| 261 | 1 | 54 |
| 下载次数 | 被引频次 | 阅读次数 |
针对现有的电动汽车充电量负荷预测模型准确性低、稳定性差等问题,提出了一种基于图卷积神经网络(GCN)和小生境免疫—狮子算法(NILA)改进的预测模型,实现电动汽车充电量的准确预测。同时,本研究还设计三相交流充电桩控制系统,其中包括了三相交流充电桩与控制状态检测的硬件电路。实验结果表明,本研究NILA-GCN模型在电动汽车充电量负荷预测中具有较好的准确性,预测误差范围控制在[0.23%,2.86%]。
Abstract:Aiming at the problems of low accuracy and poor stability of the existing electric vehicle charging load forecasting models, an improved forecasting model based on graph convolutional network(GCN) and Niche immune lion algorithm(Nila) is proposed to realize the accurate prediction of electric vehicle charging. At the same time, the control system of three-phase AC charging pile is designed, including the hardware circuit of three-phase AC charging pile and control state detection. The experimental results show that the Nila-GCN model has good accuracy in electric vehicle charging load forecasting, and the prediction error range is controlled at range of [0.23%, 2.86%].
[1] 林越,刘廷章,陈一凡,等.基于AP-HMM混合模型的充电桩故障诊断[J].广西师范大学学报(自然科学版),2018,36(1):25-33.
[2] 蒋林洳,万伟江,丁霄寅,等.一种基于直接蒙特卡洛法的电动汽车充电负荷模型[J].供用电,2018,35(4):20-25.
[3] 罗江鹏,张玮,王国林,等.基于出行链数据的电动汽车充电需求预测模型[J].重庆理工大学学报(自然科学),2020,34(6):1-8.
[4] 李怡霖.新能源汽车充电管控平台数据挖掘研究[D].大连:大连理工大学,2019.
[5] YANG D S,PANG Y H,ZHOU B W,et al.Fault Diagnosis for Energy Internet Using Correlation Processing-Based Convolutional Neural Networks[J].IEEE Transactions on Systems,Man,and Cybernetics:Systems,2019,49(8):1739-1748.
[6] 郑了了.网约车出行:现实困境与实现路径:基于H市的实证研究[J].山西高等学校社会科学学报,2020,32(9):43-51.
[7] 段宗涛,张凯,杨云,等.基于深度CNN-LSTM-ResNet组合模型的出租车需求预测[J].交通运输系统工程与信息,2018,18(4):215-223.
[8] 郭小帆,马天男,王超,等.计及电动汽车充电站接入的配电网网架规划优化研究[J].智慧电力,2019,47(6):64-70.
[9] 张海燕,王东,樊平,等.基于加权Voronoi图和GA-PSO的电动汽车充电站最优规划[J].电网与清洁能源,2019,35(5):77-82.
[10] KIM S,LEE G,KWON G Y,et al.Deep Learning Based on Multi-Decomposition for Short-Term Load Forecasting[J].Energies,2018,11(12):3433.
基本信息:
中图分类号:TP183;U491.8
引用信息:
[1]张谢,陈朔,王尉,等.基于图卷积神经网络的充电量预测模型[J].微型电脑应用,2023,39(02):45-49.
基金信息:
国网安徽省电力有限公司信息化项目(2010003-AFW)
2023-02-20
2023-02-20