Prediction by a Genetic Algorithm of the Fiber-Matrix Interface Damage for Composite Material. Part 1. Study of Shear Damage in Two Composites T300/914 and PEEK/APC2

We have conducted research on the effect of fiber arrangement on the optimization of the damage to the interface, recently published. We develop in this paper the evolution of the damage in the shearing direction by genetic algorithm. The results obtained by varying the shear rate of the interface o...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Datum:2014
Hauptverfasser: Mokaddem, A., Alami, M., Doumi, B., Boutaoush, A.
Format: Artikel
Sprache:English
Veröffentlicht: Інститут проблем міцності ім. Г.С. Писаренко НАН України 2014
Schriftenreihe:Проблемы прочности
Schlagworte:
Online Zugang:http://dspace.nbuv.gov.ua/handle/123456789/173128
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Назва журналу:Digital Library of Periodicals of National Academy of Sciences of Ukraine
Zitieren:Prediction by a Genetic Algorithm of the Fiber-Matrix Interface Damage for Composite Material. Part 1. Study of Shear Damage in Two Composites T300/914 and PEEK/APC2 / A. Mokaddem, M. Alami, B. Doumi, A. Boutaoush // Проблемы прочности. — 2014. — № 4. — С. 123-129. — Бібліогр.: 18 назв. — англ.

Institution

Digital Library of Periodicals of National Academy of Sciences of Ukraine
id irk-123456789-173128
record_format dspace
spelling irk-123456789-1731282020-11-24T01:26:49Z Prediction by a Genetic Algorithm of the Fiber-Matrix Interface Damage for Composite Material. Part 1. Study of Shear Damage in Two Composites T300/914 and PEEK/APC2 Mokaddem, A. Alami, M. Doumi, B. Boutaoush, A. Научно-технический раздел We have conducted research on the effect of fiber arrangement on the optimization of the damage to the interface, recently published. We develop in this paper the evolution of the damage in the shearing direction by genetic algorithm. The results obtained by varying the shear rate of the interface of the two materials studied T300 and PEEK, show a good agreement between our genetic approach and the mathematical models presented by Cox and Weibull. Для композитных материалов, армированных волокнами, актуальной является оптимизация размещения последних, что обеспечивает минимальное повреждение на стыке волокон и матрицы. Исследована кинетика развития повреждений в направлении действия максимальных сдвиговых напряжений с помощью генетического алгоритма, ранее разработанного авторами. Для двух композитов типа T300 и PEEK с использованием предложенного генетического подхода получены расчетные данные для различных скоростей сдвиговых деформаций на стыке волокон и матрицы, которые хорошо согласуются с полученными на основании математических моделей Кокса и Вейбулла. В дальнейших исследованиях планируется изучение влияния термомеханического нагружения на сдвиговое повреждение зоны стыка волокон и матрицы. Для композитних матеріалів, армованих волокнами, актуальною є оптимізація розміщення останніх, що забезпечує мінімальне пошкодження на стику волокон і матриці. Досліджено кінетику розвитку пошкоджень у напрямку дії максимальних зсувних напружень за допомогою раніше розробленого авторами генетичного алгоритму. Для двох композитів типу Т300 і PEEK з використанням запропонованого генетичного підходу отримано розрахункові дані для різної швидкості зсувних деформацій на стику волокон і матриці, які добре узгоджуються з отриманими на основі математичних моделей Кокса і Вейбулла. У подальших дослідженнях планується вивчення впливу термомеханічного навантаження на зсувне пошкодження зони стику волокон і матриці. 2014 Article Prediction by a Genetic Algorithm of the Fiber-Matrix Interface Damage for Composite Material. Part 1. Study of Shear Damage in Two Composites T300/914 and PEEK/APC2 / A. Mokaddem, M. Alami, B. Doumi, A. Boutaoush // Проблемы прочности. — 2014. — № 4. — С. 123-129. — Бібліогр.: 18 назв. — англ. 0556-171X http://dspace.nbuv.gov.ua/handle/123456789/173128 539.4 en Проблемы прочности Інститут проблем міцності ім. Г.С. Писаренко НАН України
institution Digital Library of Periodicals of National Academy of Sciences of Ukraine
collection DSpace DC
language English
topic Научно-технический раздел
Научно-технический раздел
spellingShingle Научно-технический раздел
Научно-технический раздел
Mokaddem, A.
