State Assessment Method for Power Secondary Equipment Based on Digital Twin Technology

Power system secondary equipment is a critical component for ensuring the stable operation of the grid. However, traditional maintenance modes primarily rely on fault-based and periodic maintenance, which suffer from unpredictability, over-maintenance, or under-maintenance. With the rapid development of smart grids, the operational data of secondary equipment has become multi-source, heterogeneous, and rapidly updated, necessitating a technical approach capable of real-time perception, dynamic analysis, and accurate prediction of equipment states. To address this, this paper proposes a state assessment method that integrates digital twin technology with interval-extreme learning machines. Interval-type principal component analysis is employed to reduce the dimensionality of massive operational data and screen key assessment indicators. Combined with a particle swarm optimization-extreme learning machine algorithm, a state assessment model based on operational feature clustering is established to achieve accurate state assessment of secondary equipment. A digital twin system is then designed based on this model. Simulation results demonstrate that this method significantly improves assessment accuracy, providing a feasible technical solution for the preventive maintenance of power secondary equipment.