Control-Oriented Closed-loop Identification Modeling Of High-Speed Trains

As a very important means of transport,high-speed railways has a lot of economic advantages such as high speed,low energy consumption,great carrying capacity,safety and comfortable,so high-speed tailways naturally become the world’s priority to the development of green transport.However,the train speed continue to improve with the progress of science and technology and the interaction between the various parts of the train also increased,for example,wheel and rail,bogie and internal parts.With the significant deterioration of the dynamic effects of the entire system of the train,this puts a higher demand for the safety and stability of the train.So,in order to make high-speed train running more secure and promote the better development of high-speed railway,we should urgently carry out some research on the high-speed train dynamics model which can better control and optimize the high-speed train.First of all,this paper is based on physical principles and established on high-speed train dynamics model.By classifying,separating,simplifying and reestablishing the established high-speed train dynamics model,we can built a dynamic model which describes the single-particle vehicle dynamics model and the dynamic model of the main components in order to get difference equation and transfer function by mathematical processing of the established dynamics model.At the same time,we can analyze the parameters influencing the dynamic accuracy of the dynamic model to verify the rationality of the selected parameters and to know that the determined parameters meet the requirements of the high-speed train dynamics model.Secondly,the design of data collection program.According to the data needed to identify the high-speed train dynamic model data,we can carry out data collection program design.The interface software for collecting data is created with LabView which can offer clear drawing interface,huge computing power and professional image processing.And then,we can dispose and stimulate the data collected by using the programming advantages of MATLAB.The sensor is selected for the data to be collected with the description of the working principle.The data collected by the experiment are analyzed,denoised and filtered.Finally,the high-speed train dynamics model is identified by the maximum likelihood open-loop identification method and the closed-loop frequency domain identification method,and the comparison data between the identification data and the actual value in the model are obtained.The comparison between the two different identification methods and the comparison between the identification value and the true value can be used to understand the degree of influence of the parameters on the model in the model,which can provide the theoretical basis for accurate modeling.The data fitting is carried out,and the validity of theidentification method is evaluated by the comparison of the data fitting chart and the identification curve of the identification simulation and the degree of the degree of the curve fitting.Then,the convergence speed and the influence degree of the parameters are analyzed The Closed-loop identification of the established model for the design of the accuracy of the controller provides a guarantee for the closed-loop identification of the transfer function of the adaptive PID or other controller design,with the controller to assess the accuracy of identification accuracy.The results show the validity of the proposed identification method and the accuracy of the train dynamics model.