基于TOPSIS和递归等权法的中长期负荷组合预测Medium and Long-term Load Combination Forecasting Based on TOPSIS and Recursive Equalization
李辉,陈耀,丁杰,赵爱芳,朱珞敬
LI Hui,CHEN Yao,DING Jie,ZHAO Aifang,ZHU Luojing
摘要(Abstract):
中长期负荷预测是电力系统规划的重要环节。针对单一负荷预测模型预测精度不足、适用性有限等缺点,采用TOPSIS和递归等权法建立组合模型。以单一预测模型各个年份拟合的相对误差绝对值的倒数为属性集,单一预测模型的种类为方案集,通过TOPSIS法对单一模型的拟合精度进行排序和筛选,确定参与组合的模型;利用递归等权法确定参与组合模型的权重。通过对某省全社会用电量进行预测,并与单一预测模型进行误差比较,结果表明所提组合模型具有较高的实用价值。
Medium and long-term power load is a key link of power system planning. In view of low accuracy and limited applicability of single load forecasting model, technique for order preference by similarity to an ideal solution(TOPSIS) and recursive equalization are used to build combination model. Reciprocal of fit relative error absolute value in each year of various single forecasting models are taken as an attribute set and categories of various single forecasting models are taken as a scheme set. TOPSIS is used to sort and filter for prediction accuracy of single forecasting models, and the models of participation are identified. Then, the recursive equalization is use to determine the weights of the various single models. The high practical value of combination model is verified through social power consumption forecasting in a province and comparison of error with single forecasting model.
关键词(KeyWords):
TOPSIS;递归等权法;中长期电力负荷;组合预测
TOPSIS;recursive equalization;medium and long-term power load;combination forecasting
基金项目(Foundation):
作者(Author):
李辉,陈耀,丁杰,赵爱芳,朱珞敬
LI Hui,CHEN Yao,DING Jie,ZHAO Aifang,ZHU Luojing
DOI: 10.19585/j.zjdl.201803006
参考文献(References):
- [1]毛李帆,江岳春,龙瑞华,等.基于偏最小二乘回归分析的中长期电力负荷预测[J].电网技术,2008,32(19):71-77.
- [2]李辉.用马尔可夫法改进残差GM(1,1)模型中长期负荷预测[J].陕西电力,2017,45(5):75-77.
- [3]杨胡萍,李辉.基于Adaboost的改进多元线性回归算法中长期负荷预测[J].太原理工大学学报,2017,48(5):818-821.
- [4]王栋,邵常宁,费建平,等.基于大客户和地方电厂数据管理预测平台的负荷还原预测方法研究[J].浙江电力,2015,34(12):11-15.
- [5]李春祥,牛东晓,孟丽敏.基于层次分析法和径向基函数神经网络的中长期负荷预测综合模型[J].电网技术,2009,33(2):99-104.
- [6]蒋燕,王少杨,封芸.基于递归等权组合模型的中长期电力负荷预测[J].电力系统及其自动化学报,2012,24(1):151-155.
- [7]牛东晓,曹树华,卢建昌,等.电力负荷预测技术及其应用[M].北京:中国电力出版社,2009.
- [8]贺辉.电力负荷预测和负荷管理[M].北京:中国电力出版社,2012.
- [9]张栋梁,严健,李晓波,等.基于马尔可夫链筛选组合预测模型的中长期负荷预测方法[J].电力系统保护与控制,2016,44(12):63-67.
- [10]温青,张筱慧,杨旭.基于负荷误差和经济发展趋势的组合预测模型在中长期负荷预测中的应用[J].电力系统保护与控制,2011,39(3):57-61.
- [11]杨胡萍,李辉,占思凯,等.基于理想点多属性决策算法的中长期电力负荷组合预测[J].燕山大学学报,2017,41(1):63-67.
- [12]孙广强,姚建刚,谢宇翔,等.基于新鲜度函数和预测有效度的模糊自适应变权重中长期电力负荷组合预测[J].电网技术,2009,33(9):103-107.
- [13]陈毅波,郑玲,姚建刚.基于粗糙集理论与D-S证据理论改进的多元回归负荷预测方法研究[J].电力系统保护与控制,2016,44(6):62-67.
- [14]张锦爱.基于法的电网规划方案评估[J].安徽电力,2013,30(3):32-34.
- [15]康重庆,夏清,刘梅.电力系统负荷预测[M].北京:中国电力出版社,2007.
- [16]雷涛,吕勇.马淑慧.基于改进灰色理论的主动配电网中长期负荷控制[J].电网与清洁能源,2016,32(9):22-28.
- [17]李辉.改进残差GM(1,1)模型在中长期负荷预测中的应用[J].广东电力,2017,30(9):81-85.