Nodal voltage regulation employing an indirect self-tuning approach

The design of a indirect self-tuning power system nodal voltage regulator is presented. A recursive least-squares (RLS) algorithm with variable forgetting factor estimates at each sampling period the parameters of a suitable discrete-time linear model that describes in an approximated way the system dynamics from the regulation node. The proposed model comprises both an unknown bias representing the no-load voltage and the noise mainly due to voltage signal measurement and commutation in the electronic actuator. The regulator polynomials are designed according to the pole-assignment technique that requires the solution of the Diophantine equation. In presence of power system variations, the polynomial coefficients are on-line update on the basis of the estimated parameters. The performance; of the proposed indirect self-tuning control scheme has been analyzed by an accurate numerical time simulation study in the case of step change in the set-point, unknown load variations and unexpected line opening