DETERMINATION OF THE TECHNICAL CONDITION OF THE ELECTRIC MOTOR UNDER INCOMPLETE INITIAL DATA CONDITIONS

Abstract

The article analyzes the damageability of electric motors at agricultural processing enterprises in Ukraine. Modern microprocessor systems are being introduced in order to automate the process of determining the technical condition of electric motors. Their software implements complex and informative methods for determining the current state of electric motors based on the analysis of the current values of diagnostic parameters. The current values of not all diagnostic parameters of electric motors are known at the time of determining the technical condition. Therefore, their predicted values are applied. The authors propose to evaluate the current technical condition of electric motors by analyzing the value of the coefficient of its residual life. The values of this coefficient vary from 1 p.p. (the engine is fine) to 0 p.p. (the engine is defective.) An example shows the use of software, namely, Anfis editor of the Fuzzy Logic Tool Box application of the Matlab application package, for creating a mathematical model of the residual life coefficient of an electric motor. It is noted that the error in training the model based on 582 considered variants of combinations of diagnostic parameters and the corresponding values of the residual life of the electric motor does not exceed 0.00288 p.p. (0.2%), and for test voters – 4.1%. Taking into account the large number of mutually influential diagnostic parameters of electric motors, in order to simplify the assessment of the current technical condition, we propose to use the integral diagnostic parameter  the coefficient of residual life. At the same time, it is difficult to determine the technical condition of a working electric motor, because some diagnostic parameters can be measured in a disconnected or disassembled electric motor (for example, measuring the insulation resistance of the stator winding with a megaohmmeter or measuring the diameter of the rotor shaft under the bearing in order to identify the cause of vibration). Therefore, determined at the rate of the process, under such conditions, the technical state of the operating electric motor is predictable  fuzzy. To determine it, one should use the methods and means of neuro-fuzzy modeling. Therefore, the problem of improving the quality of operation of electric motors in agricultural production by improving the quality of their diagnostics is relevant and of great national economic importance.