Probabilistic coordinated design of multi-band power system stabilizers using the unscented transformation

This paper introduces an optimization-based approach for the probabilistic design of multi-band power system stabilizers. The primary aim is to minimize the sum of gains, thereby reducing the control effort. The design constraints focus on ensuring that the probabilities of security and stability, as determined from the perspective of angular small-signal stability, meet the user-specified confidence level. Differential evolution algorithm is employed to solve this optimization challenge, yielding high-quality results within a shortened computational timeframe. To decrease computational demands while preserving accuracy further, the unscented transformation is utilized to conduct the necessary probabilistic small signal stability analysis within the solution process. Application of this approach to a Brazilian test system demonstrates its promising capabilities, particularly in comparison with Monte Carlo simulation under scenarios with load uncertainties.