证据理论在中期负荷预测中的应用  

Application of Evidential Theory in Middle-term Load Forecasting

作　　者：倪明[1] 高晓萍[1] 单渊达[1] 

机构地区：[1]东南大学电气工程系

出　　处：《中国电机工程学报》1997年第3期199-203,共5页Proceedings of the CSEE

基　　金：国家自然科学基金

摘　　要：负荷预测中遇到的一个比较大的问题是影响负荷的因素很多，而且其中有很多不确定的因素。为此，本文提出了利用证据理论进行负荷预测的方法。本文中将影响负荷的因素分为三类：一类是负荷本身的历史数据；二类是可定量的相关因素；三类是难以定量的不确定因素。本文中就仅考虑一、二类因素和同时考虑三类因素两种情况，给出了利用证据理论进行负荷预测的推理网络和实现方法。算例表明。本文提出的方法预测精度较高，是一种将经验同数学模型有机结合的理想方法。A problem encountered in the load forecasting is that there are a lot of factors influencing the load and many of them are uncertain. In order to solve this problem,this paper proposes a new method of Evidential Theory,In this paper,the factors which affect the load are divided into three kinds: the first is the historical data of load itself.the second is the certainly associated factors ,the third is the uncertainly associated factors.For two conditions-only considering the first and the second kinds of factors and considering all of the three kinds of factors-this paper presents the reasoning networks and implementing methods respectively.The results of simulation show that this method is accurate enough and is an ideal way to combine the experience and mathematical models.

关 键 词：证据理论 电力系统 负荷预测 

分 类 号：TM715[电气工程—电力系统及自动化]