Research On The Design And Application Of Intuitionistic Fuzzy Neural Network

Intuitionistic fuzzy sets are one of the most influential expansion and development for the Zadeh fuzzy set theory. It considers three parameters at a time: degree of membership, non-membership and hesitation. Thus, it is more flexible and practical than the traditional fuzzy sets in dealing with fuzziness, uncertainty and so on. Combing the learning ability of artificial neural network with the uncertainty reasoning ability of intuitionistic fuzzy logic, a intuitionistic fuzzy neural network can be built. Power transformer is the main equipment of electric power system, and it has important practical significance to research on the technology of power transformer fault diagnosis. However, to combine various kinds of intelligent methods to power transformer faults diagnosis has become an inevitable trend.Based on the above, this paper studies the design of intuitionistic fuzzy neural network and its application to power transformer fault diagnosis. Firstly, it introduces the fuzzy neural network technology and intuitionistic fuzzy set theory and analyzes the structure and learning algorithm of fuzzy neural network based on T-S model. On this base, it realizes the design of the intuitionistic fuzzy neural network through the form of intuitive synthesis, which is also tested with UCI standard data set. Then, the paper introduces the DGA technology of power transformer fault diagnosis and selects fault data sample. In view of this, a fault diagnosis system based on intuitionistic fuzzy neural network is established, which shows the specific implementation process and data results. Finally, the results of this system are compared with the traditional fuzzy neural network, proving that the design of this intuitionistic fuzzy neural network for fault diagnosis is feasible, and the diagnostic accuracy is higher than traditional fuzzy neural network and ANFIS (adaptive neural fuzzy inference system).