Optimal control of switched systems arising in fermentation processes (Q2250285)

In this book, some mathematical models arising in fermentation processes are presented. They are in the form of a nonlinear multistage system, switched autonomous system, time-dependent switched system, state-dependent switched system, multistage time-delay system, or switched time-delay system. On the basis of these dynamical systems, optimization problems including optimal control problems and optimal parameter selection problems are considered. The existence of optimal controls and optimization algorithms for the above mentioned optimization problems are discussed. The book is organized as follows. Chapter 1 briefly introduces the switched system, optimal control and fermentation process, and reviews their literature. Results from measure theory and functional analysis are stated without proofs in Chapter 2. In Chapter 3, some results in constrained mathematical programming are reviewed. Chapter 4 presents a short course in optimal control theory. From Chapter 5 onward, the main attention is focused on the optimal control of switched systems arising in fermentation processes. This starts with optimal control of a nonlinear multistage system, which is a degenerate switched system since the switching law is decided a-priori. Compared with existing systems, the proposed system is much closer to the actual fermentation process. A global optimization algorithm based on the control parameterization concept and the improved particle swarm optimization algorithm is constructed to solve the optimal control problem. Numerical results show that the concentration of target product concentration at the terminal time is increased considerably compared with experimental results. In Chapter 6, a switched autonomous system with variable switching instant to model the constantly fed-batch process is proposed. Taking the switching instants as the control function, an optimal control problem to optimize the fermentation process is formulated. By introducing a time-scaling transform, the optimal control problem is translated into an equivalent one with parameters and fixed switching instants. A computational approach to seek the optimal switching instants is developed. In Chapter 7, a time-dependent switched system, in which the feeding rate is the control function and the switching instants are the optimization variables, is proposed to formulate the fed-batch fermentation process. A constrained optimal control problem involving the time-dependent switched system is then presented. To seek the optimal control and the optimal switching instants, the control parameterization enhancing transform together with the constrained transcription technique is used in order to convert the constrained optimal control problem into a sequence of mathematical programming problems. Numerical results show that the target product concentration at the terminal time can be increased compared with previous results. In Chapter 8, considering the hybrid nature in the fed-batch fermentation process, a state-based switched system to model the fermentation process is proposed. A constrained optimal switching control model is then presented. Because the number of the switchings is not known a-priori, the above optimal control problem is reformulated as a two-level optimization problem. An optimization algorithm is developed to seek the optimal solution on the basis of a heuristic approach and the control parameterization method. In Chapter 9, considering the microbial metabolism mechanism -- i.e. the production of new biomass is delayed by the amount of time it takes to metabolize the nutrients -- in the fed-batch fermentation process, a multistage time-delay system to formulate the process is proposed. In view of the effect of time delay and the high number of kinetic parameters in the system, the parametric sensitivity analysis is used to determine the key parameters. An optimal selection model is presented, and a global optimization method is developed to seek the optimal key parameters. In Chapter 10, taking the mass of target product per unit time as the performance index, a constrained optimal control model with free terminal time to optimize the production process is formulated. Using a time-scale transformation, the optimal control problem is equivalently translated into one with fixed terminal time. A computational approach is then developed to seek the optimal control and the optimal terminal time. In Chapter 11, taking the switching instants and the terminal time as the control variables, a free terminal time-delayed optimal control problem is proposed. Using a time-scaling transformation and parameterizing the switching instants into new parameters, an equivalent optimal control problem is presented. A numerical solution method is developed to seek the optimal control strategy by the smoothing approximation method and the gradient of the cost functional together with that of the constraints. The book is addressed to graduate and postgraduate students as well as to control engineers and researches.