Applsci 13 01155 v2

applied
sciences
Article
Implicit to Explicit Algorithm for ABAQUS Standard
User-Subroutine UMAT for a 3D Hashin-Based Orthotropic
Damage Model
M. R. T. Arruda 1, * , M. Trombini 2 and A. Pagani 2
1 CERIS-IST-ID, Universidade de Lisboa, 1049-001 Lisboa, Portugal

2 POLITO MUL2, Politecnico di Torino, 10129 Torino, Italy
* Correspondence: mario.rui.arruda@tecnico.ulisboa.pt
Abstract: This study examines a new approach to facilitate the convergence of upcoming user-
subroutines UMAT when the secant material matrix is applied rather than the conventional tangent
(also known as Jacobian) material matrix. This algorithm makes use of the viscous regularization
technique to stabilize the numerical solution of softening material models. The Newton–Raphson
algorithm predictor-corrector of ABAQUS then applies this type of viscous regularization to a UMAT
using only the secant matrix. When the time step is smaller than the viscosity parameter, this type
of regularization may be unsuitable for a predictor-corrector with the secant matrix because its
implicit convergence is incorrect, transforming the algorithm into an undesirable explicit version
that may cause convergence problems. A novel 3D orthotropic damage model with residual stresses
is proposed for this study, and it is analyzed using a new algorithm. The method’s convergence is
tested using the proposed implicit-to-explicit secant matrix as well as the traditional implicit and
explicit secant matrices. Furthermore, all numerical models are compared to experimental data. It
was concluded that both the new 3D orthotropic damage model and the new proposed time step
algorithm were stable and robust.
Keywords: user subroutine UMAT; implicit to explicit; orthotropic damage; 3D Hashin failure;
composite GFRP
Citation: Arruda, M.R.T.; Trombini, M.;
Pagani, A. Implicit to Explicit
Algorithm for ABAQUS Standard
User-Subroutine UMAT for a 3D
Hashin-Based Orthotropic Damage 1. Introduction
Model. Appl. Sci. 2023, 13, 1155. The need for new 3D orthotropic damage models is primarily due to their use in
https://doi.org/10.3390/ simulating 3D connections in composite structures [1–3]. To predict the behavior of exterior
app13021155 profiles in hybrid beams and novel carbon-reinforced fiber aluminum laminates, a number
Academic Editor: Avik Mahata of 2D orthotropic damage models have occasionally been used in conjunction with shell
elements [4,5]. Other authors, however, discovered limitations in the accuracy of numeri-
Received: 29 December 2022 cally simulating 3D GFRP connections using the classical 2D formulation with continuum
Revised: 13 January 2023
shell elements [6], related to either pilling, shear-out, or bearing due to the incorrect stress
Accepted: 13 January 2023
distribution in the thickness in shell formulation for non-linear analysis.
Published: 15 January 2023
Since the start of the 2000s [7], a more comprehensive 3D orthotropic damage formula-
tion has been sought after due to the engineering industry’s rapid adoption of composite
structures. The need for structural design has also grown among structural engineers in
Copyright: © 2023 by the authors.
the field of composites [8]. In all circumstances, the problem must be described in 3D, and
Licensee MDPI, Basel, Switzerland. the accuracy of the computed stress level relies on whether the problem is being examined
This article is an open access article at the micro, meso, or macro scale [9].
distributed under the terms and For the structural designer, applying a complete 3D orthotropic damage model still
conditions of the Creative Commons poses significant issues such as (i) full 3D formulation with explicit material parameters and
Attribution (CC BY) license (https:// (ii) simple and clear guidelines for direct application. The scientific community mostly uses
creativecommons.org/licenses/by/ these damage orthotropic damage models to confirm experimental campaigns and adopted
4.0/). material properties, therefore these two factors are crucial for the structural designer [10].
Appl. Sci. 2023, 13, 1155. https://doi.org/10.3390/app13021155 https://www.mdpi.com/journal/applsci

Appl. Sci. 2023, 13, 1155 2 of 24
Unfortunately, academic researchers are not using composite orthotropic damage models
without a current experimental campaign to arrive at better design solutions, despite the
fact that this is a fundamental requirement for engineering in the composites industry given
the absence of structural codes for these materials.
There are two main goals for this project. The first is to analyze a novel 3D orthotropic
damage model’s numerical effectiveness using the Hashin failure criterion, using classical
MTL damage progression [11] with energy regularization [12]. Experimental data are
compared and used to validate this orthotropic damage model. The second study looks at the
new ITE-UMAT algorithm, which speeds up the convergence of the UMAT subroutine when
the secant matrix rather than the traditional tangent matrix is employed in the predictor-
corrector. The algorithm is compared with fully implicit and explicit numerical results.
According to the authors’ best knowledge, the use of a 3D orthotropic damage model
with residual stresses using viscous regularization and the secant matrix for the predictor-
corrector algorithm has not been fully studied and validated. This work plans to fill this
gap in knowledge.
2. Recent Advances in 3D Orthotropic Damage Models

Since the 2000s, several 3D orthotropic damage models have been proposed, which
is primarily due to the scientific community’s availability of commercial finite element
software used for implementing new material constitutive relations [13,14].
Due to the composite material’s ply-by-ply behavior and the formulation’s simplifica-
tion, basic 2D orthotropic damage models were employed for shell elements prior to the
widespread adoption of 3D orthotropic damage models [15]. These 2D orthotropic dam-
age models were sufficient to simulate beams [4], columns [16], and composite sandwich
panels [17], however, with significant constraints when modelling 3D beam-to-column
connections [6].
The difficulty in evolving from a 2D to a 3D failure model is primarily related to the
first being assembled and tested using only plane analysis [18], as this leads to different
results when a formulation is then extended into the third dimension [19]. In addition, the
3D failure criterion and damage evolution may be significantly influenced by the kind of
orthotropic composite material being researched. The use of the 3D Hashin failure criterion
is not novel and has been suggested for shell elements [20] and solid elements [21,22],
in any case, without any application for the interlaminar failure. An established 3D
method based on the combination of several elliptic surfaces for the adoption of separated
failure criteria [23] presented encouraging results in simulating the connection of three bolt
joints for T300/5228A composite material, with discontinuous damage evolution. For the
stochastic analysis of peek laminates, a few different iterations of these elliptic surfaces were
used [24]. For composite fiber metals, the fiber and matrix 3D failure criteria are based on a
square root sum of the total strain [25], but with a cohesive interlaminar failure criterion. A
variation of this last method was applied in FRP composites [26,27], but with restrictions
on the failure criterion for material orientation. One of the first successful applications
for a full 3D analysis with Larc05 was presented in the early work of [28], with variation
proposed by [29]. Due to its simplicity of implementation, some 3D failure criteria with
simple degradation models [30] in terms of elastic parameters have also been successful
and continue to be popular among structural engineers. One of the best examples of a fully
3D Hashin failure criterion using simplified degradation models was suggested by [31] for
bolted connections. Recently, the Hou criteria were successfully used to simulate drilling
damage, based on an improved 3D elliptic failure surface [32]. The NU failure criterion was
proposed and implemented in a three-point bending load to improve the matrix damage
evolution due to increased shear [33]. Recently, using an invariant-based concept to study
the bearing in composite connections showed excellent findings [34,35].
Appl. Sci. 2023, 13, 1155 3 of 24
3. 3D Orthotropic Stiffness Formulation

The orthotropic formulation is described using ABAQUS standard notation for both
stress and strain tensors [36]. The compliance and stiffness matrices are used to calculate
the elastic constitutive relation for stress and strain (1).
ε i = Hij σj [ Hd ]−1 = [Cd ] σi = Cij ε j (1)
with:
σ1 = σ11 ; σ2 = σ22 ; σ3 = σ33 ; σ4 = τ12 ; σ5 = τ13 ; σ6 = τ23
(2)
ε 1 = ε 11 ; ε 2 = ε 22 ; ε 3 = ε 33 ; ε 4 = γ12 ; ε 5 = γ13 ; ε 6 = γ23
To present a 3D orthotropic stiffness that is both coherent and thermodynamically
acceptable, it is necessary to assemble the damage compliance matrix Hd exclusively using
the diagonal term (8) to account for the damage behavior [29,37]. It is then possible to
assemble the damaged stiffness matrix Cd , by inverting the previous compliance matrix (10).
It is important to state the relevance of Γd (13), because this term ensures that no transversal
stress occurs when subjected to uniform longitudinal stress. Just as in the previous 2D
Hashin formulation [15], the tension and compression damage (dt , dc ) are activated using
the corresponding effective stress space (3). For this work, damage variables are associated
with the fiber d f , the matrix dm , and the interlaminar di .
σi
σ̂i = (3)
(1 − d i )
where:
d1 = d f ; d2 = dm ; d3 = di ; d4 = ds12 ; d5 = ds13 ; d6 = ds23

d f t i f σ̂1 ≥ 0
df = (4)
d f c i f σ̂1 < 0

dmt i f σ̂2 + σ̂3 ≥ 0
dm = (5)
dmc i f σ̂2 + σ̂3 < 0

dit i f σ̂3 ≥ 0
di = (6)
dic i f σ̂3 < 0



 d s12 = 1 − 1 − d f t 1 − d f c (1 − dmt )(1 − dmc )

ds13 = 1 − 1 − d f t 1 − d f c (1 − dit )(1 − dic ) (7)


ds23 = 1 − (1 − dmt )(1 − dmc )(1 − dit )(1 − dic )

The main advantage is that, by design, this matrix verifies thermodynamic admissibil-
ity principles when using Gibbs free energy (14) [38,39].
 H11 
1− d f H12 H13 0 0 0

 H21 H22 
 1− d m H23 0 0 0 

H H33
 31 H32 1− d i 0 0 0 
[ Hd ] =  (8)

 0 H44 
 0 0 1−ds12 0 0 

 0 H55
 0 0 0 1−ds13 0 

H66
0 0 0 0 0 1−ds23
1
  ν12
 H11 =  H12 = H21 = − E1

 E1   H44 = G12
1 ν13
H13 = H31 = − E
H22 = H = G13 (9)
E2
1  H = H = − ν23
1  55
H66 = G23

H = 
33 E3 23 32 E2
Appl. Sci. 2023, 13, 1155 4 of 24
C11 C12 C13 0 0 0

 
C21 C22 C23 0 0 0 
 
C31 C32 C33 0 0 0 
[Cd ] = Γd   (10)
 0
 0 0 C44 0 0 
 0 0 0 0 C55 0 
0 0 0 0 0 C66
with: 


 C11 = 1 − d f E1 [1 − (1 − dm )(1 − di )ν23 ν32 ]

 h i
C22 = (1 − dm ) E2 1 − 1 − d f (1 − di )ν13 ν31





 h i

C33 = (1 − di ) E3 1 − 1 − d f (1 − dm )ν12 ν21
(11)



 C12 = C21 = 1 − d f (1 − dm ) E1 [ν21 + (1 − di )ν31 ν23 ]

C13 = C31 = (1 − di )(1 − dm ) E1 [hν31 + (1 − dm )ν21 ν32 ] i





 C23 = C32 = (1 − dm )(1 − di ) E2 ν32 + 1 − d f ν31 ν12


 C44 = (1 − ds12 ) G12 /Γd
C = (1 − ds13 ) G13 /Γd (12)
 55
C66 = (1 − ds23 ) G23 /Γd
 −1

 Γd = 1 − Γf mν12 ν21 − Γmi ν23 ν32 − Γf i ν13 ν31 − 2 × Γf mi ν12 ν23 ν31

Γ = 1 − d f (1 − d m )


 fm


Γmi = (1 − dm)(1 − di ) (13)




 Γf i = 1 − d f (1 − di )


 Γ = 1 − d (1 − d )(1 − d )
f mi f m i
1
Ψ (σ, d, z) = {σ}t [Cd ]{σ} (14)
2
with:  
6 3y1
Hii σi2
Ψ (σ, d, z) =
1
2
 ∑1 − d i
+ ∑ 2 × Hij σi σj 

(15)
i =1 i,j=1y2
where:
σ1 = σ11 ; σ2 = σ22 ; σ3 = σ33 ; σ4 = τ12 ; σ5 = τ13 ; σ6 = τ23
d1 = d f ; d2 = dm ; d3 = di ; d4 = ds12 ; d5 = ds13 ; d6 = ds23
When calculating thermodynamic forces with this orthotropic damaged stiffness,
these are always positive (16), validating the second principle of thermodynamics (17) and
demonstrating the damage model is thermodynamically admissible.
∂Ψ 1 Hii σi2
Yi = = ≥0 (16)
∂di 2 (1 − d i )2
⇒ ∑Y δd ≥ 0
i i (17)
i
4. Adopted 3D Hashin-Based Damage Model

