基于蝙蝠—粒子群算法优化BP神经网络的电动汽车锂电池故障诊断
Fault Diagnosis of Lithium Battery for Electric Vehicle Based on BP Neural Network Optimized by Bat-particle Swarm Optimization Algorithm
乔维德 1陆超 2袁桂芳3
作者信息
- 1. 无锡开放大学 科研与发展规划处,江苏 无锡 214011
- 2. 无锡开放大学 教务处,江苏 无锡 214011
- 3. 江苏联合职业技术学院无锡机电分院 自动化工程系,江苏 无锡 214028
- 折叠
摘要
针对电动汽车锂电池故障诊断问题,在分析锂电池故障特征与故障原因的基础上,建立电动汽车锂电池故障诊断模型.该模型包括锂电池故障样本采集处理、BP 神经网络、故障特征编码输出及故障类型诊断.采取蝙蝠—粒子群算法优化BP神经网络初始结构参数,利用改进BP 算法和故障样本训练并测试BP神经网络.仿真实验结果表明:相比BP 算法、遗传算法、粒子群算法,蝙蝠—粒子群算法优化BP神经网络的故障诊断准确性最高、训练时间最短、训练误差最小.
Abstract
In order to diagnose the fault of lithium battery of electric vehicle,on the basis of analyzing the fault characteristics and causes of lithium battery failure,a fault diagnosis model of lithium battery for electric vehicle was established,which included lithium battery fault sample collection and processing,BP neural network,fault feature coding output and fault type diagnosis.The bat-particle swarm optimization algorithm was used to optimize the initial structure parameters of the BP neural network,and the improved BP algorithm and fault samples were used to train and test the BP neural network.The simulation results show that compared with the BP algorithm,genetic algorithm and particle swarm algorithm,the bat-particle swarm optimization BP neural network has the highest fault diagnosis accuracy,the shortest training time and the smallest training error.
关键词
锂电池/BP神经网络/蝙蝠—粒子群算法/故障诊断Key words
lithium battery/BP neural network/bat-particle swarm optimization algorithm/fault diagnosis引用本文复制引用
出版年
2024
NETL
NSTL
万方数据