constitutive modeling of anisotropic hyperelastic materials by polyconvex strain energy functions
Constitutive Modeling(ppt)
Response of nonlinear elastic models is similar to the proverbial “… the last straw which broke the camel’s back.”
Elastoplasticity
Certain basic steps are followed in formulating all constitutive models based on plasticity. One must define: • yield criterion/surface • plastic potential/flow rule • hardening rule
dε = dε + dε
• For example in 1D:
e
p
1 1 dε1 = dε + dε = + d σ1 E H
e 1 p 1
where E=elastic Young’s modulus, and H=plastic modulus
Hardening rule
• Describes the evolution of the yield surface during loading.
Summary of elements of elastoplasticity:
Additivity: Elasticity:
dε = dε e + dε p
dσ = D : dε
e
p
e
∂g (σ, q) Flow rule: dε = ϕm (σ, q) = ϕ ∂σ
Hardening rule:
Example of a failure surface: Mohr-Coulomb:
光学模型介绍英文作文
光学模型介绍英文作文英文:Optical model is a mathematical model that describesthe behavior of light in various media, such as air, water, and glass. It is widely used in the field of optics to predict and analyze the propagation, reflection, refraction, and absorption of light.The optical model is based on the principles of geometrical optics and wave optics. Geometrical optics assumes that light travels in straight lines and obeys the laws of reflection and refraction at the interface between two media. Wave optics, on the other hand, considers light as a wave that can diffract, interfere, and undergo polarization.One of the most important parameters in the optical model is the refractive index, which is a measure of how much a material slows down the speed of light. Therefractive index is different for different materials and can be used to calculate the angle of refraction when light passes through a medium.Another important parameter is the absorption coefficient, which measures how much light is absorbed by a material per unit distance. This is important for designing optical devices such as filters and lenses.In addition to these parameters, the optical model also considers factors such as the thickness of the medium, the angle of incidence, and the polarization state of light. By combining all these factors, the optical model can accurately predict the behavior of light in various scenarios.中文:光学模型是一种描述光在各种介质中行为的数学模型,例如空气、水和玻璃。
岩土材料固有各向异性的模拟
第23卷 第10期岩石力学与工程学报 23(10):1604~16072004年5月 Chinese Journal of Rock Mechanics and Engineering May ,20042003年10月26日收到初稿,2003年12月25日收到修改稿。
作者 余天堂 简介:男,32岁,2000年于河海大学工程力学系水工结构专业获博士学位,现任副教授,主要从事计算力学方面的研究工作。
E-mail :tiantangyu@ 。
岩土材料固有各向异性的模拟余天堂(河海大学土木工程学院 南京 210098)摘要 给出了岩土材料的一种各向异性模型,并用一个各向异性参数来描述岩土材料的固有各向异性。
各向异性参数、单轴抗压强度和单轴抗拉强度是分布函数,其分布可以用微结构张量和加载方向表示。
建立了一个既能描述岩土材料的固有各向异性又能反映开裂和屈服2种破坏模式的破坏准则。
数值模拟结果表明,该模型能有效地描述岩土材料的固有各向异性。
关键词 岩土力学,破坏准则,各向异性,微结构张量分类号 TU 411.3 文献标识码 A 文章编号 1000-6915(2004)10-1604-04MODELING OF INHERENT ANISOTROPY FOR GEOTECHNICALMATERIALYu Tiantang(College of Civil Engineering ,Hohai University , Nanjing 210098 China )Abstract A model of anisotropy for geotechnical material is presented. The inherent anisotropy of this kind of material is described with an anisotropy parameter. A microstructure tensor and loading direction are incorporated in the distribution functions of anisotropy parameter ,uniaxial compressive strength and uniaxial tensile strength. A failure criterion describing the fracture and yielding for geotechnical material is set up. Numerical simulation results show that the inherent anisotropy of geotechnical material can be efficiently described with the model. Key words rock and soil mechanics ,failure criterion ,anisotropy ,microstructure tensor1 引 言由于在岩土体内部存在层理、片理等,或者,在某一方向有非常发育的节理系统,从而形成某种程度的成层现象,并沿着层面方向和垂直层面方向具有不同的弹性常数,因此,岩土体具有很强的固有各向异性。
Handbookofmaterialsbehaviormodels
HANDBOOK OF MATERIALS BEHAVIOR MODELSCONTENTSForeword(E.van der Giessen)Introduction(J.Lemaitre)Chapter1Background on mechanics of materialsChapter2Elasticity,viscoelasticityChapter3Yield limitChapter4PlasticityChapter5ViscoplasticityChapter6Continuous damageChapter7Cracking and fractureChapter8Friction and wearChapter9Multiphysics coupled behaviorsChapter10Composite medias,biomaterialsChapter11GeomaterialsCHAPTER1Background on mechanics of materials1.1Background on modelingJ.Lemaitreiii Contents1.2Materials and process selectionY.Brechet1.3Size effect on structural strengthZ.BazantCHAPTER2Elasticity,viscoelasticity2.1Introduction to elasticity and viscoelasticityJ.Lemaitre2.2Background on nonlinear elasticityR.W.Ogden2.3Elasticity of porous materialsN.D.Cristescu2.4Elastomer modelsR.W.Ogden2.5Background on ViscoelasticityK.Ikegami2.6A nonlinear viscoelastic model based onfluctuating modesR.Rahouadj,C.Cunat2.7Linear viscoelasticity with damageR.SchaperyCHAPTER3Yield limit3.1Introduction to yield limitsJ.Lemaitre3.2Background on isotropic criteriaD.Drucker3.3Yield loci based on crystallographic textureP.Van HoutteContents iii3.4Anisotropic yield conditionsM.Zyczkowski3.5Distortional model of plastic hardeningT.Kurtyka3.6A generalised limit criterion with application tostrength,yielding and damage of isotropic materialsH.Altenbach3.7Yield conditions in beams,plates and shellsD.DruckerCHAPTER4Plasticity4.1Introduction to plasticityJ.Lemaitre4.2Elastoplasticity of metallic polycrystals by theself-consistent modelM.Berveiller4.3Anisotropic elasto-plastic model based oncrystallographic textureA.M.Habraken,L.Ducheˆne,A.Godinas,S.Cescotto4.4Cyclic plasticity model with non-linear isotropicand kinematic hardening-No LIKH modelD.Marquis4.5Multisurface hardening model for monotonic andcyclic response of metalsZ.Mroz4.6Kinematic hardening rule with critical state ofdynamic recoveryN.Ohno4.7Kinematic hardening rule for biaxial ratchettingH.Ishikawa,K.Sasaki4.8Plasticity in large deformationsY.F.Dafalias4.9Plasticity of polymersJ.M.Haudin,B.Monasse4.10Rational phenomenology in dynamic plasticityJ.R.Klepaczko4.11Conditions for localization in plasticity andrate-independent materialsA.Benallal4.12An introduction to gradient plasticityE.C.AifantisCHAPTER5Viscoplasticity5.1Introduction to viscoplasticityJ.Lemaitre5.2A phenomenological anisotropic creep model forcubic single crystalsBertram,J.Olschewski5.3Crystalline viscoplasticity applied to single crystalG.Cailletaud5.4Averaging of viscoplastic polycristalline materialswith the tangent self-consistent modelA.Molinari5.5Fraction models for inelastic deformationJ.F.Besseling5.6Inelastic compressible and incompressible,isotropic,small strain viscoplasticity theory basedon overstress(VBO)E.Krempl,K.Ho5.7An outline of the Bodner-Partom(BP)unifiedconstitutive equations for elastic-viscoplastic behaviorS.Bodner5.8Unified model of cyclic viscoplasticity based on thenon-linear kinematic hardening ruleJ.L.Chaboche5.9A model of non-proportional cyclic viscoplasticityE.T anaka5.10Rate-dependent elastoplastic constitutive relationsF.Ellyin5.11Physically-based rate and temperature dependantconstitutive models for metalsS.Nemat-Nasser5.12Elastic-viscoplastic deformation of polymersE.M.Arruda,M.BoyceCHAPTER6Continuous damage6.1Introduction to continuous damageJ.Lemaitre6.2Damage equivalent stress-fracture criterionJ.Lemaitre6.3Micromechanically inspired continuous models ofbrittle damageD.Krajcinovic6.4Anisotropic damageC.L.Chow,Y.Wei6.5Modified Gurson modelergaard,A.Needleman6.6The Rousselier Model for porous metal plasticityand ductile fractureG.Rousselier6.7Model of anisotropic creep damageS.Murakami6.8Multiaxial fatigue damage criteriaD.Sauci6.9Multiaxial fatigue criteria based on amultiscale approachK.Dang Van6.10A probabilistic approach to fracture in highcycle fatigueF.Hild6.11Gigacycle fatigue regimeC.Bathias6.12Damage mechanisms in amorphous glassypolymers:crazingR.Schirrer6.13Damage models for concreteG.Pijaudier-Cabot,J.Mazars6.14Isotropic and anisotropic damage law of evolutionJ.Lemaitre,R.Desmorat6.15A two scale damage model for quasi brittle andfatigue damageR.Desmorat,J.LemaitreCHAPTER7Cracking and fracture7.1Introduction to cracking and fractureJ.Lemaitre7.2Bridges between damage and fracture mechanicsJ.Mazars,G.Pijaudier-Cabot7.3Background on fracture mechanicsH.D.Bui,J.B.Leblond,N.Stalin-Muller7.4Probabilistic approach to fracture:the Weibull modelF.Hild7.5Brittle fractureD.Franc¸ois7.6Sliding crack modelD.GrossContents vii7.7Delamination of coatingsH.M.Jensen7.8Ductile rupture integrating inhomogeneitiesin materialsJ.Besson,A.Pineau7.9Creep crack growth behavior in creep ductile andbrittle materialsT.Yokobori Jr.7.10Critical review on fatigue crack growthT.Yokobori7.11Assessment of fatigue damage on the basis ofnon-linear compliance effectsH.Mughrabi7.12Damage mechanics modelling of fatiguecrack growthX.Zhang,J.Zhao7.13Dynamic fractureW.G.Knauss7.14Practical applications of fracture mechanics-fracture controlD.BroekCHAPTER8Friction and wear8.1Introduction to friction and wearJ.Lemaitre8.2Background on friction and wearY.Berthier8.3Models of frictionA.Savkoor8.4Friction in lubricated contactsJ.Freˆne,T.Ciconeviii Contents 8.5A thermodynamic analysis of the contact interfacein wear mechanicsH.D.Bui,M.Dragon-louiset,C.Stolz8.6Constitutive models and numerical methods forfrictional contactM.Raous8.7Physical models of wear,prediction of wear modesK.KatoCHAPTER9Multiphysics coupled behaviors9.1Introduction to multiphysics coupled behaviorJ.Lemaitre9.2Elastoplasticity and viscoplasticity coupledto damageA.Benallal9.3A fully anisotropic elasto-plastic-damage modelS.Cescotto,M.Wauters,A.M.Habraken,Y.Zhu9.4Model of inelastic behavior coupled to damageG.Z.Voyiadjis9.5Thermo-elasto-viscoplasticity and damageP.Perzyna9.6High temperature creep-deformation andrupture modelsD.R.Hayhurst9.7A coupled diffusion-viscoplastic formulation foroxidasing multi-phase materialsE.P.Busso9.8Hydrogen attackE.van der Giessen,S.Schlo¨gl9.9Hydrogen transport and interaction with materialdeformation:implications for fractureP.Sofronis9.10Unified disturbed state constitutive modelsC.S.Desai9.11Coupling of stress/strain,thermal andmetallurgical behaviorsT.Inoue9.12Models for stress-phase transformation couplingsin metallic alloysS.Denis,P.Archambault,E.Gautier9.13Elastoplasticity coupled with phase changesF.D.Fisher9.14Mechanical behavior of steels during solid-solidphase transformationsJ.B.Leblond9.15Constitutive equations of shape memory alloyunder complex loading conditionsM.T okuda9.16Elasticity coupled with magnetismR.Billardon,L.Hirsinger,F.Ossart9.17Physical ageing and glass transition of polymersR.Rahouadj,C.CunatCHAPTER10Composite media,biomechanics10.1Introduction to composite mediaJ.Lemaitre10.2Background on micromechanicsE.van der Giessen10.3Non linear composites-secant methods andvariational boundsP.Suquet10.4Non local micromechanical modelsJ.Willis10.5Transformationfield analysis of composite materialsG.Dvorak10.6A damage mesomodel of laminate compositesdeve`ze10.7Behavior of ceramix-matrix composites underthermomechanical cyclic loading conditionsF.A.Leckie,A.Burr,F.Hild10.8Limit and shakedown analysis of periodicheterogeneous mediaG.Maier,V.Carvelli,A.T aliercio10.9Flow induced anisotropy in shortfiberscompositesA.Poitou,F.Meslin10.10Elastic poperties of bone tissueG.A.Cowing10.11Bimechanics of soft tissueS.C.HolzapfelCHAPTER11Geomaterials11.1Introduction to geomaterialsJ.Lemaitre11.2Background of the behaviour of geomaterialsF.Darve11.3Models for geomaterialsN.D.Cristescu11.4Behaviour of granular materialsI.Vardoulakis11.5Micromechanically-based constitutive model forfrictional granular materialsS.Nemat-NasserContents xi11.6Linear poroelasticityJ.W.Rudnicki11.7Non linear poroelasticity for liquid non saturatedporous materialsO.Coussy,P.Dangla11.8An elastoplastic constitutive model for partiallysaturated soilsB.A.Schreffler,L.Simoni11.9Sinfonietta classica a strain-hardening model forsoils and soft rocksR.Nova11.10A generalized plasticity model for dynamicbehaviour of sand including liquefactionM.Pastor,O.C.Zienkiewicz,A.H.C.Chan11.11A critical state bounding surface model for sandsM.T.Manzari,Y.F.Dafalias11.12Lattice model for fracture analysis of brittledisordered materials like concrete and rockJ.G.M.van Mier。
A constitutive model of cyclic plasticity
A constitutive model of cyclic plasticityFusahito Yoshida*Department of Mechanical Engineering,Hiroshima University 14-1,Kagamiyama,Higashi-Hiroshima 739-8527,JapanReceived in ®nal revised form 28September 1999AbstractThis paper addresses a constitutive model of cyclic plasticity with special emphasis on the yield-point phenomena.In order to point out the deformation characteristics of a mild steel,four types of experiments were conducted,i.e.uniaxial tension at several crosshead speeds,cyclic straining,and stress-and strain-controlled ratchetting.A viscoplastic constitutive model of cyclic plasticity is proposed on the premise that the phenomena of sharp yield point and the subsequent abrupt yield drop result from rapid dislocation multiplication and the stress-dependence of dislocation velocity.Besides,cyclic plasticity behavior,such as the Bauschinger e ect,cyclic hardening/softening characteristics and ratchet-strain accumulation,is described by some kinematic and isotropic hardening rules.The cyclic stress±strain responses predicted by this model agree well with the corresponding experimental results.#2000Published by Elsevier Science Ltd.All rights reserved.Keywords:A.Dislocations;B.Constitutive behaviour;B.Cyclic loading1.IntroductionAnnealed steels and some bcc metals,as well as some aluminium alloys,exhibit a sharp yield point and the subsequent abrupt yield drop followed by the yield plateau (in other words `lower yield ')in their stress±strain curves of uniaxial tension,as schematically illustrated in Fig.1.The plastic deformation from the yield point (A in Fig.1)to the end of the yield plateau (B in Fig.1)is caused by the Lu ders band propagation (hereafter,we call the strain from A to B `Lu ders strain'),where the plastic deformation is completely inhomogeneous.After the Lu ders bands coverthe International Journal of Plasticity 16(2000)359±/locate/ijplas0749-6419/00/$-see front matter #2000Published by Elsevier Science Ltd.All rights reserved.P I I :S 0749-6419(99)00058-3*Tel.:+81-824-24-7541;fax:+81-824-22-7193.E-mail address:yoshida@mec.hiroshima-u.ac.jp (F.Yoshida).whole specimen,the deformation becomes homogeneous,and the ¯ow stress rises as the material hardens (from B to C in Fig.1).Such yield-point phenomena and the inhomogeneous plastic deformation in metals have been intensively investigated in the ®eld of metal physics (e.g.Cottrell and Bilby,1949;Lomer,1952;Hahn,1962;Conrad,1963;Bullen et al.,1964;Stein and Low,1966;Phillips and Armstrong,1969;Hall,1970;Imamura et al.,1971;Fujita and Miyazaki,1978;Hu,1983;Neuhauser and Hampel,1993).Among these studies,it is worth noting that Hahn (1962)proposed a uniaxial constitutive model for iron and related bcc polycrystalline metals based on Johnston and Gilman's (1959)concept that the abrupt yield drop is a consequence of rapid dislocation multi-plication and the stress-dependence of dislocation velocity.After hisproposition,360 F.Yoshida /International Journal of Plasticity 16(2000)359±380Shioya and Shioiri (1976)extended the model to a multiaxial one,which also inclu-ded their original proposal for the criterion for the production of the mobile dis-location density,and they calculated the inhomogeneous plastic deformation appearing in bar torsion of a mild steel.Itoh et al.