It is important to note that there is no scientific consensus regarding the true formu-
lation of the 3D Hashin failure criterion. This is because the true cross-failure criterion
between intralaminar and interlaminar is still heavily debated in the scientific commu-
nity [24,31]. When the damage evolution laws are studied [29,40], this uncertainty becomes
even more pronounced, and several assumptions must be admitted to fully verify the
thermodynamic admissibility criteria.
Appl. Sci. 2023, 13, 1155 5 of 24
Contrary to popular belief, the Hashin failure criterion began as a 3D failure criterion
for shell elements and was later simplified for 2D plane elements, though it lacked an
interlaminar failure criterion. Another limitation of this failure model is that it was designed
for unidirectional composites, but many authors continue to apply it to general lamina
composites [41,42].
4.1. Hashin 3D Failure Criterion

The Hashin 3D failure criterion in this work is based on the fusion of formulations
published by several academics [21,30,43]. These continue to acknowledge a primary plane
behavior that is consistent with the material’s primary resistance stress [6], in which its
failure criteria are assembled.
Since it is challenging to accurately describe the real behavior of composites in the 3D
space, the original failure criteria make various material and tensor field simplifications. In
this work, to extend to a fully 3D Hashin damage model, these simplifications will also be
applied and explained throughout the manuscript.
The first simplification admits that all unidirectional fibers are transversely isotropic,
in which the failure criterion depends on effective stresses (18), with the plane 2-3 rotating
around axis 1 [43]. Consequently, the failure criteria for fiber breaking are established Ff in
axis-1 and matrix cracking Fm for plane 2-3, using their respective tensorial plane stresses
(18). For this work, all the tensors are expressed in terms of effective stresses because these
failure criteria are later employed in damage models.
Ff (σ̂1 , τ̂12 , τ̂13 ) = 1.0

(18)
Fm (σ̂2 , σ̂3 , τ̂21 , τ̂31 , τ̂23 ) = 1.0
Despite the fact that the Hashin criterion is always positioned in the material principal
directions, it is preferable to rewrite the failure criterion functions in terms of stress invari-
ants using the envelope quadratic Equation (19), in order to support a more reliable failure
formulation similar to Von-Mises.
Ff ( I1 , I4 ) = A f I1 + B f I12 + D f I4 = 1.0
(19)
Fm ( I2 , I3 , I4 ) = Am I2 + Bm I22 + Cm I3 + Dm I4 = 1.0
with: 

 I1 = σ̂1
I2 = σ̂2 + σ̂3

2 − σ̂ σ̂ (20)
I = τ̂23
 3 2 3


I4 2 + τ̂ 2
= τ̂12 13

It is possible to simplify in the case of pure shear stress τij , τjl = 0, σ = 0 , for each of
the respective shear directions, and compute the values of D f , Cm , and Dm (21), for which
S L and ST are the longitudinal and shear stress, respectively.

 Cm = 12
ST
(21)
 D f = Dm = 12
S L
The second simplification is that the tensile and compressive failure criteria may
present different failure modes for both the fiber and the matrix, resulting in different
failure surface functions.
For the third simplification, it concedes that for the fiber buckling in compression (22),
there is no influence of shear stresses. Other authors have used the fiber kinking mode
to contest this last fiber simplification for 3D model compression [28]. Previous studies
have observed the deterioration of matrix shear modulus in fiber cycle compression [44,45],
but with small influence and still some uncertainties. For this reason, simplification is still
applied in this work.
Appl. Sci. 2023, 13, 1155 6 of 24
For the third simplification, only quadratic terms are admitted in the fiber failure
2
criterion to preserve the envelope ecliptic surface A f t = 0; B f t = 1/Xt . In any case, the
possibility of a fully closed surface is eliminated by the simplification of the compressive fiber
failure criterion. (23). Xt and Xc are the fiber tensile and compressive strength, respectively.
1 1
A f t = 0; B f t = Xt2
; Dft = Df = S2L
(22)
1
A f c = 0; B f t = Xc2
; Dft =0
2 2 + τ̂ 2
σ̂1 τ̂12
Ff t = Xt + S2L
13
≤ 1.0 i f σ̂1 ≥ 0
2 (23)
σ̂1
Ff c = Xc ≤ 1.0 i f σ̂1 < 0
The fourth simplification is acknowledging that the tensile matrix failure criterion only
depends on quadratic terms (19), which is in accordance with some cohesive models [46],
and promotes a closed failure surface function (24), (25). Yt and Yc are the matrix tensile
and compressive strength, respectively.
1 1 1
Amt = 0; Bmt = Yt2
; Cmt = S2T
; Dmt = Dm = S2L (24)
2 2 − σ̂ σ̂ 2 + τ̂ 2
σ̂2 +σ̂3 τ̂23 2 3 τ̂21
Fmt = Yt + S2T
+ S2L
31
≤ 1.0 i f σ̂2 + σ̂3 ≥ 0 (25)
The fifth simplification pertains to the matrix’s resistance to compression, in which

the collapse during biaxial behavior is linked to crushing when σ2 = σ3 = −σ Yc .
However, it must verify the collapse during uniform compression when σ2 = −Yc . In
these conditions, if the shear stress components are null, and we are on the onset of matrix
compressive failure, the following systems of Equation (26) are assembled.
Amc (σ̂2 + σ̂3 ) + Bmc (σ̂2 + σ̂3 )2 − σ̂2

( σ̂2 σ̂3
(
S2T
= 1.0 −2σ̂Amc + 4σ̂2 Bmc − S2T
= 1.0
2
⇔
Amc (σ̂2 ) + Bmc (σ̂2 ) = 1.0 −Yc Amc + Yc2 Bmc = 1.0
(26)
Since the system of equations has three unknowns and only two equations, these
cannot be solved directly but rather by using the assumption that in biaxial compression
σ2 = σ3 = −σ Yc , the following limit (27) can be admitted. It is possible to determine the
remaining unknown matrix compressive parameters for a closed-envelope surface using
this limit, as long as Yc > 2ST . As a result, the compressive matrix failure criterion can be
put together (28).
Yc2
1 4S2T
−1
lim ⇒ Bmc = 2 ⇔ Amc = (27)
σ→∞ 4ST Yc
2 2 2 2 2
−σ̂2 σ̂3

Yc σ̂2 +σ̂3 σ̂2 +σ̂3 τ̂23 τ̂21 +τ̂31
Fmc = 2ST −1 Yc + 2ST + S2T
+ S2L
≤ 1.0 i f σ̂2 + σ̂3 < 0 (28)
The condition of matrix tensile and compressive behavior is demonstrated in the work
of [43], and it uses proof by contradiction, in which it is the only condition that does not
violate the principal of σ̂2 ≥ 0; σ̂3 ≥ 0 in al quadrants of the space σ̂2 − σ̂3 .
In the case of the interlaminar failure criterion, a cohesive model (29) is used for
the tensile interlaminar failure criterion with the influence of the shear stresses in the
interlaminar plane (30), which is based on the initial work of [21]. For the interlaminar
compressive failure, the influence of the shear is minor, therefore, in this case, only the
Appl. Sci. 2023, 13, 1155 7 of 24
pressure stresses are not taken into account for the failure criterion (30). Zt and Zc are the
out-of-plane damage tensile and compressive thresholds, respectively.
2 2 2
σnn τnt τnl (29)
σt + sT + sL ≤ 1.0
2 2 2
σ̂3 τ31 τ32
Fit = Zt + SL + ST + ≤ 1.0 i f σ̂3 ≥ 0
2 (30)
σ̂3
Fic = Zc ≤ 1.0 i f σ̂3 < 0
4.2. Damage Evolution

For this work, the same formulation of equivalent strains and stresses used in previous
2D Hashin damage models is used [15]. Then, using the same MLT principles [11], the
equivalent stresses and strains are assembled for the 3D space, with different equivalent
stresses and strains for the fiber, matrix, and interlaminar in tensile and compressive
damage evolution [47].
The equivalent strain is calculated using equivalent principal virtual work [48], written in
the form of (31). For the particular cases of the fiber and matrix directions, these equations are
rewritten into (32) and (33). The adopted stresses and strains used to compute the equivalent
ones are associated with the effective stresses present in the respective failure criterion.
q
σeq ε eq = ∑ σij ε ij and ε eq = ∑ ε2ij (31)
 q
ε = hε 1 i2 + γ12 2 + γ2


 eq, f t 13




 ε eq, f c = h−
q ε 1 i

 ε eq,mt = hε 2 i2 + γ2 + γ2 + γ2

21 23 31
q (32)
 ε eq,mc = h−ε i2 + γ2 + γ2 + γ2 + γ2

 2 21 21 23 31

 q
2 2 2
 ε eq,it = hε 3 i + γ31 + γ32




ε eq,ic = h−ε 3 i

= hσ1 ih
√ε 1 i+2τ12 γ212 +τ13
 γ13
σ
 eq, f t 2


 hε 1 i +γ12 +γ13


σeq, f c = h−σ1 i






 σeq,mt = hσ2 ih√ ε 2 i+τ21 γ21 +τ23 γ23 +τ31 γ31


h ε 2 i2 + γ2 + γ2 + γ2


21 23 31
(33)
h−σ2 ih−
√ ε 2 i+τ221 γ212 +τ232 γ23 +2 τ31 γ31
σeq,mc =


h−ε 2 i +γ21 +γ23 +γ31





hσ ihε i+τ γ +τ γ32

σeq,it = 3 √ 3 2 31 231 32


 2
hε 3 i +γ31 +γ32





σeq,ic = h−σ1 i

A characteristic length Lc is added to the formulation to represent the constitutive law

as a stress-equivalent displacement relation, which reduces the mesh dependency that may
result from material softening [12]. The characteristic length is calculated in this study as
the square root of the numerically calculated surface area. The definitions of equivalent
displacement (34) are identical to those first discovered in the study of [49], which are
independent of the direction of the fracture. This last point is crucial in orthotropic damage
models since the damage progression will always be influenced by either the behavior
of the fibers or the matrix. It has been demonstrated in various publications that some
element dependency/sensitivity is predictable as a result of FEM with softening if the mesh
is irregular and the stress field distribution is non-uniform [12,50].
study as the square root of the numerically calculated surface area. The definitions of
equivalent displacement (34) are identical to those first discovered in the study of [49],
which are independent of the direction of the fracture. This last point is crucial in ortho-
tropic damage models since the damage progression will always be influenced by either
the behavior of the fibers or the matrix. It has been demonstrated in various publications
Appl. Sci. 2023, 13, 1155 that some element dependency/sensitivity is predictable as a result of FEM with softening 8 of 24
if the mesh is irregular and the stress field distribution is non-uniform [12,50].
The evolution law is then revised, finally taking the form embodied in Figure 1, in
whichThethe area ABClaw
evolution ⇒ is𝜀 then
(𝑑 =revised,
0) under the taking
finally curve isthe
linked
formtoembodied
the fracturein energy
Figure 1, 𝐺 in
, .
ˆ

At thethe
which moment,
area ABCnew⇒experimental
ε eq (d = 0) methodologies
under the curveare emerging
is linked to fracture
to the calculateenergy
fractureG fen-
,m .
ergy
At the[51], but for
moment, neworthotropic
experimental materials, the classical
methodologies Compact Tension
are emerging test fracture
to calculate is still sug-
en-
gested.
ergy [51], but for orthotropic materials, the classical Compact Tension test is still suggested.
Figure1.1.Constitutive
Figure Constitutiverelation
relationfor
forthe
theequivalent
equivalentstress
stressand
andequivalent
equivalentdisplacement.
displacement.
Theinitial
The initialvalues 0𝜎 and
valuesσeq and 0 𝛿
δeq are calculated
are calculated in theinonset
the onset of damage,
of damage, by analytically
by analytically solv-
solving
ing Equations
Equations (23), (28),
(23), (25), (25), and
(28),(30).
andTo (30). To verify
verify the Kuhn–Tucker
the Kuhn–Tucker equations,
equations, at the
at the begin-
ˆ curve ADC
(d) ⇒.

