ISSN 1817-2172, рег. Эл. № ФС77-39410, ВАК

Differential Equations and Control Processes
(Differencialnie Uravnenia i Protsesy Upravlenia)

The Improvement of Numerical Solutions of Ordinary Differential Equations by Genetic Transforms

Author(s):

Vladimir Nikolaevich Taran

Don State Technical University,
Technological Institute (branch) of DSTU in Azov, Rostov region, Azov
Professor, Doctor of Physical and Mathematical Sciences

vladitaran@rambler.ru

Artem Mihailovich Dolzhenko

Don State Technical University,
Technological Institute (branch) of DSTU in Azov, Rostov region, Azov

dolzhenkoartem@gmail.com

Kristina Kyastuchio Rybalko

Don State Technical University,
Technological Institute (branch) of DSTU in Azov, Rostov region, Azov

kristina.rybalko@inbox.ru

Abstract:

The article describes the modified genetic algorithm for the Cauchy problem solving. To analyze the method effectiveness we perform series of tests which results are presented in tables and figures.The comparison of solutions obtained by the offered method and classical ones (Runge-Kutta and Adams-Bachfort) has been done. The input parameters of the algorithm which give the most accurate result are determined. The application of the algorithm to the problems which do not have the analytical solution is shown. Scientific novelty of the work consists in the realization of a new numerical method for solving ordinary differential equations, which has higher class of accuracy than classical methods. The analysis of scientific works in a scope of genetic algorithms has shown that the authors method of refinement numerical solutions by genetic algorithms is unique and has not been previously described. Relevance of the method is caused by the possibility to apply the offered approach to a modification of a wide class of numerical algorithms solutions.

Keywords

References:

  1. Barricelli, Nils Aall. Symbiogenetic evolution processes realized by artificial methods. Methodos. 1957. pp. 143-182
  2. Fraser, Alex. Simulation of genetic systems by automatic digital computers. I. Introduction. Aust. J. Biol. Sci. 10. 1957. pp. 484-491
  3. Losee R. M. An introduction to genetic algorithms. Information Processing & Management. 1997. Т. 33. № 3. p. 407
  4. Schlapfer M. F. A comparison of genetic and other algorithms for the traveling salesman problem. 1998
  5. Gaspin Ch., Schiex T. Genetic algorithms for genetic mapping. Lecture Notes in Computer Science. 1998. Т. 1363. p. 145
  6. Panchenko T. V. [Genetic algorithms]. Geneticheskiye algoritmy. Astrakhan publishing house " Astrakhan University", 2007. - 87 p. (In Russ. )
  7. Mitchell Melanie A. An Introduction to Genetic Algorithms. Bradford Book. The MIT Press Cambridge, Massachusetts, London, England Fifth printing, 1999
  8. David A Coley. An introduction to genetic algorithms for scientists and engine. World Scientific Publishing Co. Pte. Ltd. 1999. 223 p
  9. Chen, C. L., Chang, M. H. An enhanced genetic algorithm. In Proc. EUFIT’93 (1993), vol. II, pp. 1105-1109
  10. Forrest S., Mitchell M. What makes a problem hard for a genetic algorithm? some anomalous results and their explanation. Machine Learning. 1993. Т. 13. № 2-3. pp. 285-319
  11. Pollak G. A. [Application of genetic algorithms for training neural networks]. Materialy 63-y nauchnoy konferentsii [Materials of the 63rd scientific conference]. South Ural State University. 2011. pp. 174-178. (In Russ. )
  12. Le K. H., Surkov N. E., Ostroukh A. V. [Genetic algorithms in problems of rational organization of information and computational processes]. Geneticheskiye algoritmy v zadachakh ratsional'noy organizatsii informatsionno-vychislitel'nykh protsessov. Automation and control in technical systems. 2014. № 4 (12). pp. 82-99. (In Russ. )
  13. Taran V. N., Dolzhenko A. M., Rybalko K. K. [Analysis of the effectiveness method clarification numerical solution of ordinary differential equations by genetic transformations]. Analiz effektivnosti metoda utochneniya chislennykh resheniy obyknovennykh differentsial'nykh uravneniy geneticheskimi preobrazovaniyami. Scientific Bulletin. 2016. № 3 (9). pp. 153-162
  14. Taran V. N., Boyko E. U., Dolzhenko A. M. [Modification of the greedy algorithm by genetic transformations]. Modifikatsiya zhadnogo algoritma geneticheskimi preobrazovaniyami. The certificate of registration of computer programs 18. 01. 2017 number 2017610869
  15. Dolzhenko A. M., Butrina E. G. [Clarification of decisions of " The traveling salesman problem" of the Genetic mutations]. Utochneniye resheniy zadachi kommivoyazhera geneticheskimi mutatsiyami. Bulletin of Perm University. Series: “Mathematics. Mechanics. Computer science”. 2013. № 2 (21). pp. 9-15

Full text (pdf)