J.R. Vázquez, P. Salmerón, F.J. Alcántara, and J. Prieto (Spain)
Power Factor Compensation and Conditioning, Power Quality, Harmonics, Adaptive control, Neural network.
In this paper, a new method to control an Active Power
Line Conditioner (APLC) using Artificial Neural
Networks (ANNs) is presented.. The APLCs are used to
compensate the generated harmonics and to correct the
load power factor in power systems with non-linear loads.
The Pulse Width Modulation control (PWM) proposed is
designed with two Neural Networks blocks. On the one
hand, adaptive networks estimate the reference
compensation currents. On the other hand, a multilayer
feedforward network works as hysteresis band
comparator. A practical case with Matlab-Simulink is
presented to check the proposed control performance.
Besides, an experimental prototype was built to test full
practical system. The results confirm the possibility and
usefulness of controlling an APLC by means of ANNs.