See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/292158367

Load Paths Visualization in Plane Elasticity Using Load Path Function Method

Conference Paper · January 2016

DOI: 10.2514/6.2016-0233

CITATIONS READS
3 563

2 authors:

Kaveh Gharibi Ali Tamijani

Embry-Riddle Aeronautical University Embry-Riddle Aeronautical University
7 PUBLICATIONS   15 CITATIONS    28 PUBLICATIONS   130 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Load path in plates and shells View project

All content following this page was uploaded by Kaveh Gharibi on 29 January 2016.

The user has requested enhancement of the downloaded file.

 AIAA 2016-0233
AIAA SciTech
4-8 January 2016, San Diego, California, USA
57th AIAA/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference

Load Paths Visualization in Plane Elasticity Using Load

Path Function Method

Ali Y.Tamijani1, Kaveh Gharibi2

The paper proposes a load path function method for load path visualization and load flow
calculation in plane elasticity problems. The load path functions are directly derived from 2D
equilibrium equations in the absence of body forces. The stresses are written in terms of the load path
functions and the load flow is calculated using the load path function contours. A set of Poisson's
equations in terms of load path functions and stress components is developed and an efficient
numerical procedure to solve them is discussed. Numerical results show the proposed load path
Downloaded by EMBRY-RIDDLE AERO UNIV. on January 29, 2016 | http://arc.aiaa.org | DOI: 10.2514/6.2016-0233

function method is able to define the load paths and obtain the load flow using an Airy stress function
for problems with available closed form solution, and also can be easily integrated into the numerical
approaches to calculated load paths for complex problems.

Nomenclature
𝛹! = Load path function in the x direction
𝛹! = Load path function in the y direction
𝜎!! = Normal stress in the x direction
𝜎!! = Normal stress in the y direction
𝜎!" = Shear stress in x-y plane
𝛷 = Weight function
∆𝐹!   = Internal load in x-direction
∆𝐹!            = Internal load in y-direction

I. Introduction
In structural mechanics, the knowledge of the load paths may provide critical insights into the performance,
functionality, and efficiency of the structure. The load paths in a structure help the designer understand how the
loads flow throughout the structure and ensure that the loads are transmitted and distributed according to the desired
structural responses. In fact, the use of loads in design optimization can be traced back to the Michell structure,
which was designed based on the total tension and total compression loads [1]. After the emergence of numerical
methods, mainly the finite element method (FEM), the design process has moved away from the calculation of
internal forces and focused to a greater extent on the approximation of stress. In most of the current design methods,
the local stresses are aggregated into a single relationship and used as a constraint or objective function. These types
of approximations cause the loss of physics in the design process, which sometimes results in difficulty in
determining the optimal design. Indeed, it has been proven that approximations based on internal forces, which are
more invariant than stresses and strains, in the design optimization process are more accurate than stress- or strain-
based approximations [2, 3]. Despite the importance of load paths in structural mechanics, the computational
structural methods (e.g., FEM, the Ritz method, and the meshless method), and as a result the commercial software
(e.g., ANSYS, MSC Nastran, and ABAQUS), do not provide information regarding the load transformation

1
Assistant professor, Department of Aerospace Engineering, Embry Riddle Aeronautical University, Daytona
Beach, Florida, 32114, USA
2
Graduate research assistant, Department of Aerospace Engineering, Embry Riddle Aeronautical University,
Daytona Beach, Florida, 32114, USA
1

Copyright © 2015 by Copyright © 2015 by Kaveh Gharibi. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.. Published by the American Institute of Aeronautics and Astron
 throughout the structure. In the following the state-of-the-art approaches that have been proposed to identify and
visualize load paths are discussed.
Principal stress trajectories were the first method used to determine the load paths in structures. By definition, the
principal stress is located at an angle at which the shear stress is zero. In this method, load paths are assumed to be
tangent to the vector field of the compressive and tensile principal stresses. Although the tension and compression
forces are constant on the principal stresses trajectories, the paths along them does not necessarily present the load
transfer from applied load to reaction points.
Another method for identifying the load paths is to use the direction and the trajectory of the stress pointing vectors
[4-7]. In this approach, the load paths are defined as curves that surround ‘load tubes’ that carry a constant load. The
continuity in the load path is ensured by using the equilibrium of forces acting on the load tube. It has been shown
that the stress applied on the walls of the force tubes does not contribute to the equilibrium, meaning that the sum of
all forces acting on the walls is equal to zero. The Runge-Kutta algorithm is used in this method to plot the contours
through the vector field.
The concept of transferred forces and potential transferred forces also is introduced to investigate load paths in
Downloaded by EMBRY-RIDDLE AERO UNIV. on January 29, 2016 | http://arc.aiaa.org | DOI: 10.2514/6.2016-0233

