文献类型：期刊文献

中文题名：基于温度区间的蓄冰空调负荷ARMAX模型

英文题名：ARMAX Model of Ice-storage Air Conditioning System Load Based on Temperature Interval

作者：徐今强[1,2];肖睿[1];黄冲[1];高日新[1];冯自平[1]

机构：[1]中国科学院广州能源研究所,广州510640;[2]广东海洋大学信息学院,湛江524088

年份：2009

卷号：31

期号：10

起止页码：109

中文期刊名：武汉理工大学学报

外文期刊名：Journal of Wuhan University of Technology

收录：CSTPCD、、Scopus、北大核心2008、CSCD_E2011_2012、北大核心、CSCD

基金：基金委-广东省联合基金项目(U0634005);国家“863”计划(2006AA05Z254);广东省科技计划项目(2006A10705002)

语种：中文

中文关键词：蓄冰空调;负荷预测;温度区间;ARMAX模型;辨识

外文关键词：ice-storage air conditioning system; load prediction; temperature interval; ARMAX model; identification

中文摘要：针对室外温度与空调负荷存在强相关的特点,提出一种新的预测次日平均负荷的ARMAX模型辨识方法。根据历史负荷和次日室外温度构建一个非线性ARMAX模型,在此基础上根据不同温度区间获得几组参数时变的线性子模型,子模型参数由在线回归算法辨识。结果表明,所提出的基于温度区间的ARMAX模型不仅比单一温度区间模型预测精度高,而且预测时不需要准确知道室外温度。

外文摘要：Ice-storage air conditioning system is an important technology, which can move the power used from day to night, the accuracy of load prediction is a key for optimizing the ice-storage air conditioning system control. In this paper, a new ARMAX model identification method for next day average load prediction is proposed considering the strong correlation between outdoor temperature and load. First a nonlinear ARMAX model for the average load is constructed taking into account the historical load and the next day outdoor temperature, based on this nonlinear model, several time-varying linear sub-models are obtained for different temperature intervals and the parameters of sub-models are identified using an on-line recursive algorithm. Obtained results show that the proposed temperature interval based ARMAX model not only is more accurate than the single temperature interval based model, but also it is not necessary to know precisely the outdoor temperature because the proposed prediction model is based on an interval for the future temperature instead of a number.