Alami, M.
Doumi, B.
Boutaoush, A.
Prediction by a Genetic Algorithm of the Fiber-Matrix Interface Damage for Composite Material. Part 1. Study of Shear Damage in Two Composites T300/914 and PEEK/APC2
Проблемы прочности
description We have conducted research on the effect of fiber arrangement on the optimization of the damage to the interface, recently published. We develop in this paper the evolution of the damage in the shearing direction by genetic algorithm. The results obtained by varying the shear rate of the interface of the two materials studied T300 and PEEK, show a good agreement between our genetic approach and the mathematical models presented by Cox and Weibull.
format Article
author Mokaddem, A.
Alami, M.
Doumi, B.
Boutaoush, A.
author_facet Mokaddem, A.
Alami, M.
Doumi, B.
Boutaoush, A.
author_sort Mokaddem, A.
title Prediction by a Genetic Algorithm of the Fiber-Matrix Interface Damage for Composite Material. Part 1. Study of Shear Damage in Two Composites T300/914 and PEEK/APC2
title_short Prediction by a Genetic Algorithm of the Fiber-Matrix Interface Damage for Composite Material. Part 1. Study of Shear Damage in Two Composites T300/914 and PEEK/APC2
title_full Prediction by a Genetic Algorithm of the Fiber-Matrix Interface Damage for Composite Material. Part 1. Study of Shear Damage in Two Composites T300/914 and PEEK/APC2
title_fullStr Prediction by a Genetic Algorithm of the Fiber-Matrix Interface Damage for Composite Material. Part 1. Study of Shear Damage in Two Composites T300/914 and PEEK/APC2
title_full_unstemmed Prediction by a Genetic Algorithm of the Fiber-Matrix Interface Damage for Composite Material. Part 1. Study of Shear Damage in Two Composites T300/914 and PEEK/APC2
title_sort prediction by a genetic algorithm of the fiber-matrix interface damage for composite material. part 1. study of shear damage in two composites t300/914 and peek/apc2
publisher Інститут проблем міцності ім. Г.С. Писаренко НАН України
publishDate 2014
topic_facet Научно-технический раздел
url http://dspace.nbuv.gov.ua/handle/123456789/173128
citation_txt Prediction by a Genetic Algorithm of the Fiber-Matrix Interface Damage for Composite Material. Part 1. Study of Shear Damage in Two Composites T300/914 and PEEK/APC2 / A. Mokaddem, M. Alami, B. Doumi, A. Boutaoush // Проблемы прочности. — 2014. — № 4. — С. 123-129. — Бібліогр.: 18 назв. — англ.
series Проблемы прочности
work_keys_str_mv AT mokaddema predictionbyageneticalgorithmofthefibermatrixinterfacedamageforcompositematerialpart1studyofsheardamageintwocompositest300914andpeekapc2
AT alamim predictionbyageneticalgorithmofthefibermatrixinterfacedamageforcompositematerialpart1studyofsheardamageintwocompositest300914andpeekapc2
AT doumib predictionbyageneticalgorithmofthefibermatrixinterfacedamageforcompositematerialpart1studyofsheardamageintwocompositest300914andpeekapc2
AT boutaousha predictionbyageneticalgorithmofthefibermatrixinterfacedamageforcompositematerialpart1studyofsheardamageintwocompositest300914andpeekapc2
first_indexed 2025-07-15T09:39:53Z
last_indexed 2025-07-15T09:39:53Z
_version_ 1837705351116357632
fulltext UDC 539.4 Prediction by a Genetic Algorithm of the Fiber–Matrix Interface Damage for Composite Material. Part 1. Study of Shear Damage in Two Composites T300/914 and PEEK/APC2 A. Mokaddem, a,1 M. Alami, b B. Doumi, c and A. Boutaous b a Houari Boumediene University of Science and Technology, USTHB, Algiers, Algeria b Mohamed Boudiaf University of Science and Technology of Oran, USTO, Oran, Algeria c University of Dr. Moulay Tahar, Saida, Algeria 1 mokaddem.allel@gmail.com ÓÄÊ 539.4 Ïðîãíîçèðîâàíèå ïîâðåæäåíèÿ êîìïîçèòà íà ñòûêå ìàòðèöû è