Journal of Electrical Engineering and Automation

Mechatronic systems are widely applied in modern industrial fields, where their reliability and service life directly affect overall operational efficiency and safety. Traditional reliability assessment and lifetime prediction methods exhibit significant limitations in adaptability, accuracy, and dynamic response under complex working conditions. By constructing a predictive model that integrates data-driven approaches with physical mechanisms, the degradation patterns of the system can be more effectively captured, thereby enhancing the accuracy and practicality of predictions. Empirical analysis based on typical application scenarios demonstrates that such models exhibit strong stability and generalization capability in real-world operations, providing technical support for intelligent maintenance and full lifecycle management of the system.

Mechatronics; Reliability analysis; Lifetime prediction; Data-driven; Complex operating conditions