In practice, load-frequency control (LFC) systems use proportional-integral (PI) controllers that are designed using a linear model. These controllers are incapable to gain good dynamical performance for a wide range of operating conditions especially in deregulated environments. Also the order of robust controllers is as high as the plant. This gives rise to complex structure of such controllers and there is some reluctancy in industry toward the use of high-order and complex controllers. In this paper, a simple intelligent approach based on reinforcement learning (RL) is proposed for the LFC problem in deregulated environments. This method does not depend on any knowledge of the system and it admits considerable flexibility in defining the control objectives. The new intelligent solution performance has been compared with a powerful robust ILMI- based controller. The resulting controller is shown to minimize the effect of disturbances for a wide range of load changes in the presence of system nonlinearities and has good ability to track the contracted and non-contracted demands.
