基于时间序列分析的建筑能耗预测方法
Building energy consumption prediction method based on time series analysis
摘要：
将时间序列分析方法应用到商业建筑能耗的分析和预测中，介绍了建筑能耗预测模型的建模方法，将逐月积温值、逐月相对湿度平均值、逐月工作日天数及逐月非工作日天数4个建筑逐月能耗的主要影响因子引入建筑能耗预测模型，根据逐月建筑能耗数据，建立基于时间序列分析并加入物理原理化处理的数据驱动模型，并进行检验与修正。应用该方法对3个商业建筑进行了实例分析，结果显示，对于逐月出租率基本不变的商业建筑，能耗预测结果较理想。
Abstract：
Applies the time series analysis method to analyses and prediction of energy consumption of commercial buildings, and presents the modeling method of the building energy consumption prediction model. Introduces the four main factors of influencing monthly energy consumption, i.e. monthly cumulative temperature value, monthly mean relative humidity value, monthly work days and monthly non-work days into the building energy consumption prediction model. According to monthly building energy consumption data, establishes a data-driven model based on time series analysis and physical principle, and the model is modified and verified. Applies the method to three commercial buildings and the results show that for the commercial building with almost constant monthly occupancy rate, the energy consumption prediction result is better agreed with the actual data.
Keywords:time series, energy consumption prediction, commercial building, monthly cumulative temperature value, monthly mean relative humidity value, monthly work days, monthly non-work days