Genetic estimation of iron losses in strip wound toroidal cores under PWM flux conditions

Recently estimation of iron losses in electrical steels have become very significant for the manufacture of well-designed magnetic devices that are to be subjected to electronically generated voltage excitations such as pulse width-modulated (PWM) voltage excitation. In this study, iron loss components were estimated by means of the genetic algorithms method and electrical steel manufacturers’ data. The results were compared with those obtained by a widely used classical method. The comparison has shown that the genetic algorithms method improved loss estimation with respect to the classical method under both sinusoidal and PWM flux conditions.