Fractal Dimension: An Index to Quantify Parameters in Genetic Algorithms

Genetic Algorithms (GAs) are direct searching methods which require little information from design space. This characteristic beside robustness of these algorithms makes them to be very popular in recent decades. On the other hand, while this method is employed, there is no guarantee to achieve optimum results. This obliged designer to run such algorithms more than one time to achieve more reliable results. There are many attempts to modify the algorithms to make them more efficient. In this paper, by application of fractal dimension (particularly, Box Counting Method), the complexity of design space are established for determination of mutation and crossover probabilities (Pm and Pc). This methodology is followed by a numerical example for more clarification. It is concluded that this modification will improve efficiency of GAs and make them to bring about more reliable results especially for design space with higher fractal dimensions.

Authors:

• Mahmoud R. Shaghaghian

Keywords:

• Genetic Algorithm
• Fractal Dimension
• BoxCounting Method
• Weierstrass-Mandelbrot function.

References:

[1] J. H. Holland, (1975), "Adaptation in natural and artificial systems",
Ann Arbor: The University of Michigan Press, 1975.
[2] J. A. Vasconcelos, J. A. Ramírez, R. H. C. Takahashi., and R. R.
Saldanha, (2001), "Improvements in Genetic Algorithms", IEEE
Transactions on Magnetics, 37(5), 2001, 3414-3417.
[3] J. Andre, P.Siarri, and T.Dognon, "An improvement of the standard
genetic algorithm fighting premature convergence in continuous
optimization", Advances in Engineering Software, 32(1), 2001, 49-60.
[4] O.Hrstka, and A.Kučerová, "Improvements of real coded genetic
algorithms based on differential operators preventing premature
convergence", Advances in Engineering Software, 35(3-4), 2004, 237-
246.
[5] Z.Ye, Z.Li, and M.Xie, "Some improvements on adaptive genetic
algorithms for reliability-related applications", Reliability Engineering
& System Safety, 2009, Article in Press.
[6] G. R. Rakhshandehroo, M. R.Shaghaghian, A. R.Keshavarzi, and N.
Talebbeydokhti, "Temporal variation of velocity components in a
turbulent open channel flow: Identification of fractal dimensions",
Applied Mathematical Modeling, 33, 2009, 3815-3824.