Satisfactory optimization method for glutamic acid fermentation process

In this paper, the dynamic model of the glutamic acid fermentation process is established by using the neural network. Combined with satisfaction optimization method, multi-objective and multi-variable optimization for the glutamic acid fermentation process is proceeded with the objects of the production rate of glutamic acid and conversion rate. The paper also set up the corresponding satisfaction function for all the objects, and constructed the satisfactory optimization model for glutamic acid fermentation process. In order to realize the whole process optimization, the trajectories of operation variables in each hour from start to finish and the beginning and end moment are treated as the optimization variables, combined with the real-coding genetic algorithm to proceed the multi-objective optimization. Although the above optimizing methods for two objectives can't guarantee to obtain the process optimal solutions, but can achieve the satisfactory results to decision maker.