Research On Fault Diagnosis And Compensation For Traction System Actuator Of High-speed Trains

High-speed train as one of the important means of transportation and freight transportation is widely used.With the rapid development of the high speed rail technology and the continuous improvement of train's operating speed,the requirements for safety and reliability are becoming higher.The traction system generating the traction/breaking force consists of rectifiers,inverters,PWMs(pulse width modulations),traction motors,and mechanical drives,etc.,among which PWMs,traction motors,and mechanical drives are considered as actuators.Actuator failures are often uncertain in patterns,amplitudes,and time instances.These failed actuators may deteriorate the train performance severely,resulting in time delay or cancellation of the other trains.Therefore,it is crucial for the traction system of high-speed trains to study the effective failure compensation technologiesThis paper takes actuators of high-speed train traction system as the research object.Considered time-varying parameters of the high-speed train dynamics model caused by the multiple operating conditions and the time-varying friction characteristics,a piecewise constant model is established to describe the train longitudinal dynamics,in which the piecewise constants and the fault patterns are unknown.Moreover,both parameterized and non-parametric actuator fault modes are developed.Considered the multiple actuators in the high-speed train traction system with the unknown failure time and locations,an adaptive technique based fault diagnosis observer is designed for different failure locations,in which the observer parameters are updated by adaptive laws.Through the match of observer and train system,the fault can be diagnosed.Based on Lyapunov functions,it is proved that the observer errors can be convergent under the matched or cannot be convergent when unmatched.To diagnosis the position of failed actuators and deal with the unknown parameters,a fault alarm scheme is proposed by using a bank of adaptive observers to match the plant model.Simulation results on a high-speed train model are presented to illustrate the performance of the developed adaptive observer based fault diagnosis scheme.Then,the adaptive fault compensation tracking problem is investigated for high-speed trains with disturbances and actuator failures in this paper.An adaptive controller is designed for the healthy system with known bounds on disturbances.For both the parameterized and non-parametric failures,the adaptive failure compensation with the adaptive laws are presented to achieve the position tracking performance of high-speed train fault system in the present of disturbances with known bounds.The adaptive failure compensation for unknown bounds on disturbances is also discussed under the parameterized failure.The stability of the corresponding closed-loop system and asymptotic tracking performance are proved via Lyapunov direct method,and validated using a high-speed train model.Furthermore,this paper also studies the fault compensation tracking problem of high speed train with uncertain input and actuator failures based on sliding-mode control.For the healthy system with known bounds on disturbances,sliding surface and the control laws are designed to ensure the tracking performance of system.Considered the bounded fault occurred,the adaptive failure compensations based on sliding-mode are designed for both fault parameters bounds known and unknown.The convergence of tracking error is proved via Lyapunov direct method.The simulation results show the effectiveness of the proposed method.At last,the paper concludes and gives prospects for the study.The innovations and limitations are proposed for further investigations.