电网技术2011，Vol.35Issue(11)：195-199,5.
应用启发式最小二乘支持向量机的中长期电力负荷预测
Medium-and Long-Term Load Forecasting Based on Heuristic Least Square Support Vector Machine
李如琦 1苏浩益 1王宗耀 2邓国良 1陈铁洲1
作者信息
- 1. 广西大学电气工程学院,广西壮族自治区南宁市530004
- 2. 南昌工程学院机械与电气工程学院,江西省南昌市330099
- 折叠
摘要
Abstract
Founded on＇ total amount of annual load, annual load growth amount, annual load growth rate, acceleration rate of annual load growth and annual gross domestic product （GDP）, a long-term load forecasting model based on heuristic least square support vector machine （LS-SVM） is built. Firstly, the low dimensional input variable space is mapping to high dimensional characteristic space by kernel function and a kernel partial least squares regression model is built to fit out power consumption per unit of GDP; then taking the power consumption per unit of GDP as heuristic operator, on the basis of historical data the total load amount in the forecasted year is reasonably assumed, and then using the heuristic operator the annual GDP corresponding to the assumed total load amount is obtained by inverse method to form expanded training samples of SVM and the extrapolation forecasting of SVM is changed into interpolation evaluation; finally, the forecasting result is solved by well-trained SVM. Calculation results of actual case show that using the proposed method the forecasting results with higher accuracy can be obtained, and the proposed method is practicable.关键词
负荷预测/支持向量机/核偏最小二乘回归/启发式算子/单位国内生产总值电耗Key words
load forecasting/support vector machine/kernel partial least squares regression/heuristic operator/electricity consumption per unit of GDP分类
信息技术与安全科学引用本文复制引用
李如琦,苏浩益,王宗耀,邓国良,陈铁洲..应用启发式最小二乘支持向量机的中长期电力负荷预测[J].电网技术,2011,35(11):195-199,5.
