Self-adaptability parallel active filter to load variations study

Due to the increasing use of power electronics equipment, the quality of energy in the electrical network is deteriorating sharply. Indeed, this equipment consumes non-sinusoidal currents, even if they are powered by a sinusoidal voltage, and behaves like generators of harmonic currents. To "clean up" the network, there are several means, among them; the parallel active filter (PAF) which is the most used means because it has proven its effectiveness for the elimination of harmonics caused by a non-linear load. However, this PAF must prove its effectiveness constantly, because the unexpected variation of the loads supplied by the electrical network can call into question its effectiveness. The main objective of this work is to confirming that PAF is a permanent filter and adapts automatically and instantaneously to any load. For this, several simulation tests were carried out under MATLAB/Simulink. The results obtained are very satisfactory, as they confirmed the robustness and self-adaptability of the filter with respect to the sudden variation in loads, the tests carried out on the PAF also show us that the response of the filter is done almost in real time.