首页 | 本学科首页   官方微博 | 高级检索  
     

基于BP网络的下一交易日无约束市场清算价格预测模型
引用本文:杨莉,邱家驹,江道灼.基于BP网络的下一交易日无约束市场清算价格预测模型[J].电力系统自动化,2001,25(19):11-14,33.
作者姓名:杨莉  邱家驹  江道灼
作者单位:浙江大学电气工程学院
摘    要:提出一种基于反向传播(BP)网络预测下一交易日无约束市场清算价格的方法。该方法去除了原始数据中的季节性趋势和增长趋势,用相关性分析技术选择输入量,遵循“重近轻远”的原则重新设计了拟合误差的代价函数,利用最新数据在线修正BP网络,并用美国加州电能交易所公布的真实数据进行仿真,结果令人满意。

关 键 词:电力市场    价格预测    无约束市场清算价格    反向传播网络    相关性分析
收稿时间:1/1/1900 12:00:00 AM
修稿时间:1/1/1900 12:00:00 AM

BPN BASED DAY-AHEAD UNCONSTRAINED MARKET CLEARING PRICE FORECASTING MODEL
Yang Li,Qiu Jiaju,Jiang Daozhuo Zhejiang University,Hangzhou ,China.BPN BASED DAY-AHEAD UNCONSTRAINED MARKET CLEARING PRICE FORECASTING MODEL[J].Automation of Electric Power Systems,2001,25(19):11-14,33.
Authors:Yang Li  Qiu Jiaju  Jiang Daozhuo Zhejiang University  Hangzhou  China
Affiliation:Yang Li,Qiu Jiaju,Jiang Daozhuo Zhejiang University,Hangzhou310 0 2 7,China
Abstract:An innovative method to forecast the day- ahead unconstrained marketclearing price U MCP based on BP network is presented.The seasonal and increasing tendency of raw data is wiped off.Correlation analysis technique is used to choose the input variables of BPN. An error performance function is redesigned to put more emphasis on the latest data. A tracking mechanism is developed by on- line adjusting weights.Finally,a sim ulation on the retrospective real data acquired from California Power Exchange web site shows considerable promising results.
Keywords:power market  price forecasting  unconstrained m arket clearing price  U MCP  back propagation network  BPN  correlation analysi
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《电力系统自动化》浏览原始摘要信息
点击此处可从《电力系统自动化》下载全文
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