基于统计特征量和支持向量机的短期售电量预测研究

Research on Short Term Electricity Sales Forecasting based on Statistical Characteristics and Support Vector Machine

售电量预测是电网建设规划的重要依据。由于售电量预测是多种因素综合作用的多变量、非线性问题,传统的方法往往难以取得满意的预测效果。本文提出了一种基于统计特征量和支持向量机的短期售电量预测方法。选用对售电量影响度高的因素及其统计量做为特征因子,运用支持向量机进行建模和预测。实验结果表明,该方法预测准确较高,有着较为广阔的应用前景。

Electricity sales forecasting is an important basis for power grid construction and planning. Because the sale of electricity is an issue of multi variable, nonlinear interaction of many factors, the traditional methods are of- ten difficult to obtain satisfactory forecasting effect. This paper presents a method to predict the short-term electrici- ty sales based on statistical characteristics and support vector machine. The selection of the sale of electricity fac- tors with high influence and its statistics as the characteristic factor, the modeling and prediction using support vec- tor machine were made. The experimental results show that the prediction accuracy of the method is high, has a broad prospect of application.