https://doi.org/10.3311/PPci.

14801
Creative Commons Attribution b |1159
Periodica Polytechnica Civil Engineering, 63(4), pp. 1159–1170, 2019

A New Nodal Stress Recovery Technique in Finite Element

Method Using Colliding Bodies Optimization Algorithm

Ali Kaveh1*, Mohammad Reza Seddighian1

1
Department of Civil Engineering, Iran University of Science and Technology, Narmak, Postal Code 1684613114, Tehran, Iran
*
Corresponding author, e-mail: [email protected]

Received: 06 August 2019, Accepted: 11 October 2019, Published online: 28 November 2019

Abstract
In Finite Element Method (FEM), the stress components are calculated within the elements firstly, and then these components are
recovered to the nodes. For the recovery process, there are several well-known methods in which the increase of their accuracy
imposes additional costs into the problem. In this paper, a new nodal stress recovery technique is proposed in which Colliding Bodies
Optimization (CBO) Algorithm fits an appropriative function for nodal stress fields. The CBO employs this function to compute the
stress components in the nodal coordinates. Therefore, a particular model to stress fields and its components will be available. It can
be considered as a connection between analytical approaches and numerical methods, providing benefits of both categories. Finally,
the accuracy, efficiency, and applicability of the new technique are investigated employing three diverse examples.
Keywords
Finite Element Method, error estimation, optimization, metaheuristics, Colliding Bodies Optimization

1 Introduction
In Finite Element Method (FEM) part of the computation
is accomplished within the elements. However, significant
entities must report as objectives in the nodes. Hence, the
data transfer from the elements to the nodes is required.
This procedure is called Recovery, that is an important
step of FEM. A node in FEM mesh is affected by several (a)

elements incident with that node, so-called as Reference

Node. These set of nodes and elements are known as a
patch, Fig. 1(a), (b).
The initial idea for the nodal stress recovery is to aver-
age the stress values within a patch and consider the results
as stress components of the reference node. This proce-
dure which is called Averaging is the simplest and the least
accurate technique to achieve the nodal stress components.
In usual cases, in which the behavior of stress fields in a (b)
patch is almost uniform, there is no obstacle to estimate Fig. 1 (a) A general FEM mesh and a patch. (b) The related patch of a
reference node (shown in red)
the stress components of a reference node in this manner.
In the mentioned conditions, the averaging technique may final results. In addition to this problem, when a numerical
propound suitable solutions. On the other hand, in problems method is employed to compute the solutions, availability
which suffer from turbulent stress fields, it is not effortless of an appropriate analytical function and/or field which can
to approximate the nodal stress components by averaging. model the behavior of a node in a specific patch is usually
This difficulty often leads to a considerable decrease in the an unattainable ambition. Accordingly, the stress recovery
accuracy of the solutions influencing the correctness of the is considered as an important part in numerical analysis.

Cite this article as: Kaveh, A., Seddighian, M. R. "A New Nodal Stress Recovery Technique in Finite Element Method Using Colliding Bodies Optimization
Algorithm", Periodica Polytechnica Civil Engineering, 63(4), pp. 1159–1170, 2019. https://doi.org/10.3311/PPci.14801
1160|Kaveh and Seddighian
Period. Polytech. Civ. Eng., 63(4), pp. 1159–1170, 2019

To validate the final results of any stress recovery tech- Anitescu et al. [13] executed a recovery based error estima-
nique, it is necessary to find out how much the solutions tion technique in high-order splines. For the low order 3D
are accurate. This procedure is usually called error esti- FEM, Sharma et al. [14] improved stress recovery. Wang
mation. Since there is no exact solution in the problems by et al. [15] employed superconvergence analysis for solving
which the relative error can be calculated, a criterion must Maxwell’s equations and Yuan et al. [16] developed a new
define to accomplish this errand. adaptive finite element method (AFEM) based on element
Generally, there are two classes of techniques to error energy projection method. This field advanced by efforts
estimation, which are divided into the Residual Based and of Patton et al. [17] on laminated composite plastic and
the Recovery Based methods. Originally, the residual based endeavors of Chi et al. [18]. Caldern et al. [19], Kim and
technique is introduced by Babuška and Rheinboldt [1] that Jang [20], Sracic and Elke [21], Boffi and Gastaldi [22],
presumes the local residuals of solutions. The employed Prato Torres et al. [23], Ganis et al. [24] and Song et al. [25]
error estimator in this method is based on the norm of energy advanced this field by essays and all of them performed
that is computed by residuals in a patch of elements. In the efforts on applications of error estimation and the AFEM.
following, this technique is continued by several research- For employing mentioned methods, the number of
ers as Ainsworth and Oden [2]. Then again, another error patch points must be equal to the unknowns, i.e., when the
estimator method suggested based on executing the recov- C1SPR in which a quadratic surface is fitted to the patch is
ery process to improve the accuracy of numerical solutions employed, the corresponding data of at least six points are
by Zienkiewicz and Zhu [3]. In this method, the approxima- required based on Khayyam1 - Pascal's triangle. It means
tion of error is defined as the difference between the recov- that at least six points must be available in the related patch.
ered and numerical solutions. They also proved that if the By increasing the necessity of more accuracy in the fitted
recovery technique will be Superconvergent, then the esti- function and/or model, the number of compulsive nodes
mator is always asymptotically exact [4, 5]. One of the most also will increase, and this informs that the problem must
important studies is carried out by Zhu and Zhang [6]. They solve by a more powerful element. This issue finally leads
proved that for each residual based estimator, there is a cor- to an impressive increment of the procedure cost. The sec-
responding recovery based process. Given that the same ond limitation of executing the prevalent methods which
thing cannot be said for the opposite situation, the recovery will be called in the following as Classic Methods is the
based methods with optimal performance does not seem to type of selected functions. In classic methods, the formu-
have an alternative residual process. Therefore, the scope of lations are based on polynomial functions, i.e., only finite
applications of recovery based methods is greater. sentences of polynomials can be selected in the solving pro-
As mentioned, in the recovery based procedure, cedure. Furthermore, since the number of points in a related
the approximation of error is defined as the difference patch of a node is finite, therefore the selection of high
between the recovered and numerical solutions. Hence, order powers in polynomial function is almost impossible.
the existence of a method to obtain recovered solutions Hence, this note informs about the third limitation of classic
is obligatory. The most straightforward approach for solu- methods. These three limitations are not all the drawbacks
tion recovery, as mentioned previously, is to average the of classic methods. When the area of an element decrease
stress values within the patch (Averaging Technique). (the phenomenon occurs in common problems and exactly
Zienkiewicz and Zhu [4, 7] presented the Superconvergent is an aim of adaptive finite element method procedure), the
Patch Recovery (SPR) technique, one of the most effec- coordinates of points in a patch will be very close together.
tive proposed methods for recovery of solutions. Since This property leads to a singularity in the formulation of the
the gradients of the finite element approximation exhibit classic methods. Therefore, for many classes of problems,
superconvergence at the Gauss integration points of iso- it is not feasible for this category of techniques to estimate
parametric elements [8–11], SPR technique will lead to an appropriate model of variables behavior.
more accurate solutions. Moslemi and Khoei [5] devel- In this paper, a new method is proposed to recover the
oped Weighted SPR by adding a parameter indicates the stress components from a patch to a reference node using
distance of a patch point from a reference node into the a metaheuristic algorithm. For this objective, the Colliding
SPR formulation. Khoei et al. [12] proposed the Modified Bodies Optimization (CBO) Algorithm is employed.
Superconvergent Patch Recovery (MSPR) technique by
further improvements on the formulations of the SPR. 1 Ghíyath al-Din Abu'l-Fath Umar ibn Ibrahim Al-Nisaburi al-Khayyam
 Kaveh and Seddighian
|1161
Period. Polytech. Civ. Eng., 63(4), pp. 1159–1170, 2019

