A new harmonic current forecasting method for HAPF of micro-grid

There are a lot of power quality problems such as a large number of harmonic, caused by quickly changing diverse types of loads and power resources in micro-grid. This paper proposed a hybrid active power filter (HAPF) harmonic current prediction method based on the hybrid intelligent prediction model of Empirical Mode Decomposition (EMD) and Intrinsic Mode Regression (SVR). This method Combine the EMD method and SVR module, First, the harmonic current is decomposed into every harmonic current using EMD method; Then using different kernel function in SVR module to predict different times harmonic currents at the next time; Finally, the predicted value of each harmonic weighted summation. Simulation results show that, this method can be well to improve the prediction performance.©2013 Universitas Ahmad Dahlan. All rights reserved.