ADRC with synthesis tuning algorithm for superheating steam temperature of CFBB

The cascade controller of active-disturbance-rejection-controller and PI controller is applied to the steam temperature system of circulating fluidized bed, which has the characters of large time-delay, fast time-varying and various disturbances. A synthesis tuning algorithm which combined the intelligent optimal method NSGA_II and numerical optimal method algorithm is adopted to tune the parameters of controller. In the computer simulation, all parameters of ADRC and PI controller is tuned with 100% load. The least square algorithm is also applied to tuning PI controller. Then plant with PI controller is looked as a generalized object and fitting function is computed. For the fitting function, multi-objective algorithm NSGA_II is applied to tuning ESO. Then NLF is tuned with non-linear least square algorithm - Levenberg-Marqudart algorithm. As the comparison algorithm, PI-PI controller is well tuned with least square algorithm. The simulation with load-varying demonstrated more robustness and the simulation with water-feed disturbance demonstrated more disturbance rejection of ADRC-PI cascade controller.