Research On Speed Ratio Control Of Metal Belt Continuously Variable Transmission Based On RBF Neural Network PID

As CVT(Continuously Variable Transmission) can adjust the speed ratio continuously, it can achieve the best match of the engine-transmission-road environment, improve economy performance and driving comfort of the car. It has becoming more popular in automobile manufacturers in recent years. So this issue selects a metal belt type continuously variable transmission as the research object, and use RBF neural network intelligent control in the speed ratio control, to improve the control effect of PID.First of all, based on consulting a large number of domestic and foreign literature, the paper discussed the development and application of CVT, research status at home and abroad. Analyzed the structure and working mechanism of metal belt type CVT. Introduced the control technology of continuously variable transmission. Studied on the speed ratio control strategy for three kinds of working condition of CVT.Then, in Matlab, the engine test data were analyzed and processed. Got the output torque, output power and fuel consumption feature model by using neural network, and got the best economy and power performance curve of the engine. Analyzed the CVT transmission system, and the transmission system’s dynamic model was established. Established the simulation platform of vehicle transmission system based on CVT in Simulink.The last, focused on researching the control algorithm of PID based on RBF neural network, and designed the RBF-PID speed ratio controller by using S-function. Then, tested the performance of conventional PID controller and RBF-PID controller under common automobile conditions, compared to conventional PID controller, RBF-PID controller improved the control precision and response speed.