Decentralized optimum power flow using evolutionary multi-objective evolutionary algorithms

This work presents the development of a computational tool for decentralized optimal power flow (OPF) solution. For this purpose, the OPF problem is decoupled into areas defining several regional OPF subproblems. The OPF is modeled as a constrained nonlinear optimization problem, non-convex, in that the active power losses and optimal dispatch of active and reactive power are minimized simultaneously. Regional OPF subproblems are solved by multiobjective evolutionary algorithm based on the Pareto theory. The proposed approach employs a diversity-preserving mechanism to overcome the premature convergence of algorithm and local optimal solutions. Fuzzy set theory is employed to extract the best compromises of the Pareto set. In addition, a hierarchical clustering algorithm is implemented for reducing Pareto set. To validate the efficiency of the model and the proposed solution technique, the results e analyses of the simulations with the RTS-96 e IEEE-354 test systems are presented.