摘要
针对电网负荷预测时点预测误差相对较大的问题,本文提出一种模糊信息粒化支撑向量机的负荷预测方法。该方法采用支撑向量机为短期负荷预测的基本算法,结合了模糊信息粒化模型,通过三角型隶属函数对选定时间窗口的历史数据进行粒化,得到该时间窗口内数据变化的最小、平均和最大值,进一步结合支撑向量机进行训练与预测,实现了电网负荷的点预测和区间预测。以西安地区日负荷历史数据为例进行了算例分析,结果表明:本文提出的方法在进行点预测时精度高,平均误差为2.24%;能够对一定时间范围内的负荷变化情况和变化趋势进行预测,负荷数据真值全部落在所得的预测区间内。本文提出的方法对电网调度计划安排工作有一定意义。
In this paper, a new power load forecasting method based on fuzzy information granulation support vector machine is proposed to solve the problem that the time point prediction error is relatively large in power load forecasting. This method adopts the basic algorithm of the support vector machine as the short-term load forecasting, and combines the fuzzy information granulation model. The triangular membership function of historical data for the selected the minimum, average and maximum values of the data changes within the time window, with further training and prediction of support vector machine, this method is able to predict the point and interval power load forecasting. Taking the historical data of daily load in Xi＇an as an example, an example analysis is carried out. The results show that the method proposed in this paper has high accuracy in point prediction, with an average error of 2.24%. This method can predict the load change and change trend in a certain time range, and the real value of the load data all falls within the predicted interval. The method proposed in this paper has a certain significance for power grid scheduling.
作者
陈本阳
张成刚
倪鸣
张蕾
CHEN Ben-yang, ZHANG Cheng-gang, NI Ming, ZHANG Lei(State Grid Shaanxi Electric Power Company, Xi ＇ an 710048, Chin)
出处
《电子设计工程》
2018年第6期56-59,共4页
Electronic Design Engineering
关键词
负荷预测
支撑向量机
模糊信息粒化
区间预测
power load forecasting
support vector machine
fuzzy information granulation
interval prediction
