Journal of Engineering Research

Multi-physical field coupling modeling for digital twins of power grid equipment is an important foundation for achieving accurate mapping and prediction of equipment operating states. Aiming at the problem of insufficient accuracy in multi-field collaborative calculation with traditional modeling methods, this paper proposes an accuracy improvement method based on the integration of high-precision physical constraints and data-driven approaches. Under the framework of finite element multi-physical field solution, this method introduces a multi-scale modeling strategy for coupling characteristics in multiple domains such as electromagnetism, heat, and force, and dynamically corrects model parameters by combining real-time sensor data. Through multi-level model verification and error feedback optimization, the cumulative error in multi-field coupling calculations is significantly reduced, and the consistency between simulation and actual measurement is improved. Experimental results show that this method achieves excellent performance in thermo-electro-mechanical coupling analysis of typical power grid equipment such as transformers and circuit breakers. It provides technical support for intelligent monitoring and preventive maintenance throughout the equipment's full life cycle, and has important engineering application value and promotion potential.

Digital twins of power grid equipment; Multi-physical field coupling modeling; Accuracy improvement; Data-driven; Dynamic parameter correction