(1992)implemented Hahn's model to a ®nite element (FE)code and simulated the nonuniform deformation pattern in mild steel plates subjected to uniaxial tension.However,to the present author's knowledge,for cyclic plasticity taking account of the yield-point phenom-ena,very limited numbers of papers on the experimental observations (e.g.Yoshida and Murakami,1995;Jiang and Kurath,1999)have been published so far,and almost none of constitutive models has ever been proposed.From the viewpoint of stress analyses in engineering design,if we intend to predict the stress±strain distributions in steel structures which are subjected to cyclic load-ing,we need a constitutive model that can properly describe the yield-point phe-nomena as well as the subsequent cyclic plasticity behavior especially within the Lu ders strain range.This is because,in most of the cases of cyclic loading,the plastic zones in the structure would be surrounded by elastic zones,and as a result,the plastic strain ranges would scarcely exceed the Lu ders strain range.Nevertheless,most of the previous investigations of models of cyclic plasticity on steels have been conducted only focusing on material behavior in the strain-hardening region (even though some steels they dealt with had apparent yield-point nature,e.g.Yoshida et al.1984;Hassan and Kyriakides,1992,1994),but not in the yield-plateau region.This paper provides some typical examples of experimental data of cyclic plasticity including the yield-point phenomena for a mild steel,i.e.uniaxial tension at several crosshead speeds,cyclic straining,and stress-and strain-controlled ratchetting.Furthermore,a viscoplastic constitutive model of cyclic plasticity is proposed on the premise that the sharp yield point and the subsequent abrupt yield drop resultfrom Fig.1.Schematic illustration of the stress±strain response showing yield-point phenomena.F.Yoshida /International Journal of Plasticity 16(2000)359±380361rapid dislocation multiplication and the stress-dependence of dislocation velocity.Besides,cyclic plasticity behavior,such as the Bauschinger e ect,cyclic hardening/softening and rate-dependent ratchetting is described by some kinematic and iso-tropic hardening rules.The performance of the model is evaluated by comparing the calculated stress±strain responses with the corresponding experimental results.2.Experimental observations2.1.Experimental proceduresFour types of experiments,i.e.uniaxial tension at several crosshead speeds,cyclic straining,and stress-and strain-controlled ratchetting were conducted (the details of the test conditions will be indicated later)using a servo-controlled testing machine.Hot-rolled 0.070%-C steel plate of 2.5-mm thickness was tested.Prior to machining the plate,it was annealed at 1000 C for an hour to obtain a specimen of grain size of 26.5m m.In order to prevent buckling,adequate specimen thickness was obtained by adhesive bond of two plates so that the thickness of the specimen was 5.0mm.The shape of the specimen is illustrated in Fig.2.The displacement in the gauge length of 14mm was measured by a clip-on extensometer.Either cross-head speed or average strain in the gauge length (hereafter called `gauge-length strain')were controlled.True-stress-controlled tests were also conducted by calculating the true-stress during the experiments.2.2.Experimental results and discussionStress±strain curves obtained from uniaxial tension tests at three di erent cross-head speeds are depicted in Fig.3.In these stress±strain curves,we observe asharpFig.2.Shape of a mild steel specimen.362 F.Yoshida /International Journal of Plasticity 16(2000)359±380yield point and the subsequent abrupt yield drop followed by the yield plateau.The yield plateau results from the formation of Lu ders-band front and its propagation.For the formation of band-structure in polycrystalline iron,Fujita and Miyazaki (1978)proposed such a mechanism that surrounding grains around each trigger grain (they call `the a ected zone'which expands to the fourth and ®fth neighboring grains when the grain size is less than 30mm or less)are deformed at the same time when the local stress reaches its critical value.The driving force of the Lu ders-band front propagation would be the long-range interaction among individual grains.Imamura et al.(1971)investigated the deformation structure of the Lu ders bands in an iron specimen by the etch-pit technique and found the strain gradient in the band fronts.Moreover,they reached an important conclusion that the dislocation pile-ups should not be a reason for the Lu ders band propagation since any dislocation pile-ups at grain boundaries had not been observed.It should be noted that,in Fig.3,the Lu ders strain 4v becomes larger with increasing crosshead speed (strain rate).This rate-dependence of the Lu ders strain 4v can be found also in some other papers,e.g.the data from Fujita and Miyazaki (1978)and Yoshida and Murakami (1995)together with the present experiments are summarized in Fig.4.Obviously,the (average)gauge-length strain rate is much lower than the local strain rate within the propagating Lu ders-band front.[Imamura et al.(1971)found that the real strain rate in the Lu ders-band front,whose width was three to four grain diameters in a steel of 40mm grain size,was two orders of magnitudes higher than the gauge-length strain rate.]Therefore,we cannot deter-mine quantitatively the actual strain-rate dependence of the length of Lu ders strain 4v directly from Fig.4.However,the results shown in this ®gure allow us to under-stand how strongly the Lu ders strain 4v depends on the gauge-length strainrate.Fig.3.Stress±strain curves of mild steel obtained from uniaxial tension tests conducted at various cross-head speeds.F.Yoshida /International Journal of Plasticity 16(2000)359±380363Stress±strain responses during a cyclic straining of the strain range of 0.005and 0.01at gauge-length stain rate of 10À4s À1are illustrated in Fig.5(a)and (b).In the case of the strain range of 0.01the material slightly strain-hardens,whereas cyclic non-hardening (slightly softening)was observed in the case of strain range of 0.005.This is the well-known strain-range dependence of cyclic hardening behavior,and the reasonable modelling for it has been given by the concept of non-(isotropic)-hardening strain range (Chaboche et al.,1979;Ohno,1982).The cyclic hardening/softening behavior also depends on materials,e.g.cyclic softening is more dominant on 0.039%-C steel (refer to Yoshida,1999)compared to the present 0.070%-C steel.The Bauschinger e ect is clearly seen even in a small cyclic strain range like 0.005[in Fig.5(a)].The Bauschinger e ect would be attributed to the short-range dislocation interactions (the interactions of dislocations with other dislocations,interstitials,precipitates and other barriers),as well as the long-range interaction among indivi-dual grains (e ects of internal stress ®eld induced by grain constraint and elastic±plastic inhomogeneity).Moreover,it should be noted that in Fig.5(b)the cyclic hardening (i.e.the isotropic hardening)begins after a non-hardening strain cycle,which corresponds to the stress±strain response in monotonic tension that the strain hardening begins after a certain non-hardening region (the yieldplateau).Fig.4.The e ect of gauge-length strain rate on the Lu ders strain.364 F.Yoshida /International Journal of Plasticity 16(2000)359±380Cyclic hardening behavior of the material is clearly shown in the stress±strain curves registered in a strain-controlled ratchetting (see Fig.6).In this test,the gauge-length strain was controlled in such a way that a cyclic strain range Á4of 0.007and a mean strain increment 4of 0.002(the de®nitions of the cyclic strain range Á4and the mean strain increment 4in this type of experiment are illustrated in Fig.6)were prescribed in each strain cycle,where the gauge-length strain rate was 10À4.The stress range is found to be almost constant for the ®rst few cycles after yielding,and after that,it increases with increasing number of strain cycles.The maximum stress for a given mean strain is slightly lower than the ¯ow stress in uniaxial tension.This result suggests that the isotropic hardening evolution takes place only in each ratchet-strain increment,but not during the strain cycling (within hysteresis loops).Therefore,from this experimental result,it is possible to decompose the strain hardening into the kinematic and isotropic hardening contributions.At the physical level,the kinematic hardening after the yield plateau would be mainly due to the long-range dislocation interactions,and the evolution of the isotropic hardening would be associated with the interactions of dislocations with orientation-indepen-dent defects and the growth of cells and subgrains (Stou er and Dame,1996).The e ect of stress rate on the ratchet-strain accumulation was observed in a stress-controlled ratchetting (see Fig.7).In this experiment,the stress was controlled in such a way that the maximum stress was equal to 250MPa (almost the same stress level as the lower yield stress),while the minimum stress was 0MPa (i.e.the stress ratio R s 'min a'm x 0)and the stress rate was 100MPa s À1for the ®rst 20cycles and 1MPa s À1for the subsequent stress cycles.During the fast stress cycles (at 100MPa s À1),almost no ratchet-strain accumulation was found,whereas during the subsequent slow stress cycles (at 1MPa s À1)the ratchetting strain accumulated rapidly until the shakedown took place.The ratchet-strain accumulation anditsFig.5.Stress±strain responses during cyclic straining:(a)strain range of 0.005;(b)strain range of 0.01.F.Yoshida /International Journal of Plasticity 16(2000)359±380365Fig.6.Stress±strain response during strain-controlled ratchetting,where the gauge-length strain was controlled in such a way that a cyclic strain range of 0.007and a mean strain increment of 0.002were given in each strain cycle at strain rate of 10À4s À1.Fig.7.Stress±strain response during strain-controlled ratchetting,where the stress was controlled in such a way that the maximum stress of 250MPa and minimum stress of 0MPa (i.e.the stress ratio R s 'min a'm x 0)were kept constant,while the stress rate was 100MPa s À1for the ®rst 20cycles and the 1MPa s À1for the subsequent stress cycles,respectively.366 F.Yoshida /International Journal of Plasticity 16(2000)359±380shakedown was also observed in the case of stress ratio R s 'min a'm x À0X6(see Fig.8,where'm x 255MPa).It is interesting that the shakedown limits of strain, for both the above two cases,is almost equal to the Lu ders strain.This result sug-gests that the shakedown may be due to rapid evolution of isotropic hardening which begins just after the yield plateau.3.Constitutive modelling3.1.Framework of viscoplasticity modelThe well-known relationship between the dislocation velocity#and the shear plastic strain rate p concerning the deformation of single crystals is given by X p b&m#Y I where b stands for the Burgers vector and&m is the mobile dislocation density.The dislocation velocity#strongly depends on the resolved shear stress(,and several types of equations for the relationship between#and(have been proposed(e.g. Hahn,1962;Conrad,1963;Suezawa et al.,1979;Hull and Bacon,1984;Francois et al.,1998).One of the equations has the form:#(À(D(h i nY PFig.8.Stress±strain response during strain-controlled ratchetting,where the maximum stress of255MPa and minimum stress ofÀ153MPa(R s 'min a'm x À0X6).F.Yoshida/International Journal of Plasticity16(2000)359±380367where ( denotes the interaction stress acting on a moving dislocation;D (and n are the drag stress and stress-rate sensitivity exponent,respectively.From Eqs.(1)and (2)a constitutive equation is derived asX p b &m (À( D (h i n Q For polycrystals,the following constitutive model is proposed by analogy with the above equation for single crystals.The yield function f is expressed by the kinematic and isotropic hardening asf 32 s À X s À À Y 0 R 2Y R where s denotes the stress deviator; the kinematic hardening internal stress (the back stress);Y 0the initial yield stress;and R the isotropic hardening internal stress.The viscoplastic constitutive equation is given by 4"X b &m M '"À Y 0 R D ()n Y 4X p 3 s À 2'"4"X Y '" 32 s À X s À r Y W b b b a b b b Y S where 4X p and 4"X are the plastic strain rate and the e ective plastic strain rate,and M is the Taylor factor (M 2X 76according to TBH polycrystal plasticity calculation).In the present model,the yield-point phenomena are attributed to the rapid dis-location multiplication and the stress-dependence of dislocation velocity.The mobile dislocation density &m is expressed as a fraction f m of the total dislocation density &t :&m f m &t X T The total dislocation density &t is given as an empirical equation (Hahn,1962)of the e ective plastic strain 4":&t &0 C 4" Y Uwhere &0denotes the initial dislocation density (usually for metals about an order of 106cm À2)and C and a are material constants (C is in the range of 109±1011cm À2'and a 0X 7À1X 5after Hahn,1962;Kohda,1973;Hull and Bacon 1984).Even though several propositions for the evolution equations of mobile dislocation den-sity have been put forth (e.g.Kocks,1976;Yonenaga and Sumino,1978;Suezawa et al.,1979;Braasch et al.,1996;Franc ois et al.,1998),direct measurements of the fraction f m have never been reported for mild steels.As a ®rst approximation,it is given in this paper by the following phenomenological expression so as to describe the rapid dislocation multiplication with increasing plastic strain 4":368 F.Yoshida /International Journal of Plasticity 16(2000)359±380f m f m 0 f m Àf m 0 1Àexp l 4" ÈÉY Vwhere l is a material constant.In this equation,it is assumed that the fraction f m of mobile dislocations,initially equals f m 0,increases rapidly with increasing plastic strain 4"and approaches to an asymptotic limit f m (in this paper it is assumed to be 0.1after Hahn,1962).In the case when a mild steel is fully annealed,most of the dislocations are locked,consequently the initial value of the fraction f m 0would be close to zero.The dislocation multiplication,&t and &m ,assumed in the present cal-culation by Eqs.(7)and (8)is shown in Fig.9.3.2.Kinematic and isotropic hardening rulesThe evolution equation for the back stress due to the short-range dislocation interactions is given by the nonlinear kinematic hardening rule (`NLK-rule')of the Armstrong and Frederick (1966)type as suggested by Chaboche et al.