A Method of Remote Fault Diagnosis Based on Region Synchronous Enlargement of Self-Organizing Feature Map

This paper proposes a novel Self-Organizing Feature Map (SOM) model for the Remote Fault Diagnosis. The proposed method is based on Region Synchronous Enlargement (RSE). The input vector in the proposed model can be processed automatically. By using the synchronous enlargement method, only the position of the regions is needed to find central seeds. Then synchronously enlarges these seeds to find the boundary point between them. Furthermore, the method is easy to be extended to n-dimensional system. The proposed method has been successfully evaluated using twelve different datasets, and had greatly improved the rate of correct classification.