Privacy-Preserving Distributed Probabilistic Load Flow

In a multi-regional interconnected grid, the probabilistic load flow (PLF) of any region cannot be calculated individually but should consider uncertainties introduced in other areas. Accordingly, topologies, loads, and generations every are needed. Although renewable generation data could assumed as publicly known, some regional independent system operators (ISOs) still would not share important parameters with others. This motivates development privacy-preserving distributed (PPD) PLF method. The challenge is to identify mapping between flows uncertain power injections across regions without full information about entire grid. main idea this paper respectively calculate coefficient matrix constant vector mapping: for former, PPD accelerated projection-based consensus algorithm proposed; latter, average leveraged. Consequently, method derived each ISO analytically obtain its joint way sharing - key contribution paper. Experiments on 118- 1354-bus systems demonstrate that can generate same results corresponding centralized method, has satisfactory accuracy compared frequently used methods.