Performance analysis of stand-alone six-phase induction generator using heuristic algorithms

The paper exhibits the performance analysis of six-phase self-excited induction generator for stand-alone wind energy generation system. The analysis is based essentially on solving the nonlinear equivalent circuit of the SP-SEIG, which is to find the per-unit frequency and the magnetizing reactance minimizing the determinant of the nodal admittance matrix instead of solving two non-linear equations with two unknowns. Hence, the equation-solving problem is converted to an optimization problem. The obtained minimum yields the adequate magnetizing reactance and frequency which will be used subsequently to compute the self-excitation process requirements in terms of the prime mover speed, the excitation capacitance and the load impedance on the one hand and to predict the generator steady state performance parameters on the other. In this work, the analysis is performed using three different global search algorithms, the genetic algorithm (GA), the particle swarm optimization (PSO) technique and the Taguchi optimization method (TM). A study of some simulation results is carried out using MatLab to compare between these three algorithms in terms of accuracy and guaranteed convergence in finding the minimum of the admittance.