Nonlinear Finite-Dimensional Model Predictive Control of the Burgers Equation

L. Bai and D. Coca (UK)


Distributed parameter systems, nonlinear predictivecontrol, systems identification, partial differentialequations, Burgers equation.


This paper proposes a method for synthesizing reduced order nonlinear model-based predictive controllers for the Burgers equation. The predictive control law is computed based on nonlinear multi-step-ahead finite-element predictors, identified directly from experimental input output data. This means that in practice, this approach can be used even when an accurate partial differential equation (PDE) model of the process is not available. The proposed approach avoids time consuming numerical optimization algorithms associated with most common nonlinear predictive control strategies, which makes it suitable for real-time implementation. The design method can deal effectively with load disturbances and noise in a similar manner to that adopted in the classical Generalized Predictive Control framework. The method is used to synthesize and test in numerical simulations nonlinear predictive controllers for the one-dimensional Burgers equation.

Important Links:

Go Back