论文部分内容阅读
遗传算法是一种借鉴生物界自然选择和进化机制发展起来的高度并行、随机、自适应的搜索方法。MATLAB通用遗传算法工具箱GAOT本着使用群体搜索技术,将种群代表一组问题的解,通过对当前种群施加选择、交叉和变异等一系列遗传操作,从而得到新的一代种群,并逐步使种群进化到包含近似最优解状态的原则,对苯热裂解脱氢反应及产赖氨酸分批发酵动力学模型参数进行了估算。与文献中已有的结果相比,模型计算值与实验值的吻合程度相似或更优。所用方法也可用于以微分方程组为数学模型的这类过程的参数估算或寻优,从而研究者可以更加集中注意力于深入解决与数学模型本身有关的问题。
Genetic algorithm is a highly parallel, random and adaptive search method developed from the natural selection and evolutionary mechanism in the biological world. MATLAB Universal Genetic Algorithm Toolbox GAOT uses a population search technique to solve a set of problems by representing the population as a set of genetic operations by applying selection, crossover and mutation to the current population to generate a new generation of populations and gradually Evolution to include the principle of approximate optimal solution state, the thermal decomposition of benzene dehydrogenation and lysine batch fermentation kinetic model parameters were estimated. Compared with the existing results in the literature, the model calculated values agree well with the experimental values or better. The method used can also be used to estimate or optimize the parameters of such a process using a system of differential equations as a mathematical model so that the researcher can pay more attention to solving the problems related to the mathematical model itself.