To improve the overall smoothness of the connected air suspension

an improved gray wolf algorithm is proposed

which firstly uses the exponential function to improve the convergence factor nonlinearization to enhance the algorithm′s breadth search and depth search capability. Secondly

to prevent the gray wolf population diversity from being destroyed

the Chebyshev chaos mapping method is used to initialize the gray wolf population. Then the dynamic weight method is proposed to increase the weight proportion of alpha wolves in the wolf population. Finally

the optimal solution of the improved gray wolf algorithm is used as the initial weights of the BP neural network controller and the control performance of the connected air suspension is analyzed by the AMESim-MATLAB joint simulation. The research results show that the improvement rate of the algorithm for body acceleration reaches 39.5%. It effectively improves the overall smoothness of the connected air suspension.