An Output Probabilistic Constrained Optimal Control Algorithm and Its Application to AGC System

In actual industrial processes, the state and the output of the system are accompanied by randomness due to the existence of uncertainty. However, the output variables are often easier to measure than the state variables and can intuitively reflect the control effect of the system. Furthermore, the output constraints expressed as the probabilistic form are more in line with actual industry under the influence of uncertainty. An output probabilistic constrained optimal control algorithm based on model algorithm control (MAC) is proposed in this paper. The output probabilistic constraints are transformed into the form of deterministic ones, and then a quadratic programming (QP) problem is constructed by combining the deterministic constraints with the performance index function of MAC. The algorithm is supposed to ensure that the actual output satisfies the probabilistic constraints and is as close as possible to the expected value while the quadratic performance index is minimized. Finally, the algorithm is applied to automatic gauge control (AGC) system in hot strip rolling process in order to verify its correctness and effectiveness.