Fault Detection for Non-Condensing Boilers using Simulated Building Automation System Sensor Data
This work addresses energy waste and emissions in building heating systems, but it is incremental as it applies existing machine learning methods to a new simulated dataset.
The study tackled fault detection in non-condensing boilers by creating a simulated dataset of 10,000 steady-state simulations for 14 boilers and testing classification methods, with Support Vector Machines achieving over 90% accuracy but failing to generalize across boilers.
Building performance has been shown to degrade significantly after commissioning, resulting in increased energy consumption and associated greenhouse gas emissions. Continuous Commissioning using existing sensor networks and IoT devices has the potential to minimize this waste by continually identifying system degradation and re-tuning control strategies to adapt to real building performance. Due to its significant contribution to greenhouse gas emissions, the performance of gas boiler systems for building heating is critical. A review of boiler performance studies has been used to develop a set of common faults and degraded performance conditions, which have been integrated into a MATLAB/Simulink emulator. This resulted in a labeled dataset with approximately 10,000 simulations of steady-state performance for each of 14 non-condensing boilers. The collected data is used for training and testing fault classification using K-nearest neighbour, Decision tree, Random Forest, and Support Vector Machines. The results show that the Support Vector Machines method gave the best prediction accuracy, consistently exceeding 90%, and generalization across multiple boilers is not possible due to low classification accuracy.