âîëîêîí ñ ïîìîùüþ ãåíåòè÷åñêîãî àëãîðèòìà. Ñîîáùåíèå 1. Àíàëèç ïîâðåæäåíèé îò ñäâèãîâûõ íàïðÿæåíèé â äâóõ êîìïîçèòàõ T300/914 è PEEK/APC2 À. Ìîêàääåì à,1 , Ì. Àëàìè á , Á. Äîóìè â , À. Áóòàó á à Íàó÷íî-òåõíîëîãè÷åñêèé óíèâåðñèòåò èì. Õóàðè Áóìåäüåíà, Àëæèð, Àëæèð á Íàó÷íî-òåõíîëîãè÷åñêèé óíèâåðñèòåò èì. Ìóõàìåäà Áîóäèàôà, Îðàí, Àëæèð â Óíèâåðñèòåò Ìóëàé Òàõàð, Ñàèäà, Àëæèð Äëÿ êîìïîçèòíûõ ìàòåðèàëîâ, àðìèðîâàííûõ âîëîêíàìè, àêòóàëüíîé ÿâëÿåòñÿ îïòèìèçàöèÿ ðàçìåùåíèÿ ïîñëåäíèõ, ÷òî îáåñïå÷èâàåò ìèíèìàëüíîå ïîâðåæäåíèå íà ñòûêå âîëîêîí è ìàò- ðèöû. Èññëåäîâàíà êèíåòèêà ðàçâèòèÿ ïîâðåæäåíèé â íàïðàâëåíèè äåéñòâèÿ ìàêñèìàëüíûõ ñäâèãîâûõ íàïðÿæåíèé ñ ïîìîùüþ ãåíåòè÷åñêîãî àëãîðèòìà, ðàíåå ðàçðàáîòàííîãî àâòî- ðàìè. Äëÿ äâóõ êîìïîçèòîâ òèïà T300 è PEEK ñ èñïîëüçîâàíèåì ïðåäëîæåííîãî ãåíåòè- ÷åñêîãî ïîäõîäà ïîëó÷åíû ðàñ÷åòíûå äàííûå äëÿ ðàçëè÷íûõ ñêîðîñòåé ñäâèãîâûõ äåôîð- ìàöèé íà ñòûêå âîëîêîí è ìàòðèöû, êîòîðûå õîðîøî ñîãëàñóþòñÿ ñ ïîëó÷åííûìè íà îñíî- âàíèè ìàòåìàòè÷åñêèõ ìîäåëåé Êîêñà è Âåéáóëëà.  äàëüíåéøèõ èññëåäîâàíèÿõ ïëàíèðóåòñÿ èçó÷åíèå âëèÿíèÿ òåðìîìåõàíè÷åñêîãî íàãðóæåíèÿ íà ñäâèãîâîå ïîâðåæäåíèå çîíû ñòûêà âîëîêîí è ìàòðèöû. Êëþ÷åâûå ñëîâà: ñòûê, ñäâèãîâîå ïîâðåæäåíèå, âîëîêíî, ìàòðèöà, êîìïîçèò, ãåíå- òè÷åñêèé àëãîðèòì. Introduction. Nowadays the enormous use of polymer materials is attributed to their extraordinary combination of properties, low weight and ease of processing. However for improvement of some properties such as thermal and mechanical stability, large numbers of additives were added to polymeric matrix and formed polymer matrix composite [1, 2]. A composite is defined as a combination of two or more materials with different physical and chemical properties and distinguishable interface. Composite materials have a wonderful and different range of applications. Important advantages of composites over many metal compounds are high specific stiffness and specific strength, high toughness, corrosion resistance, low density and thermal insulation [1–3]. In most composite materials, one phase is usually continuous and called the matrix, while the other phase called the dispersed phase. On the basis of the nature of the matrices, composites can be classified into four major categories [3, 4]: © A. MOKADDEM, M. ALAMI, B. DOUMI, A. BOUTAOUS, 2014 ISSN 0556-171X. Ïðîáëåìû ïðî÷íîñòè, 2014, ¹ 4 123 1. Polymer matrix composite. 2. Metal matrix composite. 3. Ceramic matrix composite. 4. Carbon matrix composite [4, 5]. Polymer matrix composite can be processed at a much lower temperature, compared to other composite. Depending on the types of polymer matrices, polymer matrix composite are classified as thermosetting composites and thermoplastic composites [6, 7]. Lemaitre and Chaboche [8] consider a damaged solid in which an element of finite volume a notch large enough relative to heterogeneities is defined as follows: S is area representative volume element identified by its norm n, S e is effective resistance area (if S Se � ), and S d is damaged area, S S Sd e� � . The mechanical measurement of local damage in relation to n is then characterized by D S Sd� . If D� 0, the material is in a pristine or not damaged. If D� 1, the volume element is broken into two parts along the plane normal n. If 0 1� �D , D characterizes the state of damage defined, the macroscopic elastic behavior of the damaged material can be calculated using D through the stiffness. In this study, we present a genetic approach of shear damage to the interface of two composites materials T300/914 and PEEK/APC2, using the Weibull probability model. This model accounts for the damage of the two main constituent fibers and matrix, the results