Parameter Estimation for a Single VSC

This chapter demonstrates a non‐invasive parameter estimation (PE) method, which is able to predict simultaneously the circuit and control parameters of a voltage source converter by means of the derived closed‐loop steady‐state models. An accurate mathematical model (i.e. white‐box model) together with an artificial neural network or deep neural network (i.e. black‐box model) is a viable, time‐saving, and cost‐effective method for PE. The parameter estimator will be validated in experiment and with an offline simulator such as PSCAD/EMTDC. Moreover, the commonly used optimization technique for parameter estimation, particle swarm optimization (PSO), is also implemented to contrast the performance of the presented white‐box‐and‐black‐box model. The deviations between the measured data and the mathematical model can cause the PSO to converge to values quite different from the actual ones.