基于Granger-PSO-SVM的省域中长期电力负荷预测  

Mid-long Term Power Forecasting in Province Area Based on Granger-PSO-SVM

作　　者：王继娴 WANG Ji-xian(Department of Economic Management,North China Electric Power University(Baoding),Baoding 071003,China)

机构地区：[1]华北电力大学(保定)经济管理系,保定071003

出　　处：《价值工程》2021年第15期89-90,共2页Value Engineering

摘　　要：电力负荷预测是电网调度、规划里必不可少的环节,准确的负荷预测对区域电网安全稳定的运行以及当地经济发展都具有重要意义。影响地区电力负荷的因素因地而异,针对区域特征选取相应的影响指标会获得更加精准的预测效果,因此本文先运用格兰杰因果检验提取影响因素,接着结合粒子群优化支持向量机模型对某省全社会用电量进行训练与预测。根据MAE检验结果,本文提出的组合预测模型Granger-PSO-SVM误差较小,对今后的省域中长期电力负荷预测有一定的借鉴意义。Power load forecasting is an indispensable link in power grid dispatching and planning.Accurate load forecasting is of great significance to the safe and stable operation of regional power grids and the development of local economy.The factors affecting the regional power load vary from place to place,and the corresponding impact indicators for regional characteristics will obtain more accurate prediction results.Therefore,the Granger causality test is used to extract the influencing factors.Then this paper uses particle swarm optimization support vector machine model to train and predict the electricity consumption of a whole society in a province.MAE test standards show that the combined prediction model Granger-PSO-SVM proposed in this paper has lower error and it has certain reference significance for future mid-long term power load forecasting in the province.

关 键 词：格兰杰因果检验 电力负荷预测 粒子群优化 支持向量机 

分 类 号：TM732[电气工程—电力系统及自动化]