Possibility of Synchronous Machine Parameters Estimation via On-Line Field Tests using Genetic Algorithm in Bisotun Power Plant

Generator modelling plays an important role in system planning, operation and post-disturbance analysis. Synchronous generator parameter estimation can be performed by two general methods: Standstill and Rotating Time Domain Response (RTDR) method. Standstill testing in the time and frequency domains has gained rapid popularity, Owing to the simplicity of its implementation even on very large machines. It is still sometimes criticized for its excessively low levels of test currents, not to mention that the rotational effects of the coupling axes, the centrifugal force effects on damper windings and the magnetic saturation effects can not be observed during standstill [1,2]. These limitations triggered the emergence of running time-domain response (RTDR) testing, based on small [31 or large signal disturbances around the nominal operating point. The genetic algorithm approach as an estimator is a new approach and in this paper it is used for synchronous machine parameters identification from time-domain response data. The purpose of this work is to present a new approach based on genetic algorithm identification of the synchronous machine parameters from timedomain on-line small-disturbance test, obtained by abruptly changing the reference voltage or torque. For this purpose small and large disturbances are introduced and small disturbances is simulated as a suitable disturbance to determine the dynamic parameters of the synchronous generators in Bisotun Thermal Power Plant. This paper consists of the following parts: In section (2) Bisotun Power Plant synchronous generator characteristics and its applied model is introduced ; section (3) presents RTDR method, Small and large disturbances and introduces a full algorithm for on-line methods, in section (4) the results of simulati on of small disturbances for Bisotun Power Plant generator is shown; section (5) outlines the applied genetic algorithm; the full algorithm of on-line method is simulated in section (6) and finally estimated and simulated results is shown for comparison.

کلیدواژه ها:

Synchronous Generator ، Parameter Estimation ، On-Line Test ، Genetic Algorithm