enhance the global search capability of the WOA. The effectiveness of the WOA-BPNN-based force–current model is verified using the experimental sample set. Finally, experimental results demonstrate significant motion performance improvements over the conventional model and the unoptimized BPNN-based model.
- 14 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Electronics
- Vol. 73 (3) , 3675-3685
Piecewise Nonlinear Force–Current Model of Long-Stroke Double-Sided Linear Switched Reluctance Motors Using WOA-BPNN Algorithm
To improve the motion performance of long-stroke double-sided linear switched reluctance motors, a piecewise nonlinear force–current model is proposed by integrating the backpropagation neural network (BPNN) with the improved whale optimization algorithm (WOA). WOA features a few parameters and strong global search capability, overcoming the limitations of BPNNs in optimizing their structure and parameters, and improving the accuracy of the BPNN-based model. In long-stroke applications, an uneven double-sided air gap is inevitable, which intensifies the nonlinear relationship between the electromagnetic force and current. To address this issue, a segmentation strategy is employed to divide the long stroke into multiple short-stroke regions, based on the thrust force characteristic obtained from finite element analysis and the actual air gap distribution. This strategy enables the development of an accurate, highly nonlinear piecewise force–current model for long-stroke applications. In each short-stroke region, a nonlinear force–current model represented by a BPNN is developed. A WOA is then designed to optimize both the structure and parameters of the BPNN-based model. Moreover, a refined convergence factor incorporating nonlinear elements is introduced to enhance the global search capability of the WOA. The effectiveness of the WOA-BPNN-based force–current model is verified using the experimental sample set. Finally, experimental results demonstrate significant motion performance improvements over the conventional model and the unoptimized BPNN-based model.
- Open Access
Design and Analysis of a High-Speed Slotless Permanent Magnet Synchronous Motor Considering Air-Gap Airflow
The air-gap airflow significantly influences the performance of high-speed slotless permanent magnet synchronous motors (HSSPMSM), yet this critical factor is frequently overlooked during the design phase, resulting in performance deviations. This paper presents the design and multi-physical analysis of a 10 kW/40,000 rpm HSSPMSM, explicitly accounting for air-gap airflow effects. A comprehensive coupling model integrating electromagnetic, thermal, mechanical, and airflow fields is established to guide the motor design. Based on this analysis, the motor dimensions and parameters are determined, and a prototype is fabricated. Experimental validation demonstrates that the developed HSSPMSM successfully meets the design specifications. Considering air-gap airflow can obtain more accurate thermal design results with an accuracy improvement of 6.8% compared to not considering air-gap airflow. The close agreement between the simulated and measured performance confirms the effectiveness of the proposed design methodology that incorporates airflow effects.
Open AccessDesign and Analysis of a High-Speed Slotless Permanent Magnet Synchronous Motor Considering Air-Gap Airflow
The air-gap airflow significantly influences the performance of high-speed slotless permanent magnet synchronous motors (HSSPMSM), yet this critical factor is frequently overlooked during the design phase, resulting in performance deviations. This paper presents the design and multi-physical analysis of a 10 kW/40,000 rpm HSSPMSM, explicitly accounting for air-gap airflow effects. A comprehensive coupling model integrating electromagnetic, thermal, mechanical, and airflow fields is established to guide the motor design. Based on this analysis, the motor dimensions and parameters are determined, and a prototype is fabricated. Experimental validation demonstrates that the developed HSSPMSM successfully meets the design specifications. Considering air-gap airflow can obtain more accurate thermal design results with an accuracy improvement of 6.8% compared to not considering air-gap airflow. The close agreement between the simulated and measured performance confirms the effectiveness of the proposed design methodology that incorporates airflow effects.
- Open Access
- 12 May 2025
- Published by MDPI AG in Biomimetics
- Vol. 10 (5) , 311
Obstacle Feature Information-Based Motion Decision-Making Method for Obstacle-Crossing Motions in Lower Limb Exoskeleton Robots
To overcome the problem of insufficient adaptability to the motion environment of lower limb exoskeleton robots, this paper introduces computer vision technology into the motion control of lower limb exoskeleton robots and studies an obstacle-crossing-motion method based on detecting obstacle feature information. Considering the feature information of different obstacles and the distance between obstacles and robots, a trajectory planning method based on direct point matching was used to generate offline adjusted gait trajectory libraries and obstacle-crossing gait trajectory libraries. A lower limb exoskeleton robot obstacle-crossing motion decision-making algorithm based on obstacle feature information is proposed by combining gait constraints and motion constraints, enabling it to select appropriate motion trajectories in the trajectory library. The proposed obstacle-crossing-motion method was validated at three distances between the obstacle and the robot and with the feature information of four obstacles. The experimental results show that the proposed method can select appropriate trajectories from the trajectory library based on the detected obstacle feature information and can safely complete obstacle-crossing motions.
- 12 May 2025
- Published by MDPI AG in Biomimetics
- Vol. 10 (5) , 311
Open AccessObstacle Feature Information-Based Motion Decision-Making Method for Obstacle-Crossing Motions in Lower Limb Exoskeleton Robots
To overcome the problem of insufficient adaptability to the motion environment of lower limb exoskeleton robots, this paper introduces computer vision technology into the motion control of lower limb exoskeleton robots and studies an obstacle-crossing-motion method based on detecting obstacle feature information. Considering the feature information of different obstacles and the distance between obstacles and robots, a trajectory planning method based on direct point matching was used to generate offline adjusted gait trajectory libraries and obstacle-crossing gait trajectory libraries. A lower limb exoskeleton robot obstacle-crossing motion decision-making algorithm based on obstacle feature information is proposed by combining gait constraints and motion constraints, enabling it to select appropriate motion trajectories in the trajectory library. The proposed obstacle-crossing-motion method was validated at three distances between the obstacle and the robot and with the feature information of four obstacles. The experimental results show that the proposed method can select appropriate trajectories from the trajectory library based on the detected obstacle feature information and can safely complete obstacle-crossing motions.