Model Validation and Calibration of Large Power Systems Using Hierarchical Topological Cuts

This paper presents a methodology for validating and calibrating dynamic models in large interconnected power systems using synchronized measurements. The approach combines topological cuts at several hierarchical levels with Hybrid Dynamic Simulation (HDS), allowing the systematic validation of network partitions or indication of partitions with model inaccuracies. Discrepancy indicators computed from Phasor Measurement Units (PMU) data are employed for both validation and detection of modeling inaccuracies. Once discrepancies are located, a calibration procedure is applied, which is formulated as an optimization problem and solved using Genetic Algorithms. The methodology is assessed on a test system using synthetic measurements and on the Brazilian Interconnected Power System (BIPS) using real disturbance records. The results indicate the effectiveness of the proposed methodology for model validation, the identification of inaccuracy sources, and model calibration.