High-speed EMU Predictive Control Method Based On Multiple Model

The high-speed EMU as a new way of transportation giving people travel brings a lot of convenience.In recent years,our country's high-speed rail industry gets rapid development,and high-speed operation total mileage has reached 20000 kilometers,and high-speed EMU speed also reached more than 300 km/h.But with the high-speed EMU running speed faster and faster,the traditional control method will exist many defects,and seriously affected the safety and efficiency of high-speed EMU operation.Therefore,in order to make high-speed EMU operate safely and effectively,and need to study new modeling and control methods to adapt to high-speed EMU operation.The running environment of high-speed EMU is extremely complex,and the running process contains various random and unknown state.As a result,the modeling and optimization control methods of high-speed EMU are more complicated than ordinary low-speed trains,besides,the modeling and optimization control methods for low-speed trains can't meet the operation requirements of high-speed EMU.As the running environment of high-speed EMU is complex and changeable,the operation process of high-speed EMU multiple model method is put forward in this paper to establish its dynamic model,and the implicit generalized predictive adaptive control method is used to realize speed tracking control for high-speed EMU.Specific studies are as follows:1.The nonlinear characteristic in the operation process of high-speed EMU is palpable,and the faster the speed of high-speed EMU operation process,the more obvious the nonlinear features will be captured,which requires more accurate high-speed EMU modeling and optimization control method.Firstly,the practical operation data of CRH380 AL in Xuzhou east–Taian is collected and processed,then,the data is divided into two parts according to the traction and braking force,which includes traction data and braking data,moreover,the number of traction models and braking models is determined with brake model fuzzy c-means clustering algorithm combined with DB cluster index,and the recursive least squares method is utilized to identify each sub models of traction and braking traction,and set up the braking model;At last,the rationality of the established model is verified with practical data.2.As the complexity and heavy calculation of ordinary generalized predictive control algorithm,therefore,this paper designs an implicit generalized predictive adaptive controller to control the process of practical operation of high-speed EMU.The algorithm can directly identify the parameters in the control rate,which could avoid solving Diophantine equation,and greatly simplifies the calculation as well as saves the operation time.3.Taking CRH380 AL as the research object,and using the proposed modeling and control method to realize the speed tracking control of high-speed EMU,besides,we also verify the validity of the proposed modeling through simulation experiment.The simulation results show that the method can track the given target velocity effectively,and velocity error is within the allowed range,and satisfy the speed limit conditions,In conclusion,the safety,accurate,comfortable,and efficient operation of high-speed EMU is implemented.