V. Ya. Bespalov , V. L. Maximkin, A. V. Antonenkov

The process of any electric motor-s work can be described by a system of differential equations. The way to solve a simple system with constant parameters is not difficult. But often some of these parameters are changed in random way. The main idea of our study is to find the method that allows to solve effectively the system of differential equations. An induction motor (IM) operations with random load was taken as a subject of research. The main problem here is a difficulty to study the thermal state of IM. Nowadays one of the up-and-coming technologies is artificial neural networks (ANNs) that have already been successfully applied in many areas, in particularly, in the field of electromechanics. Neural networks, with their remarkable ability to derive meaning from complicated or uncertain data, can be used to extract patterns and identify trends that are too complex to be described by computer techniques. A trained neural network can be thought of as an "expert" in the category of information it has been given to analyze. The ability of ANNs to learn by example makes them very flexible and powerful tool. Furthermore, there is no need to devise an algorithm in order to solve a specific task; i.e. there is no need to understand the internal mechanisms of that task. They are also very well suited for real time systems because of their fast response and computational times which are due to their parallel architecture. Feed-forward ANNs is the best architecture for our task that allow signals to travel one way only - from input to output. There is no feedback loops, i.e. the output of any layer does not affect the same layer. Feed-forward ANNs tend to be straight forward networks that associate inputs with outputs. They are extensively used in pattern recognition technique. Conducted research allows us to say that developed mathematical model based on ANNs gives satisfactory results. The possibility of using ANNs for the problems of electromechanics was demonstrated. The application of ANN is of interest as from the view point of good results and of using up-to-dated scientific achievements as well.