| Title |
Building Digital Twins Based In-situ Fault Detection Method |
| Authors |
Byeongjun Jang ; Jeyoon Lee ; Sungmin Yoon |
| DOI |
https://doi.org/10.6110/KJACR.2026.38.9.509 |
| Keywords |
건물 디지털 트윈; 냉난방공기조화시스템; 센서 이상 탐지; 가상 현장 중심 보정 Building digital twins; Heating; ventilation; and air conditioning systems; Sensor fault detection; Virtual in-situ calibration |
| Abstract |
As global demand for carbon neutrality rises, fault detection and diagnosis (FDD) in building HVAC systems has become crucial for energy-efficient operation. However, traditional FDD models are often built with manually tuned parameters and do not adequately capture the unique operational characteristics of individual buildings. This study addresses these limitations by proposing a novel method that combines an in-situ modeling approach with a performance evaluation process for fault detection. In this method, fault detection models are automatically generated using site-specific operational data and are continuously calibrated and updated throughout the building's life cycle within a digital twin framework. To validate the effectiveness of this approach, comparative experiments were conducted between models generated by our proposed method and traditional parameter-based models across various fault scenarios. The results show that the models produced by our method consistently demonstrated superior detection performance in all scenarios. These findings indicate that our proposed method not only improves fault detection performance but also supports more adaptive and energy-efficient building operations over time. |