A New Strategy for Parameter Estimation of Dynamic Differential Equations Based on NSGA II
A new strategy for parameter estimation of dynamic differential equations based on nondominated sorting genetic algorithm II (NSGA II) and one-step-integral Treanor algorithm is presented. It is adopted to determine the exact model of catalytic cracking of gas oil. Compared with those conventional methods, for example, quadratic programming, the method proposed in this paper is more effective and feasible. With the parameters selected from the NSGA II pareto-optimal solutions, more accurate results can be obtained.
KeywordsSequential Quadratic Programming Nondominated Sorting Nondominated Sorting Genetic Algorithm Nondominated Front Dynamic Differential Equation
Unable to display preview. Download preview PDF.
- 1.Jiang, A.P., Shao, Z.J., Qian, J.X.: Optimization of Reaction Parameters Based on rSQP and Hybrid Automatic Differentiation Algorithm. Journal of Zhejiang University (Engineering Science) 38, 1606–1610 (2004)Google Scholar
- 4.Goldberg, D.E.: Genetic Algorithm in Search, Optimization and Machine Learning [M]. Addison-Wesley, Reading (1989)Google Scholar
- 5.Deb, K., Agrawal, S., Pratap, A., et al.: A Fast Elitist Nondominated Sorting Genetic Algorithm For Multi-objective Optimization: NSGA II [A]. In: Proc of the Parallel Problem Solving from Nature VI Conf. [C], Paris, pp. 849–858 (2000)Google Scholar
- 6.Xu, S.L.: Common Algorithm Set by FORTRAN. Tsinghua University Press (1995)Google Scholar
- 7.Froment, G.F., Bischoff, K.B.: Chemical Reactor Analysis and Design. Wiley, New York (1979)Google Scholar