Application of a computer simulation technique for optimal control of a nonlinear system

To improve biological system performances and productivity significantly, control, including optimal control, should be implemented online. The dynamics of biological processes are highly nonlinear and key variables are difficult to measure. As an aid to contributing solutions to the twin difficulties of complex dynamics and measurement difficulty, a relatively simple mathematical model has been developed for the purpose of evaluating optimal controllers. A simulation of a fed-batch fermentation process is developed which includes the nonlinear dynamics of the process. In order to solve the state equations for the nonlinear systems, a classical first-order Euler algorithm and a fourth-order Runge-Kutta algorithm for the integration of the nonlinear differential equations were developed. The state models have been expanded to include the various nonlinearities