Application Of Optimal Control On Both Fermentation Process And Ice Storage Engineering

This dissertation investigates the application of optimal control on both microbial produc-tion of1,3-propanediol and ice storage engineering.Firstly, we consider the fed-batch culture by Klebsiella pneumoniae for production1,3-propanediol, which is controlled by feed-forward control, coupling with alkaline fed-batch. Control strategy based on numerical simulation with pH coupled control is designed to maximum the production. Secondly, uncoupling with alka-line fed-batch.pH value measured automatically with transmitter and logic control is taken into consideration. Finally, on the basis of multistage growing characteristic of bacterium and the concentration of CO2,the process of microbial fermentation is divided into two phases,which are logarithmic growth and stable phase.The optimal control strategy of fed-batch culture is designed with feedback control.The research is supported by the National Natural Science Foundation of China, the Major State Basic Research Development Program of China and the National High Technology Research and Development Program of China. In addition, optimal control strategy about ice storage engineering on air conditioning system is also considered. The optimal control on the ice storage air conditioning system is accord with the national re-quirement of energy saving and emission reduction.In this paper, a nonlinear and non-smooth mixed integer optimization model is built and an optimization algorithm is constructed to seek the optimal solution.Numerical results are implemented on a factual ice storage air condi-tioning system, which has got a good profit.Ice storage air conditioning system produces ice at night and uses it during peak periods for cooling.It is so called"cutting peak and filling trough".This reduces energy usage during the peak periods.So it can balance the load demand of electrical power network in the city. The main contributions obtained in this dissertation are summarized as follows.1.A fed-batch culture with coupled feeds of glycerol and alkali is considered. In order to the factual experiments.On the basis of the numerical solution,which is studied by Chongyang Liu,we propose an optimal control strategy of feeding instants under fixed feeding rate.Correspondingly, an automatic control system is designed according to the process requirement of the glycerol feed-batch culture production of1.3-propanediol, which involves the design of control circuit, flow control software and upper computer integrated control system.2.Based on the microbial production of1,3-propanediol in fed-batch culture with uncou-pling feeds of glycerol and alkali and the optimized numerical result of the dynamic system in microbial process, we explore an automatic control system to implement the optimal control strategy, in which fed-batch of glycerol is controlled with feed-forward control and alkalinity is controlled according to the pH transmitter. The designs of control diagram, PLC control program and supervisory computer control system are involved. Then the feedback control of glycerol pump is discussed.The fed-batch process is di-vided into logarithmic growing phase and stable phase according to the concentration of CO2.And the control strategy of fed-batch culture is divided into two subprograms. The control box, PLC programmed and supervisory computer control system are designed on the existing fermenter in the laboratory. The feedback control program,which was tested and modified for many times,has been running successfully.3.We considered the optimization of an multi-period and multi-equipment ice storage sys-tem.Taking the dynamic process of the ice storage system as constraint, a nonlinear and non-smooth mixed integer optimization model is built, with the running state of equip-ments as discrete optimal variables, the flow rate of fluid in the equipments as continuous optimal variables,and the total system running cost of the system as target function.The existence of solutions to the optimization problem is explored. Due to the finiteness of the admissible set of the discrete variables, the optimization problem is equivalently divided into a finite number of sub-optimization problems,which are of the form of linear pro-gramming with continuous variables to be optimized.An optimization algorithm is then constructed.Numerical results are implemented on an factual ice storage air conditioning system, revealing that both the power consumption at peak and the running cost for user are reduced considerably. So the mathematical model and optimization algorithm is valid.