基于神经网络的蓄能器耐久在线智能测试方法

打开文本图片集

中图分类号：U461.7+1 文献标志码：B DOI：10.19710/J.cnki.1003-8817.20250219

Fused Neural Network-Based Online Intelligent Testing Method for Accumulator Durability

Tao Xinyang， Zhou Yuting，Meng Kai， Han Wenbin， Liu Yuan，Fan Xue （BYD Auto Industry Company Limited， Shenzhen 518118）

Abstract：Toaddresstheproblemsof longcycletimein traditionalaccumulator durabilitytestsand low eficiencyinperformancetesting，ahybridreal-time lifepredictionmodel basedonLong Short-TermMemory （LSTM） andGeneticAlgorithm-Back PropagationNeural Network（GA-BP）isproposed.Thismodelcanmonitor the performance statusof specimens inreal time andrealize inteligent durability testing basedonthe integrationof MATLABandLabVIEW platforms.Testresultsshowthatthefailureidentificationpositionsofthismethodare consistent with the measured results，the full-cycle performance prediction error is less than 15% ，and the test timeis saved by an average of approximately 10% ：

Keywords：Neural network，Accumulator，Real-timemonitoring，RUL prediction

1前言

随着汽车行业竞争日益激烈，产品开发节奏持续加快，对缩短生产周期提出了更高要求。（剩余9402字）