2009 · 26 citations · 3 references
ArchitectureComputational EngineeringGenetic EngineeringOptimizationReconfigurable ComputingCombinatorial OptimizationCrossover OperatorsEvolutionary ComputationEvolutionary OptimizationGenetic AlgorithmsPopulation-based SearchDistribution Feeder ReconfigurationDistribution Network ReconfigurationEfficiencyGenetic Algorithm OptimizationGenetic Algorithm Parameter TuningNetwork ReconfigurationPopulation-based OptimizationReconfiguration
An improved genetic algorithm with infeasible solution disposing of distribution network reconfiguration is proposed in this paper. Decimal encoding is improved and size of chromosome equals to the count of tie switches. Not only the discrimination of infeasible solution but also its corrective disposition under complex loops is given. In the evolution course, the ratio of feasible solution is 1, the valuable information that reflects the evolution is preserved mostly, and the problem of low efficiency and local optimum is avoided. Make the crossover and mutual operator adjusted adaptively and nonlinearly, enhances the convergence rate. The reconfiguration results of 69 bus system shows the feasibility and validity of the proposed method.
3