Stochastic optimal power flow for hybrid AC/DC grids considering continuous non-Gaussian uncertainty

The integration of renewable energy sources (RES) and the increasing adoption of High Voltage Direct Current (HVDC) transmission are reshaping modern power systems. Whereas RES introduce complex uncertainties in power system operation, often contributing to grid congestion, HVDC technology provides flexibility for effective congestion management. Accurately leveraging this flexibility through optimal scheduling of HVDC converters and minimizing expected RES curtailment requires optimization frameworks that simultaneously account for (i) continuous non-Gaussian uncertainty, (ii) hybrid AC/DC grid compatibility, and (iii) RES curtailment. However, existing Stochastic Optimal Power Flow (SOPF) models do not combine all three dimensions because their interaction significantly increases the nonlinearity. To bridge this gap, this paper introduces a Polynomial Chaos Expansion based chance-constrained SOPF framework that integrates these three dimensions within a single model, paving the way for reliable and cost-efficient hybrid AC/DC grid operation under high RES penetration. The effectiveness of the proposed framework is demonstrated through four case studies on 5-bus, 67-bus, 118-bus, and 588-bus hybrid AC/DC test systems. Results show that by accurately capturing interactions between input uncertainty, HVDC converter set-points, and RES curtailment, the proposed framework minimizes expected RES curtailment and operational costs, leading to significant financial savings under high RES penetration. In addition, the framework is shown to maintain computational scalability on large-scale systems while preserving modeling accuracy under continuous non-Gaussian uncertainty.