Data-based Iterative Learning Control With Applications In Automatic Train Operation

As a way of transportation with high efficiency,environmental protection and energy efficient,more and more scholars pay attention to the research of train.In order to ensure the punctual,safe and energy-efficient operation of the train,the automatic train control(ATC)system has been configured on the high-speed train,urban rail and other modern trains.The ATC system includes automatic train op-eration(ATO),automatic train protection(ATP),and automatic train supervision(ATS).As the core component of ATC,one of the most core functions of ATO is to control the train to run according to the expected speed track and the relevant information provided by ATP and ATS.According to the characteristics of train repetitive operation,this paper applies the iterative learning control(ILC)to ATO system,and proposes two ILC control schemes to train speed tracking and energy-efficient control,and analyzes their convergence in theory,and then verifies the effectiveness and robustness of the two ILCs proposed in this paper through simu-lation and comparative analysis.The main work and innovation of this paper are summarized as follows:(1)Considering the restriction of energy consumption,the force between car-riages and the time-varying resistance,an improved model free adaptive iterative learning control(iMFAILC)is designed.The iMFAILC is obtained by minimizing the modified input criterion function by introducing an energy consumption term into the input criterioin function of the prototype MFAILC.When the train runs in a strict repetitive operation mode,i.e.with the same control task and operation environment,iMFAILC can realize the energy-efficient control and speed tracking of the train,which provide a systematic way of balancing the trade-off between the tracking performance and energy consumption.At the same time,the theoretical analysis proves the convergence of the proposed method on the iteration axis.(2)Considering the non-repetitive interference of high-speed train caused by weather,an improved model free adaptive iterative learning control with feedback control(iMFAILC-FC)algorithm is designed.In the process of train operation,the repetitive characteristics can be learned by the feedforward controller(iMFAILC)algorithm,while the non-repetitive interference can be identified and corrected by the feedback controller.This paper also analyzes the stability of the iMFAILC-FC in theory.(3)Taking the multi-point mass model of train as the controlled object,the simulation of the iMFAILC controller and the iMFAILC-FC controller are carried out.At the same time,The effectiveness and robustness of the proposed methods are demonstrated comparing with the prototype MFAILC controller,P-type ILC controller and model predictive control(MPC).The simulation results show that the iMFAILC controller can achieve the speed tracking and energy-efficient of the train simultaneously,and the iMFAILC-FC can effectively suppress the non-repetitive interference of the train in the operation process.