Simulation analysis of quarter car active suspension control based on QBP-PID

Abstract

In order to further improve the stability and comfort of automobile active suspension, a BP neural network controller based on Q-learning algorithm optimization (QBP-PID) is proposed. QBP-PID uses BP neural network to adjust the PID gain, introduces the optimal strategy of Q-learning to correct the weight momentum factor, and optimizes the key weights in the neural network, so that the controller has better learning ability and online correction ability. A quarter suspension simulation model with random road excitation as the system input is established in Simulink software. The root mean square of body acceleration and tire dynamic displacement are used as the evaluation indexes of active suspension performance. The simulation results show that compared with the traditional passive suspension, PID control suspension and BP-PID control suspension, the active suspension using QBP-PID control algorithm can significantly improve the driving stability and comfort of the vehicle.