ning of eachofincremental
beginning process,process,
each incremental the evolution law is updated
the evolution law istoupdated
curve ADC to ⇒ ε eq
Appl. Sci. 2023, 13, 1155 9 of 2
𝜀 (𝑑)
In [11],
. it was shown that the tensile and compressive damage variables are estimated
using In
Equation (35) for
[11], it was a linear
shown thatsoftening
the tensile evolution dependingdamage
and compressive on the fiber or matrix
variables are evolution.
estimated
using Equation (35) for a linear eq softeningeq evolutionu
depending on the fiber or matrix evo-
lution. δij =𝛿Lc ε ij𝛿 and − 𝛿 δeq = 2G f /σeq (34)
,
𝑑 = 𝑖𝑓 𝛿 ≤ 𝛿 ≤ 𝛿 (35
𝛿 𝛿= 𝐿 𝛿𝜀 0 − 𝑎𝑛𝑑
𝛿
u 𝛿 = 2𝐺 /𝜎 (34)
δeq, δeq − δeq
0 u
d =
In this work, the possibility of i f δeq
the equivalent ≤ δeqstresses
≤ δeq presenting residual (35) stress i
δeq δequ − δ0
eq
also adopted, which is important primarily when compression is high and very presen
in the structural
In this response.
work, the Inofaddition,
possibility to promote
the equivalent stresses some numerical
presenting residualstability, it is recom
stress is also
adopted,
mendedwhich
to useis tensile
important primarily
residual when
stress compression
that may varyisfromhigh and
1% very
to 10% present in initial
of its the peak
structural response. In addition, to promote some numerical stability, it is recommended
value. The adopted residual equivalent stress formulation is presented in Figure 2. A new
to use tensile residual stress that may vary from 1% to 10% of its initial peak value. The
damage function is assembled during the residual equivalent stress segment after point D
adopted residual equivalent stress formulation is presented in Figure 2. A new damage
(37).
function is assembled during the residual equivalent stress segment after point D (37).
Figure2.2.Constitutive
Figure Constitutiverelation with with
relation residual stress for
residual the equivalent
stress stress and equivalent
for the equivalent stress anddisplacement.
equivalent displace
ment.
𝛿
⎧𝜎 = (1 − 𝑑 )𝑘 𝛿 = (1 − 𝑑 ) 𝜎 = %𝜎
⎪ 𝛿
⇒𝛿 =𝛿 −% 𝛿 −𝛿 (36
⎨ 𝛿 𝛿 −𝛿
⎪(1 − 𝑑 ) =
Appl. Sci. 2023, 13, 1155 9 of 24
D

r = (1 − d ) k δ D = (1 − d ) δeq σ0 = %σ0
 σeq

D eq eq D δ0 eq eq
eq D u u 0
0
δeq ( u −δ D
δeq ) ⇒ δeq = δeq − % δeq − δeq (36)
eq
 (1 − d D ) =

D u − δ0
δeq ( δeq eq )
r = %σ0
σeq = (1 − dr )k eq δeq = σeq eq
σ0
⇔ (1 − dr ) δ0eq δeq = %σeq
0
(37)
eq
0
δeq
⇔ dr = 1 − % δeq
4.3. Viscous Regularization

Even when only the energy regularization described in the previous section is used,
some issues may arise: (i) Loss of positive definiteness of the tangent stiffness; (ii) some
damage localization; and (iii) some numerical convergence issues. To overcome these
issues, a simple/efficient procedure is to implement viscous regularization.
Damage localizes in a narrow band when the tangent stiffness tensor is not positive
definite, and the numerical solution depends on the numerical discretization (decreasing
the element size in the localized zone decreases the computed energy dissipated). Therefore,
the structural response is not objective because it does not converge to a single solution
with mesh refinement. Standard Newton–Rapshon approaches also have convergence
issues when the tangent stiffness is not positively definite [52].
As mentioned above, the viscous regularization approach can help to solve some of
these convergence problems. This algorithm is built on an artificial Duvaut–Lions viscosity
model [53]. More details on the theoretical background of this method can be found in the
initial published work of [54]. This algorithm improves convergence by making the tangent
stiffness matrix of the softening material positive definite for small enough time increments.
In this regularization procedure, the incremental Equation (38) provides a definition for the
viscous damage variable.
1
δdv = (d − dv ) × δt (38)
η
where d is the damage variable determined as previously explained, dv is the regularized
viscous damage variable, and η is the viscosity stabilization factor. The viscous and non-
viscous stresses are written as (39) and (40), respectively.
{σv } = [Cdv ]{ε} (39)
{σ} = [Cd ]{ε} (40)

The damaged elastic matrix, Cdv , is computed using the viscous damage variables for
each failure mode. The rate of convergence of the model in the softening regime is typically
improved by using viscous regularization with a small value for the viscosity parameter
(small compared to the characteristic time increment), without degrading the quality of the
numerical solution. The fundamental notion is that the viscous system’s solution relaxes
to that of the inviscid case as η → 0 , where t represents time. For the time integration of
the internal variables, an algorithm must be implemented. The internal variables can be
updated using a backward-Euler approach [55], resulting in the viscous damage variable
(41), by merging it with the preceding expression (38).
1 t+∆t
dtv+∆t = dtv + ∆dtv ⇒ dtv+∆t = dtv + d − dtv+∆t × ∆t (41)
η
Appl. Sci. 2023, 13, 1155 10 of 24
Now, it is possible to calculate the viscous damage while keeping the implicit time integra-
tion (42), utilizing data from the past viscous damage and the current non-viscous damage.
η ∆t
dtv+∆t = dtv + dt+∆t (42)
η + ∆t η + ∆t
It is important to guarantee that the level of viscosity is not that high, which is
performed indirectly by comparing the level of elastic damaged energy (43) and the viscous
energy dissipation (44), in which the first must always override the second. If the time step
∆t if very small, it is possible that the first term of the sum in (44) dominates the second,
providing an increase in viscous energy dissipation, which might compromise the accuracy
of the results.
1 h n t+∆t
∑
i
Edt+∆t = Edt + σv,i + σv,it
∆ε i (43)
2 i =1
1 h n t+∆t
∑
i
Evt+∆t = Evt + σv,i + σv,i t
− σit+∆t − σit ∆ε i (44)
2 i =1
5. Proposed Predictor-Corrector Implicit to Explicit

Many authors have used UMAT user-subroutines to implement new materials in
ABAQUS without resourcing a fully Jacobi stiffness matrix (aka tangent matrix), using only
a secant matrix for the ABAQUS predictor-corrector, with viscous regularization.
As referenced before, viscous regularization has only been demonstrated to be a fully
implicit method when a tangent matrix is applied to the predictor-corrector algorithm [54,56].
The use of a secant matrix with viscous regularization lacks the theoretical background to
prove its implicit convergence. It can be depicted in expression (42) that the if the time step
is larger than the viscosity parameter ∆t η, then the level of regularization is low enough,
and the implicit standard convergence can be admitted since the viscous damage is always
related to the current time step as the non-viscous damage. However, if the time step is
lower than the viscosity parameter ∆t η, then the viscous damage will always be related
to the previous time step, therefore prompting an explicit time step integration.
Conflicts may arise in this last scenario because the strain increment corresponds to a
full implicit integration strategy (45). These inconsistencies happen when damage follows
an explicit time step update dtv+∆t ' dtv , and the strain increment follows a full implicit
integration scheme. This can cause some convergence problems, especially if the time step
is very small ∆t η. Furthermore, when the time step is very small since the viscous
damage solution overrides the non-viscous damage solution, the viscous damage energy
tends to grow (44), indirectly providing extra energy that is not real.
ε t+∆t = ε t + ∆ε t+∆t (45)
To solve this problem, a new implicit-to-explicit algorithm is proposed. The funda-

mental basis guarantees that when the time step is large, the damage evolution follows
an implicit time step integration, but when the time step is smaller than the viscosity
parameter, the damage evolution switches to an explicit time step integration. This is an
optimization of the predictor-corrector of ABAQUS, which states that when the time step
is large a full implicit formulation is used, but when the predictor-corrector algorithm of
ABAQUS presents complications in converging or reducing the time step, then an explicit
formulation is used to suppress the converging difficulty. The basis of this formulation
is using the increment correction parameter v linked to the portion of the non-viscous
damage variable dt+∆t in expression (46).
ε t+∆t = ε t + v × ∆ε t+∆t (46)
in which:
∆t
v= (47)
∆t + η
Appl. Sci. 2023, 13, 1155 11 of 24
It is important to point out that when v = 1.0 the material stiffness and stress update
during the time integration in the UMAT follows a full implicit formulation; in contrast, when
v = 0.0, it follows a full explicit time integration for the material stiffness and stress update.
Using this algorithm, it is guaranteed that when performing the viscous regularization,
if the predictor-corrector uses large time steps, then both the viscous and non-viscous
solution portion of expression (42) will contribute to the implicit time step integration.
However, when the time step is small enough, then the increment correction parameter
tends to be small, close to an explicit integration, and the influence of the non-viscous
solution overrides the viscous solution, therefore using only the previous damage variable.
This way there is no mixed implicit time step integration, with previous step damage
variables, as suggested before, perhaps compromising the convergence and accuracy of the
numerical solution.
To better understand the variation of v with ∆t for a fixed η = 10−5 , it is depicted in
Figure 3a, in which it is possible to see that even for small times steps from ∆t = 10−3 to
∆t = 10−5 , the value of v > 50%, therefore garanting some influence of the ∆ε t+∆t in the
final strain increment ε t+∆t . This is close to an implicit behavior but does not guarantee it.
Appl. Sci. 2023, 13, 1155 As depicted in Figure 3b using the logarithmic scale, only when ∆t < 10−6 is the value 12 of of 25
v < 10%, being close to an explicit form, but still with some influence of ∆ε t+∆t in the final
strain increment ε t+∆t .
(a) (b)
Figure 3. Variation of the correction parameter with the time step increment. (a) Normal Scale; (b)
Figure 3. Variation of the correction parameter with the time step increment. (a) Normal Scale;
logarithmic Scale.
(b) logarithmic Scale.
It is
In important
conclusion, to
thepoint out that
algorithm tendsexpression (46) is only
to be more implicit whenused
∆t > to update
η, and the
on the damage
other
variables; however,
hand, more explicit when ∆t stress
for the < η. Inupdate,
any case,expression (45) is non-problematic,
this last situation permanently used sinceeven
for for
explicit integration
very small (if not,
time steps 10−internal
∆t the 6 , an explicit solution may still provide good and accurate
forces would not be in equilibrium).
numerical results. It is important to state that this algorithm admits the damage variable
from the current t+∆t is always saved for the next time step, otherwise viscous
time step dand
6. Numerical Investigation Validation
damage and non-viscous damage would never be updated during the incremental processes.
In Itthis section, the
is important results
to point outfrom the experimental
that expression campaign
(46) is only of [57–59]
used to update with GFRP
the damage
specimens,
variables;beams,
however,andforconnections are tested
the stress update, and compared
expression with the proposed
(45) is permanently used even damage
for
model and the stabilization algorithm. Since most material parameters,
explicit integration (if not, the internal forces would not be in equilibrium). including fracture
energy, are included in these experimental results, it is essential to validate numerical
6. Numerical
models in orderInvestigation
to accuratelyand Validation
simulate the post-peak behavior in materials with softening.
In this section, the results from
Most of these experimental campaigns have the experimental
also beencampaign
verified of [57–59]
with with GFRP
previous numerical
specimens, beams, and connections
tests in the works of [60,61]. are tested and compared with the proposed damage
model and the stabilization algorithm. Since most material parameters, including fracture
For numerical tests, a total time analysis of 1.0 s was admitted, due to the non-linear
energy, are included in these experimental results, it is essential to validate numerical
analysis being static, and a maximum time step increment of 0.001 s, which is 100 times
models in order to accurately simulate the post-peak behavior in materials with softening.
larger than the chosen
Most of these viscosity
experimental parameter.
campaigns have also been verified with previous numerical
tests in the works of [60,61].
6.1. Compact Tension Test
The compact tension test is a traditional test to assess fracture energy, usually in iso-
tropic materials [62], and was recently extended to also estimate the fracture energy in
orthotropic GFRP specimens [58]. The full details of the experimental campaign for the
Appl. Sci. 2023, 13, 1155 12 of 24
For numerical tests, a total time analysis of 1.0 s was admitted, due to the non-linear
analysis being static, and a maximum time step increment of 0.001 s, which is 100 times
larger than the chosen viscosity parameter.
6.1. Compact Tension Test

