Improved Genetic Algorithm-based Optimal Power Flow Problem

The paper proposes a genetic algorithm to solve the problem of optimal power flow. To overcome the premature, an improved genetic algorithm is proposed. The improved genetic algorithm introduces a differentiated mechanism of population, and adopts a dynamic adaptive cross, mutation operator. Combined with an improved heuristic crossover method and the chaotic variation degenerated by iteration, the improved genetic algorithm designs a mechanism to make it out of the local optimal. The results show that the improved algorithm avoids the phenomenon of "premature" to some extent, and enhances the convergence. The improved algorithm is better than the standard genetic algorithm and adaptive genetic algorithm in the solution precision and the searching speed.Focused on minimizing the active power loss of power system, the inequality constraints of active power, reactive power and voltage are added to the objective function as a penalty, and then the corresponding fitness function is constructed. In order to meet the equality constraints of basic flow equations, the individuals must calculate the flow, and the individual fitness will be gradually modified by flow calculation. Use standard genetic algorithm, adaptive genetic algorithm and improved genetic algorithm to simulate IEEE14 and IEEE118 node standard test system. Experimental results show that the improved genetic algorithm avoid falling into local optimum in the optimization process. The solution accuracy is greatly enhanced, and the computing time is greatly reduced. Experiments demonstrate the effectiveness of the algorithm.