MVC architecture based neuro-fuzzy approach for distribution feeder reconfiguration for loss reduction and load balancing

This paper presents a feeder reconfiguration algorithm for the line loss reduction and feeder load balancing with minimum consumption of time. The proposed algorithm efficiently utilizes a heuristic based fuzzy strategy and constrained fuzzy operation along with back propagation neural network. This approach reduces the computation cost making it suitable for online application. A layered MVC architecture with strict top-down dependency is proposed to decrease software couplings. A new network configuration is obtained through the proposed algorithm, which line achieves loss reduction and feeder load balance at the same time. The effectiveness of the proposed approach is demonstrated by employing the feeder switching operation scheme to a distribution system. The desired switching operations can be fulfilled in a very efficient manner as indicated from the results.