Application Of The Cascade Generalized Predictive Control In The Thermal Processes

Generalized predictive control (GPC) is widely used in process control due to its low requirement on the accuracy of system models, strong robustness, and ease to implement on computers. This thesis proposes a cascade generalized predictive control algorithm based on a thorough analysis of GPC. The inner and outer controllers of a cascade system are replaced by one cascade generalized predictive controller. The method is applied to the control of main steam temperature, which is known to have large time constant and delay. Simulations show that the method can achieve good control performance.With the development of image processing technology, the feedback signal of radiant energy which could reflect the status of the furnace combustion is used in the boiler-turbine coordinated control system (CCS) to decrease the time constant. This thesis proposes a multivariable cascade generalized predictive decentralized control algorithm, and applies it in the coordinated control system. Simulations results show that the algorithm can improve the system stability and robustness, and achieve good control performance.