Research On Intelligent Fault Diagnosis Of Connection Looseness Of Lithium Ion Battery System Based On Information Physics Fusion

The transportation industry is accelerating into the new energy field as the global energy crisis continues to deteriorate and non-renewable resources are nearing exhaustion.As one of the three key core technologies of electric vehicles,the battery management system has developed rapidly,and the security module has received much attention.This paper studies lithium ion battery modeling and parameter identification,battery management system sampling time model,lithium ion battery system connection loose fault diagnosis,and centralized battery management system.The main research contents are as follows:1.the working principle and characteristics of lithium-ion battery are studied and analyzed.The first-order RC model is selected as the research object,and the recursive least squares method with forgetting factor is used for parameter identification.The model after identification is ideal.Based on the identified RC model,a simulation model of single cell connection looseness in the battery pack is constructed,which lays a foundation for the later fault diagnosis algorithm.2.Introducing the concept of time model in information physical fusion system,this paper constructs the time model in battery management system and proposes the alignment algorithm based on correlation coefficient.The alignment algorithm aligns the single voltage data and current data sampled by the battery management system on the time axis.Experiments have been carried out.The experimental results show that the alignment algorithm can effectively prevent the fault diagnosis algorithm from misdiagnosis due to the offset of data sampling and that the algorithm is of great significance to improve the accuracy of battery management system parameter estimation.3.Based on the correlation coefficient method,this paper proposes a fault diagnosis algorithm for connection loosening of lithium ion power battery system.The mathematical theory is used to prove the universality of the correlation coefficient method in the field of connection loose fault diagnosis,and a diagnostic algorithm based on calculation window is proposed.The sensitivity,effectiveness,applicability and versatility of the algorithm are analyzed by model simulation.Finally,The experiment results indicate that the algorithm can accurately and timely alarm and locate different loose connection faults without hardware redundancy,and can overcome the problems of false alarm and alarm failure caused by battery inconsistency.With generality and applicability,the algorithm fully meets the real-time requirements of automotive applications and can be applied in the vehicle battery management system.4.At the end of this paper,a centralized battery management system is designed,and its system structure and different functional modules are set forth.Based on the centralized BMS,the vehicle test of the time model and diagnostic algorithm proposed in this paper are carried out.The system works stably and reliably,and realizes accurate real-time diagnosis of loose connection.