基于BFGO-CNN-LSTM的EV充电数据异常诊断研究
Research on Diagnosis of Electric Vehicle Charging Data Anomalies Based on BFGO-CNN-LSTM
王秋实1
作者信息
- 1. 国网铁岭供电公司,辽宁铁岭 112000
- 折叠
摘要
为应对各种扰动对电动汽车(EV)在充电过程中因故障识别精度而造成的负面影响,提出一种创新的BFGO-CNN-LSTM模型,对EV充电数据中的异常部分进行精确诊断.通过该方式,模型能充分利用EV充电数据的时空特征,极大提升EV充电数据异常检测的准确性和运算效率.
Abstract
In order to cope with the negative impact of various perturbations on fault id entification accuracy during electric vehicle(EV)charging,an innovative BFGO-CNN-LSTM model is proposed to accurately diagnose the abnormal parts of EV charging data.In this way,the model is able to fully utilize the spatio-temporal characteristics of EV charging data,which greatly enhances the accuracy and computational efficiency of EV charging data anomaly detection.
关键词
数据异常诊断/卷积神经网络/长短时记忆神经网络/电动汽车/优化算法Key words
data anomaly diagnosis/CNN/long and short-term memory neural network/EV/optimization algorithm引用本文复制引用
出版年
2024
NETL
NSTL
万方数据