Vibration control of a class of semiactive suspension system using neural network and backstepping techniques

In this paper, we address the problem of designing the semiactive controller for a class of vehicle suspension system that employs a magnetorheological (MR) damper as the actuator. As the first step, an adequate model of the MR damper must be developed. Most of the models found in literature are based on the mechanical behavior of the device, with the Bingham and Bouc-Wen models being the most popular ones. These models can estimate the damping force of the device taking the control voltage and velocity inputs as variables. However, the inverse model, i.e., the model that computes the control variable (generally the voltage) is even more difficult to find due to the numerical complexity that implies the inverse of the nonlinear forward model. In our case, we develop a neural network being able to estimate the control voltage input to the MR damper, which is necessary for producing the optimal force predicted by the controller so as to reduce the vibrations. The controller is designed following the standard backstepping technique. The performance of the control system is evaluated by means of simulations in MATLAB/Simulink.