robust control of active suspension system for a quarter rail car model using neural network based controller

Active suspensions that combine conventional mechanical structures with advanced electronics, sensors, and controllers have enabled the development of railway vehicles that can meet the new demands for higher speed, improved ride comfort, and stricter safety standards. nevertheless, these aspects are affected by low track quality or high train speed.  therefore, it is crucial to regulate the vibration of the vehicle’s suspension by using advanced control and automation techniques that can optimize the performance of a rail car suspension system. a method to improve these factors under such operating conditions is active suspension control. active suspension enables designers to achieve a comfort level that is impossible with passive suspension elements. this work introduces the mathematical model of a two-degree-of-freedom system and the implementation of a robust artificial neural network control system for the active suspension system of a rail car. the control system that is proposed comprises a robust controller, a narma-l2 controller, which is a type of neural network controller that can be used to control nonlinear systems, and a model neural network of the rail car’s suspension system. a standard pid controller is also used for comparison to control the railway vehicle’s suspension system. the simulation results indicate that the proposed control system has enhanced efficiency and a better outcome at adjusting to random track disturbances for the railway vehicle’s suspension.