基于分形特性修正气象相似日的节假日短期负荷预测方法

发布时间：2018-05-09 00:29

本文选题：节假日短期负荷预测 + 气象相似日　；参考：《电网技术》2017年06期

【摘要】：节假日负荷易受气象信息、国家调休政策等影响,预测精度较低。为解决上述问题,提出了基于分形特性修正气象相似日的节假日短期负荷预测方法。采用归一化处理和日期适当调整,解决了数据值之间的差异性、经济增长率和负荷变化趋势不一致的问题;将分形特性的自相似性引入节假日短期负荷预测,剔除气象突变带来的不良影响,在海量历史样本集中精准确定相似日查找范围;依据相似性原理,综合考虑气象、日类型等影响因素,建立曲线辨析函数计算负荷的差异系数,在特定范围内查找与待预测节假日气象最相似的一天。以南方某电网数据为实际算例进行仿真,结果表明所提出方法满足工程实际的需求,其中2015年春节期间96点日负荷准确度达97.63%。
[Abstract]:Holiday load is easy to be affected by meteorological information and national policy of adjustment and rest, and the accuracy of forecast is low. In order to solve the above problems, a short-term load forecasting method for holiday based on fractal characteristics is proposed. The problem of difference between data values, inconsistency of economic growth rate and trend of load change is solved by using normalized treatment and appropriate adjustment of dates, and the self-similarity of fractal characteristics is introduced into short-term load forecasting for holidays and holidays. After eliminating the bad influence brought by the sudden change of meteorology, we can accurately determine the search range of the similar day in the mass historical sample concentration; according to the similarity principle, considering the influence factors such as meteorology, day type and so on, the differential coefficient of load calculation by curve discriminating function can be established. Look for the day that is most similar to the holiday weather to be predicted in a specific range. Taking the data of a southern power grid as an example, the simulation results show that the proposed method meets the actual needs of the project, and the accuracy of the daily load at 96 points during the Spring Festival of 2015 is 97.63.
【作者单位】：广西电力系统最优化与节能技术重点实验室(广西大学);广西电网公司调度控制中心;
【基金】：国家自然科学基金项目(51407036) 国家重点基础研究发展计划项目(973项目)(2013CB228205)~~
【分类号】：TM715

【相似文献】