基于改进灰色模型的建筑能耗预测研究

Study on building energy consumption prediction based on the improved grey model

针对公共建筑逐月能耗呈现周期性震荡的特点,提出一种基于改进灰色模型的建筑能耗预测方法.在传统灰色模型(grey model,GM)的基础上,对建筑能耗的历史数据进行三角变换法处理,较好地解决了震荡序列的模拟和预测;引入弱化缓冲算子改善原始数据序列变化率较大时TGM模型的预测效果.将弱化TGM预测模型应用于各类建筑物的能耗预测,取得较为满意的结果.

Aimed at the public building with the characteristic of oscillations of the monthly building en- ergy consumption, this paper presents a predictive method of building energy consumption based on an improved GM method. On the basis of the traditional GM model, the trigonometric transformation on the historical data of building energy consumption well carries out to model and predicts the shock se- quences. The introduction of the weakening buffer operator improves the effect of prediction on TGM model when the presence of large fluctuations in original data sequences. The weakening TGM model is applied to predict the energy consumption of various types of buildings and the results are satisfactory.