Furthermore, an exclusive model for each stress compo- initial and the final meshes of a simple beam with a central
nent is fitted based on the particular behavior of each node. notch. As shown in Fig. 3, the stress error over the problem
In the following, the accuracy of the proposed method is domain in the most zones is more than 60 percent and in
evaluated in the first example. In the second and third the critical nodes same as restrained nodes in which there
examples, two types of FEM generally problems, i.e., is the stress intensity this error is more than 200 percent.
crack propagation and mesh adaptivity are solved in order By employing an AFEM procedure initial mesh (Fig. 2)
to be assured that the proposed method can successfully updated to an appropriate mesh (Fig. 4) in which the stress
take the place of classic methods in the Stress Recovery error over the domain in most zones is less than 5 percent,
problems and their applications. and there are high error values only in finite nodes con-
taining stress singularity in their proximity (Fig. 5).
2 Adaptive Finite Element Method The AFEM is employed widespread in various fields
An auxiliary approach for improving the accuracy of the of science including engineering, mathematics, physics,
FEM is the Adaptive Finite Element Method (AFEM) which etc. Moslemi and Tavakkoli [28] could develop a static
generally is divided into three categories H-Adaptivity, approach for error estimation in AFEM. Khoei et al. [29]
P-Adaptivity, and R-Adaptivity [26, 27]. According to this modeled cohesive crack propagation using AFEM and
approach, the initial mesh will be updated based on error advanced this issue [30] by modeling of non- planar cohe-
estimator commands, and by this trick, the total error over sive fracture crack growth via AFEM. Also, Moslemi
the problem domain will decrease. Figs. 2 to 5 indicate the and Khoei [31] employed AFEM for modeling of damage
growth and crack initiation. The applications of AFEM are
not confined to inducted instances. This subject includes
many other fields too. Bahbah et al. [32] employed AFEM
in an anisotropic framework. de-Pouplana and Oñate [33]
could combine this field with non-local damage for qua-
si-brittle materials. For compressible flows, Guaily and
Fig. 2 The simple beam with a central notch and its initial mesh
Megahed [34] executed AFEM for planar and axisymmet-
ric problems. Electromagnetics is another field in which
AFEM successfully is employed. This procedure is released
by Ledger et al. [35] by developing an HP-Adaptivity
method for scattering problems. The isogeometric analy-
sis also influenced by the adaptivity procedure. Thermo-
mechanics also is a field of science which uses adaptiv-
ity benefits. Badnava et al. [36] released an H-Adaptivity
Fig. 3 The stress error contour of Fig. 2 (based on the SPR technique) phase-field model for fractures in this concept. Soghrati
et al. [37] developed an automated model-based on the
AFEM for microstructures. Adaptive FEM is employed
for RC shells by Lackner and Mang [38]. Lackner and
Mang [39] released a paper for application of AFEM in
concrete structures. Finally, there is a review essay on
the adaptive techniques in the finite element method
Fig. 4 The updated mesh of Fig. 2 Based on AFEM
released by Chen [40].
Although, the AFEM is a helpful approach to decrease
the error of numerical analysis, however, it is an iterative
approach in which the problem must solve several times.
Obviously, this property imposes an impressive cost on
solving procedure. The other objectionable note for AFEM
is that since the AFEM executes stress recovery techniques
for achieving its final results, all the three drawbacks which
Fig. 5 The stress error contour of Fig. 4 (based on the SPR technique) mentioned for classic methods also exist for it.
1162|Kaveh and Seddighian
Period. Polytech. Civ. Eng., 63(4), pp. 1159–1170, 2019

