In terms of the multi-area optimal power flow (OPF) problem, the optimized objectives are always a fuel cost function expressed by a second-order polynomial. However, the valve-point loading effect, whose cost curve i...In terms of the multi-area optimal power flow (OPF) problem, the optimized objectives are always a fuel cost function expressed by a second-order polynomial. However, the valve-point loading effect, whose cost curve is a transcendental function formed by the superposition of the sine and polynomial function, will make the objective function non-convex and non-differentiable. Conventional distributed optimization technologies can hardly make a solution directly. Therefore, it is necessary to realize a distributed solution for multi-area OPF from another point of view. In this paper, we constitute a new double-layer optimization mechanism. The proposed distributed meta-heuristic optimization (DMHO) algorithm is put on the top layer to optimize the dispatching of each area, and in each iteration a distributed power flow calculation method is embedded as the bottom layer to minimize the mismatch of power balance. Numerical experiments demonstrate that the proposed approach not only implements a multi-area OPF distributed solution but also accelerates the convergence rate, improves the solution accuracy and enhances the robustness. In addition, a fully decentralized computation experiment is performed in an actual distributed environment to test its practicability and computation efficiency.展开更多
基金supported by the National Natural Science Foundation of China(52177087)High-end Foreign Experts Project(G2022163018L)Guangdong Basic and Applied Basic Research Foundation(2024A1515030192).
文摘In terms of the multi-area optimal power flow (OPF) problem, the optimized objectives are always a fuel cost function expressed by a second-order polynomial. However, the valve-point loading effect, whose cost curve is a transcendental function formed by the superposition of the sine and polynomial function, will make the objective function non-convex and non-differentiable. Conventional distributed optimization technologies can hardly make a solution directly. Therefore, it is necessary to realize a distributed solution for multi-area OPF from another point of view. In this paper, we constitute a new double-layer optimization mechanism. The proposed distributed meta-heuristic optimization (DMHO) algorithm is put on the top layer to optimize the dispatching of each area, and in each iteration a distributed power flow calculation method is embedded as the bottom layer to minimize the mismatch of power balance. Numerical experiments demonstrate that the proposed approach not only implements a multi-area OPF distributed solution but also accelerates the convergence rate, improves the solution accuracy and enhances the robustness. In addition, a fully decentralized computation experiment is performed in an actual distributed environment to test its practicability and computation efficiency.
