Implicit reference model and optimal aim strategy for electrical generator adaptive control

Abstract

The "steady state stability" of the French power transmission system has been greatly improved by supplying additional signals to the voltage regulator. As usual, the control system parameters, including additional signals, have been chosen in order to cover all the possible operating conditions and network configurations. These additional signals lead to an increase in the "steady state stability" domain even with a regulator having constant parameters. Better results have been obtained using several different adaptive control methods. The first method which has been studied and implemented by analog hardware was based on the hyperstability approach and introduced new concepts as series and partial reference models.In order to master completely the time constants of the closed-loop system, a second adaptive control method using a predictor and an implicit reference model has been studied and implemented numerically by micro-processors. At last, a simpler and more robust new method has been introduced : the reference model optimal aim strategy using a very simple predictor.