| Title |
Developing a DNN-Based Control Algorithm for Optimizing AHU Discharge Air and Cooling Water Temperatures in an Office Building |
| Authors |
Sang Hun Yeon ; Kwang Ho Lee |
| DOI |
https://doi.org/10.6110/KJACR.2024.36.11.535 |
| Keywords |
냉방 에너지; 심층 신경망; 에너지 시뮬레이션; 사무소 건물; 설정 온도 제어 Cooling energy; Deep neural network; Energy simulation; Office building; Set-point control |
| Abstract |
Energy consumption in modern society has been identified as one of the primary causes of global warming. In response, international agreements have been established to enhance energy efficiency and reduce energy consumption in an effort to mitigate global warming. In the building energy sector, research on energy-saving technologies utilizing ML (Machine Learning) has also been actively conducted. This study aimed to analyze energy-saving effects of controlling chilled water and AHU discharge air temperature using a Deep Neural Network (DNN). To achieve this, a coupled simulation using Energy Plus and MATLAB was established. The DNN model demonstrated a performance with a cvRMSE of 23.35%. Based on this, it was found that approximately 16% of energy could be saved compared to the base case (chilled water temperature at 32℃ and AHU discharge air temperature at 14℃). This study confirms that the proposed approach is an effective method for improving energy efficiency and saving energy in cooling systems. |