Direct Sequential Based Firefly Algorithm for the \(\alpha \)-Pinene Isomerization Problem

  • Ana Maria A. C. Rocha
  • Marisa C. Martins
  • M. Fernanda P. Costa
  • Edite M. G. P. Fernandes
Conference paper

DOI: 10.1007/978-3-319-42085-1_30

Part of the Lecture Notes in Computer Science book series (LNCS, volume 9786)
Cite this paper as:
Rocha A.M.A.C., Martins M.C., Costa M.F.P., Fernandes E.M.G.P. (2016) Direct Sequential Based Firefly Algorithm for the \(\alpha \)-Pinene Isomerization Problem. In: Gervasi O. et al. (eds) Computational Science and Its Applications – ICCSA 2016. ICCSA 2016. Lecture Notes in Computer Science, vol 9786. Springer, Cham

Abstract

The problem herein addressed is a parameter estimation problem of the \(\alpha \)-pinene process. The state variables of this bioengineering process satisfy a set of differential equations and depend on a set of unknown parameters. A dynamic system based parameter estimation problem aiming to estimate the model parameter values in a way that the predicted state variables best fit the experimentally observed state values is used. A numerical direct method, known as direct sequential procedure, is implemented giving rise to a finite bound constrained nonlinear optimization problem, which is solved by the metaheuristic firefly algorithm (FA). A Matlab™ programming environment is developed with the mathematical model and the computational application of the method. The results produced by FA, when compared to those of the fmincon function and other metaheuristics, are competitive.

Keywords

\(\alpha \)-pinene isomerization Parameter estimation Direct sequential procedure Firefly algorithm 

Copyright information

© Springer International Publishing Switzerland 2016

Authors and Affiliations

  • Ana Maria A. C. Rocha
    • 1
  • Marisa C. Martins
    • 1
  • M. Fernanda P. Costa
    • 2
  • Edite M. G. P. Fernandes
    • 1
  1. 1.Algoritmi Research CentreUniversity of MinhoBragaPortugal
  2. 2.Centre of MathematicsUniversity of MinhoGuimarãesPortugal

Personalised recommendations