Relax control is to relieve the control of the plants while CPS compliances are ensured. This paper describes an application of the Q-learning algorithm in Automatic Generation Control (AGC) to achieve relax control. As the Q-learning algorithm always pursuits the maximum reward in long term,the number of pulse reversals,the value of CPS,and the change of the power outputs are introduced as the control variables in the reward function of the Q-learning controller. To get the maximum long-term reward,Q-learning controller will try to reduce the number of pulse reversals,to ensure the CPS compliances,and to decrease the change of the power outputs. When the coefficients of the control variables are suitable,the CPS compliances are ensured,the number of pulse reversals are reduced，and the power outputs are kept changing smoothly. Cases show that the proposed controllers can obviously enhance the performance of relax control of AGC systems while the CPS compliances are ensured.
展开▼
