The Theory of Fuzzy Sets as a Means of Assessing the Periods of Service of Asynchronous Electric Motors

Abstract

In modern fuel and energy complexes, the main part of the machinery is represented by induction motors. The operating conditions and design features of engines often impede assessing their technical condition. The most common electrical faults are phase loss, intra-winding short circuits, winding short circuits to the housing, inter-winding short circuits, open rotor rods, open windings or in an external circuit. In addition to the electrical part damage, mechanical damage is also distinguished. It includes bearing damage, poor engine mounts, dynamic and static eccentricity. The paper suggest using the theory of fuzzy sets for the objective assessment of the state of an asynchronous machine. The description of objects and phenomena by fuzzy sets requires the use of linguistic variables. They are described by means of fuzzy sets in a certain domain of definition and by the name of variables. Linguistic assessment uses not only quantitative, but also qualitative data. The material obtained as a result of quantitative assessment methods needs to be used, passing it through some transformations. Information on the probability of an emergency situation can be presented in the form of a range of change, represented as a fuzzy set. As input data, this method uses data recognized by an artificial neural network, which is obtained on the basis of data on amplitudes of odd harmonics from 3-11 inclusively generated by the motor. As a result of the experiments, the range of permissible values of linguistic variables is found. It is found that the change in the harmonic composition of currents and voltages of the motor during damage of various kinds is non-linear. Using the Matlab software, the input data is processed. The actions of the algorithm lead to a quantitative assessment of the technical condition of the electric motor, provide its linguistic description.