(1979),McDowell and Moosbrugger (1987)and Chaboche and Nouailhas (1989):X i C i 23a i 4X p À i 4"X Y i 1Y 2X W In the experimental results of the stress±strain responses during cyclic straining [Fig.5(a)and (b)]and strain-controlled ratchetting (Fig.7),the Bauschinger e ect is observed even within the Lu ders strain range.It can be described by the above mentioned NLK model.Fig.9.Assumed dislocation multiplication [&0 106cm À2;C 5Â1010cm À2;a 1in Eq.(7)and f m 0 10À5;f m 0X 1;l 20in Eq.(8)].Since slip modes in the strain hardening region would be di erent from in the preceding yield plateau,an additional term of back stress evolution,which is asso-ciated with slip systems becoming active when the homogeneous plastic deformation starts after the Lu ders deformation in the yield plateau,should be introduced.It is already well known that the use of the NLK rule leads to overprediction of the ratchetting strain accumulation because the recall (or the dynamic recovery)term is too active.Therefore,for the strain-hardening region,the following nonlinear kinematic hardening rule with a threshold (`NLK-T'rule proposed by Chaboche,1991)is chosen:X C 32 4X p À1À th () 4"X Y " 32X r Y IH where <>is the MacCauley bracket,and th is the threshold for the back stress .In this equation,the recall term is active only when " b th .The kinematic hardening due to the long-range dislocation interactions is expres-sed by the linear kinematic hardening as:X vu 23H H I 4X p Y IIwhere H H I is determined from the asymptotic tangent modulus of plasticity in the uniaxial stress±strain curve.Finally,the kinematic hardening is expressed as a summation of the above back stress components:1 2 vu for the yield pl te u r nge Y 1 a 2 vu for the str in h rdening r nge X' IP In order to describe the strain range dependence of cyclic hardening,the evolution of nonisotropic hardening surface g de®ned in the plastic strain space (Chaboche et al.,1979;Ohno,1982)was introduced:g 23 4p À X 4p À À 240Y IQ where and are the center and the radius of the surface.The evolution equations for and are given asX 1ÀC xr À4"X 4p À nd X C xr À4"X Y where À 23 4"X 4p À X 4X p Y nd W b a b Y IR C xr is a material parameters controlling the evolution rate.The isotropic hard-ening rule is given by the equation:R X m Q ÀR 4"X when g 0 nd d g d 4p X 4X p b 0Y R X 0when g `0or d g d 4p X 4X p 40Y W b a b Y ISwhere Q stands for the asymptotic value of the isotropic hardening stress R ,and m is a material parameter controlling the rate of its evolution.Since the isotropic hard-ening does not take place during the Lu ders band propagation,the following restriction for the evolution of the isotropic hardening stress is also imposed:R X 0when 4"`4v X IT As mentioned in the Section 2.2,the Lu ders strain 4v strongly depends on the strain rate 4"X ,therefore it is given by the equation:4v 4v 0 A log 104"X Y IUwhere 4v 0stands for the Lu ders strain at 4"X 1and A is a material constant.The drag stress D is assumed to be a function of the isotropic hardening stress R asD D 0 Q R D ÀÀD 0 Y D 0b D Y IV where D 0and D are the initial (at R 0)and the asymptotic (at R Q )values of the drag stress.This equation for the drag stress evolution allows us to describe the weaker rate-dependence at the strain hardening region compared to at the yield plateau,as well as the characteristic behavior of parallel stress±strain curves observed in the uniaxial tension tests at several strain rates.The stress±strain response in a material element during uniaxial tension and sub-sequent compression,as well as the evolution of the back stress and the isotropic hardening stress R ,calculated by this model are depicted in Fig.10.All the material parameters in this constitutive model for the mild steel,determined by stress-strain curve ®tting,are listed in Table 1.It is found from this ®gure that the model can describe the yield-point phenomena,as well as the rapid strain±hardening which begins just after the yield plateau and the Bauschinger e ect in the strain reversal.4.Veri®cation of the model by numerical simulationThe above constitutive model had been implemented in a FE code of in®nitesimal elasto-viscoplasticity using the linear triangular elements (FE formulation has been presented in a reference by Itoh et al.,1992).The numerical simulation was per-formed,using a FE mesh shown in Fig.11,for the stress±strain responses corre-sponding to the aforementioned experiments.Table 1Material parameters used in the calculationDislocation multiplicationKinematic hardening Isotropic hardening b =2.5Â10À9cm a 1=30.0Y 0=70.0MPa &o =1.0Â106cm À2a 2=10.0MPa,C 2=4000.0C xr =0.06C =3.0Â1011cm À2a =15.0MPa,C =330.0 0=0.002a =1.0 th =15.0MPa Q =120.0MPa f mH =1.0Â10À5H H I =1100MPa m =70.0f m =0.1D 0=160.0MPa l =20.0D a =80.0MPa Lu ders strain Elastic constants 4vH 0X 032A 0X 0025'when 4"X4À8s À1Young's modulus=190.0GPa Poisson's ratio=0.34vH 0X 012A 0'when 4"X <10À8s À1Rate sensitivity exponentn =25Fig.10.Calculated stress±strain response in a material element during uniaxial tension at strain rate of 0.1s À1and the subsequent compression,as well as the evolution of the back stress and the isotropic hardening stress R .10À8s À1The calculated stress±strain curves in uniaxial tension (corresponding to the experiments shown in Fig.3)is illustrated in Fig.12.In this ®gure,the sharp yield-point followed by the abrupt yield drop and the subsequent yield plateau,as well as the rate-dependence of ¯ow stress (especially in the yield plateau)and of the Lu ders strain,are well reproduced.The e ective plastic strain distributions at several stages of tensile deformation are illustrated in Fig.13(a)±(c).From these ®gures,it is found that the plastic strain appears ®rst at the ®llet and near the center of the specimen,then spreads over the specimen at relatively early stage of deformation.In these ®gures,the calculated plastic strain distribution is not homogeneous,but strong plastic strain localization,usually appearing in experiments,is not found.There would be several reasons for that,e.g.from the viewpoint of constitutive modelling,the assumed yield drop (i.e.the di erence between upper and lower yield stresses)was not large enough to predict the strong strain localization.It does not meanthe Fig.11.FE mesh for numerical simulation.present type of constitutive model cannot reproduce the strain localization and the Lu ders band propagation.In fact,Itoh et al.(1992)predicted strong strain locali-zation in the simulation of uniaxial tension,using almost the same model in terms of yield-drop modelling,but expressing larger yield drop.Another problem would be that the present numerical simulation used a quarter geometry model of the specimen,which obviously cannot predict properly the Lu ders band propagating from the one end of the specimen to the other end.Furthermore,FE mesh employed in the present work was not ®ne enough to properly resolve the Lu ders band front [the minimum element size in the FE mesh was 0.5mm in nodal distance which corresponds to 20grains of the material (26.5mm in diameter)].However,the above-mentioned fact that the calculated plastic strain covers the whole gauge sec-tion of specimen at relatively early stage of deformation [see Fig.13(c)]is not only attributed to problems in modelling and the discretization.Similar phenomenon is also found in some of the experimental results of tension tests on iron and steel that plastic strain spreads wide area of specimens even though the average gauge-length strain at that time is still quite small (e.g.refer to Imamura et al.,1971;Fujita and Miyazaki,1978).In spite of such problems in modelling and the numerical technique which should be improved in future,as far as the macroscopic stress±strain responses are con-cerned,we can conclude from the following results of numerical simulation that this constitutive model will well simulate the actual material behavior of cyclic plasticity.The predictions for the cyclic straining,the strain-controlled ratchetting and the stress-controlled ratchetting,which corresponds to the experimental results depicted in Figs.5±7,are shown in Figs.14±17.The slight softening appearing in the calcu-lation of the cyclic straining for the strain range of 0.005[in Fig.14(a)]resultsfromFig.12.Calculated stress±strain curves in uniaxial tension for the cases shown in Fig.3.the e ect of dislocation multiplication [increase of &m in the constitutive equation (5)],but not the contraction of the yield surface.In contrast,slight cyclic hardening is predicted for the case of the strain range of 0.01[in Fig.14(b)]by the evolution of isotropic hardening.The cyclic hardening characteristics in the strain-controlled ratchetting are well described by this model,although the predicted peak stresses for tension side are slightly larger than the experimental results.This model can also well simulate the stress-controlled ratchetting including the stress-rate e ect (in Fig.15)and Fig.13.Calculated e ective plastic strain distributions at several stages of deformation.。
INRIA Sophia-Antipolis
PROGRESS TOWARDS AN ELECTROMECHANICAL MODEL OF THE HEARTFOR CARDIAC IMAGE ANALYSISM.Sermesant,Y.Coudière,H.Delingette,N.AyacheINRIA Sophia-AntipolisProjet Epidaure2004,route des Lucioles06902Sophia-Antipolis,FranceABSTRACTWe present a3D numerical model of the heart ventricleswhich couples electrical and biomechanical activities.Wehave adopted a simpler formulation of previously publishedmodels to achieve a trade-off between biological accuracyand computational efficiency.To carry out computations,the FitzHugh-Nagumo equations are solved along with aconstitutive law based on the Hill-Maxwell rheological law.Ultimately,the parameters of this generic model will be ad-justed from the actual patient’s ECG and from the cardiacmotion measured in4D cardiac images.By including bio-logical and physical a priori knowledge,we expect to ex-tract quantitative ventricular function parameters from timesequences of cardiac images1.1.INTRODUCTIONThe clinical motivation of this work is the quantitative mea-surement of important ventricular function parameters fromcardiac images,like the ejection fraction,the myocardiumthickness and the local strain and stress.Those parametersare useful to detect ischemic zones,measure the pathologyextent and control the therapy effectiveness.The key idea of this paper is to build a“Beating HeartModel”which contracts under electrical excitation.Byfit-ting this model to cardiac image sequences we intend to re-cover those ventricular function parameters.The knowledgeof the heart function has greatly improved at the nanoscopic,microscopic and mesoscopic scales during the last decades,thus a global integrative work of this organ becomes con-ceivable[2].Although this is a work in progress,we be-lieve that the proposed framework should allow a better un-derstanding of the heart behavior by measuring quantitiessuch as stress that are beyond the pure geometric descrip-tion currently achieved and it should help interpret cardiacimages.Our“Beating Heart Model”is based on mathematicalsystems of(non-linear)partial differential equations,set onIn the deformable model framework,a model evolves under the influence of two energies:an External Energy which makes the modelfit the images and an Internal En-ergy which acts as a regularization term and can include a priori information(shape,physical properties,motion,...)Fig.2.Electromechanical model in a4D ultrasound image.In our approach,the computation of this External En-ergy at a surface vertex depends not only on the vertex loca-tion but also on its normal direction.Different type of forces may be applied depending on the image modality.We chose to combine intensity and gradient information with a region-based approach[7]applied to the intensity profile extracted at each vertex in its normal direction.It consists in defin-ing a region with a range of intensity values and thenfind-ing its boundary by looking at the voxels of high gradient value.The extent of the intensity profile is decreased in the coarse-to-fine process.Then,we apply a force which is proportional to the distance to the closest boundary point of the image from the considered point of the mesh surface.The volumetric nature of our model strongly decreases the importance of the image outliers in the motion estima-tion since it strongly constrains the geometric(for instance the thickness of the myocardium wall)and physical behav-ior.The Internal Energy corresponds to an electromechan-ical“Beating Heart Model”which is described in the next three sections:first,the anatomical data necessary to build the model,then the electrical model used to compute the action potential wave propagation andfinally the mechani-cal model used to compute the contraction triggered by the action potential wave.3.ANATOMICAL MODELTo build our model we need data regarding both the3D ven-tricular geometry and the musclefiber directions.Indeed, the anisotropy created by thosefibers intervenes in both the electrical wave propagation and the mechanical contraction. There are different ways to obtain thosefiber directions. We are currently using data from a dissected canine heart available from the Bioengineering Research Group2of the University of Auckland,New Zealand and from reduced-encoding MR diffusion tensor imaging(dtMRI)[8].In order to complete our anatomical model we also need data about the electrical network:the Purkinje network lo-as in[10],or as the result of some equilibrium equations that govern the conducting continuum,like in the so-called bidomain model[12].The anisotropy of the ventricles is taken into account through the diffusion tensor:,in a local orthonormal basis where is parallel to the fiber.is a scalar conductivity and the anisotropy ratio between the transverse and the axial conductivities.4.3.Results of the wave propagationSimulated isochrones of activation are presented(Fig.4),af-ter a wave was initialized at the apex,using a crude approx-imation of the Purkinje network and a slightly anisotropic diffusiontensor.Fig.4.Isochrones of activation(computed with a slightlyanisotropic diffusion tensor:)We can simulate different singularities that may corre-spond to pathologies by changing the conduction parame-ters(Fig.5),for instance introducing a strong conductivityanisotropy.Fig.5.Apparitions of activation singularities with a highlyanisotropic diffusion tensor.This time-dependent computed potential can then be usedas an excitation entry to the system describing the mechan-ical behavior of the myocardium.5.MECHANICAL MODELThe myocardium is composed of musclefiber bundles spi-raling around the two ventricles.It is a nonlinear viscoelas-tic anisotropic active material.The qualitative behavior ofthe electromechanical coupling is a contraction for a pos-itive action potential and an active relaxation for a nega-tive one.Moreover,the action potential also modifies thestiffness of the material.The model introduced in[13,14]by Bestel,Clément and Sorine captures this behavior.Theglobal muscle model is based on the Hill-Maxwell rheolog-ical law which includes a contractile element,a serieselement and a parallel element,as shown on Fig.6and is detailed in[15].EpFig.6.