3 Colliding Bodies Optimization algorithm Genetic Algorithm (GA), Particle Swarm Optimization
The classic methods for finding solving procedures and effi- (PSO), Ant Colony Optimization (ACO), Colliding Bodies
cient remedies in Structural Mechanics problems are finite. Optimization (CBO) and many other instances.
Also, the proposed procedures by them require complemen- The CBO Algorithm is developed by Kaveh and
tary data, as mentioned in Section 1. One of the appropri- Mahdavi [48] in 2014. The base of this optimization algo-
ate alternative methods instead of the problem solving pro- rithm is the collisions between two bodies which are gov-
cedure is Artificial Intelligence (AI). In the scope of this erned by the law of momentum and energy. If the studied
paper, Artificial Intelligence can employ for: system is isolated, then the total momentum of the sys-
A. Recovering the stresses from a patch to a specified tem will also be conserved. This property leads to equality
node. of momentum of all objects before and after the collision
B. Fitting an appropriative stress function based on the if there is no net external force acting upon the objects.
exclusive behavior of the patch node. These details can be expressed by:
The AI methods can solve broad fields of prob-
m1v1 + m2 v2 = m1v1' + m2 v2' , (1)
lems without mentioned limitations for classic methods.
One of the most popular and efficient implements of AI where v1 and v2 are the initial velocities of the first and the
methods are Metaheuristics Algorithms. Metaheuristics second bodies before the collision, v '1 and v '2 are the final
Algorithms start problem-solving from several points, and velocities of the first and the second bodies after collisions
if the employed method is commensurate to problem con- and, m1, m2 are the mass of two bodies, respectively.
ditions, it can solve problem rapidly and more accurate According to the conservation of the total kinetic
and comfortable. These benefits are alongside two main energy the Eq. (2) can be expressed.
traits of them namely the self- governing state from com-
plementary data, and they limitation less property. 1 1 1 1
m1v12 + m2 v22 = m1v '12 + m2 v '22 + Q , (2)
Nowadays, the metaheuristics are employed widely in 2 2 2 2
various fields of science and are executed for several objec- where Q is the loss of kinetic energy during the collision
tives same as optimization, modeling, etc. This Mega-Trend (Fig. 6).
also contains Structural Engineering. Kaveh et al. [41] In a one dimensional collision, the final velocities can
applied the ant colony algorithm on the optimization of the be calculated using Eqs. (3) and (4).
topology of structures. Kaveh and Talatahari [42] designed
(m1 − ξ m2 )v1 + (m2 + ξ m2 )v2
the skeletal structures using the imperialist competitive v1' = , (3)
m1 + m2
algorithm. Rahami et al. [43] combined force method and
genetic algorithm to optimize size, geometry, and topol-
(m2 − ξ m1 )v2 + (m1 + ξ m1 )v1
ogy of trusses. Kaveh and Zolghadr [44] employed the v2' = , (4)
m1 + m2
magnetic charged system search for structural optimiza-
tion. Lebaal [45] developed a robust model optimizer based where ξ is the Coefficient of Restitution (COR) of the two
on metaheuristic approaches. Powder compaction design colliding bodies defined as Eq. (5).
procedure is one of the most important and famous prob-
lems in computational mechanics. In this filed, also the
metaheuristics are applied, and this trick leads to relevant
results. New efforts and advances in metaheuristics and
their applications can be found in Kaveh [46, 47].
Generally, nature attempts to perform optimally. It
achieves this by a trial and error procedure and spending
time as the cost. Since the physical rules describe nature
treatment, therefore, the laws of physics can be consid-
ered as an optimizer. The employing of their approach
can be useful for solving complicated problems. This fact Fig. 6 The collision between two bodies. (A) Before the Collision and
is the base of many metaheuristics algorithms same as (B) after the collision
 Kaveh and Seddighian
|1163
Period. Polytech. Civ. Eng., 63(4), pp. 1159–1170, 2019

v2' − v1'
ξ= (5)
v2 − v1

In Colliding Bodies Optimization (CBO) Algorithm,

each solution contains some variables which are consid-
ered as a Colliding Body (CB). There are two categories
of objects which are called Massed Objects. Mass objects Fig. 7 The terms of Khayyam- Pascal's triangle
are divided into Stationary Objects and Moving Objects.
As a result of objects motions, the collisions occur between If the function is defined in two- dimensional coordi-
pairs of objects. These collisions lead to (i) improve the nates, the Eq. (7) can be approximated for it and its terms
positions of moving objects and (ii) to push stationary achieve from Khayyam-Pascal's triangle2 (Fig. 7).
objects toward more appropriate positions. After the colli-
z = f ( x, y ) = a00 + a10 x + a01 y + a20 x 2 + a11 xy + .. (7)
sion, the positions, the velocities and the masses of bodies
will be updated based on new conditions. This procedure It is possible to define each stress components of nodes
will repeat until the appropriate solution will be achieved. as a function of node coordinates. This function is similar
The CBO algorithm is employed for widespread structural to what defined as Eq. (7). The most important note is that
engineering problems. Kaveh and Ardalani [49] optimized there is no limitation to select the number of terms or max-
the cost and CO2 emission of reinforced concrete frames imum power. The authors believe that hitherto this benefit
using CBO. Optimal nonlinear analysis and design of dou- there is not in any other methods which presented for this
ble-layer grids are advanced by Kaveh and Moradveisi [50] objective. The last barrier in finding suitable function is to
also using CBO. Optimal large-scale space steel frames [51], calculate coefficients. The CBO duty is precisely this pro-
structural optimization of jacket supporting structures for cedure. CBO searches and finds the best coefficients based
offshore wind turbines [52] and continuous and discrete on each node behavior. Since the patch coordinates and
optimization for design [53] also are problems which are their available stress values must satisfy fitted function,
solved using CBO. There are MATLAB codes, outlines, therefore, this information, i.e., Gauss points coordinates,
and examples of CBO in [54] for enthusiasts. and their calculated stress value from FEM can be used to
balance coefficients in order to decrease the error of the
4 Smart nodal stress recovery fitted function as much as possible. The CBO is employed
The main objective of this study, as described in Section 1, based on following Cost Function.
is to present an accurate, more comfortable, and more n
importantly, an efficient method to recover stress compo- E = ∑ ( f ( xi , yi ) − σ i ) , (8)
nents from a patch into a reference node. The Proposing i =1

