A Combined GA-ANN Strategy for Solving Optimal Power Flow with Voltage Security Constraint

This paper presents an strategy approach to solve the optimal power flow (OPF) problem for reactive power dispatch which generally requires many power flow calculations. Artificial neural networks are employed to learn in an offline mode and substitute the role of power flow in the OPF which is formulated as a mix integer nonlinear optimization with network loss minimization as the objective. This strategy is shown later in this paper that it helps improve the computational efficiency while slightly deteriorating the quality of solution. Simulation results reveal that the proposed method can speedup the computing procedure for 5 time faster than the conventional OPF while sacrificing a little accuracy. The line (L) indicator is taken into account as the constraint to ensure feasibility of optimal control variables in terms of voltage security margin. Genetic algorithm (GA) is employed as the optimization tool. The effectiveness of the method is verified on IEEE 30-bus system and compared with the conventional OPF solution where power flow is used.