Chiller systems in buildings take up the highest proportion of electricity consumption and provide the major energy management opportunities. Medium to large scale chiller systems have many components which are controlled automatically under building management systems. Trend logs are increasingly provided to measure and verify system performance. Association rule mining (ARM) is a rule-based machine learning method to examine interesting features in a big data set. Compared with sophisticated models, ARM is easily applied to analyze which operating conditions give a good or poor coefficient of performance. This study will share a case study on applying ARM to evaluate energy improvement measures of an existing chiller system serving an institutional building.
Period
20 May 2022
Event title
Technical Talk on Application of Artificial Intelligence (AI) and Big Data Analysis for Chiller Plant Optimization