(Left)Hill-Maxwell rheological model.(Right)Simplified rheological model.In[14],Bestel,Clément and Sorine model the contrac-tile element,controlled by the action potential,as fol-lows:around10minutes whereas other existing electromechan-ical models take several hours of computation.We are in-vestigating optimization and a parallel implementation to reduce the computation time to allow an interactive use of this model in time series of cardiac images.The long term goal is to use this model to extract quan-titative parameters of the ventricular function from cardiac images.We can also simulate electrical wave singularities that may correspond to pathologies or test different loca-tions of electrical onset,useful for the implantation of pace-makers,for instance.Moreover,we will be able to simulate the influence of those electrical differences on the contrac-tion.We believe that this beating heart model could help understand the consequences of local electrical or mechan-ical failures on the global motion.Additional images and videos are available on the web3.Globally,recent measurements of the electrical activ-ity,fiber directions and motion reconstruction(from tagged MRI)on the same heart should help adjust the different pa-rameters of the model[16].We will work during the com-ing months on validating this method with an entire dataset coming from the same heart.Finally,a more accurate model may be considered,and different numerical techniques exist:the complete simula-tion process can be improved,always keeping in mind the balance between the complexity of the models(more equa-tions but more realistic)and the efficiency of the numerical methods(accuracy and speed).7.ACKNOWLEDGEMENTSThis work is a part of the multidisciplinary project ICEMA4 (standing for Images of the Cardiac ElectroMechanical Ac-tivity),which is a collaborative research action between dif-ferent INRIA projects and Philips Research France[17].8.REFERENCES[1]M.Sermesant,Y.Coudière,H.Delingette,N.Ayache,andJ.A.Désidéri,“An electro-mechanical model of the heart for cardiac image analysis,”in Medical Image Comput-ing and Computer-Assisted Intervention(MICCAI’01).2001, vol.2208of Lecture Notes in Computer Science(LNCS),pp.224–231,Springer.[2] A.McCulloch,J.B.Bassingthwaighte,P.J.Hunter,D.Noble,T.L.Blundell,and T.Pawson,“Computational biology of the heart:From structure to function,”Progress in Biophysics& Molecular Biology,vol.69,no.2/3,pp.151–559,1998. [3] A.F.Frangi,W.J.Niessen,and M.A.Viergever,“Three-dimensional modeling for functional analysis of cardiac im-ages:A review,”IEEE Trans.on Medical Imaging,vol.1, no.20,pp.2–25,2001.[4]X.Papademetris,A.J.Sinusas,D.P.Dione,and J.S.Dun-can,“Estimation of3D left ventricle deformation from。
A constitutive model for the anisotropic elastic-plastic deformation of paper and paperboard
A constitutive model for the anisotropicelastic–plastic deformation of paper and paperboardQingxi S.Xia,Mary C.Boyce *,David M.ParksDepartment of Mechanical Engineering,Centre of Materials Science and Technology,Massachusetts Institute of Technology,77Massachusetts Avenue,Cambridge,MA 02139-4307,USAReceived 20June 2001;received in revised form 19February 2002AbstractA three-dimensional,anisotropic constitutive model is presented to model the in-plane elastic–plastic deformation of paper and paperboard.The proposed initial yield surface is directly constructed from internal state variables comprising the yield strengths measured in various loading directions and the corresponding ratios of plastic strain components.An associated flow rule is used to model the plastic flow of the material.Anisotropic strain hardening of yield strengths is introduced to model the evolution in the yield surface with strain.A procedure for identifying the needed material properties is provided.The constitutive model is found to capture major features of the highly anisotropic elastic–plastic behavior of paper and paperboard.Furthermore,with material properties fitted to experimental data in one set of loading directions,the model predicts the behavior of other loading states well.Ó2002Published by Elsevier Science Ltd.Keywords:Paper;Paperboard;Plasticity;Yield;Yield surface;Constitutive behavior;Anisotropic1.IntroductionPaper and paperboard are two of the most commonly utilized materials in nearly every industry.Paper is formed by draining a suspension of fibers in a fluid through a filter screen to form a sheet of pulp fibers.Paperboard is in general composed of several pulp fiber sheets bonded by starch or adhesive material,and is usually a multi-layered structure.Schematics of typical paper and paperboard macrostructure and micro-structure are shown in Fig.1,which also depicts the common nomenclature for the three orthogonal di-rections of paper and paperboard.‘‘MD’’refers to the machine (rolling)direction,and ‘‘CD’’refers to the cross or transverse direction.The machine and cross directions form the plane of the structure,and ZD refers to the out-of-plane (or through-thickness)direction.Due to the continuous nature of the paper-making process,fibers are primarily oriented in the plane;furthermore,within the plane,fibers are more highly oriented in the MD than the CD.In this paper,to simplify notation,the 1-direction is used to represent the MD,the 2-direction for ZD and the 3-direction forCD.International Journal of Solids and Structures 39(2002)4053–/locate/ijsolstr*Corresponding author.Tel.:+1-617-253-2342;fax:+1-617-258-8742.E-mail address:mcboyce@ (M.C.Boyce).0020-7683/02/$-see front matter Ó2002Published by Elsevier Science Ltd.PII:S 0020-7683(02)00238-XThe preferential fiber orientation results in highly anisotropic mechanical behavior,including aniso-tropic elasticity,initial yielding,strain hardening and tensile failure strength:•Anisotropic elastic constants for paper and paperboard have been measured by several investigators (e.g.,Mann et al.,1981;Castegnade et al.,1989;Persson,1991;Koubaa and Koran,1995).Their data show that the through-thickness moduli are at least two orders of magnitude less than the in-plane mo-duli.In-plane data of Persson (1991)and Stenberg et al.(2001a)show that moduli in the MD are 2–4times greater than those of the CD.•Persson’s (1991)data on paperboard also shows that the initial yield strength ($proportional limit)in the through-thickness direction is two orders of magnitude lower than the in-plane initial yield strength values.Stenberg’s data (Stenberg et al.(2001a,b);Stenberg,2001)on multi-layer paperboard and single-layer pulp shows similar results.Within the plane,these data show that the initial yield strength of paper and paperboard in the MD is typically greater than that in the CD by a factor of 2–4.Stenberg’s data (Stenberg,2001)also show an asymmetry in the initial yield strength for in-plane tension and compres-sion in both the machine and cross directions.•The in-plane tensile stress–strain curves of Persson (1991)and Stenberg (2001)show substantial strain hardening in which the yield strength increased by more than a factor of two after a strain of less than 5%.The in-plane strain hardening is also highly anisotropic.The percentage strain hardening achieved in MD tension is greater than that obtained in CD tension.•The Persson (1991)and Stenberg (2001)data also show that the in-plane tensile failure strength in the MD is greater than that of the CD by a factor of 2–4.•Biaxial failure stress loci (Gunderson,1983;deRuvo et al.,1980;Fellers et al.,1981)show substantially different failure strengths in machine and cross directions.These data also show that failure tends to be dominated by one or the other of these two directions when subjected to combined loading in both di-rections.Some material models have been proposed to describe the mechanical behavior of paperboard.These models fall into roughly three categories:network models,laminate models,and anisotropic models of the yield surface and/or the failure surface.Perkins and Sinha (1992)and Sinha and Perkins (1995)described a micromechanically based network model for the in-plane constitutive behavior of paper.A meso-element was constructed to represent theFig.1.Schematics of paper and paperboard macrostructure and microstructure.4054Q.S.Xia et al./International Journal of Solids and Structures 39(2002)4053–4071Q.S.Xia et al./International Journal of Solids and Structures39(2002)4053–40714055 microstructure of thefibrous paper network.The mechanical response of the meso-element depends on the fiber properties and properties of the inter-fiber bonds.They found the inelastic behavior of the inter-fiber bonds to play a crucial role in the overall in-plane inelastic behavior of paper.Stahl and Cramer(1998)also developed a network model for low densityfibrous work models can incorporate microlevel mechanisms,such as inter-fiber interaction and bonding.While these models begin to elucidate the un-derlying mechanisms of deformation,they do not provide a continuum-level description of paper or paperboard.Page and Schulgasser(1989)and Schulgasser and Page(1988)developed models of paperboard based on classical laminate theory.While this type of model can predict the elastic response well,it was not extended to capture the anisotropic yielding and subsequent strain hardening response.Gunderson(1983),Gunderson et al.(1986),deRuvo et al.(1980)and Fellers et al.(1981)each used the Tsai–Wu quadratic yield condition to model the failure loci they obtained experimentally.The quadratic nature of this type of model has many shortcomings when applied to paper and paperboard.Experimental data showed the biaxial failure locus to be distinctly non-quadratic.Arramon et al.(2000)developed a multidimensional anisotropic strength criterion based on Kelvin modes that captures the non-quadratic failure envelope.They applied the model to form a strength envelope for paperboard by constructing tensile and compressive modal bounds.However,these efforts only acted to studyfinal failure and did not attempt to study initiation of yield or subsequent strain hardening.In this research,a general three-dimensional constitutive model of the anisotropic elastic–plastic be-havior of paper and paperboard is proposed.The initial elastic behavior is modelled to be linear and or-thotropic.The onset of plasticflow is captured by a non-quadratic yield surface.The yield surface is taken to evolve anisotropically with a scalar measure of plastic strain,with plasticflow modelled using an as-sociatedflow rule.The model is detailed in the following sections,and numerical results are compared to experimental data.2.Experimentally observed behavior2.1.Elastic–plastic behavior of TRIPLEX paperboardDepending onfiber type,fiber density and the chemical/mechanical treatment,the elastic–plastic be-havior of different types of paper and paperboard differs in detail.However,general characteristics of the response remain similar.In this contribution,the anisotropic elastic–plastic behavior of paper and paper-board is illustrated using TRIPLEX e1paperboard as an exemplar material.TRIPLEX e is afive-layer paperboard:the three-layer core is constructed from mechanically processed softwood pulp(commonly referred to as‘‘mechanical’’pulp),and two outer layers(sandwiching the core)are constructed from bleached kraft pulp(commonly referred to as‘‘chemical’’pulp).Stenberg et al.(2001a,b)and Stenberg (2001)conducted an extensive experimental investigation documenting the stress–strain behavior of TRIPLEX e.Note that the outer chemical pulp layers are typically stiffer and stronger than the inner mechanical layers;however,these layers cannot be separately produced for individual evaluation.There-fore,the behavior of TRIPLEX e material will be presented in terms of its effective composite behavior.In principle,testing and modeling can be applied separately for each distinct paper lamina.The experimental results of Stenberg et al.(2001a,b)and Stenberg(2001)are reviewed below.1TRIPLEX e is a trademark of STORA-ENSO,Finland and Sweden.2.1.1.In-plane behaviorThe in-plane uniaxial tensile stress–strain curves for the MD,the CD and an orientation 45°from the MD are plotted together in Fig.2.These stress–strain curves clearly depict the anisotropic in-plane elastic,initial yield and strain hardening behavior.There is a factor of 2–3difference in the modulus and initial yield strength between MD and CD.Hardening achieved in MD (flow strength increases by 300%over a strain of 2%)is higher than that in CD (flow strength increases by 200%over a strain of 5%).MD–CD shear properties are deduced from the 45°test result.In-plane lateral strain (CD)vs.axial strain data for MD-tension and similar data for CD-tension were also recorded by Stenberg (2001),corresponding to their in-plane stress strain curves.Upon subtracting the respective elastic strain components,the lateral plastic strains for both the MD and CD tension cases are computed and shown vs.the respective axial plastic strains in Fig.3.These two curves indicate that for both test orientations,the ratio between lateral plastic strain and axial plastic strain is nearly constant until final fracture.This data provides information for later construction of the plastic flowrule.4056Q.S.Xia et al./International Journal of Solids and Structures 39(2002)4053–4071Tensile loading/unloading/reloading data (Persson,1991;Stenberg,2001)show that after various amounts of plastic strain,upon unloading,the elastic tensile modulus is nearly unaffected by plastic strain,consistent with traditional models of elasto-plasticity.Fig.4shows the in-plane compression stress–strain curves for the MD and CD directions.Note that global specimen buckling was constrained in these tests.These data show that compressive yield is an-isotropic.Furthermore,a comparison of Figs.2and 4shows a yield strength difference between tension and compression,with the compressive yield strengths being smaller than those in tension by 65%and 25%,for MD and CD,respectively.The anisotropic in-plane elastic–plastic properties obtained from these tests are summarized in Table 1.2.1.2.Out-of-plane behaviorStenberg et al.(2001a,b)experimentally obtained the out-of-plane stress–strain behavior of paperboard using a modified Arcan design (Arcan et al.,1978).Nominal stress–strain curves were obtained for TRI-PLEX e under various through-thickness loading conditions.Representative ZD tensile and through-thickness shear (ZD–MD)stress–strain curves obtained by Stenberg et al.(2001a,b)are shown in Fig.5.The stress measure is force per unit initial cross-sectional area;the x -ordinate is the nominal strain,defined as the relative normal/shear separation of the top and bottom surfaces of the laminate,divided by the initial laminate thickness.In the tensile curve,at the earliest stage ofdeformation,the stress increases linearly with strain,exhibiting a composite modulus of E 0ZD ¼18:0MPa.The stress–strain relation shows a small amount of pre-peak non-linearity before reaching a peak stress of 0.4MPa.After the peak,the stress–strain curve exhibits pronounced softening.Features similar to those of the through-thickness tensile curve are observed for the through-thickness shear curve.Thecomposite Table 1Experimental results of uniaxial tensile tests (Stenberg,2001)Elastic modulus(GPa)Poisson’s ratio Tensile yield strength (MPa)Plastic strain ratio,d p ?