Neuro-fuzzy approach to identification of electromagnetic fields of electrostatic discharge

This paper proposes and shows the application of the method of neuro-fuzzy inference with Gaussian membership functions for the identification of electrostatic discharge (ESD) electromagnetic fields in order to improve the electromagnetic safety of electronic and electrical devices used in the electric power industry and electrical engineering. The results of adaptive neuro-fuzzy inference system (ANFIS) modeling are presented, showing the effectiveness of the proposed method, and a comparative analysis with the traditional finite element method (FEM) is given. A feature of the evaluation of electromagnetic fields and currents by the method of neurofuzzy logical inference is the possibility of using not only numerical but also linguistic data, as well as solutions provided by neural networks, instantaneous, consume less memory, and have a compact structure compared to the FEM.