Multivariable Control for Tracking Optimal Profiles in a Nonlinear Fed-Batch Bioprocess Integrated with State Estimation

PANTANO NADIA; FERNANDEZ MARIA CECILIA; ROSSOMANDO FRANCISCO; OSCAR A. ORTIZ; GUSTAVO SCAGLIA

INDUSTRIAL & ENGINEERING CHEMICAL RESEARCH

AMER CHEMICAL SOC

Año: 2017

0888-5885

This paper aims to solve the problem oftracking optimal profiles for a nonlinear multivariable fedbatchbioprocess by a simple but efficient closed-loop controltechnique based on a linear algebra approach. In the proposedmethodology, the control actions are obtained by solving asystem of linear equations without the need for statetransformations. The optimal profiles to follow are directlythose corresponding to output desired variables, therefore,estimation of states for nonmeasurable variables is consideredby employing a neural networks method. The efficiency of theproposed controller is tested through several simulations, including process disturbances and operation under parametricuncertainty. The optimal controller parameters are selected through the Montecarlo Randomized Algorithm. In addition, proofof convergence to zero of tracking errors is analyzed and included in this article.