=d p k Compressive yield strength (MPa)MD5.60.3712.0)0.57.3CD2.00.12 6.5)0.133 5.045° 4.18.0Q.S.Xia et al./International Journal of Solids and Structures 39(2002)4053–40714057transverse shear modulus is observed to be G 12¼34MPa,and the peak shear stress is 1.1MPa.The through-thickness ZD–CD shear stress–strain curve shows similar features.Through-thickness tensile/shear testing conducted within a scanning electron microscope (Dunn,2000)on the same material reveals the nucleation of multiple inter-laminar microcracks near the peak stress,followed by their growth and coalescence,resulting in the observed softening.Similar results have been obtained for tests involving combined through-thickness tension and shear (Dunn,2000;Stenberg et al.(2001a,b)).Therefore,the observed peak stress and subsequent softening for through-thickness loading are due to delamination of the paperboard and will not be further considered here.This inelastic through-thickness behavior is modelled in a separate related study (Xia et al.,in preparation).It is also observed that the amount of lateral (in-plane)strain generated during the through-thickness tensile loading is negligible,indicating that Poisson’s ratios m 21and m 23are close to zero.Through-thickness compression loading/unloading stress–strain curves were also obtained by Stenberg (2001).Up to a nominal compressive strain of 3%,the compressive stress increases linearly with strain.With larger strains,the stress starts to increase exponentially with strain.The data also show that only a small amount of permanent deformation is remained after unloading from a peak strain level of more than 20%,indicating non-linear elastic ZD compressive behavior up to moderately large strains.These obser-vations of through-thickness compressive behavior will be incorporated into the modeling work.The anisotropic linear elastic out-of-plane properties are summarized in Table 2.3.Constitutive modelA three-dimensional,finite deformation constitutive model for paper and paperboard pulp layers is proposed.The model will take the in-plane behavior to be elastic–plastic and the out-of-plane behavior to be elastic.Due to the assumption of elastic out-of-plane behavior,the application of the model alone will be limited to predominant in-plane loading.However,when the model is combined with interlaminardeco-Table 2Elastic out-of-plane properties (Stenberg et al.(2001a,b))Elastic modulusE 0ZD (MPa)Poisson’s ratio m 21Poisson’s ratio m 23Shear modulus G 12(MPa)Shear modulus G 23(MPa)Stiffening constant a 18.0)0.0055)0.003534.026.0 5.44058Q.S.Xia et al./International Journal of Solids and Structures 39(2002)4053–4071hesion models(Xia et al.,in preparation),a general-purpose tool is achieved for modeling behavior under significant out-of-plane loading,such as occurs during converting processes and the in-service behavior ofa broad class of paper and paperboard products.3.1.Stress–strain relationshipFirst,the total deformation gradient F at a material point within a lamina is multiplicatively decom-posed into an elastic part and a plastic part:F¼F e F p;ð1Þwhere F p represents the accumulation of inelastic deformation.Although the maximum in-plane strain level in traditional applications of paper sheets is small,we adopt the presentfinite deformation formulation so that the model can be easily applied to applications such as paperboard converting processes,which generally involvefinite rotations of paperboard layers and may exhibit moderately large,but highly lo-calized in-plane strains.The evolution of F p is given by_F p¼L p F p;ð2Þwhere L p is the plastic velocity gradient,which will be defined by theflow rule.The elastic strain is obtained by using the elastic Green strain measure:E e¼12F eT F eÂÀI Ã;ð3Þwhere I is the second-order identity tensor.The second Piola–Kirchoffstress measure,T,relative to the plastically deformed configuration F p,is then calculated using the linear relationship: T¼C E e½ ;ð4Þwhere C is the fourth-order stiffness tensor,which is taken to be orthotropic.Values of the components of C are defined by the orthotropic elastic moduli.To model the through-thickness non-linear elastic com-pressive stress–strain relationship,the through-thickness engineering elastic constant,E ZD,is taken to be an exponential function of the ZD strain under compression as follows:E ZD¼E0ZD expðÀaE e22Þ;ð5Þwhere E0ZD is the initial elastic modulus in ZD,E e22is the ZD elastic Green strain component,and a is aconstant determined byfitting the compressive through-thickness stress–strain curve;its value is listed in Table2.The stiffness tensor C under ZD compression is in turn determined assuming constant Poisson’s ratios.The Cauchy stress,T,is calculated from its relation to the second Piola–Kirchoffstress by T¼ðdet F eÞF eÀ1TF eÀT:ð6Þ3.2.Yield conditionThe through-thickness strengths(tensile,compressive,shear)of paper and paperboard materials are typically two orders of magnitude lower than those observed in the plane.Therefore,the through-thickness stress components play little role in the inelastic deformation and failure of paperboard under in-plane loading.2Furthermore,from investigation of the mechanisms of deformation and failure of paperboard2Under significant in-plane compression or very large through-thickness compressive strain,this may not be strictly true.However, for the present,these scenarios will not be considered further.Q.S.Xia et al./International Journal of Solids and Structures39(2002)4053–40714059under through-thickness loading (when through-thickness compression is not dominant),it is clear that the majority of through-thickness inelastic deformation occurs in the form of interlaminar microcracking and delamination of the discrete pulp layers,as opposed to inelastic through-thickness deformation distributed quasi-homogeneously within laminae.Thus we can assume that only the in-plane stress components will drive the in-plane inelastic deformation of the pulp layers.Additionally,in classical metal plasticity and in plasticity-based models of yielding in polymers,the deviatoric stress is taken to drive yield because the underlying deformation mechanisms are governed by shear (for example,dislocation glide in crystalline metals).However,in the case of paper,there is no evidence that yield and subsequent plastic flow are driven by deviatoric stress.Micromechanically,yielding is governed by various forms of inter-fiber interactions.Based on these considerations,the total stress will be taken to drive the yield condition described in this research.In order to experimentally define an in-plane yield surface for paper,multi-axial data is required.Al-though the anisotropy of the yield surface is well-recognized,due to numerous studies of the uniaxial behavior in different directions,such as that reported in Fig.2,a literature search reveals virtually no data on the initial and evolving multi-axial yield surfaces of paper and paperboard.However,several researchers have obtained biaxial failure surfaces of paperboard under combinations of MD and CD normal stress.Fig.6shows a representative biaxial failure locus (deRuvo et al.,1980).Similar data have been reported in work by Fellers et al.(1981)and Gunderson (1983).One common feature observed in these failure loci is that the data points tend to form prominent sub-groups,with each sub-group lying on a nearly straight line.For example,for low,but non-zero values of MD stress,failure occurs at nearly the CD uniaxial tensile failure strength;similarly,for low,but non-zero values of CD stress,failure occurs at roughly the MD uniaxial tensile failure strength.This observation suggests that the experimental biaxial failure locus can be well captured by a set of straight lines in two-dimensional biaxial stress space and can be generalized to planes in three-dimensional space.Karafillis and Boyce (1993)and Arramon et al.(2000)developed yield surface and failure surface models,respectively,which capture this non-quadratic feature.In stress space,these lines or planes can be defined by their minimum distance to the origin,together with their corre-sponding normal directions.Given that a comprehensive set of experimental data is generally unavailable (and indeed,is a challenging task to obtain)to determine the full surface,we hereby assume that the yield surface exhibits the same characteristic features observed in the failure surface.Therefore,the yield surface is taken to be constructed of N sub-surfaces,where N K is the normal to the K th such surface,defined in the material coordinates formed by MD,CD and ZD.S K is the equivalent strength of the K th sub-surface,defined by the distance from the origin to each sub-surface,following its normal direction.Thus,the following form of yield criterion isproposed:Fig.6.Biaxial failure surfaces (deRuvo et al.,1980).4060Q.S.Xia et al./International Journal of Solids and Structures 39(2002)4053–4071fðT;S KÞ¼X NK¼1vKTÁN KS K!2kÀ1;ð7Þwhere v K is the switching controller withv K ¼1;if TÁNi>0;0;otherwise;&ð8ÞT is the second Piola–Kirchoffstress measure relative to the F p configuration,and2k is an even integer.N K is the normal of the K th sub-surface,defined relative to the material coordinates.A schematic of a four sub-surface system for biaxial loading,with zero in-plane shear stress,is shown in Fig.7.The normals and corresponding sub-surface strengths are illustrated.The parameter2k is taken to be equal to or larger than4,indicating a non-quadratic yield surface.Fig.8shows the effect of different2k values in controlling the shape of the yield surface in the biaxial stress plane for this simplified four sub-surface system.Higher2k-values give rise to sharper corners between adjacent sub-surfaces and reduce the curvature over increasing central portions of each sub-surface.A schematic of a six sub-surface yield surface,with non-zero in-plane shear stress T13,is shown in Fig.9.Thisfigure graphically illustratestheFig.8.Effect of constant2k on the shape of yield surface.Q.S.Xia et al./International Journal of Solids and Structures39(2002)4053–40714061normals and corresponding equivalent strengths of the sub-surfaces.For this yield surface,the six normals are taken to be of the following form:N K¼X3i;j¼1N Kije i e j;K¼I;...;VIð9ÞandN K ji ¼N Kij;K¼I;...;VI;ð10Þwhere e i are the basis vectors for the material coordinates formed by the MD,CD and ZD.Here,the sub-surface normal index K ranges over the six Roman numerals I–VI.Because out-of-plane stress components are assumed to have no effect on the plastic deformation within a lamina,components of N K involving the2-direction(i.e.,the ZD direction)are set to be zero.This results in elastic through-thickness behavior,as proposed.Determination of each non-zero entry of these matrices will be discussed in Section3.3.3.3.Flow ruleThe plasticflow rule is defined asL p¼D p¼_ c K;ð11Þwhere L p is the plasticflow rate and D p is the symmetric part of L p.For paper and paperboard,the in-plane plastic strains(even at failure)are small;therefore we take the skew part of L p to vanish,or W p¼0,as a simplification.K is theflow direction,and_ c is the magnitude of the plastic stretching rate.K is a second-order tensor with unit magnitude:K¼b K=k b K k;ð12Þwherek b K k¼ffiffiffiffiffiffiffiffiffiffiffiffiffib KÁb Kp¼ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiX3i;j¼1b K ij b K ijv uu t:ð13ÞFig.9.Schematic of yield surface for general in-plane loading,showing six sub-surfaces.4062Q.S.Xia et al./International Journal of Solids and Structures39(2002)4053–4071In Fig.3,the in-plane lateral plastic strain vs.axial plastic strain data showed that the ratio between these two plastic strain components is nearly constant for both the MD and CD simple tension cases.These ratios are taken to define the normal directions of the two respective sub-surfaces of the tensile quadrant of the biaxial yield surface,in the absence of shear stress.Thus,the plasticflow of the material is taken to follow an associatedflow rule:b K¼ð14ÞWith the yield condition defined in Eq.(7),b K can be further calculated asb K¼o T ¼X NK¼12k K2kÀ1KvKN KS K;ð15ÞwhereK K¼TÁN KS K:ð16ÞFor the six sub-surface yield surface shown in Fig.9,assuming an associatedflow rule,the sub-surface normals N K,K¼I;...;VI are defined using the corresponding plastic strain ratios.For example,for sub-surface I(one)in Fig.9,the plastic strain ratio from the MD simple tension test is nearly constant at)0.5. The two non-zero components of N I are then determined by solving the following two equations: N I3311¼À0:5ð17Þand(to make a unit normal)ðN I11Þ2þðN I33Þ2¼1;ð18Þwhich gives N I11¼2=ffiffiffi5pand N I33¼À1=ffiffiffi5p.Similarly,the plastic strain ratio from the CD simple tension testis nearly constant atÀ2=15,giving N II11¼À2=ffiffiffiffiffiffiffiffi229pand N II33¼15=ffiffiffiffiffiffiffiffi229p.With appropriate experimentaldata,similar calculations can determine the normals of each of the sub-surfaces.However,currently,there is no experimental data for the plastic strain ratios for compression in either the machine or cross directions. For the four sub-surface biaxial yield surface shown in Fig.7,the normals for the two sub-surfaces in the compressive quadrants,IV and V,are assumed to have normal directions parallel to those of corresponding tensile sub-surfaces,I and II,respectively,as seen in thefigure,but with generally differing strength levels. For the normal of the sub-surface representing yielding under positive pure shear stress(T13¼T31¼0;T ij¼0otherwise),the two non-zero components are N III13and N III31.Due to the symmetry of shear stress,N III 13¼N III31ð19ÞandðN III13Þ2þðN III31Þ2¼1:ð20ÞThus we have N III13¼N III31¼ffiffiffi2p=2,and the normal of the other sub-surface representing yielding undernegative pure shear stresses is taken to be N VI¼ÀN III.In summary,the components of the normals for the six sub-surfaces determined in the material coordinates are given in Table3.3.4.Strain hardening functionsTo capture the anisotropic strain hardening observed for the in-plane behavior,the equivalent strengths, S K,of each sub-surface are taken to evolve with the accumulated equivalent plastic strain,i.e.。
各向异性板料屈服轨迹的研究
16
图4 2024·O铝合金板三种艟变下不同理论屈服准则与试验屈服点之间的误差 Fig.4 Difference between experimental yield points of 2024—0 aluminum alloy sheet
Key words:biaxiai tensile test:cruciform specimen=stress-strain CUFVe;yield locus
近年来,各向异性对板料塑性变形行为的影响引起国内外学者的广泛注意,其影响作用主要表现在, 各向异性使板料在不同方向上的力学性能产生差异,对板料的屈服行为包括初始屈服和后继屈服均有显 著影响,继而影响板料的本构关系。
假设试验点坐标为(x。,J,。),试验点和理论屈服轨迹曲线的法向距离为d,,则
万:窆下善≮
(1)
川4(x?+y?)