are compared with those obtained by the Cox model. Reminder on Analytical Models. Modeling of the Interface. D12 and D22 internal variables of damage [9–11], D12 is the shear and D22 for transverse traction. Once the elastic deformation energy of damage is established, the dual variables YD12 and YD22 , variables of damage D12 and D22, and evolution variables are Y Y bYeq D D� � 12 22 . Laws of damage are � � � � 12 22 12 22 � � � � f Y h Y i ( ), ( ). (1) The interface has the following behavior: (i) linear elastic, brittle, tensile, transverse; (ii) elastic, damageable, shear. Two new damage variables are introduced: �12 and �22. The same way as the matrix, the elastic strain energy and damage variables associated YD12 and YD22 are defined and, finally, the laws of evolution [9–11]. For a single fiber surrounded by matrix, many analytical solutions have been proposed. One of the first [12] provides the shape of the shear stress along the fiber length as the form [13, 14] � � � E a th l f 2 2( ). (2) To simplify calculation, we put: 2 2 2 � G E r R r m f f fln( ) , A. Mokaddem, M. Alami, B. Doumi, and A. Boutaous 124 ISSN 0556-171X. Ïðîáëåìû ïðî÷íîñòè, 2014, ¹ 4 where Gm is shear modulus of the matrix, E f is the Young modulus of the fiber, � is deformation, a is radius of the fiber, R is distance between fibers, and � is shear stress of the interface. These variables related to the components of a composite material (fiber and matrix) are all taken into account through formula (2). These variables allow us therefore to appreciate the resulting sets of genetic algorithm. Model Based on the Statistical Approach. Damage to the matrix, when the stress is uniform, is given by [15]: D Vm m m T m mm � � � �� � � � � � � � � � � 1 0 exp , � � � (3) where � is applied stress, � m T is heat stress, Vm is the ratio volume of the matrix, and � 0m and mm are the Weibull parameters. After creation of a crack, a fragment of length L will give rise to two fragments of size L L� 1 and L XL X2 1� �( ) (X being a random number between 0 and 1) [16, 17]. At each crack up a fiber, a fiber–matrix debonding length 2l will occur with a corollary decrease of creating a new crack in part because the matrix unloaded. At each increment of stress, the break is calculated. All blocks which break reaches 0.5 give rise to new cracks. [16, 17]. A broken fiber is discharged along its entire length. That is to say, it can not break at once. The rupture follows a law similar to that described for the matrix D A Lf f eq f f mf � � � � � � � � � � � � � � 1 0 exp , max� � (4) where � max f is the maximum stress applied and Leq is the length of the fibers would have the same break in a consistent manner. Numerical Simulation by the Genetic Algorithm (GA). Development. Our work involves modeling the damage in shear D12 interface fiber matrix of a composite material. To do so, we chose to use a genetic optimization which allows us to see the evolution of the damage along the fiber. This approach is based on the probabilistic model of Weibull. The principle begins with a random generation of an initial population and the choice of Dm and D f as two random variables, then it is necessary to change this population (numbering 100 with a maximum generation equal to 50 as a stop criterion) by a set of genetic operators (selection, crossover and mutation) and each time is varied the shear rate of the interface to determine the damage D12. The population is composed of the genes on chromosome represent the following variables: the shear rate � which is between 80 and 120 N as defined by the maximum value of stress tests, the Dm and D f . These three variables