A collaborative optimized genetic algorithm based on regulation mechanism of neuroendocrine-immune system

ABSTRACT

In this paper, an improved collaborative optimized genetic algorithm (CGA) inspired from the modulation mechanism of neuroendocrine-immune system is presented. The CGA has several features as follows. The first is that the parent individuals are not involved in the copy process. The second is that more excellent individuals may be produced due to the adaptive crossover and variation probability based on the hormone modulation. In order to examine its performance, firstly, two typical test functions are selected as the simulation models. Then CGA is applied to an intelligent controller based on the modulation of epinephrine (EIC). The simulation results show that the CGA has quicker convergence rate and higher searching precision than that of immune genetic algorithm and normal genetic algorithm, and the EIC optimized has satisfactory control performance

REFERENCES

• 1: Ding Y.S., Natural Computation and Network Intelligence {M}, Shanghai Jiaotong University Press, (Jan. 2008)
• 2: Kim B-Y., Nam G. J., Ryu H. S. and Lee J. W., Optimization of filling process in RTM using genetic algorithm {J}. Korea-Australia Rheology Journal, 12(1): 83--92(2000).
• 3: Su X. H., Yang B., and Wang Y.D., A Genetic Algorithm Based on Evolutionarily Stable Strategy {J}. Journal of Software, 14(11): 1863--1868(2003).
• 4: Khaled B. and Faouzi T., Genetic algorithm for the design of a class fuzzy controller: an alternative approach {J}. IEEE Trans. on Fuzzy Systems, 8(4): 398--405(2000).
• 5: Fleming P. J. and Purshouse R. C., Evolutionary algorithms in control systems engineering: a survey {J}. Control Engineering Practice, 10(11): 1223--1241(2002).
• 6: Lam H. K., Ling S. H., F. Leung H. F., et al. Optimal and stable fuzzy controllers for nonlinear systems subject to parameter uncertainties using genetic algorithms {J}. IEEE Transactions on Industrial Electronics, 51(1): 172-- 182(2004).
• 7: Farhy L. S., Modeling of oscillations of endocrine networks with feedback {J}. Methods Enzymol, 384: 54--81(2004).
• 8: Brazzini B., Ghersetich I., Hercogova J., and Lotti T., the Neuro-immuno-cutaneousendocrine network: relationship between mind and skin, Dermatologic Therapy, 16: 123--131(2003).
• 9: Payne J. K., A neuroendocrine-based regulatory fatigue model, BIOLOGICAL RESEARCH FOR NURSING, 6(2): 141--150(2004).
• 10: Ding Y.S., Computation Intelligence: Theory, techniques and applications, Scientific Press, (Aug. 2004)
• 11: Liu B, Ding Y. S., A novel intelligent controller based on hormone modulation of neuroendocrine system {J}, computer simulation, 23(2): 129--132(2006).
• 12: Ding Y.-S., Liu B., Ren L.-H., Intelligent decoupling control system inspired from modulation of the growth hormone in neuroendocrine system, Dynamics of Continuous, Discrete & Impulsive Systems, Series B: Applications & Algorithms, 14(5): 679--693(2007).

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.8 Problem Solving, Control Methods, and Search

General Terms:
Algorithms, Design

Keywords:
crossover, genetic algorithm, optimized control, variation