Identifying important variables of energy use in low energy office building by using multivariate analysis

Abstract

The aim of the study was to indentify driving variables that contributed to energy use in a low energy office building by integrating building energy management system (BEMS) and energy use data. To take a further step towards zero emission buildings, it is necessary to identify what contributes the most to building energy use. Further, the idea was to encourage a smart use of BEMS data for energy use analysis. Principal component regression and partial least squares regression were used for the data analysis. Databases of 76 and 41 variables respectively, which included occupancy level, control signals, and water and air temperatures, were used to explain heating, electricity, and fan energy use. Variable contributions to the principal components were used to simplify the model and to find the most important variables. In this way, energy use was defined indirectly by using available variables in BEMS. The approach was tested on a low energy office building. The results showed that important variables were different for different months in the case of heating energy use. The total electricity and fan electricity use could be defined with the same variables in different months. The total electricity use could be defined by using occupancy level and input fan signals. The suggested approach could be used by building operators to identify opportunities for decreasing energy use and for energy use estimation when data are lost due to data transmission issues or other problems. A relationship between building information and energy use was established.

The performance and analysis of office building energy consumption in the west of Inner Mongolia Autonomous Region, China

“…If the main factors affecting the BEC can be identified, it means that a key point to reduce BEC is found, which is very helpful to energy-saving work. At present, many domestic and foreign scholars have studied the relationship between public building energy use and its influence factors [29][30][31][32][33][34][35][36][37][38][39][40].…”

Complex building’s energy system operation patterns analysis using bag of words representation with hierarchical clustering

“…Faults can also be detected in buildings with machine learning algorithms using the information from the installed electricity consumption meters as shown in (Figueiredo et al 2005), Domínguez et al (2013). There are different parameters available that can be useful for the prediction of electricity consumption for each HVAC component; multivariate analysis can be used to calculate these parameters (Djuric and Novakovic 2012).…”

Prediction of Space Heating Energy Consumption in Cabins Based on Multivariate Regression Modelling

“…It has the capability to address the complexity present in the building heating systems. PLSR has been used by Djuric and Novakovic and Olofsson et al to estimate the energy use of buildings [15], [16]. The study in [15] aims to identify driving variables of energy utilization in a low energy office building located in Trondheim, Norway by integrating BEMS and energy usage data.…”

“…PLSR has been used by Djuric and Novakovic and Olofsson et al to estimate the energy use of buildings [15], [16]. The study in [15] aims to identify driving variables of energy utilization in a low energy office building located in Trondheim, Norway by integrating BEMS and energy usage data. They have related building information with the building energy consumption using multivariate data analysis techniques such as Principal Component Regression (PCR) and PLSR.…”