Induction generator model parameter estimation using improved particle swarm optimization and on-line response to a change in frequency

An induction generator (IG) is preferred to a synchronous generator in many renewable energy applications. In order to achieve proper control of an induction generator it is important to have accurate knowledge of its model parameters. In this paper, an Improved Particle Swarm Optimization (IPSO) approach is used to estimate the model parameters of an IG. The IPSO is executed based on the response of the active and reactive power flows associated with the IG to a change in the frequency of the external system, which the IG is connected to. This change in frequency is applied when the IG is operating in steady state, to represent the scenario where the IG parameters must be estimated on-line, and during a large disturbance to the system equilibrium. This approach is in contrast to others in the literature that estimate the parameters of an induction machine based on its start-up behavior, or the results of mechanical tests. Therefore, this approach should offer benefits when the parameters of the IG being modeled may vary over time and need to be estimated on-line.