The compact tension test is a traditional test to assess fracture energy, usually in
isotropic materials [62], and was recently extended to also estimate the fracture energy in
orthotropic GFRP specimens [58]. The full details of the experimental campaign for the
tensile compact test can be found in [58], including the test setup and description of the
imposed displacement.
The specimen geometry, test setup, and supporting conditions are all depicted in
Figure 4. For the adopted mesh, to reduce the computational cost, the zone outside the
notch was simulated with linear orthotropic GFRP behavior that already existed in the
ABAQUS material library. The extra part of the notch was simulated with the proposed
UMAT, in which the fracture will take place. These two used different meshes, since the
zone in the notch needs an extra refinement, and later were connected using tie constraints.
The circular supports were modelled with linear isotropic steel also presented in the
Appl. Sci. 2023, 13, 1155
ABAQUS material library (ES = 195 GPa, ν = 0.3). The interaction between the supports 13 of 25
and the GFRP specimen was simulated with hard contact formulation, and with a friction
coefficient of 0.2.
(a) (b) (c)

Figure 4. 4.
Figure Geometry
Geometryand
andtest
testsetup
setup for theclassical
for the classicalcompact
compact tension
tension testtest
(CT)(CT)
[58].[58]. (a) Geometry
(a) Geometry [mm];
[mm]; (b) experimental layout; (c) laboratory conditions.
(b) experimental layout; (c) laboratory conditions.
ToToeven
evenfurther
further reduce thecomputational
reduce the computational cost,
cost, only
only halfhalf
of theofthickness
the thickness was simu-
was simulated,
lated, therefore
therefore transversal
transversal symmetry
symmetry conditions
conditions were implemented
were implemented in Figure 5.inWhen
Figure
the5.mesh
When
thewas generated,
mesh it imposeditnear-cubic
was generated, imposedshapes for allshapes
near-cubic solid finite
for elements,
all solid and several
finite meshes
elements, and
were tested
several meshes to were
prevent any to
tested mesh dependency.
prevent any mesh dependency.
A full 3D non-linear analysis was performed using the ABAQUS standard with C3D8
solid finite elements using full integration. As a result of the structural response presenting
softening during the last experimental campaign, displacement control was used when
applying a vertical load. All the material properties in plane 1-2 of the GFRP are in Tables 1
and 2, and these were obtained in the experimental campaign of [58]. For the interlaminar
material properties, it was admitted these were the same as the matrix properties, due to
some isotropic behavior in the transverse direction.
Three different tests were performed to evaluate the numerical accuracy of the new
proposed algorithm: The first test consists of a full implicit material integration with
correction parameter v = 1.0; the second test consists of a full explicit material integration
considering v = 0.0; the third test uses ITE with the correction parameter of expression
(47). It is important to point out that the UMAT was used for the GFRP using the ABAQUS
(a) (b) (c)

(a) (b) (c)
Figure 4. Geometry and test setup for the classical compact tension test (CT) [58]. (a) Geometry
[mm]; (b) experimental layout; (c) laboratory conditions.
Appl. Sci. 2023, 13, 1155 13 of 24
To even further reduce the computational cost, only half of the thickness was simu-
lated, therefore transversal symmetry conditions were implemented in Figure 5. When
the mesh was
standard, so the governing generated,
system it imposed
was solved near-cubic
using the implicitshapes for all solid finite
predictor-correct elements, and
algorithm
several
for the three tests. meshes were tested to prevent any mesh dependency.
(a) (b) (c)

Figure 5. Adopted 3D mesh (a), magnification of the proposed UMAT area (b), and symmetry
Figure 5. Adopted 3D mesh (a), magnification of the proposed UMAT area (b), and symmetry
conditions (c) for the CT test.
conditions (c) for the CT test.
Table 1. Material properties forelements
solid finite the pultruded GFRP
using full material used
integration. As ain the CT
result tests.
of the structural response present-
ing softening during the last experimental campaign, displacement control was used
E1 [MPa] E2 [MPa] E3 [MPa] G12 [MPa] G13 [MPa] G23 [MPa]
when applying a vertical load. All the material properties in plane 1-2 of the GFRP are in
29,600 11,900
Tables 1 and 2, and11,900 2900 in the experimental
these were obtained 3000 3000of [58]. For the
campaign
ν12 [−] interlaminar
ν13 [−] material
ν23 properties,
[−] itXwas admitted these were] the same
Yt as the] matrix prop-
t [MPa] Xc [MPa [MPa
erties, due to some isotropic behavior in the transverse direction.
0.27 0.27 0.30 323 426 71
Yc [MPa] Table 1. Material
Zt [MPa ] properties
Zc [MPafor
] the pultruded
S [MPaGFRP
] material
S [MPa used
] in the CT ηtests.
[−]
L T
121 E71[MPa] E121

[MPa] E 64
[MPa] G [MPa]67 G [MPa]
10−5 G [MPa]
29,600 11,900 11,900 2900 3000 3000
ν [−] ν [−] ν [−] X [MPa] X [MPa] Y [MPa]
Table 2. Fracture energy for
0.27the pultruded GFRP material
0.27 used in the CT
0.30 323tests. 426 71
Y [MPa] Z [MPa] Z [MPa] S [MPa] S [MPa] η [−]
Gft [MPa.mm] Gfc [MPa.mm] Gmt [MPa.mm121] Gmc [71
MPa.mm] 121Git [MPa.mm64 ] Gic [67
MPa.mm] 10−5
100 100 20 20 20 20
By observing Figure 6, it is important to realize that all three tests provided the
same structural response, even during the softening branch. The stiffness is the same in
all three tests, but the maximum force is slightly higher in the second test (explicit) as
expected. All numerical models present a higher maximum force than the experimental
campaign, but this was also observed by other authors [63]. Interlaminar properties
are supposed to be the same as the matrix properties, and this assumption can create
slight gaps between numerical and experimental results. Moreover, boundary conditions of
simulations may contain minor discrepancies compared to the experimental test, generating
small differences in the structural response. The final matrix damage field distribution
for all three numerical tests is also presented in Figure 7a, in which it is not possible to
determine the differences between each of the material time integration methods. Figure 7b
also shows the vertical stress distribution and matrix damage for a specific vertical strain.
Finally, Figure 8 compares the final matrix damage field distribution using full implicit
material integration, explicit and the ITE algorithm. For the matrix damage evolution, all
three tests present the same evolution with the vertical strain, but for the vertical stress
evolution during the peak, there is a clear difference between the implicit and explicit
material integration (first and third tests, respectively). However, for the evolution of
the vertical stress and vertical strain, there is almost no visible difference between the
implicit and ITE material integration, concluding that the ITE promotes the same accuracy
as the full implicit formulation. Nevertheless, explicit integration generates noticeable
Figure 8 compares the final matrix damage field distribution using full implicit material
integration, explicit and the ITE algorithm. For the matrix damage evolution, all three tests
present the same evolution with the vertical strain, but for the vertical stress evolution
during the peak, there is a clear difference between the implicit and explicit material inte-
gration (first and third tests, respectively). However, for the evolution of the vertical stress
Appl. Sci. 2023, 13, 1155 and vertical strain, there is almost no visible difference between the implicit and ITE ma- 14 of 24
terial integration, concluding that the ITE promotes the same accuracy as the full implicit
formulation. Nevertheless, explicit integration generates noticeable discrepancies in the
vertical load distribution and stress behavior compared to implicit and ITE analyses. Con-
discrepancies in the vertical load distribution and stress behavior compared to implicit and
sequently, the ITE algorithm is the most cost-effective while maintaining great numerical
ITE analyses. Consequently, the ITE algorithm is the most cost-effective while maintaining
accuracy.
great numerical accuracy.
3.5
Experimental SP1
3 Experimental SP2
Implicit
2.5
ITE Algorithm
Reaction [kN]
2 Explicit
1.5
0.5
0
Appl. Sci. 2023, 13, 1155 0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 15 of 25
Displacement [mm]
Appl. Sci. 2023, 13, 1155 15 of 25
Figure 6.6.Vertical
Figure Verticalforce vs.vs.
force displacement in the
displacement inload supports
the load for thefor
supports CT the
test.CT test.
1
70
0.9 1
0.8 70
0.9 60
0.7 0.8 60
50
Matrix Damage
σ22σ22
0.6 0.7
50
Matrix Damage
0.6 40
Stress
0.5 Implicit Implicit

40
Stress
0.5 Implicit Implicit

0.4 Explicit 30 Explicit
Vertical
0.4 Explicit 30 Explicit

0.3 ITE Algorithm ITE Algorithm
Vertical
0.3 ITE Algorithm 20 ITE Algorithm

20
0.2
0.2
10
10
0.1 0.1
0 0 00
0 0 0.05 0.05 0.1 0.1 0.15 0.15 0.20.2 0.25
0.25 00 0.040.04 0.08 0.08 0.12 0.12 0.16 0.16 0.2 0.2
Vertical
Vertical ℰ22 ℰ22
StrainStrain Vertical ℰ22 ℰ22
StrainStrain
Vertical
(a) (a) (b)(b)

Figure
Figure 7. Tensile
7. Tensile matrix damage vs.
matrix vs. vertical strain (a) and vertical stress vs. vertical strain (b) for (b)
the
Figure 7. Tensile matrixdamage verticalstrain
damage vs. vertical strain(a)(a)
andand vertical
vertical stress
stress vs. vertical
vs. vertical strain strain
(b) for the for the
CT test.
CT CT
test.test.
(a) (b) (c) (d)

(a) (b) field distribution for (a) implicit, (b)(c)
Figure 8. Final matrix damage (d)CT
explicit, and (c) ITE for the
Figure 8. Final matrix damage field distribution for (a) implicit, (b) explicit, and (c) ITE for the CT
test, (d) matrix damage scale.
Figure
test, 8.
(d)Final matrix
matrix damage
damage scale. field distribution for (a) implicit, (b) explicit, and (c) ITE for the CT
test,6.2.
(d) matrix damage scale.
6.2. Three-Point Bending
Three-Point Bending Beam
Beam
The three-point
The three-point bending
bending beambeam is is an
an experimental
experimental studystudy commonly
commonly used used to to extract
extract
6.2.fracture
Three-Point Bending
properties fromBeam
materials. The experimental setup details are presented in [59],
properties from materials. The experimental setup details are presented in [59], in
inThe
which athree-point
which a bilinear
bilinear bending
softening
softening beam
relation
relation is an the
experimental
describes
describes the mechanical
mechanical study commonly
behavior
behavior of damaged
of damaged used pul-
to extract
pultruded
truded
glass
fracture glass fiber-reinforced
fiber-reinforced
properties frompolymer polymer
(GFRP)
materials. The (GFRP) composites.
composites.
experimental onBased
Based setup on the
the details
work work
of are
[61], of material
the [61], thein [59],
presented
material
properties properties
adopted adopted
for the for the
numerical numerical
test are test
listed are
in listed
Table 3.
in which a bilinear softening relation describes the mechanical behavior of damaged in Table
Table 4 3. Table
contains 4
the contains
fracture pul-
truded glass fiber-reinforced polymer (GFRP) composites. Based on the work com-
the fracture
energies energies
adopted in adopted
this in thisFor
framework. framework.
the sake For
of the sake
clarity, of clarity,
tensile and tensile andvalues
compressive of [61], the
pressive
of fracturevalues of fracture
energies energies
for the fiber for the fiber
are estimated, whileareanestimated, whileofan
average value average
bilinear value of
behavior is
material properties adopted for the numerical test are listed in Table 3. Table 4 contains
bilinear
taken intobehavior is taken
consideration forinto
theconsideration
matrix. Moreover, for the matrix. Moreover,
interlaminar interlaminar
fracture energy frac-
is assumed
theto fracture
ture
beenergy energies
the same as theadopted
is assumed in same
this framework.
to beproperty.
matrix the as the matrix For the sake of clarity, tensile and com-
property.
pressive values of fracture energies for the fiber are estimated, while an average value of
bilinear
Table behavior is taken for
3. Material properties into
theconsideration
pultruded GFRP for the used
material matrix.
in theMoreover, interlaminar
three-point bending beam. frac-
ture energy
E [MPa]is assumed to
E [MPa]be the same as
E [MPa] the matrix
G [MPa]property.
G [MPa] G [MPa]
30,000 10,800 10,800 2000 2000 2000
Appl. Sci. 2023, 13, 1155 15 of 24
Table 3. Material properties for the pultruded GFRP material used in the three-point bending beam.

