A review of data-driven fault detection and diagnostics for building HVAC systems

Zhelun Chen, Zheng O’Neill, Jin Wen, Ojas Pradhan, Tao Yang, Xing Lu, Guanjing Lin, Shohei Miyata, Seungjae Lee, Chou Shen, Roberto Chiosa, Marco Savino Piscitelli, Alfonso Capozzoli, Franz Hengel, Alexander Kührer, Marco Pritoni, Wei Liu, John Clauß, Yimin Chen, Terry Herr
Applied Energy, Volume 339, 2023, 121030
https://doi.org/10.1016/j.apenergy.2023.121030

Abstract
With the wide adoption of building automation system, and the advancement of data, sensing, and machine learning techniques, data-driven fault detection and diagnostics (FDD) for building heating, ventilation, and air conditioning systems has gained increasing attention. In this paper, data-driven FDD is defined as those that are built or trained from data via machine learning or multivariate statistical analysis methods. Following this definition, this paper reviews and summarizes the literature on data-driven FDD from three aspects: process, systems studied (including the systems being investigated, the faults being identified, and the associated data sources), and evaluation metrics. A data-driven FDD process is further divided into the following steps: data collection, data cleansing, data preprocessing, baseline establishment, fault detection, fault diagnostics, and potential fault prognostics. Literature reported data-driven methods used in each step of an FDD process are firstly discussed. Applications of data-driven FDD in various HVAC systems/components and commonly used data source for FDD development are reviewed secondly, followed by a summary of typical metrics for evaluating FDD methods. Finally, this literature review concludes that despite the promising performance reported in the literature, data-driven FDD methods still face many challenges, such as real-building deployment, performance evaluation and benchmarking, scalability and transferability, interpretability, cyber security and data privacy, user experience, etc. Addressing these challenges is critical for a broad real-building adoption of data-driven FDD.ndoor thermal comfort plays a crucial role in enhancing the quality of life in residential and work environments. However, existing thermal comfort models often rely on complex measurements or require a large number of personal thermal votes, which limits their practical application. To address these challenges, this study develops a hybrid thermal comfort model aimed at reducing the measurement burden and personal response while improving the accuracy of personalized thermal comfort prediction. The proposed hybrid model combines a mathematical model with machine learning techniques, integrating the generalization ability of the mathematical model and the self-learning capabilities of machine learning. Data were collected from an experiment conducted in the climate controlled chamber in an office building with 12 subjects. By monitoring only wrist skin temperature, indoor air temperature, and their temporal variations, the proposed model significantly simplifies the measurement. In the absence of available training data, the mathematical model can be used independently, improving prediction accuracy by 21.11% on median and up to 44.45% over the PMV model. In a 5-fold cross-validation with 45 data points per subject, the hybrid model outperforms the standalone machine learning model by up to 24.45%. The model demonstrates robust performance with limited training data across various metrics and scenarios, highlighting its potential for practical application in building environments.