Current-Transformer Saturation Detection With Genetically Optimized Neural Networks

Abstract

Application of the genetic algorithm for the optimization of the artificial-neural-network (ANN)-based current-transformer (CT) saturation detector is presented. To determine the most suitable ANN topology for the CT state classifier, the rules of evolutionary improvement of the characteristics of individuals by concurrence and heredity are used. The proposed genetic optimization principles were implemented in MATLAB programming code. The initial as well as further consecutive network populations were created, trained, and graded in a closed loop until the selection criterion was fulfilled. Various aspects of genetic optimization have been studied, including ANN quality assessment, versions of genetic operations, etc. The developed optimized neural CT saturation detectors have been tested with ATP-generated signals, proving better performance than traditionally used algorithms and methods