method must be limitless, which can apply for any prob- where n is the number of points in the considered patch,
lem in front of mentioned properties. For this objective, σ is the calculated stress by FEM, f(x, y) is the result of the
the modeling of an appropriate function for each node pro- fitted function or recovered stress and E is the total error
prietary behavior based on CBO algorithm is employed. between fitted function and calculated stresses over the
If the fitted function can be modeled in addition to recover patch or Cost Function.
the nodal stress components, an appropriate primarily
function for each node is calculated as well and this is a
2 The pattern of numbers which is known as Pascal's triangle, has been
connection between analytical and numerical approaches,
discovered before his time. The main formula of Pascal's triangle is due
i.e., the researcher by using this procedure can utilize both to an Indian mathematician Acharya Pingala. At around 953–1029, the
analytical and numerical methods benefits. Persian mathematician Al-Karaji published a book containing the first
According to the Taylor series, it is possible to approx- description of Pascal's triangle. It was later repeated by another Persian
imate the behavior of a function by a series of polynomi- poet-astronomer-mathematician Umar Khayyam. Also, this triangle has
been utilized in other fields at around the same time. Jia Xia (1010–1070)
als as presented in Eq. (6) for a one dimensional function.
and Yang Hui (1238–1298) in China, Petrus Apianus (1495–1552) and
f ( x) = a0 + a1 x + a2 x 2 + a3 x 3 + ... (6) Michael Stifel in Germany and some other mathematician are among the
scientists who discover this triangle and developed its properties almost
independently.
1164|Kaveh and Seddighian
Period. Polytech. Civ. Eng., 63(4), pp. 1159–1170, 2019

The coefficients are managed by decreasing cost func- 1 3 5

PS a 2 a 2 a 2
tion due to CBO procedures. The computations will be fin- KI = [ 2. 9   − 4. 6   + 21.8  
W  W  W 
3

ished if the cost function is equal to zero or very close to it. BW 2

(9)
After the described procedure, an appropriate function
7 9
a 2 a 2
−37.6   + 38.7   ],
for each stress components (for example, in-plane prob- W  W 
lems σ x, σ y and τ xy) is provided in which the proprietary
behavior of each node is modeled. Therefore, by substi- where P is a concentrated load, S, W, B are length, height,
tuting the coordinates of the reference node the recovered and thickness of the beam respectively. Here, a is the
values of its stress components can be calculated. So, both length of the notch or initial crack. Details of Eq. (9) can
objective of this study is achieved. The efficiency of the be found in Fracture Mechanics References.
proposed method will be more tangible by studying three The SIF of the problem can be calculated analytically
varied examples presented in Section 5. by substituting the described parameters into Eq. (9). For
calculating the SIF by a numerical approach the following
5 Numerical examples procedure can be utilized.
The intelligent nodal stress recovery technique is intro- For static loading, the strain energy can be calculated by
duced in previous sections. In order to evaluate the pro-
1
posed algorithm, two vivificates must be done. The first is U= PV , (10)
2
the survey of accuracy and the second is the applicability.
For these objectives, three examples are designed which where P, V and U represent the concentrated load, relative
the first focuses on first aim and two others investigate the vertical displacement, and the strain energy, respectively.
second purpose. If a crack propagates with a length of δa and the difference
between strain energy in initial and final states is equal to
5.1 Calculation of stress intensity factor ∆U, then the Strain Energy Release Rate can be obtained as
In order to survey the accuracy of the proposed algorithm,
∆U
it is logical to compare its results with the outcomes of G= . (11)
δa
classic methods. As mentioned in Section 1 the classic
methods are not exact and generally are based on approx- Finally, the SIF is calculated by
imation. Therefore, although the results comparing two
categories is a common approach, however, there is a bet- K I = EG , (12)
ter way for examination. If a problem that can be solved where E is the elasticity modulus.
either by an analytical method or a numerical approach is In the described procedure, all the variables are avail-
selected, then comparing results will be more and more able except the vertical displacement. This unknown can
logical. Also, this approach can verify the accuracy of the be calculated by FEM. Now SIF can be obtained by a
proposed method more apparent and appropriate. numerical approach and then compared with the SIF cal-
One of the essential parameters in Fracture Mechanics culated from the analytical method.
is the Stress Intensity Factor (SIF) which is represented In order to find out the accuracy of the proposed algo-
in opening mode problems (mode I) by K I. In some prob- rithm in a simple beam with initial notch having a length
lems, this parameter can be calculated either by analytical of 4 mm, the crack is propagated until a length 15 mm in
or by numerical methods. For a simple beam with a central 10 steps. The details of each step are shown in Table 1.
notch, as shown in Fig. 8, K I is calculated using Eq. (9). In each step, in the beginning, the related SIF is calcu-
lated by the analytical approach. Once the mesh is solved
by FEM, the nodal stresses are calculated by the proposed
algorithm. After that the mesh is updated based on the
recovered stresses. After that, the problem is solved again
based on the new mesh. The vertical displacement of con-
centrated load point is substituted in numerical formula-
tion, and the corresponding SIF is computed. The final
Fig. 8 Simple beam with the central notch results of each step are compared and illustrated in Fig. 9.
 Kaveh and Seddighian
|1165
Period. Polytech. Civ. Eng., 63(4), pp. 1159–1170, 2019