30,000 10,800 10,800 2000 2000 2000
ν12 [−] ν13 [−] ν23 [−] Xt [MPa] Xc [MPa] Yt [MPa]
0.24 0.24 0.4 323 423 37
Yc [MPa] Zt [MPa] Zc [MPa] SL [MPa] ST [MPa] η [−]
80 37 80 33 33 10−5
Table 4. Fracture energy for the pultruded GFRP material used in the three-point bending beam.
Gft [MPa.mm] Gfc [MPa.mm] Gmt [MPa.mm] Gmc [MPa.mm] Git [MPa.mm] Gic [MPa.mm]
130 130 3.84 3.84 3.84 3.84
Appl. Sci. 2023, 13, 1155 16 of 25
The computational cost is reduced by considering only one-quarter of the beam, thus
requiring symmetry conditions, as illustrated in Figure 9. Regarding the boundary conditions,
behavior, in contrast to the region of the notch where the damage occurs, which is 16
Appl. Sci. 2023, 13, 1155
mod-of 25
the beam is simply supported, and the vertical displacement is applied through isotropic
elled using almost cubic C3D8R solid finite elements with reduced integration and hour-
steel support (Es = 210 GPa, ν = 0.3), according to [61]. The part of the beam far from the
glass control, as shown in Figure 10. Tie constraints allow the connection between all parts
notch proposes abehavior,
coarse discretization, and it is modelled with linear orthotropic behavior, in
of the beam. in contrast to the region of the notch where the damage occurs, which is mod-
contrast to the region of the notch where
elled using almost cubic C3D8R thesolid
damage occurs, with
finite elements which is modelled
reduced using
integration almost
and hour-
cubic C3D8R solidglassfinite elements
control, as shownwith reduced
in Figure 10. Tieintegration andthe
constraints allow hourglass
connectioncontrol,
betweenasall shown
parts
in Figure 10. Tieofconstraints
the beam. allow the connection between all parts of the beam.
(a) (b)
(a) (b)
Figure 9. Geometry (a) and symmetric boundary conditions (b) of the three-point bending beam.
Figure
Figure 9. Geometry
9. Geometry (a)(a)
andand symmetric
symmetric boundary
boundary conditions
conditions (b)(b)
of of
thethe three-point
three-point bending
bending beam.
beam.
(a) (b)
Figure 10. Adopted 3D mesh (a) and magnification of the proposed UMAT area (b) for the three-
point bending beam.
(a) (b)
A full 3D non-linear analysis was performed using the ABAQUS standard, without
Figure considering residual stress and with ofthethe
same time stepUMAT
restrictions as the previous nu-
Figure 10. Adopted 3Dmesh
10. Adopted 3D mesh(a)(a)and
andmagnification
magnification of the proposed
proposed UMAT area area(b)
(b)
forfor
thethe three-
three-point
point bending beam.merical model. This framework proposes a full implicit material integration as the first
bending beam. analysis, followed by an ITE test with the correction parameters defined in (47) as the
second test.
AA full 3D3D
full non-linear
non-linear analysis
The response of was
analysis thewas performed
performed
structure using
in terms thethe
using
of ABAQUS
the force vs.standard,
ABAQUS
reaction standard,
CMOD without
(crack without
mouth
considering residual stress and with the same time step restrictions as
considering residual stress and with the same time step restrictions as the previousasnumeri-
opening displacement) is compared with the experimental data the
included previous
in [59], nu-
illus-
merical model. This trated in Figure 11.proposes
framework By considering bothimplicit
a full full implicit and ITE tests,
material Figure 12b
integration asexhibits the
the first
evolution of principal stress over two different integration points, identified in Figure 12a,
analysis, followed by an ITE test with the correction parameters defined in (47) as the
in the band where the damage propagates. A detailed view of damage propagation is
second test. provided in Figure 13; specifically, the matrix is shown in Figure 13a, the fiber in Figure
The response13c, of the structure
and tensile in terms
damages can beof the reaction
recognized, whileforce
Figurevs.
13bCMOD (crack
shows shear 1-2 mouth
damage
progression.
opening displacement) Finally, Figure
is compared with14the compares the final distribution
experimental of tensile
data included matrixasdamage
in [59], illus-
using full implicit material integration and the ITE algorithm. The numerical assessment
Appl. Sci. 2023, 13, 1155 16 of 24
cal model. This framework proposes a full implicit material integration as the first analysis,
followed by an ITE test with the correction parameters defined in (47) as the second test.
The response of the structure in terms of the reaction force vs. CMOD (crack mouth
opening displacement) is compared with the experimental data included in [59], as illus-
trated in Figure 11. By considering both full implicit and ITE tests, Figure 12b exhibits the
evolution of principal stress over two different integration points, identified in Figure 12a,
in the band where the damage propagates. A detailed view of damage propagation is
provided in Figure 13; specifically, the matrix is shown in Figure 13a, the fiber in Figure 13c,
and tensile damages can be recognized, while Figure 13b shows shear 1-2 damage progres-
sion. Finally, Figure 14 compares the final distribution of tensile matrix damage using full
implicit material integration and the ITE algorithm. The numerical assessment correctly
simulates the experimental campaign, even though some differences in the softening region
can be detected. In particular, the numerical model contains an average value of matrix
fracture energy in contrast to the bilinear behavior. Furthermore, the reduced integra-
tion results in a smaller characteristic length. In addition, small inconsistencies between
Appl. Sci. 2023, 13, 1155 experimental and numerical boundary conditions could induce some variabilities. 17 of 25 The
Appl. Sci. 2023, 13, 1155
principal stress and strain obtained from the full implicit case and ITE are coincident, as 17 of 25
shown in Figure 12b. Therefore, the ITE algorithm simulates the mechanical response of the
response
beam with of the
the same
beam accuracy
with the same
as theaccuracy as theanalysis,
full implicit full implicit
withanalysis, with theofad-
the advantage requiring
vantage
significantly response
of requiring
less of the beam
significantly
computation less with
time. The the
computationsamedamage
matrix accuracy
time. as the
Theover
matrix full implicit
damage
integration over analysis,
pointinte- with the ad-
2 evolves
gration
more rapidly vantage
point 2than
evolves
point of
more requiring significantly
rapidly than
1. However, less computation
point 1. However,
by looking time.
by looking
at Figure 13c, The
there isatan matrix
Figure damage over inte-
13c, in fiber
increase
there is anthat gration
increase point
in fiber 2
damage evolves more
that causes rapidly
the shear than point
damage that1. However,
at point by
1 to exceedlooking
that at Figure 13c,
damage causes the shear damage at point 1 to exceed at point 2 from the cross
there is an increase in fiber damage
at point 2 from the cross point, as highlighted in Figure 13b. that causes the shear damage at point 1 to exceed that
point, as highlighted in Figure 13b.
at point 2 from the cross point, as highlighted in Figure 13b.
4.5
4.5
4
4
3.5
3.5
3
Force [kN]
3
2.5
Force [kN]
2.5
2 Experimental 1
2 Experimental 2 Experimental 1
1.5
Experimental 3 Experimental 2
1.5
1 Experimental 4 Experimental 3
1 Implicit Experimental 4
0.5 ITE algorithm Implicit
0.5 ITE algorithm
0
0 0.2 0 0.4 0.6 0.8 1 1.2
0 CMOD
0.2 [mm] 0.4 0.6 0.8 1 1.2
CMOD [mm]
Figure
Figure11.
11.Vertical
Verticalload vs.vs.
load CMOD
CMODfor for
implicit and and
implicit ITE for
ITEthe
forthree-point bending
the three-point beam. beam.
bending
Figure 11. Vertical load vs. CMOD for implicit and ITE for the three-point bending beam.
50
45 50
40 45
Principal stress [MPa]
35 40
Principal stress [MPa]
30 35
25 30
20 25
15 Implicit
20 - point 1
Implicit - point 2 Implicit - point 1
10 15
ITE algorithm - point 1 Implicit - point 2
5 10
ITE algorithm - point 2 ITE algorithm - point 1
0 5 ITE algorithm - point 2
0 0.05 0 0.1 0.15 0.2
0 Principal strain
0.05[-] 0.1 0.15 0.2
(a) (b) Principal strain [-]

(a) (b)
Figure 12. Identification of integration points 1 and 2 (a) and distribution of principal stress vs. prin-
cipal strain (b) for Figure
the three-point bending of
12. Identification beam.
integration points 1 and 2 (a) and distribution of principal stress vs. prin-
Figure 12. Identification of integration points 1 and 2 (a) and distribution of principal stress vs.
cipal strain (b) for the three-point bending beam.
principal strain (b) for the three-point bending beam.
1 1
0.9 1 0.9 1
0.8 0.9 0.8 0.9
ITE algorithm - point 1
damage 1-2
0.7 0.8 0.7 0.8 ITE algorithm - point 1

ix damage

age 1-2
0.6 0.7 ITE algorithm - point 2 0.6 0.7 ITE algorithm - point 2 ITE algorithm - point 1
amage
0.5 0.6 ITE algorithm - point

0.52 0.6 Cross point ITE algorithm - point 2
0
0 0.05 0.1 0.15 0.2
Principal strain [-]
(a) (b)
Appl. Sci. 2023, 13, 1155 17 of 24
Figure 12. Identification of integration points 1 and 2 (a) and distribution of principal stress vs. prin-
cipal strain (b) for the three-point bending beam.
1 1
0.9 0.9
0.8 0.8
Shear damage 1-2

0.7 0.7 ITE algorithm - point 1
Matrix damage
0.6 ITE algorithm - point 2 0.6 ITE algorithm - point 2

0.5 0.5 Cross point
0.4 0.4
0.3 0.3
0.2 0.2
0.1 0.1
0 0
0 0.025 0.05 0.075 0.1 0.125 0.15 0 0.025 0.05 0.075 0.1 0.125 0.15
Appl. Sci. 2023, 13, 1155 Principal strain [-] Principal strain [-] 18 of 25
Appl. Sci. 2023, 13, 1155

(a) (b) 18 of 25
0.9
0.8 ITE algorithm - point 1
0.70.9 ITE algorithm - point 2

Fibre damage
0.8
0.6
0.7 ITE algorithm - point 2
0.5
Fibre damage
0.6
0.4
0.5
0.3
0.4
0.2
0.3
0.10.2
00.1
0 0.025 0.05 0.075 0.1 0.125 0.15
0
0 0.025 Principal
0.05 0.075strain 0.1
[-] 0.125 0.15
Principal strain [-]
(c)
(c)
Figure 13. Tensile matrix damage vs. principal strain (a), shear damage vs. principal strain (b), and
Figure
Figure
tensile 13. Tensile
13.
fiber Tensile
damage
matrix
matrix
(c)
damage
damage
vs.
vs.
vs. principal
principal principal
strain
strain
strain
over
(a),
(a), shear
shear damage
integration damage
points
vs.
vs.
1
principal
principal
and 2 with
strain
strain
ITE
(b),
(b), and
and
algorithm for
tensile
tensile fiber
fiber damage
damage(c)
(c)vs.
vs.principal
principalstrain
strainover
overintegration
integrationpoints 1 and
points 2 with
1 and ITEITE
2 with algorithm for the
algorithm for
thethree-point
three-point bending
bending beam.beam.
the three-point bending beam.
(a) (b) (c)

(a) (b)
Figure 14. Final tensile matrix damage distribution (c)three-point
for (a) implicit and (b) ITE for the
Figure 14. Final tensile matrix damage distribution for (a) implicit and (b) ITE for the three-point
bending test; (c) matrix damage scale.
Figure 14. test;
bending Final(c)tensile
matrix matrix
damage damage
scale. distribution for (a) implicit and (b) ITE for the three-point
bending test; (c) matrix damage scale.
6.3. Double-Lap Test
The double-lap test is a standard experiment used to assess the bolt connection be-
tween structural elements. The full details of the experimental campaign and test setup
areThe
fullydouble-lap
described intest is a standard
the work of [57], in experiment used to
which it is possible assess thethe
to determine bolt connection
applied load be-
tween structural elements. The full details of the experimental campaign
using impose displacements. Some of the material properties adopted for the numericaland test setup
Appl. Sci. 2023, 13, 1155 18 of 24

The double-lap test is a standard experiment used to assess the bolt connection between
structural elements. The full details of the experimental campaign and test setup are fully
described in the work of [57], in which it is possible to determine the applied load using
impose displacements. Some of the material properties adopted for the numerical test are
shown in Table 5, and these were obtained in the previously referenced work while others
were estimated. Due to some findings’ fluctuations, the fracture energy was calculated
using the same values as in the earlier example in Table 2. For this numerical test, the bolt
edge distances (cover) of 15 mm and 70 mm were simulated since these clearly show a
difference between a shear-out and bearing collapse, respectively, as shown in Figure 15.
Table 5. Material properties for the pultruded GFRP material used in the double-lap tests.

