Fault Diagnosis And Software Development Of Li-ion Battery System

Under the background of global energy shortage and increasingly serious environmental problems,new energy vehicle technology has become the current research hotspot,and has been rapidly developed under the support of strong national policies.With the continuous development of new energy vehicle technology,the related technologies of electric vehicles and vehicle power batteries are becoming more and more mature,which makes the production and sales of electric vehicles continue to rise.But in recent years,more and more researchers focus on the safety of electric vehicles because of the frequent spontaneous combustion accidents of electric vehicles.Most of the safety problems of electric vehicles are directly related to the power battery system.Based on the actual operation data of electric vehicles,the following aspects for the fault diagnosis of lithium-ion battery system will be mainly studied in this thesis:(1)The vehicle operation data used in this thesis is preprocessed,the null and zero values in the data are eliminated,and the data of vehicle charging section is extracted for battery diagnosis.The fault tree analysis method is used to analyze the fault of battery system,focusing on the fault types and causes of battery body and battery management system.Combined with the results of fault tree analysis,the fault types of battery system are preliminarily diagnosed,so as to avoid the misjudgment of the cell battery caused by the system fault.(2)The characteristic parameters which can represent the change of battery characteristics are constructed,and the evolution rule of the characteristic parameters is analyzed.The characteristic parameters include voltage characteristic parameters such as voltage change rate,voltage deviation degree,voltage range and temperature characteristic parameters such as temperature change rate,maximum temperature value and range.The change of characteristic parameters with operation time is analyzed,and the change of characteristic parameters of all vehicles is statistically analyzed,and the safety threshold range of each characteristic parameter in the safe operation of vehicles is proposed,which provides the basis for fault pattern recognition of abnormal cell battery.(3)The algorithm of outlier detection is used to evaluate the consistency of cell battery.Two methods,distance based outlier detection and density based outlier detection,are selected to calculate the Euclidean distance between the charging voltage curves of single battery and the local outlier factor(LOF)of cell battery.By setting the appropriate threshold value to judge the abnormal cell,the results show that the two methods can detect the inconsistent cells in the battery pack,and the fault mode of abnormal cell is identified.The results show that the main causes of battery abnormality are undervoltage,low SOC and internal short circuit.(4)In order to realize the engineering application of battery system fault diagnosis of electric vehicle,the GUI interface is designed by using Python language and Py Qt5 library,and the fault diagnosis software of lithium-ion battery system is developed.By inputting test data,the software can run well.