Optimization section reinforced concrete beam using genetic algorithms

Authors

  • A. V. Putintseva Department of information systems in construction, Rostov State University of Civil Engineering, Russian Federation

Keywords:

evolutionary method, search for a global solution, Genetic algorithm, reinforced concrete beam, parameter optimization, cross section

Abstract

Purpose. In the construction industry in the design of structural elements of the main task is to reduce the cost of construction and ensuring its strength. One solution to this problem is the use of evolutionary methods. The purpose of this article is to explore the methods on the basis of which is the design of software for optimization of cross-section of the reinforced concrete beam. In addition, the author analyzes the strategies for the implementation of the software and evaluated the admissibility of the application of the developed method to the projected object. Methodology. The main objective of the study is to build mathematical optimization model cross-section beams. The author gives a comparison of possible methods of solving the task. The article describes the study of the functions of the genetic algorithm. Also are described fitness function, mutations and formation of a new generation. The developed method is implemented in a software product. Findings. On the basis of genetic algorithm been developed optimization method structural element. The calculations in created software product and compared the results with the results of calculations on a set of rules. The results showed the effectiveness of using genetic algorithm to optimize the parameters of the cross-section of a reinforced concrete beam. With a small number of parameters, the results become known for the small amount of time. To solve such a problem full brute force would take significantly more time. Originality. Search for the best solutions from the set of possible solutions is improved by the application of evolutionary methods. Genetic algorithms are a good solution to the task multicriteria. Many methods, such as method gradient descent, stops when reaching a local solution, and genetic algorithm provides a global solution to the task, not stopping on local solutions. This is the main advantage of the method. Practical value. Considered in the article evolutionary process and its software implementation provide an opportunity to carry out the optimization the parameters of the cross section of reinforced concrete beams. In article gives an overview of information on the use of genetic algorithm to optimize the parameters a cross-section of the beams

Author Biography

A. V. Putintseva, Department of information systems in construction, Rostov State University of Civil Engineering

student

References

Algoritmy. Metody. Iskhodniki. [Algorithms. Methods. Sources]. - Available at: http://algolist.manual.ru/ (Accessed 22 August 2015).

Gladkov L.A., Kureychik V.V., Kureychik V.M. Geneticheskie algoritmy [Genetic algorithms] – Moscow, FIZMATLIT Publ., 2006. 320 p.

Mandrikov E.A., Kulev V.A. Primenenie raspredelennykh vychisleniy dlya avtomaticheskoy generatsii konechnykh avtomatov s ispolzovaniem geneticheskikh algoritmov [The use of distributed computing for the automatic generation of finite automata using genetic algorithms] // Sbornik dokladov XI Mezhdunarodnoy konferentsii po myagkim vychisleniyam i izmereniyam [Proceedings of the XI International Conference on Soft Computing and Measurements] (SCM'2008). St. Petersburg: SPbSU Publ. 2008, pp 255-260.

Rutkovskaya D., Pilinskiy M., Rutkowskiy L. Neyronnye seti, geneticheskie algoritmy i nechetkie sistemy [Neural networks, genetic algorithms and fuzzy systems]. – Moscow, Hotline Telecom Publ., 2006. 452 p.

Simbirkina V. Optimalnoe proektirovanie stroitelnykh konstruktsiy s pomoshchyu geneticheskogo algoritma [Optimal design of building structures using genetic algorithm] / Mezhdunarodnaya nauchno-tekhnicheskaya konferentsiya “Prostranstvennye konstruktivnye sistemy zdaniy i sooruzheniy, metody rascheta, konstruirovaniya i tekhnologiya vozvedeniya”[International Scientific Conference "The spatial design of buildings and structures, methods of calculation, design and construction of the technology"], 2008.

SP 52-101-2003 Betonnye i zhelezobetonnye konstruktsii bez predvaritelnogo napryazheniya armatury [Set of rules 52-101-2003 Concrete and reinforced concrete structures without prestressing] - Moscow: 2004, p. 72. Available at:http://files.stroyinf.ru/data1/41/41734/ (Accessed 02 October 2015).

Serpik I.N., Alekseytsev A.V., Levkovich F.N., Tyutyunnikov A.I. Strukturno-parametricheskaya optimizatsiya sterzhnevykh metallicheskikh konstruktsiy na osnove evalyutsionnogo modelirovaniya [Structural-parametric optimization of core metal structures on the basis of evolutionary modeling] // Izvestiya vuzov. Stroitelstvo. -Proceedings of the universities. Building,2005, no. 8, pp. 16-24.

Ferreira C., “Function finding and the creation of numerical constants in gene expression programming,” In J. M.Benitez, O. Cordon, F. Hoffmann, and R. Roy, eds, Advances in Soft Computing: Engineering Design and Manufacturing, SpringerVerlag Publ., 2003, pp. 257-266.

Ferreira, C., “Gene expression programming: A new adaptive algorithm for solving problems,” Complex Systems, 13 (2): 2001, pp. 87-129.

Ferreira, C., “Genetic representation and genetic neutrality in gene expression programming,” Advances in Complex Systems, 5 (4): 2002, pp. 389-408.

NeuroProject. – Режим доступа: http://www.neuroproject.ru/. - Загл. С экрана. - Проверено: 22.08.2015.NeuroProject. - Available at: http://www.neuroproject.ru/ (Accessed 22 August 2015).

Issue

Section

Computer systems and information technologies in education, science and management