计及复杂气象影响的含光伏电源的母线峰值负荷预测  

Peak load prediction for electrical bus with photovoltaic power supply including complex meteorological influence

作　　者：朱赫炎 于长永 贺庆奎 宋坤 刘靖波 许言路 卢天琪 黄南天[2] Zhu Heyan;Yu Changyong;He Qingkui;Song Kun;Liu Jingbo;Xu Yanlu;Lu Tianqi;Huang Nantian(State Grid Liaoning Electric Power Company Limited Economic Research Institute,Shenyang 110015,China;School of Electrical Engineering Northeast Electric Power University,Jilin 132012,China)

机构地区：[1]国网辽宁省电力有限公司经济技术研究院,辽宁沈阳110015 [2]东北电力大学电气工程学院,吉林吉林132012

出　　处：《可再生能源》2020年第6期804-810,共7页Renewable Energy Resources

基　　金：辽宁省电网公司科学技术项目(LNDL2019-02PT-GC);国家自然科学基金(51307020)。

摘　　要：光伏电源并网后,配电网母线负荷波动会更复杂,峰值负荷预测更加困难。为提高母线峰值负荷预测精度,文章提出了计及复杂气象影响的母线峰值负荷预测方法。首先,根据不同气象日下累积的历史数据,通过条件互信息分析母线峰荷数据与高维气象、社会等特征间相关性,获得特征重要度排序;其次,在条件互信息降低潜在特征集合特征间冗余性基础上,针对不同气象日,以改进粒子群优化极限学习机预测精度为决策变量,开展针对性前向特征选择,确定不同最优特征子集;最后,根据最优特征子集,针对性建立不同气象日下母线峰值负荷最优预测模型。以某地区实际含高渗透率光伏电源母线负荷开展实验,证明所提方法可有效提高母线峰荷预测精度。When photovoltaic distributed generator is connected to bus load,bus load fluctuation becomes more complex and peak load prediction becomes more difficult.In order to improve the accuracy of peak load forecasting for electrical bus,a peak load forecasting method is proposed under complex weather conditions in this paper.Firstly,according to the accumulated historical data under different meteorological days,the correlation characteristics of low redundancy between peak load data for electrical bus and high-dimensional meteorological,social and other features were analyzed with conditional mutual information to obtain the feature importance ranking.Then,on the basis of reducing the redundancy among the features of the potential feature set by means of conditional mutual information,targeted forward feature selection was carried out to determine different optimal feature subsets by improving the prediction accuracy of improved particle swarm optimization extreme learning machine as the decision variable for different weather types.Finally,according to the optimal feature subset,the optimal model of peak load prediction for electrical bus under different meteorological days is established.Carrying out the experiment with the bus load of photovoltaic distributed generator in a certain area,the results show that the new method can effectively improve the accuracy of the peak load prediction of electrical bus.

关 键 词：母线峰值负荷预测 特征选择 条件互信息 极限学习机 改进粒子群 

分 类 号：TK51[动力工程及工程热物理—热能工程]