A digital twin-based framework for reliability estimation in ultrasonic guided wave structural health monitoring systems with temperature variations

The reliability estimation of ultrasonic guided wave structural health monitoring (SHM) systems is crucial yet challenging due to the need to represent practical conditions that evolve over a system’s lifespan. These conditions include various variables and uncertainties in measurements, which must be captured through experimental data collection or numerical data generation. This paper proposes and validates a novel digital twin-based framework that provides tailored reliability estimations throughout an SHM system’s lifecycle, accounting for specific measurement conditions, particularly temperature variations. The framework develops a digital twin model reflecting real-time measurement conditions using installed SHM measurements, from which key simulation parameters affected by environmental temperatures are updated. Laboratory experiments demonstrate the model’s ability to generate high-fidelity guided wave signals, effectively capturing temperature effects and noise levels. The framework’s reliability estimation performance is verified by comparing probability of detection (POD) analyses with the established superposition method. The paper further illustrates the framework’s capability to provide more accurate reliability assessments compared to traditional model-assisted POD methods by offering progressive, time-evolving performance evaluations that effectively account for dynamic temperature variations and changes in transducer performance over time. This approach represents a significant advancement in SHM reliability estimation, providing a feasible solution with improved accuracy and adaptability to real-world conditions through the dynamic development of the digital twin model.