Data-Driven Arbitrary Polynomial Chaos-Based Probabilistic Load Flow Considering Correlated Uncertainties

Generalized polynomial chaos (gPC) expansion-based probabilistic load flow (PLF) has attracted significant attention due to its accuracy and computational efficiency. However, when dealing with correlated uncertainties, the Nataf transformation is often applied to gPC, which may result in an underestimated probability of extreme events. To overcome this problem, this letter explores the application of data-driven arbitrary polynomial chaos expansion to addressing the dependent uncertainties in PLF. The effectiveness of the proposed method effectiveness is verified through case studies.