期刊文献+

基于相似日搜索的空调短期负荷预测方法 被引量:1

Short-term load forecasting of air-conditioning systems using similar day searching
原文传递
导出
摘要 针对新建楼宇空调系统做短期负荷预测工作时,缺少负荷预测所需的数据,难以实现空调系统优化节能的问题,提出一种基于相似日搜索的空调短期负荷预测方法———相似日搜索算法(SASD).算法首先通过分析空调负荷特性,定义日特征向量,构造日特征矩阵,缩小相似日的搜索范围;然后基于温度、湿度和风力3种天气影响因子,计算相似日的体感温度值;接着根据模糊思想选择正确的最终相似日判定因子,搜索得到最终相似日集合;最后通过判定选择面积中心法作为预测方法,实现工作日的负荷精确预测.仿真结果和实际预测效果表明:SASD可以精确预测空调负荷值,且在不同地区及不同时期具有一定的通用性. Owing to lack of the historical data,it is hard to forecast the short term loads of air-conditioning systems installed in new building and to implement energy-saving control.Thus,a search algorithm based on similar days(SASD) for forecasting the short term loads of air-conditioning systems was proposed.In this method,the air-conditioning load characteristic was analyzed,the day character vector was defined,the day character matrix was built to reduce the search range of similar days,and the similar days′ apparent temperature was computed based on three weather influence factors,including the temperature,humidity and wind.Then,the final-judging factors of similar days was selected according fuzzy thinking to reach the final similar days set to realize the workday load forecasting through selecting the center of area method finally.The simulation result and actual forecasting effectiveness show that this method can forecast the air-conditioning system load accurately and have some universality in different areas or times.
出处 《华中科技大学学报(自然科学版)》 EI CAS CSCD 北大核心 2011年第12期76-80,共5页 Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金 重庆市重大科技专项资助项目(CSTC 2008AB6115)
关键词 空调系统 负荷预测 相似日 体感温度 特征向量 特征矩阵 air-conditioning system load forecasting similar days apparent temperature character vector character matrix
  • 引文网络
  • 相关文献

参考文献11

  • 1康重庆,夏清,张伯明.电力系统负荷预测研究综述与发展方向的探讨[J].电力系统自动化,2004,28(17):1-11. 被引量:502
  • 2Aifur Rahman, Rahul Bhatnagar. An expert system based algorithm for short term load forecast[J]. IEEE Trans Power Systems, 1988, 3(2): 392-399.
  • 3Tsakoumis A C, Fessas P, Mladenov V M, et al. Application of chaotic time series for short-term load prediction[J]. WSEAS Trans on Systems, 2003, 2 (3) : 517-523.
  • 4Xiao Xinping College of Scienced, Wuhan University of Technologyl 430063, P R. China Deng Julong Dept. of Control, Huazhong University of Science and Technology, Wuhan 430074,P. R. China.A New Modified GM (1,1) Model: Grey Optimization Model[J].Journal of Systems Engineering and Electronics,2001,12(2):1-5. 被引量:12
  • 5Yoo H, Pimmel R L. Short term load forecasting using a self-supervised adaptive neural network[J]. IEEE Trans Power Systems, 1999, 14(2): 779-784.
  • 6Topalli A K, Erkmen I. A hybrid learning for neural networks applied to short term load forecasting[J]. Neurocomputing, 2003, 51 (7) .. 495-500.
  • 7莫维仁,张伯明,孙宏斌,胡子珩.短期负荷预测中选择相似日的探讨[J].清华大学学报(自然科学版),2004,44(1):106-109. 被引量:87
  • 8吕伟林.体感温度及其计算方法[J].北京气象,1998,(1).
  • 9Kawashima M, Dorgan C E, Mitchell J W. Hourly thermal load prediction for the nest 24 hours by ARIMA, EWA, LR, and an artificial neural net- work[J]. ASHRAE Transactions, 1995, 101(1): 186-200.
  • 10Bouslama F, Ichikaw A. Fuzzy control rules and their natural controllers[J]. Fuzzy Sets and Sys- tems, 1992, 48(1): 65-86.

二级参考文献73

共引文献596

同被引文献11

引证文献1

二级引证文献21

;
使用帮助 返回顶部