堆叠式LSTM组合模型的充电站用电量预测方法  

Power consumption prediction method for charging stations using LSTM combined model

作　　者：王彩玲[1] 丁当 Wang Cailing;Ding Dang(School of Computer Science,Xi'an Shiyou University,Xi'an,Shaanxi 710065,China)

机构地区：[1]西安石油大学计算机学院,陕西西安710065

出　　处：《计算机时代》2025年第1期1-4,共4页Computer Era

摘　　要：随着电动汽车的普及,充电站对电力需求预测的精确性日益提高。本文设计了堆叠式LSTM模型,使用预处理过的某电动汽车充电站用电量数据,对比分析传统模型和LSTM模型在不同评估指标上的表现,验证所提出模型的优越性;还对多层堆叠式LSTM模型进行训练和测试,分析不同层数LSTM模型的性能,实验结果表明,三层堆叠式LSTM模型优于其他模型,能够显著提高用电量预测的准确度。With the popularization of electric vehicles,the accuracy of charging stations for power demand prediction is increasing.In this paper,a stacked LSTM model is designed to use preprocessed electricity consumption data of an electric vehicle charging station to compare and analyze the performance of the traditional model and the LSTM model on different evaluation indexes,and verify the superiority of the proposed model;a multilayer stacked LSTM model is also trained and tested to analyze the performance of LSTM models with different numbers of layers,and the experimental results show that the three-layer stacked LSTM model is superior to other models and can significantly improve the accuracy of electricity consumption prediction.

关 键 词：用电量预测 长短期记忆网络 卷积神经网络-长短期记忆网络 堆叠式LSTM模型 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]