Development of an Intelligent System for Predicting the Reliability of Electric Motors

Abstract

The uninterrupted operation of electric motors has the greatest impact on the operational reliability of the electrical engineering complex, on which the successful functioning of all the technological equipment of an enterprise, and as a result, the enterprise output, primarily depends. The article describes the development of a method of on-line intelligent diagnostics of electric motors which is based on a combination of statistical analysis of the operation of an electric motor over the period of operation and the current measurement of certain parameters of its operation. Using the analysis of statistical data obtained in production, the sensitivity of the predicted reliability of the selected input parameters of the predictive model was evaluated. The optimization of the number of incoming parameters is carried out using the genetic algorithm method, performs a large number of experiments with various combinations of input data. The article also describes the synthesis of a neural network, is the main part of an intelligent system for predicting the reliability of an electric motor. An intelligent system for predicting the reliability of an electric motor has been developed that will allow real-time detection of defects in electrical equipment and monitoring the condition of machines, thereby predicting the duration of their operation.