Model-based fault detection for an electrohydraulic braking system with cooperative regenerative braking management

This paper presents a model-based fault detection method for an electrohydraulic braking system with cooperative regenerative braking management. The regenerative braking system can save energy and improve endurance in electric vehicles. However, failure of the regenerative system may result in the system being unable to fully convert the recovered energy into stored electrical energy, with a large portion of the energy being dissipated as heat. Meanwhile, the motor braking torque output is limited, and the range of battery state of charge is affected. In this paper, a Carsim/Simulink co-simulation was developed to model the operating conditions and failure modes of the regenerative braking system in an electric vehicle. A three-stage dynamic equation was derived to model the piston movement in the wheel cylinder of the electrohydraulic braking system. A vector control using coordinate transformation was developed for the permanent magnet synchronous motor to control the current and torque. Since various system components have been built, the advantage of model-based methods can perform predictive maintenance compared with the current corrective and preventive maintenance. Therefore, this paper proposed a three-layer systematic model (torque distribution layer, execution layer, and actuation layer) of electric vehicle regenerative braking system to detect and diagnose various fault conditions. Various parameters were compared with the normal braking mode in order to provide systematic guidance on fault detection.