[Thesis Statements] Artificial Neural Network based Transformer Protection

1. Neural networks are vital in accordance to transformer protection considering that these are dynamical systems that compute functions that best capture the statistical regularities in data: their study inevitably brings together concepts from dynamical systems theory, computation theory, and statistics.

2. The processing component of a neural networks is an algorithm (or set of differential equations) through activation patterns input to the network are translated into activation patterns that contain the net's output. The computational and dynamical outlooks are likely to address this element most, since the input/output function calculated is of main concern to the computational side, and the dynamics by which it is computed is of essential attention to the dynamical viewpoint.

3. The artificial neural network in accordance to transformer protection consists in: blocking several neural cells by the measure of corresponding damaged areas’ complete disabling; accounting of load redistribution between completely enabled and partially disabled areas, and also the corresponding elements’ threshold values specification, that serving to adequate representation of increasing data massif onto simulated objects’ real operational parameters with the neural model.