Performance evaluation of GA and RL based PSS of a multi-machine system

摘要

One of the main problems that lead to instability in a power system is the low frequency oscillation caused by swinging generator rotor. These oscillations will lead to loss of synchronism, limit the power transfer capability and also lead to the islanding of inter connected power system. So it is very important to damp out these oscillations, power system stabilizer (PSS) is a device which is commonly used for damp out these oscillations. A power system stabilizer uses angular speed as the input and produce stabilizing signal to damp out these oscillations. The most commonly used power system stabilizer is the conventional power system stabilizer, which is designed by using the theory of phase compensation in the frequency domain and it uses a lead-lag network in its design. This paper focus on the design of PSS for a Single machine connected to infinite bus system (SMIB) and Multi-machine power system having 3 generator and 9 bus system. For optimizing the gain of PSS, Genetic algorithm and Reinforcement learning techniques are used. The simulations are carried out using state space modelling technique.