An Artificial Neural Network Developed in MATLAB-Simulink for Reconstruction a Distorted Secondary Current Waveform. Part 1

Recently, there has been an increased interest in the use of artificial neural networks various branches electric power industry including relay protection. Аrtificial are one fastest growing areas intelligence technology. аrtificial engineering, Existing microprocessor-based protection devices a traditional digital signal processing monitored signals which is reduced to multiplying values successive samples current and voltage by predetermined coefficients order calculate their RMS values. In this case, calculated often do not reflect real processes occurring protected electrical equipment due to, for example, transformer saturation because DC component presence fault current. When saturated, its secondary waveform characteristic non-periodic distorted form, significantly differs from primary (true) waveform, causes underestimation value compared true value. turn, trip time delay or even operation failure. The conjunction with provides different approach functioning both measuring logical parts devices, increases speed reliability such comparison implementation. A possible application purposes occurrence detection type identification, distortion restoration up value, undistorted sections during saturation, abnormal operating modes detection, magnetizing inrush. article describes detail stages practical implementation MATLAB-Simulink environment example restore saturation.