Real-time prediction of sub-item building energy consumption based on PCA-AR-BP method

2018 3rd Asia Conference on Power and Electrical Engineering

能源学;电工学

Qian, Qing^1 ; Tang, Guizhong^1,2 ; Zhang, Guangming^1,2

School of Electrical Engineering and Control Science, Nanjing University of Technology, Nanjing

211800, China^1

Jiangsu Province Green Building Engineering Technology Research Center, China^2

关键词: Building energy consumption;    Current energy;    Energy consumption prediction;    Prediction accuracy;    Prediction model;    Real-time prediction;    Special energy consumption;    Total energy consumption;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/366/1/012043/pdf
DOI  :  10.1088/1757-899X/366/1/012043

来源: IOP

【 摘 要 】

In this paper, a new method for real-time prediction of building energy consumption is proposed, this method solves the problem that the kinds of energy consumption are not distinguished and the prediction accuracy is low in the current energy consumption prediction algorithms. This paper divides the total energy consumption into four sections. Firstly, three main influencing factors of building energy consumption are extracted using PCA to realize real-time prediction; Secondly, the method of lighting energy consumption prediction based on time series analysis is constructed, the lighting energy consumption of the building is predicted in real time. Finally, the energy consumption prediction model based on BP network is established to predict the air conditioning, power and special energy consumption of the building. The experimental results show that the prediction model can predict energy consumption in every part of a building more accurately and effectively.