Abstract—The objective of this work is to make use of
conventional response surface methodologies and basic
elements from metaheuristic algorithms in the design of
influential variables for engineering systems. A method of
steepest ascent and its integrated approaches with simulated
annealing, firefly and ant colony optimisation algorithms, are
compared on a simulated continuous stirred tank reactor or
CSTR with various levels of signal noise. These metaheuristics
contain the complicatedness in terms of their parameters. An
additional series of computational experiments were conducted
and analysed in terms of the minimax and mean squared error
performance measures including Taguchi’s signal to noise ratio.
Proper levels of these parameters are analysed to recommend
the best parameter choices. On the experimental results of all
the algorithms with the preferable levels of parameters, the
method of steepest ascent seems to be the most efficient on the
CSTR surface at the lower levels of noise. However, the
integrated approaches with all simulated annealing, firefly and
ant colony optimisation elements work well when the standard
deviation of the noise is at higher levels. Although the average,
the standard deviation of the greatest actual concentration of
the product and percentage of sequences ended at the optimum
from the integrated algorithms with simulated annealing and
ant colony optimisation seem to be better, they need more
average design points, especially with ant colony optimisation
element, to converge to the optimum when compared.
Index Terms—Simulated annealing, firefly, ant colony
optimisation; steepest ascent; continuous stirred tank reactor.
P. Luangpaiboon is an Associate Professor, the Industrial Statistics and
Operational Research Unit (ISO-RU), Department of Industrial Engineering,
Faculty of Engineering, Thammasat University, 12120, THAILAND.
[Phone: (662)564-3002-9; Fax: (662)564-3017; e-mail:
lpongch@engr.tu.ac.th].
Cite:Pongchanun Luangpaiboon, "Continuous Stirred Tank Reactor Optimisation via Simulated Annealing, Firefly and Ant Colony Optimisation Elements on the Steepest Ascent," International Journal of Machine Learning and Computing vol. 1, no. 1, pp. 58-65, 2011.