This is the authors' abstract. We don't add key points for this paper.
To mitigate the potential loss of computational accuracy in the Reduced‐Order Model (ROM) due to modal changes during transformer operation, this paper proposes a dynamic updating method for the ROM. This method enables the model to dynamically adjust and adapt to system changes. When transformer operating conditions change, new snapshot data is employed to update the original snapshot matrix, while the POD modes are updated by integrating matrix low‐rank decomposition with the Singular Value Decomposition (SVD) results of the original snapshot matrix—thus avoiding the need of SVD for the new snapshot matrix. By incorporating discrete measurement data from the winding temperature, the modal coefficients are solved in real‐time based on Gappy POD, facilitating the construction of the dynamic ROM. The proposed method was validated using a simulation model of 110 kV transformer windings. The results demonstrates that the maximum error in updating the POD modes is only 3.60 × 10 −6 , with a single update requiring approximately 0.12s. Furthermore, the dynamic ROM reduces the maximum error by 1.78 K. Without considering the snapshot matrix formation time, the average computation time for each time step is about 0.02s. This study presents a novel solution for the dynamic application of the ROM in the transformer temperature field.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Liu et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: