Neural Network Technology for Technical Assessment States of Power Transformers

This article describes the development and testing of a methodology for creating an artificial neural network for assessing the technical condition of power transformers and its application to improve the efficiency of a hydroelectric power station. Based on the analysis of scientific and technical documentation, a sample was developed with scores for each parameter of the power transformer components. The developed technique based on an artificial neural network was tested using the example of finding the most significant parameters of a power transformer. It has been established that the results of neural network modeling are fully consistent with the operational results. The prospects of using neural network technologies to assess the technical condition of power transformers are shown.