21,300 2900 2900 3000 2500 2500
Appl. Sci. 2023, 13, 1155 ν12 [−] ν13 [−] ν23 [−] Xt [MPa] Xc [MPa] 19 of
Yt [MPa ] 25
0.28 0.28 0.30 334.0 316.0 29.0
Yc [MPa] Zt [MPa] Zc [MPa] SL [MPa] ST [MPa] η [−]
Y [MPa] Z [MPa] Z [MPa] S [MPa] S [MPa] η [−]
51.9
51.9 29.0
29.0 51.9
51.9 52.4
52.4 33.8
33.8 10−
10
5
—5
Figure 15. Test setup for the loading conditions and impose displacement for the double-lap test,
Figure 15. Test setup for the loading conditions and impose displacement for the double-lap test, and
and used bolt edge distance (cover) for the circular opening [57].
used bolt edge distance (cover) for the circular opening [57].
and C3D8R solid finite elements using full and reduced integration and hourglass control
and C3D8R solid finite elements using full and reduced integration and hourglass control
to evaluate any mesh sensitivity due to the high variability of the stress field. The time
to evaluate any mesh sensitivity due to the high variability of the stress field. The time step
step restrictions are the same as the previous numerical models.
restrictions are the same as the previous numerical models.
For
For the GFRP plate,
the GFRP plate,in
inorder
ordertotoreduce
reducethe thecomputational
computational cost,
cost, part
part of the
of the plate
plate near
near the
the opening
opening waswas modelled
modelled with with the UMAT,
the UMAT, but forbutthefor thepart,
other other anpart, anorthotropic
elastic elastic orthotropic
material
material
from thefrom
ABAQUSthe ABAQUS library
library was used.was
Forused. For the
the steel bolt,steel bolt, an isotropic
an isotropic elastoplastic
elastoplastic material
from the ABAQUS library was used (Es = 195 GPa, ν = 0.3, f y = 300 MPa, f u = 𝑀𝑃𝑎,
material from the ABAQUS library was used (𝐸 = 195 𝐺𝑃𝑎, 𝜈 = 0.3, 𝑓 = 300 𝑓 =
691 MPa).
691 𝑀𝑃𝑎). Again, to reduce the computational cost, only one-quarter of the
Again, to reduce the computational cost, only one-quarter of the structure was simulated, structure was
simulated, which one-half
which represents represents one-half
of the GFRPofplate
the GFRP
and one plate
steeland oneassteel
plate, shownplate,
in as shown
Figure 16, in
in
Figure 16, in which the imposed vertical displacement is inserted into the
which the imposed vertical displacement is inserted into the bottom of the steel plate. Only bottom of the
steel plate.
the ITE Only thewas
algorithm ITEemployed
algorithmin was
thisemployed
example,inwith thisthe
example,
primary with
goaltheofprimary goal
numerically
of numerically evaluating the simulation of the bearing collapse in the contact
evaluating the simulation of the bearing collapse in the contact between the steel bolt and between
the
the steel
GFRPbolt and the GFRP plate.
plate.
material from the ABAQUS library was used (𝐸 = 195 𝐺𝑃𝑎, 𝜈 = 0.3, 𝑓 = 300 𝑀𝑃𝑎, 𝑓 =
691 𝑀𝑃𝑎). Again, to reduce the computational cost, only one-quarter of the structure was
simulated, which represents one-half of the GFRP plate and one steel plate, as shown in
Figure 16, in which the imposed vertical displacement is inserted into the bottom of the
Appl. Sci. 2023, 13, 1155 steel plate. Only the ITE algorithm was employed in this example, with the primary 19
goal
of 24
of numerically evaluating the simulation of the bearing collapse in the contact between
the steel bolt and the GFRP plate.
(a) (b) (c) (d)

Figure 16. Adopted 3D mesh for 70 mm cover (a) and 15 mm cover (b) and symmetry conditions
Figure 16. Adopted 3D mesh for 70 mm cover (a) and 15 mm cover (b) and symmetry conditions for
for 15 mm cover (c) and 70 mm cover (d) for the double-lap test.
15 mm cover (c) and 70 mm cover (d) for the double-lap test.
To correctly simulate the bearing, the residual compressive stress after the maximum
peak must also be simulated. Two values of compressive residual stress were considered,
namely, 15%, as estimated from the work of [60], and 80%, which is approximately the relation
between the bearing resistance and compressive resistance according to [64], when using the
experimental values form [57]. In addition, to improve convergence, the tensile residual stress
was 1% and 10%, respectively. A sensitivity test showed that residual tensile stress below 15%
had no influence on the maximum numerical resistance and its softening branch.
The structural response of the numerical models and its comparison with the experi-
mental campaign is exhibited in Figure 17 for both 15 mm and 70 mm cover, with 15% and
80% residual compressive strength, for both reduced (RED) and full (FULL) integration.
The vertical force was measured in the steel plate, and the vertical displacement is the
relative movement between the steel plate and the GFRP plate (AB distance Figure 15). As
expected, due to the extra flexibility of the experimental test setup, the numerical model
is less flexible than the experimental one. This is a common problem when comparing
numerical vs. experimental structural responses [65]. The magnitude of the stiffness in the
numerical model was dependent on the adopted final geometry and support conditions,
therefore due to its unknown variability, it is not expected to reach the same value as the one
obtained in the experimental campaign. For the 15 mm cover, there is a clear overestimation
of the maximum reaction for both reduced and full integration, which is associated with
the tensile matrix resistance and longitudinal 1-2 shear resistance being overestimated.
Furthermore, the adopted value of the fracture energy was estimated, not measured, in an
experimental campaign. However, for the 70 mm cover, the maximum reaction is near the
average experimental one, since the bearing is in the collapse mode, and this is associated
with the compressive resistance, which was not estimated. It is observed that for the 15 mm
cover, there is no clear difference between adopting a residual stress of 15% or 80% since
the collapse mode is shear-out. However, for the 70 mm cover, clearly, the 80% residual
compressive stress produces better results when compared with the post-peak values of
the experimental campaign.
The final shear 1-2 damage field for all numerical models with reduced and full
integration is portrayed in Figure 18, in which it is possible to perceive that for the 15 mm
cover, it is clearly a full shear-out collapse, but for the 70 mm cover, there is a bearing
collapse followed by a partial shear-out mode that is not finalized. This is in accordance
with the findings from the experimental campaign up to a 5 mm relative displacement.
being overestimated. Furthermore, the adopted value of the fracture energy was esti-
mated, not measured, in an experimental campaign. However, for the 70 mm cover, the
maximum reaction is near the average experimental one, since the bearing is in the col-
lapse mode, and this is associated with the compressive resistance, which was not esti-
mated. It is observed that for the 15 mm cover, there is no clear difference between adopt-
Appl. Sci. 2023, 13, 1155 20 of 24
ing a residual stress of 15% or 80% since the collapse mode is shear-out. However, for the
70 mm cover, clearly, the 80% residual compressive stress produces better results when
compared with the post-peak values of the experimental campaign.
16
7 Experimental 1
Experimental 2
14
Experimental 3
6
full_15%_15
12
full_80%_15
5 Experimental 1
red_15%_15
Force [kN]
Force [kN]
10 Experimental 2
red_80%_15 Experimental 3
4
8 full_15%_70
full_80%_70
3 6 red_15%_70
red_80%_70
2 4
1 2
0 0
0 0.2 0.4 0.6 0.8 1 0 1 2 3 4 5
Displacement [mm] Displacement [mm]
(a) (b)
Appl. Sci. 2023, 13, 1155 21 of 25
Figure 17. Force vs. displacement for (a) 15 mm cover and (b) 70 mm cover for the double-lap test.
Figure 17. Force vs. displacement for (a) 15 mm cover and (b) 70 mm cover for the double-lap test.
The final shear 1-2 damage field for all numerical models with reduced and full inte-
gration is portrayed in Figure 18, in which it is possible to perceive that for the 15 mm
cover, it is clearly a full shear-out collapse, but for the 70 mm cover, there is a bearing
collapse followed by a partial shear-out mode that is not finalized. This is in accordance
with the findings from the experimental campaign up to a 5 mm relative displacement.
RED_15%_15 mm RED_80%_15 mm FULL_15%_15 mm FULL_80%_15 mm

Figure 18. Final shear damage distribution for 15 mm of cover for the double-lap test, with re-
duced/full integration (RED/FULL) with 15% and 80% residual compressive stress.
Furthermore, only for the 70 mm cover can it be seen in Figure 18 that some differ-
Furthermore, only for the 70 mm cover can it be seen in Figure 18 that some differences
ences exist in the final shear damage field distribution for 15% and 80% residual compres-
exist in the final shear damage field distribution for 15% and 80% residual compressive
sive stress. Similar to the structural response in Figure 19, there are some differences in
stress. Similar to the structural response in Figure 19, there are some differences in the
the final shear damage field distribution for reduced and full integration with 80% resid-
final shear damage field distribution for reduced and full integration with 80% residual
ual compressive stresses.
compressive stresses.
In Figure 20a, the shear stress and shear damage evolution are presented in the Gauss
point nearest the circular opening perimeter at 30◦ of the symmetry axis in the GFRP plate.
The findings are shown for both 15 mm and 70 mm covers, with residual compressive stress
levels of 15% and 80%. These findings are significant because they show that increasing the
residual compressive stress has almost no effect on the residual shear stress and that this is
more related to the level of cover, which influences the stress field indirectly. This is also
the case for the shear damage evolution displayed in Figure 20b.
Finally, to better understand that the proposed ITE algorithm does not overestimate
the viscous energy and consequently the viscous damage, the elastic damaged energy
(ESE) and the viscous dissipation energy (VSE) are exhibited in Figure 21. As expected for
the RED_15%_15, there is a big reduction of force after 0.4 s, as shown in Figure 17, and
as expected in this case, both the value of ESE and VSE increase, but after a given time
step, VSE stabilizes and ESE increases, demonstrating that the VSE does not dominate the
RED_15%_70 mm structural response.mm
RED_80%_70 For RED_80%_70, this does notmm
FULL_15%_70 occur due toFULL_80%_70
the small forcemm
reduction
during the structural response, in which VSE is always very small.
In Figure 20a, the shear stress and shear damage evolution are presented in the Gauss
point nearest the circular opening perimeter at 30° of the symmetry axis in the GFRP plate.
Furthermore, only for the 70 mm cover can it be seen in Figure 18 that some differ-
ences exist in the final shear damage field distribution for 15% and 80% residual compres-
Appl. Sci. 2023, 13, 1155 sive stress. Similar to the structural response in Figure 19, there are some differences in 21 of 24
the final shear damage field distribution for reduced and full integration with 80% resid-
ual compressive stresses.
Appl. Sci. 2023, 13, 1155 22 of 25
50 1
45 red_15%_15 0.9
40 red_80%_15 0.8
35 red_15%_70 0.7
Shear Stress Ƭ12
Shear Damage
30 red_80%_70 0.6
RED_15%_70 mm RED_80%_70 mm FULL_15%_70 mm FULL_80%_70
red_15%_15 mm
25 0.5
red_80%_15
Appl. Sci. 2023, 13,
20 1155 Figure 19. Final shear damage distribution
0.4 for 70 mm of cover for the double-lap test, with re- 22 of 25
Figure 19. Final shear damage distribution for 70 mm of cover for thered_15%_70
double-lap test, with re-
15 duced/full integration (RED/FULL) with
0.3 15% and 80% residual compressive stress.
red_80%_70
10 0.2
5
In Figure 20a, the shear stress and
0.1
shear damage evolution are presented in the Gauss
50
0
point nearest the circular opening perimeter
0
1 at 30° of the symmetry axis in the GFRP plate.
45 0 0.05 The 0.1findings are
0.15shown for0.2
red_15%_15 both 150.9 mm
0 and 70 0.05
mm covers, with
0.1 residual0.15compressive0.2
Shear Strain γ12 Shear Strain γ12
40 stress levels of red_80%_15
15% and 80%. These 0.8 findings are significant because they show that in-
35 (a)
creasing the residual compressive stress
red_15%_70 0.7 (b)on the residual shear stress
has almost no effect
Shear Stress Ƭ12
Shear Damage
30 and that this is red_80%_70