其中疗为试验点的个数。
由图1、图2可见,对于SPEN钢板,在本文的实验条件下,所研究的几个理论屈服准则中,Hosford 各向异性J茁服准则的理论轨迹与实验结果最为接近,Hiil48准则与实验结果相差最大,此外一向被视为只 适用丁.各向同性材料的Mises准则与实验结果也较为接近。其他几种各向异性屈服准则如Barlat89、Gotoh、 Hill79、Hill90得到的理论轨迹与试验点的差别介于Hill48和Hosford准则之间,而且彼此相差不大。
with existing yield criteria loci
3.结论
1)对丁.SPEN钢板来说,Hosford各向异性屈服准则得到的理论屈服轨迹与实验屈服轨迹符合得最 好,其次是Mises屈服准则,Hill48屈服准则最差。
镍基单晶合金叶片疲劳寿命预测方法研究
镍基单晶合金叶片疲劳寿命预测方法研究潘冬;杨晓光;胡晓安;石多奇【摘要】研究了3种针对镍基单晶合金各向异性低循环疲劳寿命建模的方法,分别为基于单晶合金弹性模量与晶体取向相关性的方法,与各向异性屈服函数相关的方法和传统滑移系的方法.对基于屈服函数的方法进行了修正以将其应用于单晶合金.利用公开文献中DD3单晶合金的低循环疲劳数据对修正的模型进行了验证,并对采用这3种方法得到的数据进行了比较.结果表明:修正的疲劳寿命模型和基于取向函数的寿命模型的预测结果与试验数据相比基本落在3倍分散带内,而采用基于滑移系的方法所得结果在4倍分散带内.基于屈服函数的修正模型和另外2种模型均可以较好地与3维有限元应力分析直接衔接,便于涡轮叶片结构级的寿命预测.【期刊名称】《航空发动机》【年(卷),期】2014(040)003【总页数】4页(P45-48)【关键词】镍基单晶合金;寿命预测;疲劳;各向异性;叶片;涡轮【作者】潘冬;杨晓光;胡晓安;石多奇【作者单位】中航工业燃气涡轮研究院,成都610500;北京航空航天大学能源与动力工程学院,北京100191;北京航空航天大学能源与动力工程学院,北京100191;北京航空航天大学能源与动力工程学院,北京100191;北京航空航天大学能源与动力工程学院,北京100191【正文语种】中文【中图分类】V231.95镍基单晶高温合金已经广泛应用于涡轮叶片上[1]。
由于消除了晶界,使其热、疲劳和蠕变性能得到了显著提高[2]。
但材料的各向异性特征给强度和寿命评估带来了挑战;基于各向同性的强度理论和寿命模型不足以给出符合工程精度的预测结果[3]。
且由于实际叶片处于多轴应力状态下,发展既能考虑各向异性,又能处理多轴应力的寿命模型对于单晶叶片的设计具有十分重要的现实意义。
针对单晶合金叶片高温低周疲劳损伤研究,Li S X[4]和石多奇等[5-6]通过对各向同性材料的宏观唯象疲劳寿命模型进行扩展,构造1个取向函数来考虑材料的各向异性的影响;Swanson[7]和岳珠峰等[8-9]基于材料细观塑性理论,研究材料特定滑移系的塑性滑移规律,建立晶体滑移疲劳寿命模型;Roland等[10]通过对应力-寿命法进行扩展,构建基于Hill屈服函数的等效应力考虑定向凝固合金的各向异性的影响。
涡轮叶片(发动机)形状
∙Keeping pace with new technology turbine blades, turbine buckets, w e are committed to meet the needs of the most demanding gas turbine operators.∙Gas turbine Blades, turbine buckets with f ully turbulated cooling holes.∙Small or large Z form design turbine blades, turbine buckets.∙Gas turbine blades, turbine buckets w ith peripheral cooling schemes.∙Cooling holes produced within the casting of gas turbine blades, turbine buckets.∙Our cutting edge machines capable of running over 100 HP manuf acture turbine blades and turbine vanes that reduces normal delivery cycles and costs by half.∙Turbine blades, turbine buckets w ith c omplex serration machining and modified pin s lots to reduce vibration and increase part lif e.∙Precision, super alloy investment castings developed for all row turbine blades, turbine buckets.∙Gas turbine blade, turbine bucket tip cutbacks to meet any modification requirements.∙Hydraulic f ixtures and our state of art CNC machines can produce turbine blade lengths of up to 60 inches.∙Corrosion resistant bond coating applied on turbine blades, turbine buckets to provide high adhesion and low oxide content.Ceramic top coating applied on turbine vanes, turbine nozzles for high temperature resistant thermal cycling, reduction of part strain and stress.Expert Information | Service | New | About BAM∙∙∙∙∙∙∙∙∙Working groupModelling and Simulation of Mechanical Behaviour of Materials Examples of activities of the working group∙Physically based material modelling∙Determination of material parameters, model verification∙High speed loading/cutting∙Loading in a turbine blade∙Notch in a single crystal∙Microsystem technology, flip chip∙Determination of the elastic constants of anisotropic materials Physically based material modellingHere in the case of Nickel base superalloys with a high volume fraction of the γ' precipitate phase. Nickel base superalloys, also as single crystals, are widely used for hot section of turbine blades in power plants or aero engines. A specific constitutive law has been developed and implemented in an FE code, which explicitly takes into account the intricate interactions between dislocations and precipitates.Representation of the γ' precipitates (in grey) in an octahedral {111} slip plane and of the main dislocation mechanisms of a singlecry stalline fcc superalloy:1. Filling of matrix channels with dislocation half-loops (blue)2. Shearing of matrix and precipitates by large dislocation segments (red)3. Multiple cross slip between phase boundaries, resulting in macroscopic cubic slip (purple)4. Climbing of dislocation loops around the precipitates, resulting in a recov ery of the internal stresses (green)The complex interactions between dislocations and precipitates and the stress fields generated by the interface dislocations are responsible for essential features of the macroscopic deformation behaviour of these alloys. Some of them are listed below:∙importance of cubic slip and its relation to the crystal orientation∙recovery mechanisms∙strength degradation due to directional coarsening of the precipitatesDetermination of Material Parameters, Model VerificationSimulation of the stress-response with a viscoplastic constitutive model at a non-isothermal, axial-torsional, cyclic straning for the verification of the model:Comparison of the experiment on a hollow specimen (lef t) with the numerical simulation (right), 15 min hold-times at 850 °C, material: IN738LCThe model was exclusively adapted to isothermal, uniaxial tensile, LCF and creep tests. The physical control of the optimization was realized by relating of material parameter groups to different hardening and softening phenomena in the material behaviour. One set of material parameters was determined for each testing temperature.High Speed Loading/CuttingThe behaviour of metallic materials at high loading rates is characterised primarily by a thermal softening due to the fact that the loading time is too short for a sufficient heat flow. The softening can lead to the formation of shear bands. For verification of the constitutive models (e.g. Johnson-Cook model) experiments and finite element (FE) simulations are carried out on notched flat specimens.FE simulation (ABAQUS/explicit) of the shear band f ormation in a notched f lat specimen, the stiffness of the test machine is considered by truss elementsAn industrial application of the material behaviour under high loading rates is, beside impact problems, for example the high speed cutting process with the formation of shear bands atchip segmentation. For the optimisation of the cutting process parameter, FE simulations with appropriate constitutive models for deformation and damage are required.FE simulation (ABAQUS/explicit) of the dev elopment of a segmented chipStress in a Turbine BladeViscoplastic FE-analysis (ABAQUS/standard) of a cross-section of an internally cooled turbine blade for the determination of the stress distribution in the blade:Distribution of the normal stress in the axial direction of the turbine blade (FE-model of Siemens, KWU)Notch in a Single CrystalSimulation of the damage behaviour at notches in single-crystalline superalloys under cyclic loading at high temperature:Location of macro-crack initiation at a notch in a circumf erentially notched specimen with (001)-orientation made f rom aprecipitation-hardened, f ace-centered cubic single cry stal (SC16, 950 °C, right) and comparison with the results of a FE-simulation with a cry stallographic model (lef t)Micro-System Technology, Flip-ChipSimulation of the strain situation at a crack in the interface of a solder bump, which is placed between the silicon chip and a ceramics substrate:Field of the accumulated inelastic strain in a cracked structureThe electrical failure of such an electronic structure occurs not before a large crack has been formed.Determination of the elastic constants of anisotropic materialsThe elastic constants of materials can be accurately determined from the measured resonance frequencies of freely vibrating specimens. The procedure is described in the ASTM standard E1875 for isotropic materials.The case of anisotropic materials, like, e.g. single crystals, is much more demanding. First, the number of required independent constants is higher. In addition, the eigenmodes often consist of a mixture of flexural and torsional components. Hence, the classical formulae, which have been derived under the assumption of isotropy, do not apply. As an alternative, the eigenmodes can be evaluated by the Finite Element Method for given elastic constant s.By iteratively varying the independent elastic constants until an agreement is obtained with the measured resonance frequencies, the actual elastic constants of an anisotropic material can be estimated. Eventually, the experimentally obtained resonance spectrum can be completely interpreted (see Figure below).With the apparatus available at the division, resonance measurements can be performed up to 1900 °C, allowing for the characterisation of the dependency of the elastic constants upon the temperature up to 1900 °C.Interpretation of the spectral curv e of a single cry stal superalloy with help of FEMService∙Presentations and publications can be found in the "PUBLICA" database∙Posters of the Division 5.2∙EventsBack to Division | back to working group2011-01-04topDr.-Ing.Bernard FedelichUnter den Eichen 8712205 Berlinphone:+49 30 8104-3104email:***********************SecretariatUnter den Eichen 8712205 Berlinphone:+49 30 8104-1529fax:+49 30 8104-1527How to find BA M转发至微博网易首页-新闻-体育-NBA-娱乐-财经-股票-汽车-科技-手机-女人-论坛-视频-博客-房产-家居-教育-读书-游戏-微博|免费邮箱- 通行证登录网易论坛网易首页>> 网易新闻>> 网易新闻论坛>> 网上谈兵>> 武器装备[登录] 签到帮助返回武器装备版12后页末页GO共27829浏览57回帖狙击兵等级:0级积分:3911在线:171小时发帖:1465篇TA的微博[个人信息] [发送信息] [只看楼主] [举报] [到微生活转转] 管理置顶精华推荐相关迁移加锁限制下沉修改删除加黑IP报名打印楼主2008-02-09 16:25:03涡轮轴等燃气涡轮航空发动机技术贴涡扇发动机的贴子本人以前发过了.现在本贴重点讨论涡轮喷气发动机和涡轮轴发动机.下图就是涡喷发动机:由于涡轮喷气发动机的推进效率低,能量损失大,耗油率高,因此,为提高推进效率,在带动压气机的涡轮之后,又加一个涡轮,用来带动对内外涵道气体同时进行增压的压气机(通常叫做风扇)。
