Passive filter design using genetic algorthims for adjustable speed drives

A new approach to designing improved broadband passive harmonic filters (IBFs) for such applications as direct on-line, adjustable-speed drives is proposed, which employs genetic algorithms (GAs). The method is based on frequency domain modeling of the system. The success of the method requires accurate representation of the load harmonics. By using the proposed GAs program, it becomes possible to quickly find the optimized power quality indices that will minimize filter size and harmonic resonance problems. The paper illustrates the potential performance of optimal IBFs using computer simulation to validate GA-assisted filter designs for 5.5 kW and 55 kW ASDs.