Design-based computational procedure for performance prediction and analysis of self-excited induction generators using motor design packages

A computer algorithm is explained in order to determine the performance of capacitor self-excited induction generators (SEIG) taking the machine design details as the input data. The magnetisation characteristic which forms the basis of analysis is obtained using the B/H curve of the core material and by employing a curve-fitting technique. The predicted parameter and performance characteristics of a normal induction motor operating as SEIG are compared with corresponding experimental results. Effects of variation of realistic design parameters are studied to obtain guidelines on optimum design. Typical results on a range of normal induction machines are presented to assess their suitability as SEIG.