레트로 커미셔닝 실시를 통한 HVAC 시스템 에너지 절감 효과 검증에 관한 연구

Abstract

This study presents an empirical investigation aimed at improving the energy performance of HVAC systems in actual buildings through the implementation of Retro-Commissioning (RCx). As the building sector accounts for a substantial portion of total energy consumption, and HVAC systems operate continuously throughout the year, they are key targets for optimization. The study focuses on two different buildings. One building underwent structural enhancement of its thermal source system and simulation-based quantitative analysis, while the other building employed operational data from the Building Automation System (BAS) to apply energy-saving control algorithms and analyze their effectiveness. In the first building, an outdoor air cooling system was introduced to the thermal source system serving a specialized laboratory. Instead of operating chillers, chilled water was produced by utilizing low-temperature outdoor air during winter and intermediate seasons through cooling towers and heat exchangers. A physics-based simulation model was then constructed and calibrated using actual operation data to quantify the system's effectiveness. By inputting five years of historical outdoor temperature data into the model, it was determined that the system could reduce the annual power consumption of the thermal source system by an average of approximately 52.8%. In the second building, a Fault Detection and Diagnosis (FDD) process was conducted using BAS operation data. System faults such as sensor data update errors, incorrect differential pressure valve settings, and improper PI control gain values were identified and resolved, thereby stabilizing system operation. After stabilization, several energy-saving algorithms were developed and implemented. First, an AI-based control setpoint optimization algorithm was developed to reduce energy consumption in the thermal source system, resulting in a 7.6% reduction in energy use. Next, a refrigerant evaporating temperature control algorithm, originally developed for air-source systems in prior studies, was applied to a water-source geothermal heat pump system. The algorithm dynamically adjusted the evaporating temperature based on real-time load conditions, leading to a 10.3% reduction in power consumption. Additionally, a variable air volume (VAV) control algorithm was introduced to replace the constant air volume (CAV) operation in AHU systems. By modulating fan frequency according to space demand, the fan power consumption was reduced by 36.4%. This study confirms that HVAC system energy efficiency can be significantly improved by eliminating operational faults through RCx and sequentially applying optimal control algorithms. Furthermore, the quantitative analysis and empirical results based on real data provide a practical foundation for expanding these methods to other high-energy-consuming buildings and establishing intelligent building operation strategies.