Identification of Heffron–Phillips model parameters for synchronous generators operating in closed loop

Abstract

Heffron–Phillips model of a synchronous machine is commonly used in small signal stability analysis and for off-line design of power system stabilisers. The data used to determine the parameters of this model are either hard to measure or require the machine to be taken off-line to take the measurements which, in general, is inconvenient. Identifying these parameters from online data measurements is important since it does not require any a priori knowledge of the machine data. The problem of closed-loop identification of the Heffron–Phillips model parameters is of practical importance since the data used for identification can be gathered when the machine is normally connected to the power system. The use of open-loop identification techniques using data gathered during closed-loop operation of synchronous generators leads to bias errors in the estimated parameters. Motivated by the fact that the synchronous machine model is multivariable and is well defined in a state space structure, a closed-loop subspace parameter identification technique is proposed. Consistency of the proposed approach is illustrated using Monte Carlo analysis. Comparison of the proposed method with open-loop identification technique shows the superiority of this approach.