Digital Twin Enabled Asset Anomaly Detection for Building Facility Management
Ajith Kumar, Parlikad
Jennifer Mary, Schooling
4th IFAC Workshop on Advanced Maintenance Engineering, Services and Technologies
MetadataShow full item record
Xie, X., Qiuchen, L., Ajith Kumar, P., & Jennifer Mary, S. Digital Twin Enabled Asset Anomaly Detection for Building Facility Management. 4th IFAC Workshop on Advanced Maintenance Engineering, Services and Technologies. https://doi.org/10.17863/CAM.51737
Assets play a significant role in building utilities by undertaking the majority of their service functionalities. However, a comprehensive facility management solution that can help to monitor, detect, record and communicate asset anomalous issues is till nowhere to be found. The digital twin concept is gaining increasing popularity in architecture, engineering and construction/facility management (AEC/FM) sector, and a digital twin enabled asset condition monitoring and anomaly detection framework is proposed in this paper. A Bayesian change point detection methodology is tentatively embedded to reveal the suspicious asset anomalies in a real time manner. A demonstrator on cooling pumps is developed and implemented based on Centre for Digital Built Britain (CDBB) West Cambridge Digital Twin Pilot. The results demonstrate that supported by the data management capability provided by digital twin, the proposed framework realizes a continuous condition monitoring and anomaly detection for single asset, which contributes to efficient and automated asset monitoring in O&M management.
This research that contributed to this paper was supported by the Centre for Digital Built Britain (CDBB) with funding provided through the Government’s modern industrial strategy by Innovate UK, part of UK Research & Innovation. It was also partly funded by the EPSRC/Innovate UK Centre for Smart Infrastructure and Construction (Grant Numbers EP/N021614/1 and 920035).
Technology Strategy Board (920035)
Embargo Lift Date
This record's DOI: https://doi.org/10.17863/CAM.51737
This record's URL: https://www.repository.cam.ac.uk/handle/1810/304658
All rights reserved