| Distribution network reconfiguration is an important measure to improve the operation feature as well as to raise the planning level in a distribution system. It is also one of the most important problems in distribution system automation. Distribution network reconfiguration alters the topological structures of distribution network by changing the open/closed states of the sectionalizing and tie switches. In the way to reduce network loss, insulate faults, displace load etc. It is an effective way of improving network operation structure, and increasing network dependability and security.Because the problem of Distribution network reconfiguration is very complex, a hybrid algorithm that combines genetic algorithm and Alopex has been employed in the paper. And at the same time the inheritance, the variability and the convergence criterions of the GA have also been improved. It avoids the disregard of objective function direction information in the traditional GA by the addition of Alopex perturbation, and improves the hunting optima competence in whole field and climbing competence.The load of distribution network is so variable that it could lead to two problems. First, the variation of load can result in frequent operation on switch for the system of being able to realize real time reconfiguration. This will reduce switch's using life and will increase investment in network. Second, the result of single load mode reconfiguration cannot be optimized when load is variable. So we need to reconfigure network for certain time. In this case, this paper present that reconfiguration of network can be realized in certain time by simulating the variation of load with peak load mode, general load mode and least load mode. This method can optimize distribution network in that time.This paper uses the IEEE16 and the IEEE33 distribution network to prove the rationality of these methods, Case studies are give to show that the methods are right, efficient and practical. |
PDF Full Text Request