A Fuzzy PID Controller with Neural Network Algorithm for Freight Trains’ Braking System

With the rapid development of rail transit, a large number of electrical devices have been used in trains, resulting in a sharp increase in electromagnetic (EM) energy in train space and an increasingly complex EM environment. Therefore, higher requirements have been put forward for the performance of the braking system and controller of heavy trains. By analyzing the control characteristic of braking system of a truck, the fuzzy PID control and neural network control are combined. Using the fuzzy reasoning of neural network, the operating state parameters of the system are modified and adjusted online. The controller not only has the ability of self-learning, self-adaptation, parallel processing, and pattern recognition, but also can learn and adapt to the dynamic characteristics of uncertain system, which can greatly improve the effect of fuzzy control and practical control ability. The fuzzy controller can effectively overcome the drawbacks of traditional braking system and provide a stronger support for fault simulation and diagnosis analysis.