Prediction and modeling of harmonic current behavior in grid-connected photovoltaic systems based on NARX networks

This research presents the modeling and prediction of the harmonic behavior of current in an electric power supply grid with integration of photovoltaic power by inverters. The methodology was based on the use of recurrent artificial neural networks of the nonlinear autoregressive with external input type. Data were obtained from experimental sources through the use of a test bench, measurement, acquisition and monitoring equipment. The input-output parameters for the neural network were the values of current in the inverter and in the supply grid respectively. Results shown that the neural network can capture the dynamics of the analyzed system. The generated model presented flexibility in data handling, representing and predicting the behavior of the harmonic phenomenon. The obtained algorithm can be transferred to a physical or virtual system, for the control or reduction of harmonic distortion.