基于RBF神经网络的动力电池故障诊断系统研究

Fault diagnosis system for power battery based on RBF neural network

纯电动车动力电池在性能、成本、寿命、安全性上的局限是制约电动车普及的关键问题。基于RBF神经网络,设计了一套电池故障诊断系统,对动力电池组的故障进行预防和诊断。首先,利用d SPACE中的电池模型,模拟电池的故障,进行神经网络的学习训练,然后,利用三组测试数据对故障诊断系统进行测试。测试结果显示,设计的系统可以准确诊断电池故障类型与故障级别。

The limitation of power battery in its property, cost, life and safety restricts the popularizing of the electrical vehicle. In this paper, a fault diagnosis system was designed to prevent and diagnose the fault of battery using RBF neural network. First, using the battery model in dSPACE to simulate the fault of battery for the training of RBF neural network .Then the test of the fault diagnosis system was done using three groups of data. The test results show that the system can diagnose the battery fault type and fault level accurately.