基于变权组合模型的中长期负荷概率密度预测  

Medium-and Long-term Load Probability Density Forecasting Based on Variable Weight Combination Model

作　　者：刘明 王红蕾[1] 索良泽 LIU Ming;WANG Honglei;SUO Liangze(Electrical Engineering College,Guizhou University,Guiyang 550025,China)

机构地区：[1]贵州大学电气工程学院

出　　处：《电力系统及其自动化学报》2019年第7期88-94,共7页Proceedings of the CSU-EPSA

基　　金：中国南方电网有限责任公司重点科技资助项目(066601(2016)030101XT198)

摘　　要：单一模型或固定权重组合模型进行中长期负荷预测时,可能存在预测精度不高和预测误差不稳定的问题;且由于诸多不确定因素对中长期负荷预测的影响,确定的点预测结果不能很好地反映中长期负荷的波动性。针对上述问题,提出基于变权组合模型的中长期负荷概率密度预测方法,即在传统固定权重组合模型的基础上,考虑本次的预测结果,对模型权重进行自适应调整;利用变权组合模型对中长期负荷进行区间预测,最后再用高斯核密度估计方法求取区间概率密度函数。实际算例结果表明:与常规方法相比,该方法在预测精度上有所提高,预测误差波动范围进一步减小。In the medium-and long-term load forecasting using a single or fixed weight combination model,it is possible that the prediction accuracy is poor and the prediction error is unstable. Besides,because of the influences of various uncertain factors on the medium-and long-term load forecast,the fluctuations of medium-and long-term load cannot be well reflected by the point prediction result. Considering these problems,a medium-and long-term load probability density forecasting method is proposed in this paper based on a variable weight combination model,in which the model weights are adaptively adjusted on the basis of the traditional fixed weight combination model as well as the forecasting result. Then,the variable weight combination model is used to perform interval prediction for the medium-and long-term load. At last,Gaussian kernel density estimation method is used to obtain the interval probability density function. The result of a practical numerical example proves that compared with the conventional method,the proposed method can improve the prediction accuracy to a certain degree and narrow the fluctuation range of prediction errors further.

关 键 词：中长期负荷预测 变权组合模型 区间预测 高斯核密度估计 概率密度预测 

分 类 号：TM715[电气工程—电力系统及自动化]