Volume 2: Innovative Solutions for Energy Transitions: Part I

A Novel Deep Learning Approach for Equipment Load Detection for Reducing Building Energy Demand Shuangyu Wei, Paige Wenbin Tien, John Kaiser Calautit, Yupeng Wu



The present work aims to develop a learning-based approach for a demand-driven control system which can automatically adjust the HVAC set points and supply conditions in terms of the actual requirements of the conditioned space. Internal heat gains from typical office equipment, such as computers, printers and kettle will be the focus of this paper. Due to its irregular use during scheduled heating or cooling service periods, an opportunity is offered to reduce unnecessary energy demands of HVAC systems related to the actual use of the equipment and its heat gains, i.e. over- and underutilization of equipment indicate whether indoor spaces are required to be conditioned or not. The work will be using deep learning enable cameras which can locally run trained algorithms to analyze and take action based on how equipment is utilised in a space real time. This proposed strategy automatically responds to the equipment usage for optimizing energy consumption and indoor conditions. The work will compare the performance of the developed approach with a conventional approach such as the use of static heating or cooling profiles. To highlight its capabilities, building energy simulation was used and initial results showed that while maintaining thermal comfort levels, up to 11% reduction of the energy consumption can be achieved by the proposed strategy in the comparison to conventionally-scheduled HVAC systems, while only focusing on three types of equipment.

Keywords Artificial intelligence; Built environment; Deep learning; Equipment detection; Energy savings; HVAC

Copyright ©
Energy Proceedings