A Review of Physics-Informed Machine Learning Methods with Applications to Condition Monitoring and Anomaly Detection

TL;DR

This survey catalogs physics-informed machine learning (PIML) methods applied to condition monitoring, emphasizing how known physical laws can guide data-driven models. It organizes approaches into four families: physics-embedded feature spaces, data-enhanced refinement of physical models, physics-informed regularization, and physics-guided architecture design, including PINNs, transfer learning with digital twins, and physics-aware CNNs/RNNs/GNNs/GANs. Through numerous examples, the paper highlights advantages such as data efficiency, interpretability, and better generalization, while also noting limitations like computational complexity and the imperfect enforcement of soft constraints. The work provides a foundational roadmap for researchers and industry practitioners aiming to deploy PIML in maintenance, reliability, and fault-detection tasks with real-world impact.

Abstract

This study presents a comprehensive overview of PIML techniques in the context of condition monitoring. The central concept driving PIML is the incorporation of known physical laws and constraints into machine learning algorithms, enabling them to learn from available data while remaining consistent with physical principles. Through fusing domain knowledge with data-driven learning, PIML methods offer enhanced accuracy and interpretability in comparison to purely data-driven approaches. In this comprehensive survey, detailed examinations are performed with regard to the methodology by which known physical principles are integrated within machine learning frameworks, as well as their suitability for specific tasks within condition monitoring. Incorporation of physical knowledge into the ML model may be realized in a variety of methods, with each having its unique advantages and drawbacks. The distinct advantages and limitations of each methodology for the integration of physics within data-driven models are detailed, considering factors such as computational efficiency, model interpretability, and generalizability to different systems in condition monitoring and fault detection. Several case studies and works of literature utilizing this emerging concept are presented to demonstrate the efficacy of PIML in condition monitoring applications. From the literature reviewed, the versatility and potential of PIML in condition monitoring may be demonstrated. Novel PIML methods offer an innovative solution for addressing the complexities of condition monitoring and associated challenges. This comprehensive survey helps form the foundation for future work in the field. As the technology continues to advance, PIML is expected to play a crucial role in enhancing maintenance strategies, system reliability, and overall operational efficiency in engineering systems.

Figures (22)

• Figure 1: Tallied number of literary works discussed in this review, with respect to their year of publication. Note: Literature works reviewed in 2023 were limited up until the time of writing of this survey (June 2023).
• Figure 2: General outline for the process of the generation of synthetic data via physics-based methods.
• Figure 3: Finite Element model of the structure monitored constructed to provide simulated training data for Neural Network, as demonstrated in the work of seventekidis2020structural.
• Figure 4: Data augmentation employed to incorporate simulated fault and operation data for the training process of a machine learning fault classification algorithm, adapted from hopwood2022physics.
• Figure 5: The hybrid model, featuring physics-based modeling as a basis to map the observable parameters to unobservable parameters, for input to the machine learning algorithm.