Illumination Condition Effect on the Performance of CNN based Equipment Load Detection for Energy Demand Estimation

  • YEAR
    Wei, Shuangyu
    Tien, Paige Wenbin
    Calautit, John Kaiser
    Wu, Yupeng
    2020 Conference Papers
    Acoustics, Daylighting/Lighting, Natural Ventilation, Thermal Comfort and Indoor Air Quality
    Conference Papers


The main aim of this paper is to investigate the influence of lighting conditions on the detection accuracy of the vision-based equipment load detection approach. The work will be using artificial intelligence cameras to detect equipment information in different lighting levels, employing deep learning method to analyse and generate real-time equipment usage profiles for offices which can be inputted to the demand-based building controls to increase the efficiency of heating, ventilation, and air-conditioning systems. The performance of the developed approach in various illumination conditions was compared by using a building energy simulation tool. The results showed that as compared with the conventionally-scheduled heating, ventilation, and air-conditioning systems, the system with the use of equipment usage profiles conducted by the proposed approach can achieve up to 15% reduction in energy consumption depending on the setup of the artificial intelligence-enabled camera in terms of indoor lighting levels. The finding indicates that adequate illumination level contributes to the decrease of building energy demand by achieving an effective deep learning approach.

Keywords: Built Environment; Deep Learning; Equipment Detection; Energy Savings.


To top