학술논문

Fuzzy theory-based hybrid decision-making system for rotating electrical machinery fault diagnosis
Document Type
Conference
Source
2016 International Conference on Fuzzy Theory and Its Applications (iFuzzy) Fuzzy Theory and Its Applications (iFuzzy), 2016 International Conference on. :1-6 Nov, 2016
Subject
General Topics for Engineers
Analytical models
Standards
Decision making
Orbits
Stators
Integrated circuit modeling
Language
ISSN
2377-5831
Abstract
This research aims to establish a rotating electrical machinery failure decision-making system based on fuzzy theory and combine with the characteristics of electric and vibration signals to achieve a more comprehensive omen diagnosis of rotating electrical machinery. First of all, voltage and current signals are calculated through pointer to draw into a two-dimensional characteristic diagram so as to find failure characteristics from the relationship between imbalance rate and harmonic distortion, then shaft orbit diagram is mapped through vibration signals and fractal theory is used to extract characteristic parameters of orbit so as to find failure characteristics of mechanical vibration; based on this, fuzzy theory is used to integrate the expert knowledge and a complex failure decision-making system with electrical and vibration analysis technology is designed. This research develops four common failure models and a normal model for experiment and analysis, and it is known from the results that the failure diagnosis performance is not ideal with separate electric or vibration test method, which can only distinguish a few types of failures, but if it is combined with fuzzy failure decision-making system, it will have very excellent diagnosis performance, which can accurately distinguish failure problems of the experimental models to make the rotating electrical machinery have more complete failure prediction ability and it can effectively detect anomaly characteristics in the case of failure symptoms in order to avoid more serious accidents, and it also verifies the feasibility and effectiveness of hybrid failure diagnosis system proposed in this research. In-depth research will be continued in future to explore more physical signals and its analysis methods so as to establish a more reliable rotation machine failure decision-making system to make lots of factories using rotating electrical machinery to have intelligent production capacity, and long-term operation of important units can obtain more reliable security guarantee.