MFAC Based Fault Tolerant Traction Operation Control For High-Speed Trains

The large-scale and rapid development of high-speed railway transportation has provided a strong support for China to complete the building of a moderately prosperous society in an all-round way,and is a major strategic demand for China’s economic and social development.However,with the continuous improvement of the operating speed,the traction/brake actuators of high-speed trains are always in a state of high load operation,actuator fault diagnosis and fault-tolerant traction operation control have become the current research hotspots of train control system.In addition,due to the complex and changeable operating environment of high-speed trains and external random disturbances bring great challenges to the accurate modeling of its motion mechanics,the study of data-driven fault-tolerant traction operation control method based on Model Free Adaptive Control(MFAC)is of great significance to ensure the safe,reliable and stable operation of high-speed trains.The main work of this paper includes:(1)Aiming at the problem of estimating fault degree of actuators in traction/braking system of high-speed trains,a real-time fault degree estimation model based on classification and regression tree(CART)was established to realize real-time estimation of multiplicative and additive fault degree of high-speed train,which provided a basis for subsequent design of fault tolerant controller.Compared with BP neural network and support vector machine,the simulation results show that the proposed method has faster fault estimation speed and higher estimation accuracy.(2)In order to solve the operation control problem of high-speed train with unknown fault of traction/braking actuators,a single point partial form dynamic linearization data model(SPPFDL)considering actuators failure is established.Then,a SPPFDL based model free adaptive fault tolerant control(SPPFDL-MFAFTC)method for high-speed train speed tracking is proposed,and the stability analysis is carried out.Compared with the fault-tolerant control method in the literature [68],[70],SPPFDL-MFAFTC uses the CART model to improve the estimation accuracy of fault degree,meanwhile,the partial format dynamic linearization data model takes into account the influence of the change of the control input in the fixed length sliding time window on the change of the output at the next moment,so that the proposed algorithm can obtain better control effect.The simulation results show that the SPPFDL-MFAFTC method has smaller speed tracking error,better fault-tolerant control effect and passenger experience,and effectively guarantees the safe and stable operation of trains under fault conditions.(3)Aiming at the operation control problem of high-speed train with unknown fault of traction/braking actuators,a multi-point partial form dynamic linearization(MPPFDL)data model considering actuators failure is established.Then a MPPFDL based model free adaptive fault tolerant control(MPPFDL-MFAFTC)is proposed.Compared with the SPPFDL-MFAFTC method in this paper,the MPPFDL-MFAFTC method considers the real-time multiplicative and additive fault degree of the whole train,and allocates the fault function to each carriage.Therefore,the scheme designed in this paper is a new power distribution scheme,and the convergence of MPPFDL-MFAFTC is strictly proved.The comparison result of numerical simulation with PID shows that the MPPFDL-MFAFTC method effectively overcomes the actuator failure,ensuring that the train still has strong speed tracking performance under adverse conditions.