Nonlinear predictive controller based on S-PARAFAC Volterra models applied to a communicating two tank system
Résumé
This paper proposes a new predictive controller approach for nonlinear process based on a reduced complexity homogeneous, quadratic discretetime Volterra model called quadratic S-PARAFAC Volterra model. The proposed model is yielded by using the symmetry property of the Volterra kernels and their tensor decomposition using the PARAFAC technique which provide a parametric reduction compared to the conventional Volterra model. This property allows synthesizing a new nonlinear model based predictive control (NMBPC). We develop the general form of a new predictor and so, we propose an optimization algorithm formulated as a Quadratic Programming (QP) under linear and nonlinear constraints. The performances of the proposed quadratic S-PARAFAC Volterra model and the developed NMBPC algorithm are illustrated on a numerical simulation and validated on a benchmark as a continuous Stirred Tank Reactor (CSTR) system. Moreover the efficiency of the proposed quadratic SPARAFAC Volterra model and the NMBPC approach are validated on an experimental Communicating Two Tank system (CTTS).