Parameter estimation of synchronous machines using particle swarm optimization

Synchronous machines are the most widely used electrical machine in power generation. Identifying the parameters of these machines in a non invasive way is very challenging due to the inherent nonlinearity of machine performance. This paper proposes a synchronous machine parameter identification method using particle swarm optimization (PSO) with a constriction factor. PSO is an intelligent computational method based on a stochastic search that has been shown to be a versatile and efficient tool for complicated engineering problems. A modified version of PSO allows a synchronous machine model output to be used as the objective function, thus allowing a new, more efficient method of parameter identification. This paper highlights the effectiveness of the proposed method for the identification of synchronous machine model parameters, using both simulation and manufacturers measured experimental data.