Study Of Iterative Learning Active Control On High Speed Train Body Vibration Based On Big Data

In the process of train running at high speed, in addition to the open wire running, there will have the working condition of the tunnel passing and train crossing. Strong aerodynamic load will produced when a train passing through a tunnel or crossing to another train, cause an effect to vehicle body vibration. Aerodynamic load effect on the vehicle body directly, and the action time is short with big impact, algorithms which aim at track irregularity excitation can’t control it effectively. And in engineering practice it could not have effective control on vehicle body vibration because of actuator response speed. With the advent of the era of big data, it provides a new perspective for the train vibration control.Train runs every day in the same line, under the same condition, the long-term accumulation of data has a kind of repeatability. In order to find a way to improve train’s stationarity from these characteristics of repeated, combined with iterative learning control to propose the train active control method based on big data. And analyze the control effect of train vibration control of this method through the multi-body dynamics simulation software. In details, the main research work is as follows:1. Introduce the background of train big data, in order to deal with in the quasi-periodic data that produce by train running, iterative learning control is introduced, Proposed the train iterative learning active control system based on big data. And the system implementation scheme is studied.2. Built a train multi-body dynamics model by using multi-body dynamics software UM. And connect the UM with SMULINK software, UM/SIMULINK simulation model is established.3. The computed aerodynamic load data of open wire crossing is loaded into the united simulation model of train, simulation under passive and semi-active suspension, to analyze its influence on train body vibration and the semi-active control effect. Based on big data iterative learning control method, the process of train body vibration control under train crossing aerodynamic load is given. The control effect and robustness is be analysis by simulation.4. Research on the influence on train body vibration under aerodynamic load of tunnel passing through. Big data iterative learning control method is using to control the train body vibration while train passing through a tunnel. And he control effect and robustness is be analysis by simulation.Through the simulation and analysis of high-speed train active suspension model, this paper studies the effectiveness of the vehicle body vibration control by active suspension iterate learning control method based on big data. The simulation result shows that: compared with the passive suspension, the peak acceleration of vehicle body vibration case by train crossing aerodynamic load can be reduce by about 50%; in the process of tunnel passing through the peak and RMS values of vehicle body vibration acceleration can reduce about 60%～70%, and can effectively improve the stationarity of train running.