Table 1 The details of each step of Example 1 As illustrated in Fig. 12, CBO fitted a function based
No.
a
a/W
P B W S E on patch points (shown in blue) and in the following com-
(mm) (kg.f) (cm) (cm) (cm) (kg.f/cm)
puted the stress value of reference node, i.e. left restrained
1 4 0.07 40000 5 60 240 200000 node based on its coordinates (Red point in Fig. 12). Since
2 5.22 0.09 40000 5 60 240 200000 there is stress intensity in the support regions, the fitted
3 6.44 0.11 40000 5 60 240 200000 function is got an infinite value in this region which ver-
4 7.67 0.13 40000 5 60 240 200000 ify the correction of modeling. This fitted function is in
5 8.89 0.15 40000 5 60 240 200000 power 5 or utilized 21 terms of Khayyam-Pascal's trian-
6 10.11 0.17 40000 5 60 240 200000 gle. As described in Section 3 there are no limits to select
7 11.33 0.19 40000 5 60 240 200000 terms of Khayyam-Pascal’s triangle and number of poly-
8 12.56 0.21 40000 5 60 240 200000 nomials. The found coefficients of fitted function are pre-
9 13.78 0.23 40000 5 60 240 200000 sented in Table 2.
10 15 0.25 40000 5 60 240 200000 After employing the proposed method, an appropriate
model for each node and each stress components are avail-
able, and this is the main achievement of this study. The
second benefit of the proposed method is the accuracy of it
in comparison with other classic methods. The procedure
of decreasing the cost function in order to find the best
solution by CBO is shown in Fig. 13.

Fig. 9 The comparison between the results of the analytical and

proposed algorithm

Fig. 10 The utilized mesh in FEM computation related to step 10

Fig. 12 The fitted function for the left restrained node of step 10

Table 2 The found coefficients by CBO for the fitted function of left
restrained node in step 10.
Coeffi- Coeffi- Coeffi-
Value Value Value
Fig. 11 The horizontal stress contour for step 10 cient cient cient
a00 6.624e + 05 a21 -1.303e+12 a04 0
The updated mesh for FEM computation and the hori-
a10 -3.738e+08 a12 -1.465e+11 a50 -1.187e+14
zontal stress σX contour of step 10 are illustrated in Figs. 10
a01 -3.099e+07 a03 -1.971e+09 a41 -5.002e+14
and 11.
a20 4.043e+10 a40 2.803e+13 a32 -1.487e+14
The duty of CBO, as mentioned in Section 4 is to fit an
a11 1.42e+10 a31 4.354e+13 a23 -2.306e+13
appropriate function for each node and each stress com-
a02 4.26e+08 a22 9.394e+12 a14 0
ponents of a node. This proper function for left restrained
a03 -1.695e+12 a13 5.311e+11 a05 0
node step 10 illustrated in Fig. 12.
1166|Kaveh and Seddighian
Period. Polytech. Civ. Eng., 63(4), pp. 1159–1170, 2019

As illustrated in Fig. 13, the initial population error is

equal to 600 percent, and CBO could reduce this error
or cost function to 0 by spending 1000 NFE. This com-
putational power confirms that the proposed algorithm,
alongside its mentioned benefits, is quite efficient and
economical.

5.2 Crack propagation problem

The accuracy, efficiency, and outcomes of the proposed
algorithm are studied in the first example. Another prop-
erty of the proposed algorithm is its applicability to a wide
range of numerical problems. In this example, in order to
examine this property, a simple beam with non-central
notch due to a concentrated load, and the crack route of
propagation is investigated by the proposed algorithm,
Fig. 14. Since the material is supposed to be perfectly brit-
tle thus the crack propagation criterion is considered to be
perpendicular to maximum principal stress. Therefore, the
main variables to find the route of the crack propagation
are crack tip stress components which are computed by
the proposed algorithm. The logical and the correct found
route indicate the credibility of the proposed algorithm.
The crack must propagate to the loaded node. This pro-
cedure is shown in Fig. 15.
As illustrated in Fig. 15, the route of crack propagation
is following theories of fracture mechanics. This result
informs that the recovered stress is credible.

Fig. 13 The CBO process for finding the coefficients

Fig. 14 The conditions of Example 2 Fig. 15 The steps of the crack propagation
 Kaveh and Seddighian
|1167
Period. Polytech. Civ. Eng., 63(4), pp. 1159–1170, 2019

5.3 Error estimation and remeshing mesh data based on errors, Fig. 17. This procedure is due
In order to verify the credibility of the proposed algorithm, to stress recovery, and the proposed algorithm is employed
another example is designed in which the mesh of a simple to this duty for a simple beam with three initial notches.
beam with three notches is updated by the proposed algo- The final mesh which is shown in Fig. 18 is an appro-
rithm, Fig. 16. For adaptivity procedure one should ini- priate mesh for FEM analysis with tiny error percent,
tially calculate the error of analysis. Fig. 19. This example also indicates that the intelligent
If assume σ as the exact value of stress and σ̂ as the nodal stress recovery technique which is proposed in
value of stress calculated by finite element analysis, then it this study is an efficient, economical, accurate and practica-
is possible to define the stress error as Eq. (13). ble method for improving stresses from patches to nodes and
eσ = σ − σ̂ has this potential to be employed in a wide range of finite
(13)
element problems instead of using the classic methods.
Since the exact solution for stress is not available,
replacing improved stress σ* instead of it, is required. 6 Conclusions
In this paper, a new method is presented to recover
eσ  σ * − σ̂ (14)
the stress components from a patch to its corresponding
Due to some difficulties in practice, as the singularity reference node. The proposed method utilizes a meta-
of the stress field in the crack tip (i.e., the infinite value of heuristic algorithm to fit an appropriative nodal func-
error) the utilization of the norm of error is more appropri- tion which can calculate the nodal stress components
ate. If the L2 norm of the stress is defined as Eq. (15), then for any problems. To evaluate the accuracy of the pro-
the Eq. (14) will convert to Eq. (16). posed method, a fracture mechanics benchmark problem
as the first example is solved. The stress intensity factor
e = ( ∫ eT ed Ω)
1
2 (15) is calculated using both the analytical and the proposed
Ω

