Nonlinear predictive control based on a global model identified off-line

Proceedings of the American Control Conference - Tập 5 - Trang 4197-4202 vol.5 - 2002
H. Peng1,2, T. Ozaki3,4, Y. Toyoda5, V. Haggan-Ozaki6
1Currently a visiting researcher at the Institute of Statistical Mathematics, Tokyo, Japan
2College of Information Engineering, Central South University, Changsha, China
3Institute for Statistical Mathematics, Minato, Tokyo, Japan
4The Institute of Statistical Mathematics, Tokyo, Japan
5Bailey Japan Co. Ltd., Shizuoka, Japan
6Sophia University, Tokyo, Japan

Tóm tắt

A model predictive control (MPC) strategy for the non-stationary nonlinear systems with operating point-dependent dynamics is presented. The MPC proposed does not require on-line parameters estimation, because its internal model is an off-line identified global (RBF-ARX.) model, which is a generalized ARX model with Gaussian radial basis function networks-based functional coefficients. The RBF-ARX model parameters are estimated using a quickly-convergent structured nonlinear parameter optimization method (SNPOM). The quadratic programming routines may be used to solve the MPC problem with constraints. Simulation study on a chemical process shows satisfactory modeling and control performance.

Từ khóa

#Predictive control #Predictive models #Parameter estimation #Quadratic programming #Nonlinear systems #Neural networks #Nonlinear dynamical systems #Mathematics #Sampling methods #Time varying systems

Tài liệu tham khảo

10.1016/B978-0-08-042375-3.50034-2 10.1016/0005-1098(87)90087-2 10.1080/00207170010014061 peng, 2001, Nonlinear system identification using radial basis function-based signal dependent ARX model, Proceedings of 5th IFAC Symposium on Nonlinear Control Systems St Petersburg Russia 703-708 10.1016/S0967-0661(98)00127-0 10.1080/002071798221858 10.1080/002071798221515 10.1111/j.1467-9892.1980.tb00300.x