more related to the0.6level of cover, which influences thered_15%_15
stress field indi-
Figure 20. Values in the bolt opening 0.5 at 30ᵒ for 15 mm and 70 mm of cover with reduced integration:
25 rectly. This is also the case for the shear damage evolution displayed in Figure 20b.
red_80%_15
20
(a) Shear stress vs. shear strain; (b) shear
0.4
damage vs. shear strain cover for the double-lap test.
red_15%_70
15 0.3
red_80%_70
10 Finally, to better understand0.2that the proposed ITE algorithm does not overestimate
5 the viscous energy and consequently 0.1 the viscous damage, the elastic damaged energy
0 0
0 0.05
(ESE) and0.1
the viscous
0.15
dissipation
0.2
energy
0
(VSE) are exhibited
0.05 0.1
in Figure
0.15
21. As expected
0.2
for
Strain γ12
the Shear
RED_15%_15, there is a big reduction of force after Shear
0.4 s,γas
Strain 12
shown in Figure 17, and
(a)
as expected in this case, both the value of ESE and VSE (b) increase, but after a given time
step,
Figure VSE stabilizes
20. Values in theand ESE
bolt increases,
opening at 30ᵒ demonstrating
for 15 mm and 70that mmthe VSE with
of cover doesreduced
not dominate the
integration:
Figure 20. Values in the bolt opening at 30◦ for 15 mm and 70 mm of cover with reduced integration:
structural response. For RED_80%_70, this does not occur due to the small
(a) Shear stress vs. shear strain; (b) shear damage vs. shear strain cover for the double-lap test. force reduction
(a) Shear stress vs. shear strain; (b) shear damage vs. shear strain cover for the double-lap test.
Finally, to better understand that the proposed ITE algorithm does not overestimate
700
the viscous energy and consequently the viscous damage, the elastic damaged energy
600 (ESE) and the viscous dissipation energy (VSE) are exhibited in Figure 21. As expected for
500 the RED_15%_15, there is a big reduction of force after 0.4 s, as shown in Figure 17, and
as expected in this case, both the value of ESE and VSE increase, but after a given time
Energy [mJ]
400
step, VSE stabilizes andESEESE increases, demonstrating that the VSE does not dominate the
300
structural response. ForVSERED_80%_70, this does not occur due to the small force reduction
200
100
700
0
0 0.2 0.4 0.6 0.8 1
600 Time [s]
500 (a) (b)

Energy [mJ]
400 Figure 21. Total elastic and viscous damage energy for (a) reduced integration for 15 mm cover with
Figure 21. Total elastic and
ESE viscous damage energy for (a) reduced integration for 15 mm cover
300 15% residual compressive stress; (b) reduced integration for 70 mm cover with 80% residual com-
with 15% residual compressive
VSE stress; (b) reduced integration for 70 mm cover with 80% residual
pressive stress cover for the double-lap test.
200 compressive stress cover for the double-lap test.
100
7. Conclusions
7. Conclusions
0
0.4 The primary
0 0.2
The0.6primarygoal
goalof
0.8
ofthis
1 study was to present a new 3D Hashin-based damage model
this study was to present a new 3D Hashin-based damage model
Timeresidual
with [s] stresses and
and the
with residual stresses theproposal ofof
proposal anan
implicit-to-explicit (ITE)(ITE)
implicit-to-explicit material time integra-
material time inte-
(a) (b)
gration algorithm to be used in this and future damage models. This study was applied
to a published
Figure experimental
21. Total elastic campaign
and viscous damagewith specimens
energy and beams
for (a) reduced made
integration forof
15GFRP compo-
mm cover with
sites.
15% residual compressive stress; (b) reduced integration for 70 mm cover with 80% residual com-
pressive
(1) Thestress cover for
proposed the double-lap
damage model wastest. able to correctly simulate the structural response
Appl. Sci. 2023, 13, 1155 22 of 24
tion algorithm to be used in this and future damage models. This study was applied to a
published experimental campaign with specimens and beams made of GFRP composites.
(1) The proposed damage model was able to correctly simulate the structural response
of the CT test. The results are consistent with the experiments up to the peak, while
small discrepancies can be detected in the softening behavior.
(2) The ITE algorithm was also used to simulate the three-point bending test. Slight
differences with experimental results in the post-peak region can be detected due
to the estimation of some material properties and the imperfect correspondence of
boundary conditions.
(3) The use of residual stress in the orthotropic Hashin-based damage model allowed the
correct simulation of the bearing collapse in GFRP connections. However, the effect of
residual stresses on compressive behavior has a small influence on residual shear stress.
(4) The use of the ITE algorithm was robust, and in initial tests, present the same accuracy
as a fully implicit material time integration method when using the secant method in
ABAQUS standard predictor-corrector algorithm.
Further Developments
In future research, it is expected to develop the orthotropic damage model based on
the Hashin failure criterion using residual shear stresses that are important for simulating
shear-out collapse in composite connections. In addition, residual shear stress is also
important for the correct numerical behavior of the 3D beam-to-column connection for
cycle loads and dynamic analysis [57]. Moreover, a variation of the 2D Tsai-Wu-based
damage model [61] is proposed for a 3D analysis, using a similar criterion as was used in
this work. It is expected to use these orthotropic damage models in the future design of
GFRP structures.
Author Contributions: Conceptualization, M.R.T.A., M.T. and A.P.; methodology, M.R.T.A., M.T.
and A.P.; software, M.R.T.A. and M.T.; validation, M.R.T.A., M.T. and A.P.; formal analysis, M.R.T.A.
and M.T.; investigation, M.R.T.A., M.T. and A.P.; writing—original draft preparation, M.R.T.A. and
M.T.; writing—review and editing, M.T. and A.P.; visualization, M.R.T.A., M.T. and A.P.; supervision,
M.R.T.A. and A.P. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: Not applicable.
Acknowledgments: This work was partly supported by Fundação para a Ciência e Tecnologia (FCT)
under the Transitional Standard—DL57/2016/N3/UI/CERIS/CT/165/2018.
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Martins, D.; Proença, M.; Correia, J.R.; Gonilha, J.; Arruda, M.; Silvestre, N. Development of a novel beam-to-column connection
system for pultruded GFRP tubular profiles. Compos. Struct. 2017, 171, 263–276. [CrossRef]
2. Li, B.; Gong, Y.; Xiao, H.; Gao, Y.; Liang, E. A Two-Dimensional Model for Pin-Load Distribution and Failure Analysis of
Composite Bolted Joints. Materials 2021, 14, 3646. [CrossRef] [PubMed]
3. El Kadi, M.; Van Hemelrijck, D.; Tysmans, T. Improving the Anchorage in Textile Reinforced Cement Composites by 3D Spacer
Connections: Experimental Study of Flexural and Cracking Behaviors. J. Compos. Sci. 2022, 6, 357. [CrossRef]
4. Kong, S.Y.; Wong, L.S.; Paul, S.C.; Miah, M.J. Shear Response of Glass Fibre Reinforced Polymer (GFRP) Built-Up Hollow and
Lightweight Concrete Filled Beams: An Experimental and Numerical Study. Polymers 2020, 12, 2270. [CrossRef] [PubMed]
5. Hu, H.; Wei, Q.; Liu, B.; Liu, Y.; Hu, N.; Ma, Q.; Wang, C. Progressive Damage Behaviour Analysis and Comparison with 2D/3D
Hashin Failure Models on Carbon Fibre–Reinforced Aluminium Laminates. Polymers 2022, 14, 2946. [PubMed]
6. Girão Coelho, A.M.; Toby Mottram, J.; Harries, K.A. Finite element guidelines for simulation of fibre-tension dominated failures
in composite materials validated by case studies. Compos. Struct. 2015, 126, 299–313. [CrossRef]
Appl. Sci. 2023, 13, 1155 23 of 24
7. Correia, J.R.; Martins, D.; Gonilha, J.; Arruda, M.; Andre, C.; Nascimento, J.; Branco, F. Clickhouse project an all composite
emergency housing system. In Proceedings of the Conference on Advances in Composite Materials and Structures, Instambul
CACM; 2015. Available online: https://www.researchgate.net/publication/329427145_CACM2015_Title_Clickhouse_project-
an_all-composite_emergency_housing_system (accessed on 28 December 2022).
8. Arruda, M.R.T.; Lopes, B. Pre-design guidelines for GFRP composite sandwich panels. Eng. Solid Mech. 2020, 8, 169–186.
[CrossRef]
9. Arteiro, A.; Catalanotti, G.; Reinoso, J.; Linde, P.; Camanho, P.P. Simulation of the Mechanical Response of Thin-Ply Composites:
From Computational Micro-Mechanics to Structural Analysis. Arch. Comput. Methods Eng. 2019, 26, 1445–1487. [CrossRef]
10. Chowdhury, U.; Wu, X.-F. Cohesive Zone Modeling of the Elastoplastic and Failure Behavior of Polymer Nanoclay Composites. J.
Compos. Sci. 2021, 5, 131. [CrossRef]
11. Matzenmiller, A.; Lubliner, J.; Taylor, R.L. A constitutive model for anisotropic damage in fiber-composites. Mech. Mater. 1995, 20,
125–152. [CrossRef]
12. Lopes, B.; Arruda, M.R.T.; Almeida-Fernandes, L.; Castro, L.; Silvestre, N.; Correia, J.R. Assessment of mesh dependency in the
numerical simulation of compact tension tests for orthotropic materials. Compos. Part C Open Access 2020, 1, 100006. [CrossRef]
13. ABAQUS. Abaqus Unified FEA-3DEXPERIENCE R2018; Systèmes, D., Ed.; 3DS-SIMULIA: Rue Marcel Dassault, Vélizy-
Villacoublay, France, 2018.
14. ANSYS. ANSYS Structural Mechancis; ANSYS Inc.: Zanker Road, San Jose, CA, USA, 2015.
15. Lapczyk, I.; Hurtado, J.A. Progressive damage modeling in fiber-reinforced materials. Compos. Part A Appl. Sci. Manuf. 2007, 38,
2333–2341. [CrossRef]
16. Kang, I.-K.; Kim, S.-H. Compressive Strength Testing of Hybrid Concrete-Filled Fiber-Reinforced Plastic Tubes Confined by
Filament Winding. Appl. Sci. 2021, 11, 2900. [CrossRef]
17. Arruda, M.R.T.; Castro, L.M.S.; Ferreira, A.J.M.; Martins, D.; Correia, J.R. Physically non-linear analysis of beam models using
Carrera Unified Formulation. Compos. Struct. 2018, 195, 60–73. [CrossRef]
18. Nahas, M. Survey of Failure and Post-Failure Theories of Laminated Fiber-Reinforced Composites. J. Compos. Technol. Res. 1986,
8, 138–153.
19. Kaddour, A.; Hinton, M. Maturity of 3D failure criteria for fibre-reinforced composites: Comparison between theories and
experiments: Part B of WWFE-II. J. Compos. Mater. 2013, 47, 925–966. [CrossRef]
20. Hashin, Z. Fatigue Failure Criteria for Unidirectional Fiber Composites. J. Appl. Mech. 1981, 48, 846–852. [CrossRef]
21. Camanho, P.P.; Matthews, F.L. A Progressive Damage Model for Mechanically Fastened Joints in Composite Laminates. J. Compos.
Mater. 1999, 33, 2248–2280. [CrossRef]
22. Hühne, C.; Zerbst, A.K.; Kuhlmann, G.; Steenbock, C.; Rolfes, R. Progressive damage analysis of composite bolted joints with
liquid shim layers using constant and continuous degradation models. Compos. Struct. 2010, 92, 189–200. [CrossRef]
23. Cheng, X.; Wang, S.; Zhang, J.; Huang, W.; Cheng, Y.; Zhang, J. Effect of damage on failure mode of multi-bolt composite joints
using failure envelope method. Compos. Struct. 2017, 160, 8–15. [CrossRef]
24. Naderi, M.; Khonsari, M.M. Stochastic analysis of inter- and intra-laminar damage in notched PEEK laminates. Express Polym.
Lett. 2013, 7, 383–395. [CrossRef]
25. Linde, P.; Pleitner, J.; Boer, H.; Carmone, C. Modelling and Simulation of Fibre Metal Laminates. In Proceedings of the 2004
ABAQUS Users’ Conference, Boston, MA, USA; 2004. Available online: https://www.researchgate.net/publication/266268763_
Modelling_and_Simulation_of_Fibre_Metal_Laminates (accessed on 28 December 2022).
26. Wang, Y.; Tong, M.; Zhu, S. Three Dimensional Continuum Damage Mechanics Model of Progressive Failure Analysis in Fibre-
Reinforced Composite Laminates. In Proceedings of the Structures, Structural Dynamics, and Materials Conference, Palm Springs,
CA, USA, 4–7 May 2009.
27. Liu, Y.; Zwingmann, B.; Schlaich, M. Nonlinear Progressive Damage Analysis of Notched or Bolted Fibre-Reinforced Polymer
(FRP) Laminates Based on a Three-Dimensional Strain Failure Criterion. Polymers 2014, 6, 949. [CrossRef]
28. Gutkin, R.; Pinho, S.T. Practical Application of Failure Models to Predict the Response of Composite Structures. In Proceedings of
the 18th International Conference on Composite Materials, Lisbon-Portugal, Portugal; 2012.
29. Warren, K.C.; Lopez-Anido, R.A.; Vel, S.S.; Bayraktar, H.H. Progressive failure analysis of three-dimensional woven carbon
composites in single-bolt, double-shear bearing. Compos. Part B Eng. 2016, 84, 266–276. [CrossRef]
30. Olmedo, Á.; Santiuste, C. On the prediction of bolted single-lap composite joints. Compos. Struct. 2012, 94, 2110–2117. [CrossRef]
31. Mandal, B.; Chakrabarti, A. Simulating Progressive Damage of Notched Composite Laminates with Various Lamination Schemes.
Int. J. Appl. Mech. Eng. 2017, 22, 333–347. [CrossRef]
32. Wang, G.-D.; Melly, S.K. Three-dimensional finite element modeling of drilling CFRP composites using Abaqus/CAE: A review.
Int. J. Adv. Manuf. Technol. 2018, 94, 599–614. [CrossRef]
33. Liu, H.; Liu, J.; Ding, Y.; Zhou, J.; Kong, X.; Harper, L.T.; Blackman, B.R.K.; Falzon, B.G.; Dear, J.P. Modelling damage in
fibre-reinforced thermoplastic composite laminates subjected to three-point bend loading. Compos. Struct. 2020, 236, 111889.
[CrossRef]
34. Camanho, P.P.; Arteiro, A.; Melro, A.R.; Catalanotti, G.; Vogler, M. Three-dimensional invariant-based failure criteria for
fibre-reinforced composites. Int. J. Solids Struct. 2015, 55, 92–107. [CrossRef]
Appl. Sci. 2023, 13, 1155 24 of 24
35. Zhuang, F.; Arteiro, A.; Furtado, C.; Chen, P.; Camanho, P.P. Mesoscale modelling of damage in single- and double-shear
composite bolted joints. Compos. Struct. 2019, 226, 111210. [CrossRef]
36. Barbero, E.J. Finite Element Analysis of Composite Materials using ABAQUS; CRC Press: London, UK, 2013.
37. Melro, A.R. Analytical and Numerical Modelling of Damage and Fracture of Advanced Composites. In Ph.D. Thesis; FEUP: Porto,
Portugal, 2011.
38. Lemaitre, J. A Course on Damage Mechanics; Springer: Berlin/Heidelberg, Germany, 1992.
39. Lemaitre, J.; Benallal, A.; Billardon, R.; Marquis, D. Thermodynamics and phenomenology. Contin. Art Sci. Model. Mater. Behav.
Solid Mech. Its Appl. 2002, 76, 209–223.
40. Wang, C.; Roy, A.; Silberschmidt, V.V.; Chen, Z. Modelling of Damage Evolution in Braided Composites: Recent Developments.
Mech. Adv. Mater. Mod. Processes 2017, 3, 15. [CrossRef]
41. Barbero, E.J.; Cosso, F.A.; Roman, R.; Weadon, T.L. Determination of material parameters for Abaqus progressive damage analysis
of E-glass epoxy laminates. Compos. Part B Eng. 2013, 46, 211–220. [CrossRef]
42. Barbero, E.J.; Cosso, F.A. Determination of material parameters for discrete damage mechanics analysis of carbon-epoxy laminates.
Compos. Part B Eng. 2014, 56, 638–646. [CrossRef]
43. Hashin, Z. Failure Criteria for Unidirectional Fiber Composites. J. Appl. Mech. 1980, 47, 329–334. [CrossRef]
44. Mulhern, J.F.; Rogers, T.G.; Spencer, A.J.M.; Hill, R. A continuum model for fibre-reinforced plastic materials. Proc. R. Soc. London.
Ser. A. Math. Phys. Sci. 1967, 301, 473–492. [CrossRef]
45. Rosen, B.W. Mechanics of Composite Strengthening. Fibre Composite Materials, American Society of Metals, Chapter 3. Available
online: https://www.scirp.org/(S(i43dyn45teexjx455qlt3d2q))/reference/ReferencesPapers.aspx?ReferenceID=485824(accessed
on 28 December 2022).
46. Camanho, P.P.; Davila, C.G. Mixed-Mode Decohesion Finite Elements for the Simulation of Delamination in Composite Materials.
NASA/TM 2002, 211737, 20020053651.
47. Rahimian Koloor, S.S.; Karimzadeh, A.; Yidris, N.; Petrů, M.; Ayatollahi, M.R.; Tamin, M.N. An Energy-Based Concept for Yielding
of Multidirectional FRP Composite Structures Using a Mesoscale Lamina Damage Model. Polymers 2020, 12, 157. [CrossRef]
48. Lemaitre, J.; Desmorat, R. Engineering Damage Mechanics; Springer: Berlin/Heidelberg, Germany, 2005.
49. Bazant, Z.P.; Oh, B.H. Crack band theory for fracture of concrete. Mater. Struct. 1983, 16, 155–177.
50. Oliver, J. A consistent characteristic length for smeared cracking models. Int. J. Numer. Methods Eng. 1989, 28, 461–474. [CrossRef]
51. Forero, J.A.; Bravo, M.; Pacheco, J.; de Brito, J.; Evangelista, L. Fracture Behaviour of Concrete with Reactive Magnesium Oxide as
Alternative Binder. Appl. Sci. 2021, 11, 2891. [CrossRef]
52. Borst, R.; Crisfield, M.A.; Remmers, J.J.C.; Verhoosel, C.V. Nonlinear Finite Element Analysis of Solids and Structures; Wiley: United
Kingdon London, 2012.
53. Duvaut, G.; Lions, J.L. Les Inequations en Mecanique et en Physique; Dunod: Paris, France, 1972.
54. Geers, M.G.D.; Brekelmans, W.A.M.; de Borst, R. Viscous Regularization of Strain-Localisation for Damaging Materials. Dordrecht;
Springer: Berlin/Heidelberg, Germany, 1994; pp. 127–138.
55. Dunne, F.; Petrinic, N. Introduction to Computational Plasticity; Oxford University Press: Oxford, NY, USA, 2005; Volume 1.
56. Chaboche, J.L.; Feyel, F.; Monerie, Y. Interface debonding models; a viscous regularization with a limited rate dependency. Int. J.
Solids Struct. 2001, 38, 3127–3160. [CrossRef]
57. Martins, D.; Gonilha, J.; Correia, J.R.; Silvestre, N. Exterior beam-to-column bolted connections between GFRP I-shaped pultruded
profiles using stainless steel cleats. Part 1: Experimental study. Thin-Walled Struct. 2021, 163, 107719. [CrossRef]
58. Almeida-Fernandes, L.; Silvestre, N.; Correia, J.R.; Arruda, M.R.T. Fracture toughness-based models for damage simulation of
pultruded GFRP materials. Compos. Part B Eng. 2020, 186, 107818. [CrossRef]
59. Liu, W.; Feng, P.; Huang, J. Bilinear softening model and double K fracture criterion for quasi-brittle fracture of pultruded FRP
composites. Compos. Struct. 2017, 160, 1119–1125. [CrossRef]
60. Almeida-Fernandes, L.; Silvestre, N.; Correia, J.R.; Arruda, M. Compressive transverse fracture behaviour of pultruded GFRP
materials: Experimental study and numerical calibration. Compos. Struct. 2020, 247, 112453. [CrossRef]
61. Arruda, M.R.T.; Almeida-Fernandes, L.; Castro, L.; Correia, J.R. Tsai–Wu based orthotropic damage model. Compos. Part C Open
Access 2021, 4, 100122. [CrossRef]
62. Chell, G.G.; Worthington, P.J. The determination of fracture toughness of a tough steel from invalid compact tension specimens of
varying width and thickness. Mater. Sci. Eng. 1976, 26, 95–103. [CrossRef]
63. Xiong, Z.; Zhao, C.; Meng, Y.; Li, W. A damage model based on Tsai–Wu criterion and size effect investigation of pultruded GFRP.
Mech. Adv. Mater. Struct. 2022, 1–15. [CrossRef]
64. CNR-DT-205/2007; Guide for the Design and Construction of Structures made of FRP Pultruded Elements. Advisory Committee
on Technical Recommendations for Construction: Italy Rome, 2008.
65. Martinavičius, D.; Augonis, M.; Rui Tiago Arruda, M. Experimental and Analytical Study on Local Buckling Behavior of the
Concrete-filled Thin-walled Welded Steel Columns. Period. Polytech. Civ. Eng. 2020, 64, 917–927. [CrossRef]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.

