НЕЙРОКОМПЬЮТЕРНОЕ УПРАВЛЕНИЕ ПРОПУСКНОЙ СПОСОБНОСТЬЮ КАБЕЛЬНЫХ СЕТЕЙ ПОСРЕДСТВОМ УЧЕТА И КОНТРОЛЯ ИХ ПАРАМЕТРОВ

Abstract

The article discusses a neurocomputer system for predicting the resource of a power cableline (РCL) using neural network technologies. A hardware modular implementation of aneurocomputer (NC) implemented on the basis of FPGA was selected. To solve the problem ofpredicting thermal processes of РCL, it was decided to use a NeuroMatrix NM6404 digitalneurochip with a variable structure due to their high performance compared to power consumption,a high degree of versatility. To predict the temperature conditions of the РCL, an artificialneural network (INS) was developed to determine the current temperature regime for the currentcarryingcore of the РCL. The architecture of the INS for the implementation of the NC of the SCLtemperature prediction system has been selected, which allows for long-term prediction of РCLtemperatures in real time. The choice of the activation function of the INS neurons for the implementationof the NC of the SCL temperature prediction system, which allows for a long-term forecastof SCL temperatures without increasing the error with an increase in the forecast range. Theproposed neural network algorithm that predicts the characteristics of the electrical insulation ofthe РCL, based on the sliding window method for predicting time series, was tested on a controlsample of experimental data not included in the sample for training the INS. Experimental studiesof the proposed adaptive forecasting method have been carried out, namely, an adaptive algorithmhas been developed and the prediction of thermal processes in the isolation of the SCL from theload current has been performed. Analysis of the results showed that the longer the aging time, thegreater the temperature difference between the original and aged sample. When analyzing thedata obtained, it was determined that the maximum deviation of the data obtained from the INSduring the experiment from the data in the training sample was less than 3%, which is quite acceptablefor this study result. It is shown that the developed methods and algorithms are elementsof an integrated power grid management system, and the developed adaptive NC model makes itpossible to assess the current state of insulation and predict the remaining life of the РCL.