基于Kmeans-ADE-LSTM模型的电动车直流充电桩充电效率研究

张煌辉; 方杰; 张杰梁; 叶熙领; 金涛; 邵海明; 肖蕾

张煌辉,方杰,张杰梁,叶熙领,金涛,邵海明,肖蕾.基于Kmeans-ADE-LSTM模型的电动车直流充电桩充电效率研究[J].电测与仪表,2026,63(4):89-98.
zhanghuanghui,Fang Jie,Zhangjieliang,Yexiling,Jintao,Shaohaiming,Xiaolei.Research on Charging Efficiency of Electric Vehicle DC Charging Station Based on Kmeans- ADE-LSTM Model[J].Electrical Measurement & Instrumentation,2026,63(4):89-98.

基于Kmeans-ADE-LSTM模型的电动车直流充电桩充电效率研究

Research on Charging Efficiency of Electric Vehicle DC Charging Station Based on Kmeans- ADE-LSTM Model

DOI：10.19753/j.issn1001-1390.2026.04.010

中文关键词: 长短期记忆网络差分进化算法充电桩充电效率预测 Kmeans聚类

英文关键词:LSTM network, differential evolution algorithm, charging station, charging efficiency,Kmeans clustering

基金项目:国家重点研发计划资助项目（2023YFF0614803）

作者中文名	作者英文名	单位
张煌辉	zhanghuanghui	福建省计量科学研究院
方杰	Fang Jie	福建省计量科学研究院
张杰梁	Zhangjieliang	福建省计量科学研究院
叶熙领	Yexiling	福建省计量科学研究院
金涛	Jintao	福州大学
邵海明	Shaohaiming	中国计量科学研究院
肖蕾	Xiaolei	厦门理工学院

摘要点击次数: 6

中文摘要:

在电动汽车充电过程中，交直流电的转换、电能与化学能的转化等现象都会造成不可避免的电能损耗。为了以有限计量设备精准计算该损耗大小，文中提出了基于Kmeans-ADE-LSTM的充电桩效率计算理论。该方法使用记忆型神经网络对历史充电数据进行模型训练，得出直流侧电气数据与充电效率之间的模型关系，再进行新的充电数据验证。创新性地提出将所采集到的充电数据在进行效率计算之前，先对数据样本进行Kmeans聚类分析，处理后的数据再进行后续的神经网络训练。此外，为改善传统差分进化算法的变异系数选取困难问题，文中选用了高效的全局优化算法——差分进化算法，对变异系数进行自适应处理，进行LSTM的超参数最优选取；基于充电效率与时间等的关联性，结合LSTM神经网络进行充电转换效率计算。相关实验结果表明，所提出的Kmeans-ADE-LSTM充电桩充电效率预测方法具有较高的预测精度。

英文摘要:

During the charging process of electric vehicles, the conversion between AC and DC power, as well as the transformation between electrical and chemical energy, inevitably leads to electrical energy loss. To accurately calculate the magnitude of this loss with limited measurement equipment, this paper proposes a charging pile efficiency calculation theory based on Kmeans-ADE-LSTM. This method uses a memory neural network to train the model with historical charging data, deriving a model relationship between the DC side electrical data and charging efficiency, and then validates it with new charging data. Innovatively, the collected charging data is subjected to Kmeans clustering analysis before efficiency calculation, and the processed data is then used for subsequent neural network training. Additionally, to address the difficult problem of selecting the mutation coefficient in traditional differential evolution algorithms, this paper employs an efficient global optimization algorithm—the differential evolution algorithm—to adaptively process the mutation coefficient and optimally select the hyperparameters of the LSTM. Based on the correlation between charging efficiency and time, etc., the charging conversion efficiency is calculated in conjunction with the LSTM neural network. The relevant experimental results demonstrate that the proposed Kmeans-ADE-LSTM charging pile charging efficiency prediction method has a high level of predictive accuracy.

查看全文查看/发表评论下载PDF阅读器

关闭