MODEL ORDER REDUCTION OF ELECTRICAL POWER SYSTEM

Abstract

In this report the implication of model order reduction of electrical power systems is studied. With the availability of fast digital computers there as extensive research work in the area of large scale power system. These systems can be described by large number of differential equations that lead to state variable or transfer function. The complexity of modem power systems makes in many cases their exact analysis, if not a formidable, certainly a non-desirable task, due to economic and computationally difficult considerations. Thus a need exists for a systematic procedure to derive a reduced order model in the state space or transfer function form, which should provide a good approximation of the original system response to particular inputs and also may be applied to the actual design work. Choosing particular reduction method depends upon the application of the system used for controller design of any type excitation or turbine governor system of synchronous machine, while designing any controller need variables according to controller specifications we choose any one of these methods by considering nature and time domain parameters of reduced order waveform which is desirable for controller design. The model order reduction software package was written using MATLAB as the programme platform. For validation purposes, this procedure is applied to dynamic model of electrical power systems. The results of the simulations show that the reduced order responses are close enough to the original system's responses in the event when step response is applied.