Multifactor Model for Forecasting Thermal Processes in the Insulating Materials of Cable Lines

The connection of exceeding permissible temperature value of the cable, the process of intense insulation thermal wear and the reliability of power equipment is shown. Also it is shown that monitoring the insulation heating temperature values allows to obtain a reasonable estimate of power cables state of various voltage classes. A sequential two-network architecture of a neural network with a sequentially distributed structure is synthesized to determine the temperature regime of a power cable current-carrying core. A mathematical description of the neural network structural model to forecast the cable line residual life is carried out. A comparative analysis of the experimental and calculated characteristics of temperature distributions is carried out. The scope of the developed neural network for the temperature condition of a current-carrying core determining is shown, which consists in diagnosing and forecasting the electrical cable insulation resource.