Research On Wheel-rail Adhesion Control Of High-speed Train Based On Cloud Model

In recent years,with the development of the railway transportation,the safety and comfort of trains have become increasingly important.Moreover,with the long-distance travel route such as the Eurasian railway line,higher requirements are placed on the environmental adaptability of the train.In complex environment,train may occur idling or slipping,which has a negative impact on the safe,so it is necessary to study the control strategy of train adhesion.The train suspension system and the drive system have large inertia and nonlinearity,which leads to the complexity and uncertainty of the adhesion characteristics between the wheel and rail.Therefore,the adhesion system is a typical type of uncertain nonlinear system.The basic idea of the cloud model is combining ambiguity and randomness organically based on fuzzy theory and probability theory to achieve qualitative and quantitative uncertainty transformation for controlling complex nonlinear systems.The cloud model control can not only retain the unknown uncertain factors in the controlled object,but also quickly adapt to the system state and parameter changes,and significantly improve the robustness of the system.In view of this advantage,this paper applies the cloud model to the design of train adhesion control system.Through the research of cloud model control technology,the adhesion control is realized in different environments,so that the system has good control performance and prevents the train from idling or sliding.The paper expounds the current methods and characteristics of train adhesion control,analyzes the mathematical relationship between the adhesion mechanism of the train wheel and rail,adhesion and creep,and deeply studies the quantitative influence of wheel surface condition,axle weight,wheel diameter and train speed on the adhesion coefficient..On this basis,a sticking control method based on two-dimensional cloud model is proposed for the problem of stick control when the running condition of the train is uncertain.Referring to the mathematical model and algorithm of current cloud model qualitative and quantitative uncertainty transformation,the paper establishes a simplified model of the train adhesion system based on the adhesion mechanism.And uses the creep rate and creep rate variation as the input variables of the controller,torque is output variable,design adhesive controller.Finally,the simulation research of train single-axis adhesion control is carried out based on MATLAB software platform.Noise interference is introduced.The vehicle body speed and wheel cycle speed,creep rate and torque are verified under different road conditions,different loads and different initial speeds.The simulation results show that the controller based on the two-dimensional cloud model algorithm has good control effect and can be effectively used to prevent the train from idling or sliding.However,there is a jitter problem and the adhesion utilization rate is relatively low.Aiming at the problem of non-adhesive utilization of adhesive system under nonlinear and uncertain conditions,this paper proposes an adaptive cloud model based adhesion control method based on the adhesion control of two-dimensional cloud model.To begin with,the fuzzy setting is adjusted online ?K_p??K_i??K_d to realize the regulation of the adhesion parameters by the cloud model controller.In addition,noise interference is introduced to verify the changes of vehicle body speed and wheel speed,creep rate and torque under different road conditions,different load and different initial speed.The results show that this method not only effectively prevents wheel sliding and improves adhesion utilization,but also validates that the adhesion control system based on cloud model has strong anti-interference ability and the ability to deal with uncertainties.