经济新常态下基于Verhulst-SVM的中长期负荷预测模型  

Medium and long-term load forecasting model based on Verhulst-SVM under new normal economy

作　　者：张冠英[1,2] 羡一鸣 葛磊蛟 王莹[4] 赵滨滨[4] 王尧[1,2] Zhang Guanying;Xian Yiming;Ge Leijiao;Wang Ying;Zhao Binbin;Wang Yao(State Key Laboratory of Reliability and Intelligence of Electrical Equipment,School of Electrical Engineering,Hebei University of Technology,Tianjin 300132,China;Key Laboratory of Electromagnetic Field and Electrical Apparatus Reliability of Hebei Province,School of Electrical Engineering,Hebei University of Technology,Tianjin 300132,China;School of Electrical and Information Engineering,Tianjin University,Tianjin 300072,China;State Grid Tianjin Electric Power Company,Tianjin 300010,China)

机构地区：[1]省部共建电工装备可靠性与智能化国家重点实验室(河北工业大学电气工程学院),天津300132 [2]河北省电磁场与电器可靠性重点实验室(河北工业大学电气工程学院),天津300132 [3]天津大学电气自动化与信息工程学院,天津300072 [4]国网天津市电力公司,天津300010

出　　处：《电测与仪表》2019年第1期102-107,共6页Electrical Measurement & Instrumentation

基　　金：国网天津市电力公司科技资助项目(基于灵活负荷特性及其优化策略的电力需求预测研究);国家自然科学基金项目(51807134)

摘　　要：经济新常态背景下,电力系统中长期负荷预测面临着很多新问题,例如:GDP、人口等电力负荷影响因素呈"S"型曲线增长、电力负荷影响因素与电力负荷之间的不确定性增加、历史样本数量少等。为此,提出一种基于Verhulst-SVM的中长期负荷预测模型。首先,从经济新常态特征中提取影响电力负荷的主要因素,并分析各影响因素的发展趋势;然后,利用Verhulst模型对"S"型曲线增长的电力负荷影响因素进行预测,并采用支持向量机(Support Vector Machine,SVM)替代线性回归预测模型,实现小样本、高不确定性条件下中长期负荷高精度预测。最后,通过天津市2015年和2016年的负荷预测算例,验证了所提模型的精度和可靠性,可为经济新常态背景下中长期负荷预测提供借鉴。Under the background of the new normal economy of China,the medium-and long-term load forecasting of power systems faces many new problems.For example,the factors affecting the power load such as GDP and population are S-shaped curve growth,the uncertainty between power load and power load influencing factors increases,and the number of historical samples is small.To this end,a medium and long-term load forecasting model based on Verhulst-SVM is proposed in this paper.Firstly,the main factors affecting the electric load are extracted from the new normal of economy characteristics of China,and the development trend of each influencing factor is analyzed.Then,the Verhulst model is used to predict the load influencing factors which are S-shaped curve growth.And the support vector machine(SVM)is used to replace the linear regression prediction model to achieve high-precision prediction of medium and long-term load under small samples and high uncertainty.Finally,it is proved that the proposed model has high precision and reliability by predicting the power load of Tianjin in 2015 and 2016,which can provide reference for medium and long-term power load forecasting under the background of the new normal economy of China.

关 键 词：经济新常态 负荷预测 VERHULST模型 支持向量机 

分 类 号：TM715[电气工程—电力系统及自动化]