1
methods. Then, the final results are compared together,
eσ = σ * − σ̂ = ( ∫ (σ * − σ̂ )T (σ * − σ̂ )d Ω) 2 (16) which the relative error in each step was less than five
Ω
percent. Subsequently, in the second and third examples,
Since generally analysis steps are computed in elements, the efficiency and applicability of the proposed method
the whole domain Ω reduces to each element domain, and
the square of L2 the norm of the overall domain will be
obtained by summing element contributions as formulat-
ing in Eq. (17).
m 2

eσ = ∑ eσ , (17)
2

i =1 i Fig. 17 The initial mesh of Example 3

where i represent element contribution and m indicates the

total number of elements. In this manner, the new mesh
size is determined in such a way that produces uniform
error distribution.
According to the described procedure, an initial mesh
can be updated by evaluating errors and improving the Fig. 18 The final mesh of Example 3 accomplished by the proposed
algorithm

Fig. 16 The state of Example 3 Fig. 19 The stress error contours

1168|Kaveh and Seddighian
Period. Polytech. Civ. Eng., 63(4), pp. 1159–1170, 2019

are investigated. It is shown that the proposed method can crack propagation and remeshing. In both examples, the
take the place of the well-known methods same as aver- presented method led to acceptable results in an econom-
aging, SPR, etc., in the widespread FEM problems such as ical computational cost.

References
[1] Babuška, I., Rheinboldt, W. C. "A posteriori error estimates for [13] Anitescu, C., Hossain, M. N., Rabczuk, T. "Recovery-based error
the finite element method", International Journal for Numerical estimation and adaptivity using high-order splines over hierar-
Methods in Engineering, 12(10), pp. 1597–1615, 1978. chical T-meshes", Computer Methods in Applied Mechanics and
https://doi.org/10.1002/nme.1620121010 Engineering, 328, pp. 638–662, 2018.
[2] Ainsworth, M., Oden, J. T. "A unified approach to a posteriori https://doi.org/10.1016/j.cma.2017.08.032
error estimation using element residual methods", Numerische [14] Sharma, R., Zhang, J., Langelaar, M., van Keulen, F., Aragón,
Mathematik, 65(1), pp. 23–50, 1993. A. M. "An improved stress recovery technique for low-order 3D
https://doi.org/10.1007/BF01385738 finite elements", International Journal for Numerical Methods in
[3] Zienkiewicz, O. C., Zhu, J. Z. "A simple error estimator and adap- Engineering, 114(1), pp. 88–103, 2018.
tive procedure for practical engineering analysis", International https://doi.org/10.1002/nme.5734
Journal for Numerical Methods in Engineering, 24(2), pp. 337–357, [15] Wang, L., Zhang, Q., Zhang, Z. "Superconvergence Analysis and
1987. PPR Recovery of Arbitrary Order Edge Elements for Maxwell's
https://doi.org/10.1002/nme.1620240206 Equations", Journal of Scientific Computing, 78(2), pp. 1207–
[4] Zienkiewicz, O. C., Zhu, J. Z. "The superconvergent patch recovery 1230, 2019.
and a posteriori error estimates. Part 2: Error estimates and adaptiv- https://doi.org/10.1007/s10915-018-0805-8
ity", International Journal for Numerical Methods in Engineering, [16] Yuan, S., Dong, Y., Xing, Q., Fang, N. "Adaptive Finite Element
33(7), pp. 1365–1382, 1992. Method of Lines with Local Mesh Refinement in Maximum Norm
https://doi.org/10.1002/nme.1620330703 Based on Element Energy Projection Method", International Journal
[5] Moslemi, H., Khoei, A. R. "3D adaptive finite element modeling of of Computational Methods, pp. 1950–1958, 2019.
non-planar curved crack growth using the weighted superconvergent https://doi.org/10.1142/S0219876219500087
patch recovery method", Engineering Fracture Mechanics, 76(11), [17] Patton, A., Dufour, J. E., Antolin, P., Reali, A. "Fast and accurate
pp. 1703–1728, 2009. elastic analysis of laminated composite plates via isogeometric
https://doi.org/10.1016/j.engfracmech.2009.03.013 collocation and an equilibrium-based stress recovery approach",
[6] Zhu, J. Z., Zhang, Z. "The relationship of some a posteriori error Composite Structures, 225, Article number: 111026, 2019.
estimators", Studies in Applied Mechanics, 47, pp. 25–42, 1998. https://doi.org/10.1016/j.compstruct.2019.111026
https://doi.org/10.1016/S0922-5382(98)80003-2 [18] Chi, H., da Veiga, L. B., Paulino, G. H. "A simple and effective
[7] Zienkiewicz, O. C., Zhu, J. Z. "The superconvergent patch recovery gradient recovery scheme and a posteriori error estimator for the
and a posteriori error estimates. Part 1: The recovery technique", Virtual Element Method (VEM)", Computer Methods in Applied
International Journal for Numerical Methods in Engineering, 33(7), Mechanics and Engineering, 347, pp. 21–58, 2019.
pp. 1331–1364, 1992. https://doi.org/10.1016/j.cma.2018.08.014
https://doi.org/10.1002/nme.1620330702 [19] Calderón, G., Carrillo, J. C., Villamizar, J., Torres, C., Castillo, J.
[8] Chen, C. M. "Superconvergence of finite element solutions and "Adaptivity in mimetic difference schemes: one-dimensional tran-
their derivatives", Numerical Mathematics, a Journal of Chinese sient problems", Journal of Physics: Conference Series, 1159,
Universities, 3(2), pp. 118–125, 1981. Article number: 012019, 2019.
[9] Zienkiewicz, O. C., Taylor, R. L. "The finite element method, Vol. 2", https://doi.org/10.1088/1742-6596/1159/1/012019
Dynamics and non-linearity, McGraw-Hill, New York, NY, USA, [20] Kim, Y., Jang, U. G. "Element Search Methods for Large
1991. Deformation Finite Element Simulation in Geotechnical Fields",
[10] Zlámal, M. "Superconvergence and Reduced Integration in the KSCE Journal of Civil Engineering, 23(2), pp. 534–540, 2019.
Finite Element Method", Mathematics of Computation, 32(143), pp. https://doi.org/10.1007/s12205-018-1368-9
663–685, 1978. [21] Sracic, M. W., Elke, W. J. "Effect of Boundary Conditions on Finite
https://doi.org/10.2307/2006479 Element Submodeling", In: Kerschen, G. (ed) Nonlinear Dynamics,
[11] Mackinnon, R. J., Carey, G. F. "Superconvergent derivatives: A Volume 1, Conference Proceedings of the Society for Experimental
Taylor series analysis", International Journal for Numerical Methods Mechanics Series, Springer, Cham, Switzerland, 2019, pp. 163–170.
in Engineering, 28(3), pp. 489–509, 1989. https://doi.org/10.1007/978-3-319-74280-9_16
https://doi.org/10.1002/nme.1620280302 [22] Boffi, D., Gastaldi, L. "Adaptive Finite Element Method for the
[12] Khoei, A. R., Azadi, H., Moslemi, H. "Modeling of crack propaga- Maxwell Eigenvalue Problem", SIAM Journal on Numerical
tion via an automatic adaptive mesh refinement based on modified Analysis, 57(1), pp. 478–494, 2019.
superconvergent patch recovery technique", Engineering Fracture https://doi.org/10.1137/18M1179389
Mechanics, 75(10), pp. 2921–2945, 2008.
https://doi.org/10.1016/j.engfracmech.2008.01.006
 Kaveh and Seddighian
Period. Polytech. Civ. Eng., 63(4), pp. 1159–1170, 2019
|1169

