Identification and optimal output control of a laboratory power system

The paper describes the design of an optimal controller, based on output feedback, to provide integrated control of the exciter and prime mover of micro-synchronous-generator. Modelling of the control input-output measurement dynamics of the power system by an output prediction equation using least-squares analysis, is followed by multivariable control-system design based on discrete dynamic programmiing to minimise a quadratic performance index. This approach avoids any reference to the system state vector, which may contain states which are difficult to measure, and thus any need for state estimation techniques or the omission of unmeasurable states from the state feedback control law. The paper evaluates the effectiveness of this controller over a wide range of operating conditions, including dynamic and transient stability boundaries and 3-phase faults, with and without permanent line outage.