DGA fuzzy logic diagnostic method based on subordinating function

• Published in: 2020 IEEE 5th Information Technology and Mechatronics Engineering Conference (ITOEC) pp. 1381 - 1384
• Main Authors: Wang, Xuelei, Guo, Fangfang, Xu, Wei
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.06.2020
• Subjects: Cognition; Encoding; Fault diagnosis; Fuzzy logic; fuzzy multivalued decision; IEC; incidence matrix; subordinating function; Uncertainty
• DOI: 10.1109/ITOEC49072.2020.9141578

Due to the differences in the diagnostic characteristics and criteria of various methods, the conclusions are often inconsistent when using these methods to diagnose faults. when transformer faults occur, it is often a combination of multiple fault types, and the correspondence between fault types and fault symptoms is fuzzy. In order to describe the fuzzy correspondence in the complex fault diagnosis, the fuzzy mathematical theory is further introduced on the basis of the information fusion diagnosis method, and a comprehensive diagnosis method based on the information fusion and fuzzy logic is proposed. Among them, semi-cauchy lifting function was used as the subordinating function to replace the original binary function for boundary fuzzification of the criterion coding boundary. Furthermore, the coding criteria are expressed as subordinating function to form the subordinating correlation matrix. For diagnosis logic reasoning, "intersection" and "union" in fuzzy logic reasoning are respectively used to replace "and" and "or logic in binary logic to form multi-value output of fuzzy decision. The diagnosis results represent the probability of occurrence of different fault types.