Practical implementation of a three-phase active power line conditioner with ANNs technology

In this paper, a new method to control an active power line conditioner (APLC) using neural networks is presented. Actually, there is an increase of voltage and current harmonics in power systems, caused by nonlinear loads. The APLCs are used to compensate the generated harmonics and to correct the load power factor. The proposed control design is a pulse width modulation control (PWM) with two blocks that include neural networks. Adaptive networks estimate the reference compensation currents. On the other hand, a multilayer feedforward network (trained by a backpropagation algorithm) works as hysteresis band comparator. An experimental prototype was built to check the proposed control performance. The practical results confirm the possibility and usefulness of controlling an APLC by means of artificial neural networks.

I. Introduction

Frequently, ac electrical power systems include several kinds of non-linear loads (rectifiers, inverters, ac regulators,.). Thus, load currents and voltages are non-sinusoidals, and it's necessary to compensate the voltage and current harmonics. In this case, it's possible to use Active Power Line Conditioners (APLCs). The use of shunt APLCs is a method developed to suppress current harmonics and to correct power factor, especially in fast-fluctuating loads, [1]–[3]. The target is to get balanced and sinusoidal source currents by the injection of compensation currents.