Abstract
The coordination of inverse-time overcurrent protection relays (51 relays) requires careful consideration to achieve an appropriate balance of selectivity, speed, and reliability when the system is subjected to faults. This study presents an evaluation of optimization algorithms using both linear and nonlinear programming methods available in Python’s SciPy library. The setting parameters considered are the Time Dial Setting (TDS), Pickup Setting (PS), and Time-Current Characteristic (TCC) curves, which must be efficiently configured. Several solvers were analyzed, including: highs, highs-ds, highs-ipm, simplex, revised simplex, interior-point, COBYLA, SLSQP, and trust-constr. Each was evaluated based on three criteria: relay operating time, sensitivity to fault current, and computational efficiency. The results indicate that the differences in total solver execution times are minimal, on the order of 1×10⁻⁷ seconds. However, highs-ipm and COBYLA demonstrated superior performance in terms of both speed and accuracy. Moreover, the inclusion of multiple setting parameters significantly enhances the overall effectiveness of the protection system. The study concludes that the optimal solver choice depends on the trade-off between computational time and solution accuracy. This research lays a strong foundation for future work involving the application of linear and nonlinear programming techniques in relay protection settings.