A novel iterative fuzzy identification via OCA and its application to electrical distribution problem

A novel iterative fuzzy identification via Objective Cluster Analysis is proposed in this paper. The Objective Cluster Analysis algorithm is introduced and enhanced using the relative dissimilarity measure and the new consistency criterion for improving the robustness and the compactness of clustering. Then the Fuzzy c - Means clustering algorithm and the Stable Kalman Filter algorithm are respectively incorporated to identify the premise and the consequence parameters. For making the local fuzzy partitions more satisfying, the iterative fuzzy identification procedure is presented with the covering criterion to acquire the supplementary fuzzy rule prototypes. The developed approach is then applied to a case study of electrical distribution problem for relating the relationship between the village characteristics and the length of low voltage line. The results demonstrate that our method is effective.