DYNAMIC RECONFIGURATION OF DISTRIBUTION NETWORK WITH DISTRIBUTED GENERATION BASED ON CLUSTERING

摘要

With the increasing proportion of distributed generation (DG) are supplied to distribution network, the traditional distribution network reconfiguration method and model can not meet the actual needs. Taking into account the impact of distributed generation and load change on distribution network reconfiguration, a clustering technique based on load pattern index dimensions reduction is applied to the load clustering of distribution network, the particle swarm optimization based on spanning tree is proposed, which is combined with the improved adaptive step size glowworm swarm optimization to avoid the generation of infeasible solutions, an objective optimization mode which regards network loss and load balancing as the objective functions was established, which is applied to the problem of distribution network reconfiguration with distributed generation after load clustering. The results show that the accurate classification of distribution network based on the improved clustering algorithm provides a good foundation for distribution network reconfiguration. The proposed algorithm is effective and efficient for dynamic reconfiguration of distribution network.