Applsci 13 01155 v2

Uploaded by

Copyright:

Available Formats

Applsci 13 01155 v2

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Applsci 13 01155 v2

Uploaded by

Copyright:

Available Formats

applied

1 CERIS-IST-ID, Universidade de Lisboa, 1049-001 Lisboa, Portugal

Appl. Sci. 2023, 13, 1155. https://doi.org/10.3390/app13021155 https://www.mdpi.com/journal/applsci

2. Recent Advances in 3D Orthotropic Damage Models

3. 3D Orthotropic Stiffness Formulation

ε i = Hij σj [ Hd ]−1 = [Cd ] σi = Cij ε j (1)

C11 C12 C13 0 0 0

4. Adopted 3D Hashin-Based Damage Model

4.1. Hashin 3D Failure Criterion

Ff (σ̂1 , τ̂12 , τ̂13 ) = 1.0

The fifth simplification pertains to the matrix’s resistance to compression, in which

Amc (σ̂2 + σ̂3 ) + Bmc (σ̂2 + σ̂3 )2 − σ̂2

4.2. Damage Evolution

A characteristic length Lc is added to the formulation to represent the constitutive law

4.3. Viscous Regularization

{σv } = [Cdv ]{ε} (39)

{σ} = [Cd ]{ε} (40)

5. Proposed Predictor-Corrector Implicit to Explicit

ε t+∆t = ε t + ∆ε t+∆t (45)

To solve this problem, a new implicit-to-explicit algorithm is proposed. The funda-

ε t+∆t = ε t + v × ∆ε t+∆t (46)

6.1. Compact Tension Test

(a) (b) (c)

(a) (b) (c)

(a) (b) (c)

121 E71[MPa] E121

0.5 Implicit Implicit

0.5 Implicit Implicit

0.4 Explicit 30 Explicit

0.3 ITE Algorithm 20 ITE Algorithm

(a) (a) (b)(b)

(a) (b) (c) (d)

E1 [MPa] E2 [MPa] E3 [MPa] G12 [MPa] G13 [MPa] G23 [MPa]

(a) (b) Principal strain [-]

0.7 0.8 0.7 0.8 ITE algorithm - point 1

ITE algorithm - point 1

0.5 0.6 ITE algorithm - point

Shear damage 1-2

0.6 ITE algorithm - point 2 0.6 ITE algorithm - point 2

Appl. Sci. 2023, 13, 1155

0.8 ITE algorithm - point 1

0.70.9 ITE algorithm - point 2

(a) (b) (c)

6.3. Double-Lap Test

E1 [MPa] E2 [MPa] E3 [MPa] G12 [MPa] G13 [MPa] G23 [MPa]

(a) (b) (c) (d)

RED_15%_15 mm RED_80%_15 mm FULL_15%_15 mm FULL_80%_15 mm

Appl. Sci. 2023, 13, 1155 22 of 25

30 and that this is red_80%_70

500 (a) (b)

You might also like