structures [8]. To find the transferred force in an element using this method, first the displacements of the whole
structure and the load transferred to the support are calculated. The next step is to remove the element and apply the
displacements that were calculated in the first step and then calculate the load that is transferred to the support. The
difference between the transferred loads in these two steps is the load that is transferred through the element. Since
the number of analyses is equal to the number of elements in the structure in which the transferred load is of interest,
this method requires many simulations and extensive computation time to calculate the load transfer in the entire
structure. A definition for load path is also proposed based on a compliance energy that is calculated using the
transferred forces [9-11]. In this method the compliance energy is obtained according to the ratio of the reactive
force with an arbitrary point in the structure unfixed and the reactive force with the point fixed when force is applied
to the structure. Therefore, the defined compliance energy quantifies the connective strength between the loading
point and an arbitrary point.
In recent years, physical descriptions of load paths have been defined and used to synthesize optimal compliant
mechanisms [12, 13]. The load paths in compliant mechanisms are defined using geometric descriptions for the
connectivity of the point of application and point of support. Optimization of the compliant mechanism uses binary
variables to indicate the presence or absence of a load path. This method may have a limited range of applications
due to the difficulties in presenting the load paths by connectivity for continuum structures [14].
In this paper, definition and formulations for load path functions are introduced and the theoretical and
computational foundation of an efficient and robust load path algorithm based on the load path functions are
established. The proposed load path function method is able to define the load paths and obtain the load flow using
an Airy stress function for problems with available closed form solution, and also can be easily integrated into the
numerical approaches to provide load path contours and load flow for complex problems. Since the load paths are
represented using a load path function, the algorithm is capable of providing information for sensitivity analysis in
design process. Additionally, the presented method does not need starting points to define the load paths, as are
required in vector-based methods; the load paths are obtained from only a single computational analysis; and
heuristic optimization is not needed in the process of determining the load paths.

II. Methodology
Considering the 𝑥 direction load-path, there are two stress components acting on a plane that is tangent to the
paths, namely 𝜎!!  and  𝜎!" , and the load-path function in the x direction (Ψ! ) can be written as:
𝜕Ψ! 𝜕Ψ! 𝜕Ψ! 𝜕Ψ! (1)
∆Ψ! = 𝑑𝑥 + 𝑑𝑦) ,          𝑤ℎ𝑒𝑟𝑒          𝜎!! =    𝑎𝑛𝑑    𝜎!" = −
𝜕𝑥 𝜕𝑦 𝜕𝑦 𝜕𝑥
The load path is bounded by lines where there are no contributions to the force in the x direction. Thus, the
equilibrium over the tube must be satisfied in the 𝑥 direction:
𝜕𝜎!! 𝜕𝜎!" 𝜕 ! Ψ! 𝜕 ! Ψ! (2)
+ = − =0
𝜕𝑥 𝜕𝑦 𝜕𝑥𝜕𝑦 𝜕𝑥𝜕𝑦
By taking the derivative of normal stress (𝜎!! ) and shear stress (𝜎!" ) with respect to 𝑥 and 𝑦 and using the weighted
residual method, the following equation can be derived:
𝜕 ! Ψ! 𝜕 ! Ψ! 𝜕𝜎!! 𝜕𝜎!" (3)
!
+ ! 𝛷𝑑𝐴 = ( − ) 𝛷𝑑𝐴  ;    𝑤ℎ𝑒𝑟𝑒  𝛷  𝑖𝑠  𝑡ℎ𝑒  𝑤𝑒𝑖𝑔ℎ𝑡  𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛
! 𝜕 𝑦 𝜕 𝑥 ! 𝜕𝑦 𝜕𝑥

