Font Size: a A A

Research On Train Nonlinear Network Control Method Based On Neural Network Prediction

Posted on:2019-12-21Degree:MasterType:Thesis
Country:ChinaCandidate:Z L LiFull Text:PDF
GTID:2382330572969477Subject:Engineering
Abstract/Summary:PDF Full Text Request
The train control system gradually develops in the direction of networking,information and intelligence.Various disciplines are intertwined to improve the stability and reliability of the train control system.Nowadays,the train network control system has been widely used in the high-speed train control,whose main functions include receiving and sending data of various train equipments in real time,monitoring the operational status of the train in real time,and ensuring the stability and safety of the train operation system.Although the train network control system has the characteristics of simple structure and strong stability,there are still some problems,such as network delay,packet loss and packet timing disordered.These problems may greatly affect the stability and security of the train operation.At the same time,many nonlinear control systems on the train make it more difficult to control the train.Therefore,aiming at the nonlinear systems with time delay problem in train network control system,the following work has been done in this paper:(1)The concept of train network control system,the development history of this field,the research status at home and abroad,and the relevant features and application of neural network have been briefly introduced.(2)For researching on the time delay problem of the train network control system,an experimental program to measure the time delay data is designed under the actual working condition.Through the minimum mean square algorithm which is based on the AR model,an effective prediction model is established based on the data measured by the experiment.(3)Aiming at the numerous nonlinear systems on the train and the delay issue is added on this basis,propos a control method.Through the use of generalized predictive control,neural network recursive prediction,least square method identification and other methods,the nonlinear system is effectively controlled and the delay is compensated.MATLAB is used to simulate the proposed control method to verify the effectiveness of the system.An effective solution is propoesd in this paper to solve the time delay problem of nonlinear control systems on trains,which is of great significance to improve the safety and stability of the train operation.
Keywords/Search Tags:Train Network Control System, Nonlinear System, Neural Network, Time Delay
PDF Full Text Request
Related items