Three Techniques Based On Artificial Neural Networks for the Identification of Power System Harmonics A comparative Study
Résumé
A new strategy to estimate harmonic distortion from an AC line is presented for power electronic converters. An Adaptive linear neural network (ADALINE) allows the estimation of the necessary currents in order to cancel harmonics. The comparative study is carried out for three different techniques, with all of them giving the same result. The first technique is based on a Matlab implementation of the Adaline algorithm without using the toolbox. On the contrary, the Matlab toolbox is used in the second technique, and finally Simulink is used for the implementation of the third. The speed and accuracy of the ADALINE results in improved performances of the active power line filter. Thus, according to the application it is possible to choose the adequate technique for the harmonics identification.