智能材料结构中力与多物理场耦合理论及结构损伤断裂理论
智能材料结构中力与多物理场耦合理论及结构损伤/断裂理论本研究方向的成员主要基于多物理场理论和超声波相关理论技术,利用新型的微纳米级的铁电功能材料,创新地提出了针对各类材料与结构(尤其是航空材料结构)中的毫米级的损伤进行的检测和监测相关理论与技术。
团队形成了较强的凝聚力和良好的学术风气,产生了较高水平的科研成果,以下为研究骨干在近几年以来已完成的代表性阶段成果:(1)超声相控阵损伤检测中PFC换能器的研究项目负责人:骆英研究骨干:王自平、赵国旗、韩伟、虞波研究了PZT、PMN-PT等系统压电陶瓷的组成、制备技术和性能,创新地提出了用有序生长制备技术制造压电纤维复合材料及智能驱动/传感器件的新方法。
建立了一种基于新型PFC相控阵超声驱动/传感器件的超声相控阵检测系统,并成功应用于金属结构和混凝土结构的损伤检测,研制了接近国际水平的相控阵超声检测系统的原型机。
PFC片状驱动/传感器用于金属结构检测的PFC超声相控阵换能器用于混凝土结构检测的超声相控阵换能器d 、a 、N 变化时的指向性分析阵元参数变化时的波场分析基于PFC 超声相控阵驱动/传感器件的相控阵检测结果 (2)基于声发射(AE)技术的结构损伤检测方法研究骨干:骆英、顾爱军、Adudrum Marfo 、刘红光、欧晓林AE 技术作为一项独特的无损检测方法在各工程领域发挥着巨大的作用,在土木工程领域也显示出巨大的潜力。
研究了钢筋与混凝土间粘结滑移的声发射特性,并开始应用于预应力混凝土结构、钢结构、玻璃幕墙等结构的无损检测中,研究将数字图像相干法(DIC)与AE监测技术相结合,进一步验证损伤监测的准确性,为工程结构声发射检测中利用AE特性进行损伤识别奠定了基础。
CFRP碳纤维加固混凝土开孔板损伤监测桥梁声发射检测声发射与DIC检测方法比较试验装置基于Gabor小波变换理论的声发射无损检测及信号处理技术a) b) c) d)利用DIC方法测得的水平应变场判断混凝土裂缝的形成和扩展a)、b)、c),应变场的演化,d)宏观裂缝(3)基于新型应变梯度传感器的结构损伤监测技术项目负责人:骆英研究骨干:徐晨光、李康、桑胜、王晶晶、李兴家在近10年跟踪前沿研究新型铁电功能材料及智能器件的基础上,揭示微米级挠曲电材料的力/电能量转换关系,基于微米级挠曲电结构对微损伤尖端附件的应变梯度极其敏感的特性,研制用于监测结构损伤的新型应变梯度传感器,实现在线监测损伤导致的应变梯度,进而达到超前监测结构中应力集中区域损伤的萌生。
超固结土三轴率敏性特性及影响因素分析
超固结土三轴率敏性特性及影响因素分析WANG Zhichao;ZHENG Junxing;KUANG Dumin;HU Qian【摘要】为建立超固结土的弹黏塑性本构模型并预测其长期变形,研究土体不同应力历史的时间相依变形特性,采用土体率敏性三轴试验,对预加800 kPa有效应力历史的饱和黏土开展不同加载速率下的三轴剪切试验,并通过改变有效围压(50,100,200和400 kPa)、细砂含量(占比0%,10%和20%)和加载速率(0.5625%/min,0.0375%/min和0.0025%/min)等方式,深入探讨超固结比、黏性以及应变速率对土体变形特征的影响,引入能通过试验测定的应变率参数ρ来表征黏土率敏性大小.试验结果表明:归一化不排水抗剪强度随超固结比(OCR)单调递增;随着OCR 的增大,土体剪胀软化越明显;应变速率与土体的抗剪强度呈半对数线性关系;随着含砂的降低,土体的黏性增大,剪胀软化也越明显;率敏性随超固结比的增大而增强,土体黏性越大率敏性就越显著.【期刊名称】《铁道科学与工程学报》【年(卷),期】2019(016)001【总页数】8页(P92-99)【关键词】超固结土;率敏性;应变率参数;黏性【作者】WANG Zhichao;ZHENG Junxing;KUANG Dumin;HU Qian【作者单位】【正文语种】中文【中图分类】TU432随着我国交通基础设施建设的快速推进,面临着许多重超固结土沉降变形预测与控制问题,例如在城市硬质老黏土中修筑地铁以及高速公路和高速铁路高填方路堤沉降控制等复杂工程挑战。
超固结土在剪切荷载作用下会发生剪胀软化,其时间相依长期变形特性尤为复杂。
其中,基于岩土体率敏性三轴试验来建立其弹黏塑性本构模型是一种非常便捷有效的方式[1]。
Casagrande[2]最早意识到土体的率敏性现象。
随后国内外大量学者[3−4]展开了对黏土率敏性的研究,姜洪伟等[5]研究了加载速率对黏土不排水抗剪强度影响,得出黏土不排水抗剪强度跟应变速率正相关。
结构非线性分析参考文献
参考文献:1. A.M.内维尔著.李国拌,马贞勇译.混凝土的性能,北京:中国建筑工业出版社19832. P.梅泰著.祖永年,沈威,陈志潭译.混凝土的结构性能与材料,上海:同济大学出版社,19913. A.Corpinten&A.R.Ingraffea编.杨煜惠,黄政宇,万良芬等译.混凝土断裂力学.长沙:湖南大学出版社,19884. H.Kupfer,U.K.Hilsdorf,Behavior ofConcrete Under Biaxial Stresses,ACIJournal,August,19695. W.F.Chen,Evaluation of Plasticity-BasedConstitutive ModelsFor Concrete Material,SM Archives 13/1,19886. 过镇海,张秀琴.单调荷载下的混凝土应力—应变全曲线试验研究.北京:清华大学科学研究报告集第三集,19817. 朱伯龙,董振祥.钢筋混凝土非线性分析.上海:同济大学出版社,19858. 杨术秋,林泓.混凝土单轴受压应力—应变全曲线的试验研究.水利学报,1992年第6期9. 过镇海,张秀琴.反复荷载下混凝土应力—应变全曲线试验研究.北京:清华大学科学研究报告集第三集,198110. W.P.Chen,plasticity in Reinforced Concrete,McGraw-Hill Book Company,New York,198211. 过镇海,张秀琴.混凝土受拉应力—变形全曲线的试验研究.建筑结构学报,1988年第4期12. M.E.Tasuji,TheBehaviorofPlainConcreteSubjected to Biaxial stress, Dept.ofStructuralEngineering, Cornell University, Report No:360, March,197613. T.C.Y.Liu,Stress—Strain Response andFracture of Concrete inBiaxial Compression,Dept.Structure,Cornell University, Report No:339,February,197114. 宋玉普.钢筋混凝土有限元分析中的力学模型的研究.大连理工大学博士学位论文,198815. 于骁中等.混凝土的二轴强度及其在拱坝设计中的应用.水利水电科学研究院科学研究论文集,第19集(结构、材料)16. 王传志,过镇海,张秀琴.二轴和三轴受压混疑土的强度试验.土木工程学报,1987年第1期17. 董毓利.混凝土与钢纤维维混凝土本构关系和破坏准则的研究.哈尔滨建筑工程学院博士学位论文199218. O.Buyukozturk et al, Concrete in Biaxial Cyclic Compression, J. Structureal Engineering, Vol.110,No:3, March 198419. 水利水电科学研究院.混凝土的强度与破坏译文集.北京,水利出版杜,198220. 黄真.钢筋混凝土三维非线性有限元分析.天津大学博士学位论文,198921. 王竹溪.热力学(第二版).北京:高等教育出版社,196022. 范镜泓,高芝晖,非连续介质力学基础.重庆,重庆大学学报,198723. 范镜泓,材料本构关系的内时理论及其热力学基础.重庆,重庆大学学报,1985年第6期24. 陈至达,有理力学,徐州,中国矿业大学出版社,199825. 德冈辰雄著.越镇等译.理性连续介质力学入门.北京;科学出板社,198826. 杜殉.连续介质力学引论.北京;清华大学出版社,198527. 匡震邦.非线性连续介质力学基础.西安;西安文通大学出版杜,198928. 黄克智,连续介质力学.北京:清华大学出版社、北京大学出版杜,198929. 杜庆华,郑百哲,应用连续介质力学.北京;清华大学出版社,1986,30. 冯元桢著.李松年译.连续介质力学导论.北京:科学出版社,199231. 过镇海,王传志.多轴应力下混凝土的强度和破坏准则研究.土木工程学报,1991年第3期32. 夏志皋.塑性力学.上海;同济大学出版社,1991。
各向异性屈服准则的发展及应用
万方数据 万方数据 万方数据 万方数据 万方数据各向异性屈服准则的发展及应用作者:史艳莉, 吴建军, SHI Yan-li, WU Jian-jun作者单位:西北工业大学,机电学院,陕西,西,安710072刊名:锻压技术英文刊名:FORGING & STAMPING TECHNOLOGY年,卷(期):2006,31(1)被引用次数:9次1.俞汉清;陈金德金属塑性成形原理 19992.R 希尔;王仁塑性数学理论 19663.Hill R Theoretical plasticity of textured aggregates[外文期刊] 19794.Hill R Constitutive modeling of orthotropic plasticity in sheet metal[外文期刊] 1990(03)5.Hill R A user-friendly theory of orthotropic plasticity in sheet metal[外文期刊] 1993(01)6.Logan R W;Hosford W F Upper-bound anisotropic yield locus calculations assuming <111>-pencilgilde [外文期刊] 1980(07)7.Bralat F;Lian J Plastic behaviour and stretchability of sheet metals,Part Ⅰ A yield function for orthotropic sheet under plane stress conditions[外文期刊] 1989(01)8.Barlat F;Lege D J;Brem J C A six-component yield function for anisotropic materials 1991(07)9.Barlat F;Becker R C Yielding description for solution strengthened aluminum alloys sheets[外文期刊] 1997(04)10.Barlat F;Maeda Y;Chung K Yield function development for aluminum alloy sheets[外文期刊] 199711.Banabic D A new criterion for anisotropic sheet metals 199912.Banabic D;Balan T;Comsa S D Comments on a new anisotropic yield criterion[外文会议] 200013.Banabic D;Kuwabara T;Balan T An anisotropic yield criterion for sheet metals[外文期刊] 2004(0)14.Banabic D;Aretz H;Comsa D S An improved analytical description of orthotropy in metallic sheets [外文期刊] 2005(03)15.Gotoh M A theory of plastic anisotropy based on a yield function of fourth order (plane stress state) 197716.周维贤正交异性非二次屈服函数探讨 1989(07)17.Karafillis A;Boyce B C A general anisotropic yield criterion using bounds and a transformation weighting tensior[外文期刊] 199318.Banabic D Formability of Metallic Materials 200019.Banabic D;Mü ller W;K P O hlandt Experimental determination of yield loci for sheet metals 199820.吴向东不同加载路径下各向异性板料塑性变形行为的研究[学位论文] 200421.Pearce R Some aspects of anisotropic plasticity in sheet metals 1968(12)22.Woodthorpe J;Pearce R The anomalous behaviour of aluminum sheet under balanced biaxial tention[外文期刊] 1970(04)23.Lin S B;Ding J L Experimental study of the plastic yielding of rolled sheet metals with the cruciform plate specimen[外文期刊] 1995(05)24.Lege D J;Barlat F;Brem J C Characterization and modeling of the mechanical behavior and formability of a 2008-T4 sheet sample[外文期刊] 1989(07)25.Hayashida Y;Maeda Y FEM analysis of punch strengthening and cup drawing tests for aluminum,alloys using a planar anisotropic yield function 199526.Green D;Makinde A;Neale K W Experimental determination of yield surfaces and biaxial flow of 1145aluminum sheets 199827.Kuwabara T;VanBael A Measurement and analysis of yield locus of sheet aluminum alloy 6XXX 199928.倪向贵;吴恒安;王宇各向异性本构关系在板料成形数值模拟中的应用[期刊论文]-计算力学学报 2003(02)29.吴向东;万敏;周贤宾各向异性板料屈服轨迹的研究[期刊论文]-材料科学与工艺 2004(12)30.孙成智;陈关龙;林忠钦各向异性屈服准则对铝合金板成形预测精度的影响[期刊论文]-塑性工程学报 2004(03)31.Joao Pedro de Magalhaes Correia;Gérard Ferron Wrinkling Predictions in the deep-drawing process of anisotropic metal sheets[外文期刊] 2002(1-3)32.Kawaka M;Makinouchi A Plastic anisotropy in FEM analysis using degenerated solid element[外文期刊] 1996(1-4)33.沈启彧;卫原平;王玉国金属板料成形的一步有限元模拟方法[期刊论文]-上海交通大学学报 2000(10)34.刘腾喜;傅衣铭;丁皓江正交各向异性金属板料的成形极限[期刊论文]-固体力学学报 2000(02)35.中国仿真互动.利用ANSYS/LS DYNA仿真计算36.迅利科技有限公司ABAQUS公司介绍37.胡平板料冲压成形性模拟CAE软件系统KMAS1.张秀娟.刘彦奎.魏延刚.运新兵.韩锋.郑勇福.ZHANG Xiu-juan.LIU Yan-kui.WEI Yan-gang.YUN Xin-bing.HAN Feng.ZHENG Yong-fu各向异性铝合金板材本构关系及其有限元模拟[期刊论文]-大连交通大学学报2008,29(6)2.倪向贵.吴恒安.王宇.王秀喜各向异性本构关系在板料成形数值模拟中的应用[期刊论文]-计算力学学报2003,20(2)3.吴向东.万敏.周贤宾各向异性板料屈服轨迹的研究[会议论文]-20044.孙成智.陈关龙.林忠钦.赵亦希各向异性屈服准则对铝合金板成形预测精度的影响[期刊论文]-塑性工程学报2004,11(3)5.倪向贵.王宇.吴恒安.王秀喜.车玫.NI Xiang-gui.WANG Yu.WU Heng-an.WANG Xiu-xi.CHE Mei用于板料成形数值模拟的各向异性本构模型研究[期刊论文]-中国科学技术大学学报2000,30(6)1.柳玉起.徐丹.许江平改进的5参数Barlat-Lian屈服准则[期刊论文]-华中科技大学学报(自然科学版) 2008(1)2.邹修敏覆盖件的有限元和质量分析[期刊论文]-科技创业家 2012(14)3.潘秋红.董则防.黄瑶.王雷刚.万宝伟均匀化退火对纯铝铸轧-冷轧板各向异性的影响[期刊论文]-锻压技术2008(6)4.刘桂花.冯再新.贾青云.于晓东板料冲压成形过程中回弹预测及控制的研究进展[期刊论文]-热加工工艺2011(17)5.李玉强.王勇.董剑安.郭水军.须俊华.崔永生覆盖件冲压CAE分析材料模型的工程应用与发展[期刊论文]-模具技术 2009(4)6.蒋松.高锦张.贾俐俐板料多道次渐进成形下沉现象的模拟分析[期刊论文]-锻压技术 2010(2)7.沈黎萍.高锦张.蒋松.贾俐俐板料渐进成形半球形件的路径研究[期刊论文]-锻压技术 2010(6)8.曹瀚中.高锦张.肖士昌.贾俐俐渐进成形锥形件壁厚稳定区域影响因素的研究[期刊论文]-锻压技术 2012(6)9.徐梁.高锦张.贾俐俐.蒋松渐进成形直壁筒形件圆角缺陷的模拟分析[期刊论文]-锻压技术 2010(3)10.郭健.许模.宋丽娟.卢书强非均质花岗岩坝基典型地质缺陷处理研究[期刊论文]-人民黄河 2013(5)本文链接:/Periodical_dyjs200601031.aspx。
弹性力学基础-中英
The actual point of yield is often difficult to identify. A number of techniques are used to locateσy. The tangent method <or knee method> locates the yield strength at the intersection of the elastic slope and the initial portion of the plastic region <not reliably>. The preferred method is the percentage offset method where yield strength is obtained by drawing a line parallel to the initial elastic region data at 0.2% strain <0.002> offset. Where this line intersects the stress-strain curve then becomes known as the 0.2% yield strength.
