Nonlinear model predictive control of a continuous bioreactor at near-optimum conditions

The modeling and control of a nonlinear bioreactor system utilizing Volterra-Laguerre models is examined. An analytic solution to the single-input single-output (SISO) unconstrained nominal control problem is described, and a nonlinear model predictive controller (MPC) based on the analytic solution is developed which includes dynamic compensation and manipulated variable weighting. This controller avoids the entrapment in local wells seen in gradient descent nonlinear programming solutions. Additionally, extremum control is performed with no a priori knowledge of the system aside from the identified Volterra-Laguerre representation. In the presence of input magnitude constraints, the analytic controller finds the optimal input move in the feasible region. Initial extensions to the mismatch case are also examined.