Chaoxu Mu

Prof. Dr. Chaoxu Mu


Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 January 2019 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
ZGZhongke Gao
Zhongke Gao
XWXinmin Wang
Xinmin Wang
YYYuxuan Yang
Yuxuan Yang
CMChaoxu Mu
Chaoxu Mu
QCQing Cai
Qing Cai
SZSiyang Zuo
Siyang Zuo
Driver fatigue evaluation is of great importance for traffic safety and many intricate factors would exacerbate the difficulty. In this paper, based on the spatial–temporal structure of multichannel electroencephalogram (EEG) signals, we develop a novel EEG-based spatial–temporal convolutional neural network (ESTCNN) to detect driver fatigue. First, we introduce the core block to extract temporal dependencies from EEG signals. Then, we employ dense layers to fuse spatial features and realize classification. The developed network could automatically learn valid features from EEG signals, which outperforms the classical two-step machine learning algorithms. Importantly, we carry out fatigue driving experiments to collect EEG signals from eight subjects being alert and fatigue states. Using 2800 samples under within-subject splitting, we compare the effectiveness of ESTCNN with eight competitive methods. The results indicate that ESTCNN fulfills a better classification accuracy of 97.37% than these compared methods. Furthermore, the spatial–temporal structure of this framework advantages in computational efficiency and reference time, which allows further implementations in the brain–computer interface online systems.

13 April 2020 in IEEE Transactions on Power Electronics
Verified
Journal Article
Closed access
Abdul khalique JunejoAbdul khalique Junejo
Abdul khalique Junejo
WXWei Xu
Wei Xu
CMChaoxu Mu
Chaoxu Mu
MIMoustafa Magdi Ismail
Moustafa Magdi Ismail
Yi LiuYi Liu
Yi Liu
In order to enhance the speed control performance of the permanent magnet synchronous motor (PMSM) with internal and external disturbances, a new adaptive terminal sliding mode reaching law (ATSMRL) is proposed with continuous fast terminal sliding mode control (CFTSMC). Firstly, the ATSMRL is presented with the aim of reducing the input control efforts, which can dynamically adopt all positive aspects in terms of the finite time convergence, high tracking precision, and reduction of the chattering in the control input of the system. Secondly, an extended sliding mode disturbance observer (ESMDO) is designed to estimate the total disturbances of the system, and then the estimated disturbance has been brought for the feed-forward compensation technique, which would enhance the disturbance rejection ability of the system. Afterwards, the close loop stability is validated by the Lyapunov function. Finally the comprehensive numerical and experimental analyses have been carried out to demonstrate the superiority of the ATSMRL method than that of terminal sliding mode reaching law (TSMRL).

9 April 2018 in IEEE Transactions on Power Electronics
Verified
Journal Article
Closed access
WXWei Xu
Wei Xu
YJYajie Jiang
Yajie Jiang
CMChaoxu Mu
Chaoxu Mu
Frede BlaabjergFrede Blaabjerg
Frede Blaabjerg
The conventional rotor flux estimation method has issues of dc offset and harmonics, which are caused by initial rotor flux, detection errors, etc. To eliminate these defects, one improved nonlinear flux observer is proposed for sensorless control of permanent magnet synchronous machine (PMSM). First, the rotor position estimation method based on PMSM rotor flux observation is studied. Meanwhile, the limitations of the traditional rotor flux estimators, i.e., the saturation of pure integrator, phase shift, and amplitude attenuation of a low-pass filter are analyzed. Then, two novel flux observers, second-order generalized integral flux observer (SOIFO) and second-order SOIFO, are designed for the rotor flux estimation of PMSM. Based on second-order generalized integrator (SOGI) structure, the SOIFO can limit the dc component to a certain value. Furthermore, the second-order SOIFO is developed from the SOGI, which is characterized with effective dc and harmonics attenuation capability. With the second-order SOIFO, even without magnitude and phase compensation, the dc offset and harmonics of estimated rotor flux could be well eliminated. Therefore, the speed and rotor position can be estimated accurately. All the performances of the four methods are analyzed by transfer functions and Bode diagrams. Finally, the new sensorless control strategy is validated by comprehensive experimental results.

15 September 2017 in IEEE Transactions on Vehicular Technology
Verified
Journal Article
Closed access
YJYajie Jiang
Yajie Jiang
WXWei Xu
Wei Xu
CMChaoxu Mu
Chaoxu Mu
Yi LiuYi Liu
Yi Liu
To promote the drive performance of permanent magnet synchronous machine (PMSM), such as tracking accuracy of both speed and current, one improved deadbeat-based predictive current control (DPCC) scheme based sliding mode is proposed in this paper. Firstly, one novel PMSM model is derived by considering uncertainties of both parameters and external disturbances. Secondly, in order to improve the dynamic response of PMSM drive system, both sliding mode control (SMC) and DPCC are employed to control the speed and current respectively. Thirdly, a unified high-order sliding mode observer (HSMO) is designed for the estimation of disturbances and uncertainties in the speed and current loops. Furthermore, the estimated values are compensated with feedback to the designed SMC and DPCC to increase the speed robustness and current tracking accuracy. Comprehensive simulation and experiments demonstrate that the proposed control strategy is strongly robust to acute variations of load and machine parameters, and it is testified to have better speed and current tracking performance.

14 April 2017 in IEEE Transactions on Industrial Electronics
Verified
Journal Article
Open access
CMChaoxu Mu
Chaoxu Mu
YTYufei Tang
Yufei Tang
HHHaibo He
Haibo He
Randomness from the power load demand and renewable generations causes frequency oscillations among interconnected power systems. Due to the requirement of synchronism of the whole grid, load frequency control (LFC) has become one of the essential challenges for power system stability and security. In this paper, by modeling the disturbances and parameter uncertainties into the LFC model, we propose an adaptive supplementary control scheme for the power system frequency regulation. An improved sliding mode control (SMC) is employed as the basic controller, where a new sliding mode variable is specifically proposed for the LFC problem. The adaptive dynamic programming strategy is used to provide the supplementary control signal, which is beneficial to the frequency regulation by adapting to the real-time disturbances and uncertainties. The stability analysis is also provided to guarantee the reliability of the proposed control strategy. For comparison, a particle swarm optimization-based SMC scheme is developed as the optimal parameter controller for the frequency regulation problem. Simulation studies are performed on single-area and multiarea benchmark systems, and comparative results illustrate the favorable performance of the proposed adaptive approach for the frequency regulation under load disturbances and parameter uncertainties.

19 March 2019 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
CMChaoxu Mu
Chaoxu Mu
YZYong Zhang
Yong Zhang
In this paper, a learning-based robust tracking control scheme is proposed for a quadrotor unmanned aerial vehicle system. The quadrotor dynamics are modeled including time-varying and coupling uncertainties. By designing position and attitude tracking error subsystems, the robust tracking control strategy is conducted by involving the approximately optimal control of associated nominal error subsystems. Furthermore, an improved weight updating rule is adopted, and neural networks are applied in the learning-based control scheme to get the approximately optimal control laws of the nominal error subsystems. The stability of tracking error subsystems with time-varying and coupling uncertainties is provided as the theoretical guarantee of learning-based robust tracking control scheme. Finally, considering the variable disturbances in the actual environment, three simulation cases are presented based on linear and nonlinear models of quadrotor with competitive results to demonstrate the effectiveness of the proposed control scheme.