Research on Distribution Network Optimization using Improved Ant Colony Algorithm

The planning problem of electrical distribution, stated as a mixed, large scale and nonlinear optimization problem, is important in power planning. Ant colony algorithm as a new general-purpose meta-heuristic one has a good application to distribution network optimization. But the original ant colony algorithm has problems such as training inefficiency and easily falling into the local optimal minimum. To solve these problems, this paper presents improved ant colony optimization from transition probability and pheromone trail update mechanism. The transition probability control over two linkage parameters. The pheromone trail update mechanism design adaptive controls the relative importance of trail while the changes are increased. It realizes searching to the whole feasible region completely and easily falling into the local optimal minimum. This algorithm greatly improves the ability that search for the optimum result, and is favorably compared to other solution approaches such as genetic algorithms (GAs) and simulated annealing (SA) techniques. The application to the distribution network planning shows that this method is feasible and efficient