Plastic means permanent!
Plastic deformation---it is irreversible or permanent.
O
A
B
C
D
E
elastic region
yield strength 屈服应力 屈服强度
plastic region
ultimate tensile strength 抗拉强度
Tsai_Hill与Tsai_Wu屈服准则的对比研究
391
ห้องสมุดไป่ตู้
392
第六届全国爆炸力学实验技术学术会议论文集
大,这个差别来源于两个准则中应力交叉项的影响。另外,在进入塑性加载后,图 1 中根据 Tsai-Hill 准则算出的 S1 基本保持不变, 1 , 2 , 3 代表的应力线平行于静水压力线,这是与 Tsai-Hill 准则假 设静水压不影响屈服一致的。由一致性法则式(19)及式(8)看出,要在塑性加载下处处满足这两个关 系式必须恒有 d 1 d 2 d 3 ,而 dp d 1 d 2 d3 / 3 ,得出 dp d 1 d 2 d 3 ,这 就使得图 1 中三条主应力塑性加载线与静水压力线平行。而图 2 中用 Tsai-Wu 准则算出的 S1 随应变 却显著减少,甚至减到负值,三条主应力塑性加载线与静水压力线斜率不相等,发生交叉现象,这 是由于 Tsai-Wu 准则中静水压会影响屈服,应力增量方向不与静水压一致,造成应力状态最终较大 地偏离静水压力线。具体分析,由 Tsai-Wu 准则的表达式(4)和式(23)容易得到:
1
引
言
近年来,复合材料广泛应用于航空、航天等领域,成为国防工业部门的一种重要工程材料,复 合材料应用比例的多少已成为衡量航天器结构设计先进性的标志之一。由于航天器在飞行时,所处 环境复杂,可能高速碰撞、射线辐照等动载荷环境,将不可避免地引起应力波在复合材料构件中传 播,较强的应力波还可能引发材料的损伤、破坏或者断裂。由于各向异性是大多数复合材料的重要 力学特性,因而在这些材料的冲击响应问题研究中,不可避免地要用到各向异性屈服准则及在此基 础上提出的各向异性塑性本构关系,使用合适的各向异性屈服准则描述复合材料的塑性力学性能, 对正确评估航天器对应力波破坏的防护能力,提高航天器的生存能力与保持其工作性能是非常重要 的。 现在已经存在的各向异性强度准则由很多种[1],这些强度准则经常被用作屈服或破坏准则,其 中有一类是基于 von Mises 屈服准则导出的。在这一类准则中,最常用的是 Tsai-Hill 准则和 Tsai-Wu 准则。Tsai-Hill 准则早期被用来描述轧制金属板的各向异性初始屈服[2],后来和 Tsai-Wu 准则一道被 应用到复合材料力学的研究当中 [3] ,作为复合材料的屈服或破坏准则。在数值模拟方面, P. E. O’Donoghue、C. E. Anderson 等人[4][5]提出了一个基于 Tsai-Hill 准则的各向异性塑性本构关系及其计 算方法,并将其用到了大型程序 EPIC 当中。在国内,李永池 [6] 进行了以 Tsai-Hill 屈服准则和 Johnson-Cook 模型为基础的横观各向同性粘塑性本构模型的研究,并用于应力波传播的数值模拟。 由于 Tsai-Hill 准则和 Tsai-Wu 准则是两个常用的各向异性屈服准则,本文讨论了基于它们的正 交各向异性塑性本构关系及其在数值模拟中的应用方法,并对其在性质上和应用中的不同处进行比 较,探讨他们的优劣与局限性所在,以便对不同性能的各向异性材料选择适用的屈服准则。为明确 起见,若没有特别说明,本文各向异性都指正交各向异性,准则都指屈服准则。
Semi-Infinite Anisotropic Spherical Model Correlations at T = T_c
arXiv:cond-mat/9803230v4 [cond-mat.stat-mech] 26 May 1998
(Received 12 February 1998) The ordinary surface magnetic phase transition is studied for the exactly solvable anisotropic spherical model (ASM), which is the limit D → ∞ of the D-component uniaxially anisotropic classical vector model. The bulk limit of the ASM is similar to that of the spherical model, apart from the role of the anisotropy stabilizing ordering for low lattice dimensionalities, d ≤ 2, at finite temperatures. The correlation functions and the energy density profile in the semi-infinite ASM are calculated analytically and numerically for T ≥ Tc and 1 ≤ d ≤ ∞. Since the lattice dimensionalities d = 1, 2, 3, and 4 are special, a continuous spatial dimensionality d′ = d − 1 has been introduced for dimensions parallel to the surface. However, preserving a discrete layer structure perpendicular to the surface avoids unphysical surface singularities and allows numerical solitions that reveal significant shortrange features near the surface. The results obtained generalize the isotropic-criticality results for 2 < d < 4 of Bray and Moore [Phys. Rev. Lett. 38, 735 (1977); J. Phys. A 10, 1927 (1977)]. PACS numbers: , 75.10.Hk, 75.30.Pd
施耐德疏水减法模型英文原版
施耐德疏水减法模型英文原版Schneider's Dewey Reduction ModelThe Schneider's Dewey Reduction Model is an approach toproblem-solving and decision-making that allows individuals to simplify complex situations by breaking them down into manageable parts. The model is based on the work of philosopher John Dewey and was developed by psychologist Paul Schneider.The model consists of four steps:1. Identify the problem or situation: The first step is to clearly define the problem or situation that needs to be addressed. This may involve gathering information, clarifying the issue, and determining the scope of the problem.2. Break down the problem: In this step, the problem is broken down into smaller, more manageable parts. This can involve identifying key factors, trends, or patterns that contribute to the problem, as well as any underlying assumptions or biases.3. Construct possible solutions: Once the problem has been brokendown, it is easier to develop possible solutions. This can involve brainstorming ideas, considering different options, and evaluating the advantages and disadvantages of each potential solution.4. Evaluate solutions and take action: Finally, the solutions that have been developed are evaluated based on their potential effectiveness and feasibility. The best solution is then selected and implemented, with ongoing monitoring and reflection to ensure that the problem is fully resolved.The Schneider's Dewey Reduction Model is a powerful tool for tackling complex problems and making effective decisions. By breaking down the problem into manageable parts, individuals can more easily understand the situation and develop effective solutions. This approach can be used in a variety of contexts, from personal decision-making to complex organizational problem-solving.。
- 1、下载文档前请自行甄别文档内容的完整性,平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益,请联系客服反馈,我们会尽快为您处理(人工客服工作时间:9:00-18:30)。
6th European Solid Mechanics ConferenceESMC200628August–1September,2006Budapest,Hungary CONSTITUTIVE MODELING OF ANISOTROPIC HYPERELASTIC MATERIALS BY POLYCONVEX STRAIN ENERGY FUNCTIONSMikhail Itskov1,Alexander E.Ehret11Department of Continuum Mechanics,RWTH Aachen UniversityEilfschornsteinstr.18,52062Aachen,GermanyE-mail:itskov@km.rwth-aachen.deKeywords:Polyconvexity,Anisotropy,Hyperelasticity,Soft biological tissues,Calendered rubber Summary.Polyconvexity of a strain-energy function is a very important mathematical condition especially in the context of a boundary-value problem.In the present contribution,a class of polyconvex anisotropic strain-energy functions is proposed.They are given by a series represented by some convex scalar functions.Each term of this series a priori satisfies the condition of the energy and stress free natural state so that no additional restrictions have to be imposed.Special cases of the proposed hyperelastic model based on power and exponential function representations show very good agreement with experimental data.1INTRODUCTIONSoft biological tissues as well as many engineering materials,as for examplefiber-reinforced or calendered rubber-like elastomers,are characterized by strong anisotropy and are able to undergo large elastic deformations.In the constitutive modeling,these features should be taken into account in an adequate manner.Hyperelastic consti-tutive models for soft biological tissues as for example the well known Fung-elastic model[2]or the orthotropic St.Venant-Kirchhoffmodel(see,e.g.,[3])may be used to describe the material behavior.Unfortunately,these models are not elliptic and for this reason not polyconvex.Polyconvexity represents,however,a very important mathematical condition especially in the context of a boundary-value problem[1].Indeed,polyconvexity implies quasiconvexity which in turn ensures ellipticity or the so-called Legendre-Hadamard condition.It means that the acoustic tensor is positive definite so that the speed of displacement waves is always real for any direction of prop-agation.Furthermore,in combination with some continuity and growth requirements,quasiconvexity guarantees the existence of the global minimizer of the total elastic energy of the body[1].To benefit from these positive features we propose in the present contribution a class of polyconvex anisotropic strain energy functions.2ORTHOTROPIC AND TRANSVERSELY ISOTROPIC MATERIAL SYMMETRIESTo describe the material symmetry,one can make use of the so-called structural tensors defined byL i=l i⊗l i,i=1,2,3(1) for orthotropic and byL1=l1⊗l1,L2=L3=12(I−l1⊗l1)(2)for transversely isotropic material symmetry,where l i·l i=δi j denote a set of orthonormal base vectors in the principal material directions.3ANISOTROPIC POLYCONVEX STRAIN ENERGY FUNCTIONSA powerful tool to construct polyconvex functions is the following additive representation of the strain energy function W(F)W(F)=ˆW1(F)+ˆW2(adj F)+ˆW3(det F),adj F=F−1det F,(3) where F denotes the deformation gradient.Indeed,if the functionsˆW i(i=1,2,3)are convex with respect to F, adj F and det F,respectively,then the resulting strain energy function(3)is polyconvex[5].Taking into accountMikhail Itskov 1,Alexander E.Ehret 1this fact,we propose a strain energy function of the formW =14sr =1µr f r ˜I r +g r ˜Jr +h r III C ,III C =det C ,(4)where f r and g r are convex and monotonically increasing and h r are convex functions of their arguments.Theterms ˜Ir and ˜J r denote generalized invariants defined by ˜Ir =3 i =1w (r )i I i ,˜Jr =3 i =1w (r )i J i ,r =1,2,...,s ,(5)in terms of the classical orthotropic or transversely isotropic strain invariantsI i =tr(CL i ),J i =tr [(cof C )L i ],cof C =(adj C )T(6)and weight factors of the principal material directions w i ≥0,(i =1,2,3).Note that both the invariants I i and J i(6)are convex with respect to F and adj F ,respectively and III C is convex with respect to det F .The same holds forthe generalized invariants ˜Ir and ˜J r (5)as far as the weight factors w i are non-negative.It can further be shown that each term in the series (4)fulfills the condition of the energy and stress free natural state if the functions f r ,g r and h r satisfy the requirements [3,4]f r (1)=g r (1)=h r (1)=0,f ′r (1)=g ′r(1)=−12h ′r(1)=1.(7)4COMPARISON WITH EXPERIMENTAL DATAIn order to demonstrate a wide applicability of the proposed polyconvex model (4)we consider numerical exam-ples dealing with experimental data on di fferent materials.In the first example our model was fitted to uniaxial tension tests on calendered rubber sheets using power function representations for f r and g r .The second example deals with elongation tests on bovine pericardium,an important biomaterial for medical applications.Motivated by the typical exponential-like stress-strain behavior of soft biological tissues,an exponential representation was applied.In both cases,very accurate agreement with experiments has been obtained as seen in Figure 1.N o m i n a l s t r e s s t , M P aStretch λN o m i n a l s t r e s s t , M P aStretch λFigure 1:Experimental data for calendered rubber (left)and bovine pericardium (right)versus simulation by the polyconvex model (4)REFERENCES[1]Ball,J.M.(1977):Convexity conditions and existence theorems in non-linear elasticity.Arch.Rat.Mech.Anal.,63,337-403.[2]Fung,Y .C.(1993):Biomechanics.Mechanical Properties of Living Tissues.2nd Ed.,Springer,New York.[3]Itskov M.,Aksel,N.(2004):A class of orthotropic and transversely isotropic hyperelastic constitutive modelsbased on a polyconvex strain-energy function,Int.J.Solids Struct.41,3833-3848.[4]Itskov M.,Ehret,A.E.,Mavrilas D.(2005):A polyconvex anisotropic strain-energy function for soft collage-nous tissues,Biomech.Model.Mechanbiol.(in press)[5]Schr¨o der,J.,Ne ff,P.(2003):Invariant formulation of hyperelastic transverse isotropy based on polyconvexfree energy functions,Int.J.Solids Struct.40,401-445.。