Decoupling ARX time-space model based GPC for distributed parameter system

Generalized predictive control (GPC) algorithm is proposed to apply to a distributed axial dispersion chemical reactor. For a class of nonlinear distributed parameter systems described by parabolic partial differential equations (PDEs), Principal Component Analysis (PCA) method is utilized to extract the basis functions of dominant space, and the time-space decomposition is derived in terms of these basis functions to obtain the time domain outputs for system modeling. The input stimulation and time-domain outputs are used to construct the time domain decoupling ARX model, its parameters are identified by the recursive least squares (RLS) algorithm. Based on the decoupling ARX model, the MIMO system can be simplified as multiple SISO system, GPC strategy is then directed used to design the DPS controller. Simulation results show the efficiency of the proposed system modeling and control scheme.