allow us to locate the shear damage D12 interface, this one compare to damage transverse D22, and find the optimum value of the damage [18]. The evaluation of each generation is made by an objective function after the Cox and Weibull model reflecting all the variables set at the beginning of the algorithm (the mechanical properties of each constituent of the composite) by a shear damage interface. Figure 1 presents the flowchart of genetic algorithm. Simulation Results. A calculation was performed on two types of composite materials T300/914 and PEEK/APC2. We examined the variation of shear rate for different load values (�� 80, 100, and 120 N), which allowed us to calculate the shear damage of the Prediction by a Genetic Algorithm ... Part 1 ... ISSN 0556-171X. Ïðîáëåìû ïðî÷íîñòè, 2014, ¹ 4 125 126 ISSN 0556-171X. Ïðîáëåìû ïðî÷íîñòè, 2014, ¹ 4 A. Mokaddem, M. Alami, B. Doumi, and A. Boutaous Fig. 1. The genetic algorithm flowchart. a b c Fig. 2. Influence of shear stress on the damage to the interface by GA for T300/914 composite: (a) � � 80 N; (b) � � 100 N; (c) � � 120 N. interface. Figures 2 and 3 show each value of � for the level of shear damage to the interface of two materials. T300/914. Figure 2 shows that the damage D interface starts at 0.3 for �� 80 N, then increases to a maximum value of 0.7 for �� 120 N, we note the presence of a symmetry of the damage to the interface. This damage is zero in the middle of the fiber and dense at the ends. One can say that the stress concentration along the length of the fiber creates a strong degradation of the interface and which is greater at the ends relative to the center. PEEK/APC2. Figure 3 shows that the damage D interface starts this time at 0.1 for �� 80 N, then increases to a maximum value of 0.5 for �� 120 N, we note the presence of a symmetry of the damage to the interface, zero in the middle of the fiber and dense at the ends. We can say that the stress concentration along the length of the fiber creates a strong degradation of the interface most important at the ends relative with the middle; values are lower compared to those found for the T300. Conclusions. The results obtained in this study via the genetic algorithm coincide perfectly with the results at the meso level of Ladevèze, which showed that the shear rate leads to a significant reduction in the degradation of the interface. These results show that the level of damage in shear of interface is more important than the level of the transverse damage for both materials studied T300 and PEEK, which provides a good agreement between the numerical simulation and the actual behavior of the two materials. Numerical simulation shows that the PEEK is stronger than the T300. We can therefore say that the ISSN 0556-171X. Ïðîáëåìû ïðî÷íîñòè, 2014, ¹ 4 127 Prediction by a Genetic Algorithm ... Part 1 ... ba c Fig. 3. Influence of shear stress on the damage to the interface by GA for PEEK/APC2 composite: (a) � � 80 N; (b) � � 100 N; (c) � � 120 N. model is applicable to the phenomenon of damage of a unidirectional composite as a function of applied stress. Finally, we can conclude that the shear loading of interface is dominant compared to transverse loading. In future studies we’ll analyze the thermo- mechanical effect on shear damage of interface. Ð å ç þ ì å Äëÿ êîìïîçèòíèõ ìàòåð³àë³â, àðìîâàíèõ âîëîêíàìè, àêòóàëüíîþ º îïòèì³çàö³ÿ ðîçì³- ùåííÿ îñòàíí³õ, ùî çàáåçïå÷óº ì³í³ìàëüíå ïîøêîäæåííÿ íà ñòèêó âîëîêîí ³ ìàòðèö³. Äîñë³äæåíî ê³íåòèêó ðîçâèòêó ïîøêîäæåíü ó íàïðÿìêó 䳿 ìàêñèìàëüíèõ çñóâíèõ íàïðóæåíü çà äîïîìîãîþ ðàí³øå ðîçðîáëåíîãî àâòîðàìè ãåíåòè÷íîãî àëãîðèòìó. Äëÿ äâîõ êîìïîçèò³â òèïó Ò300 ³ PEEK ç âèêîðèñòàííÿì çàïðîïîíîâàíîãî ãåíåòè÷íîãî ï³äõîäó îòðèìàíî ðîçðàõóíêîâ³ äàí³ äëÿ ð³çíî¿ øâèäêîñò³ çñóâíèõ äåôîðìàö³é íà ñòèêó âîëîêîí ³ ìàòðèö³, ÿê³ äîáðå óçãîäæóþòüñÿ ç îòðèìàíèìè íà îñíîâ³ ìàòå- ìàòè÷íèõ ìîäåëåé Êîêñà ³ Âåéáóëëà. Ó ïîäàëüøèõ äîñë³äæåííÿõ ïëàíóºòüñÿ âèâ÷åí- íÿ âïëèâó òåðìîìåõàí³÷íîãî íàâàíòàæåííÿ íà çñóâíå ïîøêîäæåííÿ çîíè ñòèêó âîëîêîí ³ ìàòðèö³. 