Abstract:

An artificial intelligent power system stabilizer is designed to damp the low frequency oscillations in the range 0.2 to 2.5 Hz. The low frequency oscillations are introduced in the power system due to disturbances, usage of fast acting regulators and interconnection of different power plants with weak tie lines. The low frequency oscillations are introduced, persist and continue to grow a level of complete shutdown of the system if adequate damping is not provided. The impact of these oscillations is more in multi machine system rather than single machine system. The objective of the Power System Stabilizer is to generate a stabilizing signal, which produces a damping torque component on the generator shaft. In order to damp out oscillations different conventional controllers are used such as lead lag PSS which are slow and can be used for limited operating range. The artificial intelligent techniques have been used to damp the low frequency oscillations which are fast and can be used for all operating conditions. In this project lead- lag, ANN based power system stabilizers are designed and simulated in MATLAB environment. The ANN is trained with Error Back Propagation (EBP) training algorithm. The Artificial neural network which is used here consists one hidden layer. The speed of the generator was taken as input and the output is stabilized value of voltage. It is trained by the various pairs of inputs with respect to different types of system’s disturbances and the performances of the ANN with PSS are compared with the basic system having stabilizer as absent and also with CPSS. The result concluded that the proposed model can assure better damping of the electrical power system over a wide range of operating conditions and consequently improves the system performance dynamically.

Description:

SMIB power system, AVR, PSS, ANN