Energy fault detection for small buildings based on peer comparison of estimated operating status

Abstract As the penetration of smart meters grows, new energy conservation services using electricity consumption data have drawn much attention. Yet service frameworks for small buildings to provide alerts automatically when a failure in energy conservation is detected have not been sufficiently investigated. Thus, we propose a framework for detecting energy faults based on peer comparison of operating time estimated by self-organizing maps from smart meter data with a coarse time resolution. The framework first defines a common operating status for the business type using the results of operating time estimation from multiple buildings of the same type, and then monitors the operating status at 1-h intervals using only a fault threshold parameter specified by the user. If each estimated operating status is judged to be significantly larger than the average level for the building type, the system identifies it as an energy fault. The results of case studies using an open building energy management system dataset suggest that the proposed framework can detect failures in energy conservation caused by inappropriate HVAC management, which usually results in higher contract power consumption, or by failure to turn off appliances that unnecessarily consume energy after business hours.

[1]  Paul Raftery,et al.  Review of automated fault detection and diagnostic tools in air handling units , 2014 .

[2]  Shaolei Ren,et al.  Incentivizing Energy Reduction for Emergency Demand Response in Multi-Tenant Mixed-Use Buildings , 2018, IEEE Transactions on Smart Grid.

[3]  G. Edward Gibson,et al.  Early-warning application for real-time detection of energy consumption anomalies in buildings , 2017 .

[4]  Alvin B. Culaba,et al.  Optimal design of an integrated renewable‐storage energy system in a mixed‐use building , 2020, International Journal of Energy Research.

[5]  Barry Hayes,et al.  Non-Intrusive Load Monitoring and Classification of Activities of Daily Living Using Residential Smart Meter Data , 2019, IEEE Transactions on Consumer Electronics.

[6]  Yi Wang,et al.  Review of Smart Meter Data Analytics: Applications, Methodologies, and Challenges , 2018, IEEE Transactions on Smart Grid.

[7]  Clayton Miller.  Screening Meter Data: Characterization of Temporal Energy Data from Large Groups of Non-Residential Buildings , 2016 .

[8]  Tao Huang,et al.  Big data analytics in smart grids: a review , 2018, Energy Informatics.

[9]  Zhonghua Gou,et al.  Energy use characteristics and benchmarking for higher education buildings , 2018 .

[10]  Graeme Burt,et al.  Enhanced Load Profiling for Residential Network Customers , 2014, IEEE Transactions on Power Delivery.

[11]  G. K. Papagiannis,et al.  Load profiling in the deregulated electricity markets: A review of the applications , 2012, 2012 9th International Conference on the European Energy Market.

[12]  David M. W. Powers,et al.  Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation , 2011, ArXiv.

[13]  Felix Naumann,et al.  Data fusion , 2009, CSUR.

[14]  Thorsten Staake,et al.  Gaining IS Business Value through Big Data Analytics: A Case Study of the Energy Sector , 2015, ICIS.

[15]  Pandarasamy Arjunan,et al.  Islands of misfit buildings: Detecting uncharacteristic electricity use behavior using load shape clustering , 2021 .

[16]  Fabio Rinaldi,et al.  Building characterization through smart meter data analytics: Determination of the most influential temporal and importance-in-prediction based features , 2021 .

[17]  K. Armel,et al.  Is disaggregation the holy grail of energy efficiency? The case of electricity , 2013 .

[18]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .

[19]  Arno Schlueter,et al.  A review of unsupervised statistical learning and visual analytics techniques applied to performance analysis of non-residential buildings , 2018 .

[20]  David Irwin,et al.  WattScale: A Data-driven Approach for Energy Efficiency Analytics of Buildings at Scale , 2020, ArXiv.

[21]  Zheng Yang,et al.  DUE-B: Data-driven urban energy benchmarking of buildings using recursive partitioning and stochastic frontier analysis , 2018 .

[22]  Hidenori Komatsu,et al.  Peak demand alert system based on electricity demand forecasting for smart meter data , 2020 .

[23]  Osamu Kimura,et al.  A combination of SOM-based operating time estimation and simplified disaggregation for SME buildings using hourly energy consumption data , 2019, Energy and Buildings.

[24]  Heng Huang,et al.  Using Smart Meter Data to Improve the Accuracy of Intraday Load Forecasting Considering Customer Behavior Similarities , 2015, IEEE Transactions on Smart Grid.

[25]  Chongqing Kang,et al.  Load profiling and its application to demand response: A review , 2015 .

[26]  Alvin B. Culaba,et al.  Machine learning‐based energy consumption clustering and forecasting for mixed‐use buildings , 2020, International Journal of Energy Research.

[27]  M. Etezadi-Amoli,et al.  Smart meter based short-term load forecasting for residential customers , 2011, 2011 North American Power Symposium.

[28]  Michael Baldea,et al.  Nonintrusive disaggregation of residential air-conditioning loads from sub-hourly smart meter data , 2014 .

[29]  Da Yan,et al.  Building categorization revisited: A clustering-based approach to using smart meter data for building energy benchmarking , 2020 .

[30]  VARUN CHANDOLA,et al.  Anomaly detection: A survey , 2009, CSUR.

[31]  Srinivas Katipamula,et al.  Review Article: Methods for Fault Detection, Diagnostics, and Prognostics for Building Systems—A Review, Part I , 2005 .