Research On Fault Tolerant Control Of Sensor And Parameter Faults In Vehicle Electric Power Steering System

Electric Power Steering(EPS)system is a crucial device to assist vehicle steering and influence vehicle operating stability.The performance of EPS system directly affects the portability and driving stability of vehicle steering.However,due to the aging of components and the influence of the external environment,EPS system will inevitably suffer from faults,which will lead to accidents and casualties.Therefore,in this thesis,the EPS system is taken as the research object,aiming at sensor fault and parameter fault occur in EPS system,two different fault-tolerant control schemes are designed based on bond graph(BG)theory and sliding mode control theory to improve the safety and reliability of EPS system.Firstly,the EPS system model is built with integral causality using BG technology as the actual system,and the BG model of EPS system with differential causality is treated as a part of inverse system controller.The residual between the reference input and the actual system output signal is taken as the effort variable input of the inverse system controller.After the input of effort variable passes through the BG model of EPS system with differential causality to obtain the flow output,and the flow output is multiplied by high gain which is treated as the output of the controller.In order to improve the transient performance of the system,genetic algorithm is used to optimize the high gain parameters.Then,the closed-loop transfer function under the inverse system model is obtained from the signal flow graph,when the high gain is large enough,the transfer function tends to1.Therefore,under this condition,the performance of the fault-tolerant control system can be guaranteed even if the parameters of EPS system BG model with differential causality in the inverse system controller are inconsistent with the actual system parameters.In other words,when parameter fault occurs in the actual system,the performance of the closed-loop system remains unchanged when the gain is large enough(the output signal can still track the input signal under parameter fault),the result proves that the controller based on the inverse system theory is robust to the parameter fault of the system.Secondly,the dynamic model and two degree of freedom model of EPS system are established to design current tracking and fault-tolerant control method.When the sliding mode controller is designed,adaptive technology is used to update the sliding mode switching gain online to reduce the buffeting response.And an adaptive sliding mode controller is designed to track the target current of the assisted motor.In order to design the fault-tolerant controller,the design method of control law reconstruction is adopted,and the sliding mode controller is reconstructed when the fault occurs.In other words,a compensation control law is added to the sliding mode control law to reconstruct the novel sliding mode fault-tolerant controller.And the function of compensation control law is to compensate the system performance degradation caused by fault,which can achieve the purpose of active fault-tolerant control.Lyapunov method is used to analyze the stability of the system to ensure the feasibility of the designed fault-tolerant controller.Finally,the EPS experimental platform is built to verify the feasibility and effectiveness of the fault-tolerant controller.