Observer-based Fault Diagnosis Of Lithiumion Battery Sensors For Electric Vehicles

In recent years,the popularization of electric vehicles has become one of the important measures for countries to cope with global climate and energy crisis.The technology related to electric vehicles has also become a research hotspot in various fields.Power battery as the main energy source of electric vehicles,its running performance and safety directly affect the driving experience and safety of electric vehicles.The battery management system in the power battery system is an important component to ensure the performance and safety of the power battery pack,and the function realization of the battery management system is highly dependent on the data collection work done by the sensor.Therefore,battery sensors have an extremely important role for electric vehicles.However,due to the complex and varied operating conditions of the car,battery sensor failures often occur.In view of this,it is of great significance to fault diagnosis of lithiumion battery sensors in electric vehicles.This paper adopts a model-based method to separately propose an observer-based fault diagnosis strategy for the electric vehicle lithium-ion battery cell sensors and a fault diagnosis strategy for electric vehicle series lithium-ion battery pack sensors based on the observer and improved correlation coefficient.Finally,the effectiveness of the strategy is verified under a variety of fault conditions.The specific research work includes:For lithium-ion battery cell,the battery cell electro-thermal coupling dynamic model is constructed and the model parameters are determined.A state observer is built based on the extended Kalman filter algorithm,and fault diagnosis and separation strategy for a current sensor,a voltage sensor and a surface temperature sensor of the electric vehicle lithium-ion battery cell is proposed.Finally,the diagnosis method is tested under different fault types and sizes to verify the effectiveness of the diagnosis strategy.It is analyzed that the method has high fault diagnosis robustnessFor series lithium-ion battery pack,first,the specific model of the battery pack to be studied is selected and a test platform to collect test data is built;the first-order RC equivalent circuit model of the equivalent large cell is constructed and the model parameters are identified online.A sensor fault diagnosis and separation strategy for electric vehicle lithium-ion battery pack based on observer and improved correlation coefficient method is proposed.In order to verify the effectiveness of the proposed fault diagnosis and separation strategy,the diagnosis and separation strategy is used to test and verify the effectiveness of the fault diagnosis strategy under no sensor fault,current sensor fault and voltage sensor fault.Finally,it is analyzed that this method has two technical advantages in terms of high fault diagnosis sensitivity and high calculation efficiency.