首页 | 本学科首页   官方微博 | 高级检索  
     

一种基于径向基函数的短期负荷预测方法
引用本文:赵剑剑,张步涵,程时杰,陆俭. 一种基于径向基函数的短期负荷预测方法[J]. 电网技术, 2003, 27(6): 22-25,32
作者姓名:赵剑剑  张步涵  程时杰  陆俭
作者单位:1. 华中科技大学电气与电子工程学院,湖北省,武汉市,430074
2. 武汉市电力局,湖北省,武汉市,430074
基金项目:高等学校博士学科专项科研基金资助项目(2000048712)。
摘    要:为克服传统K均值聚类法局部寻优的缺陷,提出了基于确定性退火聚类选取径向基函数(RBF)网络隐层节点中心的方法,并采用遗传算法有效地解决了径向基函数网络的学习问题。在选择学习样本时,根据相似度方法,综合考虑了日期类型、星期类型、天气因素和曲线特性的影响。实际应用表明本方法能够改善预测精度,提高预测速度。

关 键 词:电力系统 短期负荷预测 径向基函数 K均值聚类法 遗传算法
文章编号:1000-3673(2003)06-0022-04

A DETERMINISTIC ANNEALING CLUSTERING BASED SHORT-TERM LOAD FORECASTING METHOD WITH RADIAL BASIS FUNCTION NETWORK
ZHAO Jian-jian,ZHANG Bu-han,CHENG Shi-jie,LU Jian. A DETERMINISTIC ANNEALING CLUSTERING BASED SHORT-TERM LOAD FORECASTING METHOD WITH RADIAL BASIS FUNCTION NETWORK[J]. Power System Technology, 2003, 27(6): 22-25,32
Authors:ZHAO Jian-jian  ZHANG Bu-han  CHENG Shi-jie  LU Jian
Affiliation:ZHAO Jian-jian1,ZHANG Bu-han1,CHENG Shi-jie1,LU Jian2
Abstract:To overcome the defects in local search by traditional K-means method, a method based on deterministic annealing clustering to determine the 'center' position of the hidden layer of radial basis function (RBF) network is proposed and the learning of RBF network is effectively solved by genetic algorithm. In the selection of learning samples, according to the similarity method the influences of the type of date, the type of weeks, the whether condition and the characteristics of the curves are comprehensively considered. The results of practical applications of the proposed method show that both the precision and speed of load forecasting can be improved.
Keywords:Short term forecasting  Deterministic annealing clustering  Genetic algorithm
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号