A Complete Neuromimetic Strategy for Harmonics Identification and Control of a Three-Phase Voltage Inverter used for the Power Active Filtering

Abstract

Only based on the use of neural techniques:
Artificial Neural Networks (ANNs), This paper presents a
complete neural strategy for harmonics identification and
control of a three-phase voltage inverter used for the
power active filtering (APF), so our main motivation is to
build a neural APF. This approach of compensation is
done in three neural blocks. In The first one we propose a
new neural approach based on Adalines for the online
extraction of the symmetrical voltage components, i.e.,
Phase-Locked Loop (PLL) based on the Instantaneous
Powers Theory (IPT), to recover a balanced and
equilibrated voltage system. The second block extracts the
harmonic currents with synchronized method by using
Adaline neural networks. The third block injects the
harmonic currents with opposite phase in the electrical
supply network; it uses a PI-neural controller to control
the inverter. To maintain the dc voltage capacitor
constant and compensate the inverter losses a neural
proportional integral voltage controller is used. By their
learning capabilities of ANNs, our approach is
automatically able to adapt itself to any change of the
non-linear and thus appreciably improve the performance
of traditional compensating methods. Furthermore, The
proposed neural compensation approach has been
evaluated in simulations. The results