The article is devoted to the assessment of the influence of magnetic interference, in the study of thermal fluctuation processes in the dynamic current load mode of a power cable line (SCL). On the basis of such artificial intelligence methods as neural networks and fuzzy logic, the thermal resistance of SCL insulating materials determining the throughput of the cable line of electric power systems was investigated. A comparative review of the currently existing traditional non-destructive methods for predicting thermal processes in SCR showed that most of the methods have a low prediction accuracy, as well as have a high degree of complexity and a large number of necessary computational operations to obtain the necessary data for predicting thermal processes in SCR. Also, most forecasting methods are not able to work in real time, which is an extremely significant drawback. To solve this problem, it is necessary to resort to forecasting systems that are based on artificial intelligence using machine learning methods. The method of artificial neural networks (ANN) seems to be the most promising today. The need to develop a more perfect method for analyzing the aging of SCR insulating materials is shown. The urgency of the problem of creating neural networks (NN) for assessing the throughput, calculating and predicting the temperature of SCL cores in real time based on the data of the temperature monitoring system, taking into account the change in the current load of the line and the external conditions of heat removal, has been substantiated. A neural network has been developed to determine the temperature regime of the current-carrying conductor of a power cable. A comparative analysis of the experimental and calculated characteristics of temperature distributions was carried out, while various load operating modes and functions of changing the cable current were investigated. A neural network model was developed in Matlab Simulink for predicting the temperature of a cable core. The creation, training and modeling of the neural network was carried out using the Neural Network Toolbox. The model can be used in devices and systems for continuous diagnostics of power cables by temperature conditions.
