A Real-Time Non-Invasive Anomaly Detection Technique for Cooling Systems

Abstract

The cooling systems contribute to 40% of overall building energy consumption. Out of which, 40% After identifying the anomalies, we find the cause of the anomaly. Based on the anomaly, the solution recommends a fix. If there is a technical fault, our proposed technique informs the technician regarding the faulty component, reducing the cost and mean time to repair. In the first stage, we propose a domain-inspired statistical technique to identify anomalies in cooling systems. We observe the Area Under the Curve of the Receiver Operating Characteristic ($AUC-ROC$) score of more than 0.93 in both simulation and experimentation. In the second stage, we propose using a rule-based technique to identify the anomaly’s cause and classify it into three classes. We observe an $AUC-ROC$ score of 1. Based on the anomaly classification, in the third stage, we identify the faulty component of the cooling system. We use the Nearest-Neighbour Density-Based Spatial Clustering of Applications with Noise (NN-DBSCAN) algorithm with transfer learning capabilities to train the model only once, where it learns the domain knowledge using simulated data. The overall accuracy of the three-stage technique is 0.82 and 0.86 in simulation and experimentation, respectively. We observe energy savings of up to $68\%$ in simulation and $42\%$ during experimentation.