Research Of Sensor Fault Tolerant Control Based On Analytical Reconstruction Model For EBPS

Electronic brake pedal simulator (EBPS) is the core device of pure electric vehicle braking system, which directly affects the braking performance of vehicles. The faults may reduce the reliability of the whole system which presents in a short time or slight degree in the components such as sensors in the EBPS. In some severe cases it may cause the entire vehicle brake failure and threaten the driver and the passengers’ personal safety and property safety.Now the intelligent fault diagnosis and fault-tolerant control technologies are widely applied in automobile engineering fields. However current researches about sensors in EBPS mainly focus on faults with the specific scope and single type and lack of researches about fault tolerant control for EBPS. So this dissertation focuses on improving the sensitivity and reliability of EBPS by using fault-tolerant control methods, such as multi-source information fusion, refactoring and nonlinear compensation.Firstly, considering angular displacement and pedal force sensor carried by EBPS, a sensor fault diagnosis method is proposed based on particle filter algorithm according to the mathematical model of EBPS in this paper.Secondly, an adaptive fault-tolerant control architecture model is established on the basis of based using fault reconstruction method on sensor fault diagnosis model. When the sensor failure occurs, the EBPS should realize self-regulation to maintain normal operation of the system.Finally, simulations are carried out for the established sensor fault-tolerant control model of EBPS using Matlab and FPGA, respectively.The results verify that the fault diagnosis and adaptive fault-tolerant control method can accurately detect sensor faults. And the normal output values of the fault sensor can be estimated by adaptive fault-tolerant control architecture model, which makes the EBPS still run normally. So as to further improve the reliability of the EBPS.