Key State Estimation Methods And Simulation Research On High Speed Train Adhesion Control

With the rapid development of the railway industry,the safety of high-speed train operation has become the focus of attention around the world.The operation of the train depends on the traction,but the degree of adhesion between wheel and rail limits the traction.The adhesion control systems can effectively improve the adhesion degree and adhesion utilization rate,but the acquisition of the train running state information seriously affects the performance of the adhesion control system.Because the high-speed train system is huge and the operating conditions are complex,the adhesion characteristics are easily affected by external factors,showing strong nonlinearity and easy mutation.So critical states such as the adhesion coefficient and the train speed cannot be directly obtained by detecting devices during the actual running of the train.Therefore,nonlinear system state estimation methods are used to study the critical state estimation problem on high-speed train adhesion control in this paper.It mainly includes the following contents:The CKF algorithm is an effective method for estimating the state of nonlinear systems,but the ability to track the state of the mutation is not strong.Because the rail surface adhesion coefficient is susceptible to the external environment,and it is characterized by strong nonlinearity and easy mutation,so the strong tracking theory and CKF algorithm are combined to design a rail surface adhesion coefficient estimation method based on strong tracking CKF algorithm in this paper.The simulation results show that the strong tracking CKF algorithm has better dynamic performance,stronger tracking ability and higher estimation accuracy than the traditional CKF when the system varies abruptly.The noise of high-speed train is often unknown and time-varying during actual operation,but the traditional CKF algorithm requires that noise statistics are known.In order to overcome the influence of noise on the velocity estimation of high-speed trains,the adaptive algorithm and the CKF algorithm are combined to design a high-speed train velocity estimation method based on the adaptive CKF algorithm in this paper.The simulation results show that the adaptive CKF algorithm can estimate and correct the noise statistical characteristics online,which significantly reduces the influence of noise and improves the estimation accuracy of the algorithm.The high-speed train system is complex and huge,and there are many sensors.During the operation,the rail surface condition of each wheelset may be different.A single sensor only obtains local information,which is easy to cause local data deviation and affects the estimation performance of the algorithm.Therefore,in this paper,the multi-sensor information fusion state estimation method is studied,a four-axis dynamic model of high-speed train is established and a velocity estimation method is designed based on distributed CKF algorithm.The method obtains the global uniform solution by fusing the local estimation information of multiple wheel speed sensors,which can effectively avoid the problem of reduced estimation accuracy caused by collecting local data,and greatly improve the estimation accuracy.