Definition of asynchronous motor defects based on fuzzy logic

Abstract

Relevance: there is an extension of the fuzzy inference approach for troubleshooting asynchronous machines  that are being tested during their mass production. A comparative study of the effectiveness of the three approaches based on simulation is also presented

Aim: a method of supporting the process by making decisions based on the qualification parameters of asynchronous machines, the application of functional dependencies and a self-learning approach to configure the parameters of the decision-making model is discussed. For comparison purposes, the results of the developed fuzzy diagnostic system and the traditional approach are also presented.

Methods: an appropriate approach to asynchronous machines fault diagnosis using fuzzy logic involves performing several tasks: converting measurement results (current and power in idle and short-circuit modes) from the traditional domain to the fuzzy one; data processing and decision making; and reverse transformation from the fuzzy state to the non-fuzzy one. To solve this problem, we have developed fuzzy relations between two groups of electric motor parameters. The first group includes technological factors and measured electrical parameters. The other group consists of product quality parameters.

Results: this approach allows you to determine the design and technological factors (dimensions, material) that influenced the deviations of the asynchronous machines output data.