基于KF-ESN算法的新能源汽车电池组故障在线监控系统  

New Energy Vehicle Battery Pack Fault Online Monitoring System Based on KF-ESN Algorithm

作　　者：朱布博[1] 魏秋兰[1] 孙少杰[1] 罗明[1] ZHU Bubo;WEI Qiulan;SUN Shaojie;LUO Ming(Institute of Automotive Engineering,Shaanxi Vocational and Technical College of Communications,Xi’an 710018,China)

机构地区：[1]陕西交通职业技术学院,汽车工程学院,陕西西安710018

出　　处：《微型电脑应用》2023年第9期11-15,共5页Microcomputer Applications

基　　金：陕西省教育厅专项科研计划项目(22JK0288);陕西交通职业技术学院校级科研项目(YJ20004)。

摘　　要：当前的光学超精密检测采用基于改进CNN电池组故障诊断方法受到噪声数据影响,导致故障数据监控精准度低,对此提出基于KF-ESN算法的新能源汽车电池组故障在线监控系统。使用霍尔传感器结构,检测电池组电压和电流。通过控制模块,使主机具备自动递增特性,经由SMD-140035H蜂鸣器,实现故障报警。在ESN网络中通过KF算法进行电池组故障在线估计,计算网络输出权值和误差协方差的先验值,通过目标值校正后,只需评估网络输出权值,就能得到精准故障监控系统。实验结果表明,该系统分别与标准故障电压、电流数据存在最大为0.02 V和0.01 A的误差,具有精准监控结果。Aiming at the problem that the current battery pack fault diagnosis method based on optical ultra precision detection and improved CNN is affected by noise,resulting in low accuracy of fault data monitoring,a new energy vehicle battery pack fault online monitoring system based on KF-ESN algorithm is proposed.The Hall sensor structure is used to detect the voltage and current of the battery pack.Through the control module,the host has the characteristics of automatic increment,and the fault alarm is realized through the SMD-140035h buzzer.In the ESN network,the KF algorithm is used to estimate the battery pack fault online,and the a priori values of the network output weight and error covariance are calculated.After the target value is corrected,the accurate fault monitoring system can be obtained by evaluating the network output weight.The experimental results show that the maximum errors between the system and the standard fault voltage and current data are 0.02 V and 0.01 A respectively,and the system has accurate monitoring results.

关 键 词：KF-ESN算法 新能源汽车 电池组故障 在线监控 

分 类 号：U469.72[机械工程—车辆工程]