JAUE2024-075: Development of Fault Detection System by Data Mining of BEMS Data

Abstract

Japan has declared its goal of achieving carbon neutrality by 2050 through the Paris Agreement, and efficient energy operation is required. As a result, the diffusion rate of BEMS, a building energy management system, is increasing. However, although studies have been conducted to detect building equipment defects using numerical calculations based on building modeling, they have not yet been put into practical use due to the time-consuming modeling process and the large amount of BEMS data. The objective of this study is to detect defects using a minimum amount of BEMS data. First, the causes of defects and the types of BEMS data needed to detect defects are collected from a survey of existing literature and compiled into a database. Next, multiple regression analysis will be performed on the actual data. By comparing the predicted results from the calculated regression equation with the actual data, it is determined if there is any abnormality in the building.