[23] Prato Torres, R., Domínguez, C., Díaz, S. "An adaptive finite element [36] Badnava, H., Msekh, M. A., Etemadi, E., Rabczuk, T. "An h-adap-
method for a time-dependent Stokes problem", Numerical Methods tive thermo-mechanical phase field model for fracture", Finite
for Partial Differential Equations, 35(1), pp. 325–348, 2019. Elements in Analysis and Design, 138, pp. 31–47, 2018.
https://doi.org/10.1002/num.22302 https://doi.org/10.1016/j.finel.2017.09.003
[24] Ganis, B., Pencheva, G., Wheeler, M. F. "Adaptive mesh refine- [37] Soghrati, S., Nagarajan, A., Liang, B. "Conforming to interface
ment with an enhanced velocity mixed finite element method on structured adaptive mesh refinement: New technique for the auto-
semi-structured grids using a fully coupled solver", Computational mated modeling of materials with complex microstructures", Finite
Geosciences, 23(1), pp. 149–168, 2019. Elements in Analysis and Design, 125, pp. 24–40, 2017.
https://doi.org/10.1007/s10596-018-9789-6 https://doi.org/10.1016/j.finel.2016.11.003
[25] Song, J. H., Maier, M., Luskin, M. "Adaptive finite element simu- [38] Lackner, R., Mang, H. A. "Adaptive FE analysis of RC Shells. I:
lations of waveguide configurations involving parallel 2D material Theory", Journal of Engineering Mechanics, 127(12), pp. 1203–
sheets", Computer Methods in Applied Mechanics and Engineering, 1212, 2001.
351, pp. 20–34, 2019. https://doi.org/10.1061/(ASCE)0733-9399(2001)127:12(1203)
https://doi.org/10.1016/j.cma.2019.03.039 [39] Lackner, R., Mang, H. A. "Adaptivity in computational mechanics
[26] Logan, D. L. "A First Course in the Finite Element Method", of concrete structures", International Journal for Numerical and
Cengage Learning, Boston, MA, USA, 2016. Analytical Methods in Geomechanics, 25(7), pp. 711–739, 2001.
[27] Akin, J. E. "Finite element analysis with error estimators: An intro- https://doi.org/10.1002/nag.152
duction to the FEM and adaptive error analysis for engineering stu- [40] Chen, S. H. "Computational Geomechanics and Hydraulic
dents", Elsevier, Oxford, UK, 2005. Structures", Springer Tracts in Civil Engineering, Singapore, 2019.
https://doi.org/10.1016/B978-075066722-7/50032-1 https://doi.org/10.1007/978-981-10-8135-4
[28] Moslemi, H., Tavakkoli, A. "A Statistical Approach for Error [41] Kaveh, A., Hassani, B., Shojaee, S., Tavakkoli, S. M. "Structural
Estimation in Adaptive Finite Element Method", International topology optimization using ant colony methodology", Engineering
Journal for Computational Methods in Engineering Science and Structures, 30(9), pp. 2559–2565, 2008.
Mechanics, 19(6), pp. 440–450, 2019. https://doi.org/10.1016/j.engstruct.2008.02.012
https://doi.org/10.1080/15502287.2018.1558424 [42] Kaveh, A., Talatahari, S. "Optimum design of skeletal structures
[29] Khoei, A. R., Moslemi, H., Ardakany, K. M., Barani, O. R., Azadi, H. using imperialist competitive algorithm", Computers & Structures,
"Modeling of cohesive crack growth using an adaptive mesh refine- 88(21–22), pp. 1220–1229, 2010.
ment via the modified-SPR technique", International Journal of https://doi.org/10.1016/j.compstruc.2010.06.011
Fracture, 159(1), pp. 21–41, 2009. [43] Rahami, H., Kaveh, A., Gholipour, Y. "Sizing, geometry and topol-
https://doi.org/10.1007/s10704-009-9380-1 ogy optimization of trusses via force method and genetic algorithm",
[30] Khoei, A. R., Moslemi, H., Sharifi, M. "Three-dimensional cohesive Engineering Structures, 30(9), pp. 2360–2369, 2008.
fracture modeling of non-planar crack growth using adaptive FE https://doi.org/10.1016/j.engstruct.2008.01.012
technique", International Journal of Solids and Structures, 49(17), [44] Kaveh, A., Zolghadr, A. "Magnetic charged system search for
pp. 2334–2348, 2012. structural optimization", Periodica Polytechnica Civil Engineering,
https://doi.org/10.1016/j.ijsolstr.2012.04.036 58(3), pp. 203–216, 2014.
[31] Moslemi, H., Khoei, A. R. "3D Modeling of Damage Growth and https://doi.org/10.3311/PPci.7460
Crack Initiation Using Adaptive Finite Element Technique", Scientia [45] Lebaal, N. "Robust low cost meta-modeling optimization algo-
Iranica Transaction A: Civil Engineering, 17(5), pp. 372–386, 2010. rithm based on meta-heuristic and knowledge databases approach:
[32] Bahbah, C., Mesri, Y., Hachem, E. "Interpolation with restrictions in Application to polymer extrusion die design", Finite Elements in
an anisotropic adaptive finite element framework", Finite Elements Analysis and Design, 162, pp. 51–66, 2019.
in Analysis and Design, 142, pp. 30–41, 2018. https://doi.org/10.1016/j.finel.2019.05.004
https://doi.org/10.1016/j.finel.2017.11.011 [46] Kaveh, A. "Advances in Metaheuristic Algorithms for Optimal
[33] de-Pouplana, I., Oñate, E. "Combination of a non-local damage Design of Structures", Springer, Cham, Switzerland, 2017.
model for quasi-brittle materials with a mesh-adaptive finite ele- https://doi.org/10.1007/978-3-319-46173-1
ment technique", Finite Elements in Analysis and Design, 112, pp. [47] Kaveh, A. "Applications of Metaheuristic Optimization Algorithms
26–39, 2016. in Civil Engineering", Springer, Cham, Switzerland, 2017.
https://doi.org/10.1016/j.finel.2015.12.011 https://doi.org/10.1007/978-3-319-48012-1
[34] Guaily, A., Megahed, A. "An adaptive finite element method for [48] Kaveh, A., Mahdavi, V. R. "Colliding bodies optimization: a novel
planar and axisymmetric compressible flows", Finite Elements in meta-heuristic method", Computers & Structures, 139, pp. 18–27,
Analysis and Design, 46(8), pp. 613–624, 2010. 2014.
https://doi.org/10.1016/j.finel.2010.03.001 https://doi.org/10.1016/j.compstruc.2014.04.005
[35] Ledger, P. D., Morgan, K., Peraire, J., Hassan, O., Weatherill, N. P. [49] Kaveh, A., Ardalani, S. "Cost and CO2 Emission Optimization of
"The development of an hp-adaptive finite element procedure for Reinforced Concrete Frames Using Enhanced Colliding Bodies
electromagnetic scattering problems", Finite Elements in Analysis Algorithm", Asian Journal of Civil Engineering, 17(6), pp. 831–
and Design, 39(8), pp. 751–764, 2003. 858, 2016.
https://doi.org/10.1016/S0168-874X(03)00057-X
1170|Kaveh and Seddighian
Period. Polytech. Civ. Eng., 63(4), pp. 1159–1170, 2019