1. A. P. Mouritz and A. G. Gibson, Fire Properties of Polymer Composite Materials, Springer (2006). 2. T. R. Hull and B. K. Kandola, Fire Retardancy of Polymers: New Strategies and Mechanisms, RSC Publishing, Cambridge (2009). 3. A. R. Horrocks and D. Price, Fire Retardant Materials, Woodhead Publishing (2006). 4. J. Gassan, “A study of fibre and interface parameters affecting the fatigue behaviour of natural fibre composites,” Composites: Part A, 33, No. 3, 369–374 (2002). 5. C. Rospars, E. Le Dantec, and F. Lecuyer, “A micromechanical model for thermostructural composites,” Compos. Sci. Technol., 60, No. 7, 1095–1102 (2000). 6. D. Ratna, Epoxy Composites: Impact Resistance and Flame Retardancy, in: Rapra Review Reports, Vol. 16, No. 5, Report 185 (2005). 7. M. Salavati-Niasari, M. Dadkhah, and F. Davar, “Synthesis and characterization of pure cubic zirconium oxide nanocrystals by decomposition of bis-aqua, tris-acetylacetonato zirconium (IV) nitrate as new precursor complex,” Inorg. Chim. Acta, 362, No. 11, 3969–3974 (2009). 8. J. Lemaitre and J. L. Chaboche, Mechanics of Solid Materials, Cambridge University Press (1990). 9. A. Boutaous, B. Peseux, L. Gornet, and A. Belaidi, “A new modeling of plasticity coupled with the damage and identification for carbon fibre composites laminates,” Compos. Struct., 74, No. 1, 1–9 (2006). 10. A. Boutaous, M. Elchikh, M. Abdelouahab, and A. Belaidi, “A computational strategy for the localization and fracture of laminated composites. Part 1. Development of a localization criterion adapted to model damage evolution time-delay,” Strength Mater., 43, No. 5, 519–525 (2011). 11. A. Boutaous, M. Elchikh, M. Abdelouahab, and A. Belaidi, “A computational strategy for the localization and fracture of laminated composites. Part 2. Life prediction by mesoscale modeling for composite structures,” Strength Mater., 43, No. 6, 645–653 (2011). 12. H. L. Cox, “The elasticity and strength of paper and other fibrous materials,” Br. J. Appl. Phys., 3, No. 3, 72–79 (1952). 128 ISSN 0556-171X. Ïðîáëåìû ïðî÷íîñòè, 2014, ¹ 4 A. Mokaddem, M. Alami, B. Doumi, and A. Boutaous 13. A. Mokaddem, M. Alami, and A. Boutaous, “A study by a genetic algorithm for optimizing the arrangement of fibers on damage to the fiber-matrix interface of composite material,” J. Textile Inst., 103, No. 12, 1376–1382 (2012). 14. A. Mokaddem, M. Alami, L. Temimi, and A. Boutaous, “Study of the effect of heat stress on the damage of the fibre matrix interface of a composite material (T300/914) by means of a genetic algorithm,” FIBRES & TEXTILES in Eastern Europe, 20, 6A (95), 108–111 (2012). 15. W. Weibull, “A statistical theory of the strength of materials,” Proc. Roy. Swed. Inst. Eng. Res., 151, 1–45 (1939). 16. G.-A. Lebrun, Thermomechanical Behavior and Lifetime of Ceramic Matrix Composites: Theory and Experiment, Ph.D. Thesis, University of Bordeaux I (1996). 17. N. Lissart, “Damage and failure in ceramic matrix minicomposites: experimental study and model,” Acta Mater., 45, No. 3, 1025–1044 (1997). 18. P. Ladevèze and G. Lubineau, “Pont entre les «micro» et «méso» mécaniques des composites stratifiés,” Comptes Rendus Mécanique, 331, No. 8, 537–544 (2003). Received 07. 11. 2013 ISSN 0556-171X. Ïðîáëåìû ïðî÷íîñòè, 2014, ¹ 4 129 Prediction by a Genetic Algorithm ... Part 1 ...