Active filter for power quality improvement by artificial neural networks technique

Nonlinear loads inject harmonic currents into a power system. Active filtering of electric power has become a mature technology for harmonic and reactive power compensation in two-wire (single phase), three-wire (three phase without neutral), and four-wire (three phase with neutral) ac power networks with nonlinear loads. Current harmonics are one of the most common power quality problems and are usually resolved by the use of shunt passive or active filters. A new control design using an artificial neural network (ANN) is proposed to make a conventional shunt active adaptive filter. The control design is based on ANN that use a modified Widrow-Hoff weight-updating algorithm. This ANN algorithm enhances the functionality of the shunt active filter. The proposed filter can compensate for balanced and unbalanced nonlinear load currents, adapt itself to compensate for variations in nonlinear load currents or nonlinear load types, and correct the power factor of the supply side to near unity. The integrity of the proposed scheme is fulfilled through simulation studies and the results obtained are discussed. The superiority of its performance is proved by comparison to other active filter approaches. The results show an appreciable improvement in the source current waveforms and a near unity power factor operation is achieved. The active power filter control system exhibits robustness and a high dynamic response.