[50] Kaveh, A., Moradveisi, M. "Nonlinear analysis based optimal [53] Kaveh, A., Ilchi Ghazaan, M. "Enhanced colliding bodies optimi-
design of double-layer grids using enhanced colliding bodies opti- zation for design problems with continuous and discrete variables",
mization method", Structural Engineering and Mechanics, 58(3), pp. Advances in Engineering Software, 77, pp. 66–75, 2014.
555–578, 2016. https://doi.org/10.1016/j.advengsoft.2014.08.003
https://doi.org/10.12989/sem.2016.58.3.555 [54] Kaveh, A., Bakhshpoori, T. "Metaheuristics: Outlines, MATLAB
[51] Kaveh, A., Bolandgerami, A. "Optimal design of large-scale Codes and Examples", Springer, Cham, Switzerland, 2019.
space steel frames using cascade enhanced colliding body opti- https://doi.org/10.1007/978-3-030-04067-3
mization", Structural and Multidisciplinary Optimization, 55(1),
pp. 237–256, 2017.
https://doi.org/10.1007/s00158-016-1494-2
[52] Kaveh, A., Sabeti, S. "Structural optimization of jacket supporting
structures for offshore wind turbines using colliding bodies opti-
mization algorithm", The Structural Design of Tall and Special
Buildings, 27(13), Article number: e1494, 2018.
https://doi.org/10.1002/tal.1494