时间序列方法及其在实际办公楼能耗预测中的应用-同济大学.PDF

2012 年第2 期(总第40 卷 第252 期) ■ 标准与检测 建筑节能 No.2 in 2012 (Total No.252 ，Vol.40) STANDARD TESTING doi ：10.3969/j.issn.1673-7237.2012.02.016 时间序列方法及其在实际办公楼能耗预测中的应用 1 2 2 周芮锦 ，潘毅群 ， 黄治钟 (1. 同济大学机械工程学院，上海 201804 ；2. 同济大学中德工程学院，上海 201804) 摘要： 建立了基于时间序列分析方法的建筑能耗预测模型，对上海某办公建筑逐月能耗进行预测。采用CensusX12 方法进行季节调 整，建立ARMA 时序模型，并对季节因子进行温度化处理。模拟结果表明，采用时间序列分析的方法来预测建筑能耗是一种有 效方法。 关键词： 时间序列； 建筑能耗预测； ARMA 模型 中图分类号： TU111.19 文献标志码： A 文章编号： 1673-7237(2012)02-0055-05 Time Series Analysis Method and Its Application in Energy Consumption Forecast of an Actual Office Building 1 2 2 ZHOU Rui-jin , PAN Yi-qun , HUANG Zhi-zhong (1.College of Mechanical Engineering, Tongji University, Shanghai 201804, China; 2. Sino-German College of Applied Sciences, Tongji University, Shanghai 201804, China) Abstract: Based on time series analysis method, it establishes a forecast model of building energy consumption to forecast the monthly energy consumption of an office building in Shanghai. CensusX12 method is used to do seasonal adjustment, ARMA model is established and the seasonal factor is processed with accumulated temperature. The forecast results show that time series analysis method is applicable for building energy consumption forecast. Key w ords: time series; building energy consumption forecast; ARMA Model 0 引言 建筑系统存在巨大惯性，在考虑一定随机性的基 建筑节能工作中对于建筑能耗的科学分析和合