2
 !!!! !!!"
The major issue in Eq. (3) is determining the derivative of the stress (  and   ). In order to remove the
!" !"
derivative on stress and place it on weight function (𝛷) the divergence and the Stokes’ theorem are applied to the
Eq. (3)
𝜕𝛷 𝜕Ψ! 𝜕𝛷 𝜕Ψ! 𝜕𝛷 𝜕𝛷 (4)
( + )𝑑𝐴 = (𝜎!! − 𝜎!" )𝑑𝐴
! 𝜕𝑦 𝜕𝑦 𝜕𝑥 𝜕𝑥 ! 𝜕𝑦 𝜕𝑥
The trial and test functions should satisfy the boundary conditions and have derivatives up to the order of the
differential equation (Eq. (4)). As seen in Eq. (4), there are no derivatives of normal 𝜎!! and shear stress 𝜎!" . The
aforementioned derivations removed the derivative of the stress, and, as a result, they can work with both stress-
based and displacement-based numerical methods. This is a very important feature of the proposed load-path
formulation, since most commercial simulation software programs use displacement-based numerical methods.
The weight function is a linear combination of the basis functions 𝜑! , 𝛷 = 𝑐! 𝜑! . Also the load path 𝛹 can be
approximated using the linear combination of the basis functions 𝜓! , Ψ! = 𝑑! 𝜓! .
Downloaded by EMBRY-RIDDLE AERO UNIV. on January 29, 2016 | http://arc.aiaa.org | DOI: 10.2514/6.2016-0233

III. Results and Discussion

1. Square plate under shear load
As was mentioned earlier, the stress components cannot be presented using an analytical expression for a complex
system, and the load path functions may not be derived using analytical approaches. In these problems usually
numerical approaches such as finite element methods (FEM) are utilized. The presented formulation associated with
the determination of the load paths easily can be integrated into a finite element package to provide load path
contours. To test and validate this, a one-sided, clamped, square plate subjected to shear load on the right corner
(Fig. 1) was solved using FEM and the stress components are determined. Then the stresses are substituted in Eq. (4)
and load path functions are obtained. To verify the accuracy of the proposed load-path functions, the plots of 𝜎!!
!!! !!! !!!
and    (Fig. 2 (a)), 𝜎!" and − (Fig. 2 (b)), and 𝜎!! and − (Fig. 2 (c)) were compared. The comparison of
!" !" !"
!!! !!! !!!
the results of the load path function derivatives (  , , and )  with the results of the stresses determined by
!" !" !"
the finite element method (𝜎!! , 𝜎!" , and 𝜎!! ) showed the accuracy of the approximation of Ψ!  and  Ψ! .
After verification of the new method, the load paths were determined by finding the load path functions. The load
path functions, which demonstrate the load paths for equilibrium in the 𝑥 and 𝑦 directions, are shown in Fig. 2 (d).
The load paths in the 𝑦 direction (Ψ! ) show how the concentrated load at the right corner is transferred and
distributed to the points of support on the left edge. The Ψ! contours also indicate that the load paths near the right
corner line up with 𝜎!! , since it is the dominant stress in this region, and as we get further from the point of
application, the 𝜎!" contribution increases and the orientation angle decreases. As we get closer to the top and
bottom left corners, the load paths are further under the influence of 𝜎!! and the orientation angle increases. The
load paths in the 𝑥 direction also provide physical insight into the load transfer and aid in understanding how various
resultant stresses are developed inside the structure. As can be seen in Fig. 2 (d)-  Ψ! , the load flows from one
support point to another support point. As we get further from the support edge, the load paths bend towards the
point of application at the right corner, and, at some point, they even reach the location of the concentrated load.
These two observations explain how the bending moments due to normal and shear stresses are developed inside the
structure.

P
Fig. 1 Clamped plate under shear load

3
Downloaded by EMBRY-RIDDLE AERO UNIV. on January 29, 2016 | http://arc.aiaa.org | DOI: 10.2514/6.2016-0233

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

𝝏𝚿𝒙 𝝏𝚿𝒙 𝝏𝚿𝒚
Fig. 2 (a) Comparison of and 𝝈𝒙𝒙 ; (b) Comparison of - and 𝝈𝒙𝒚 ; (c) Comparison of - and
𝝏𝒚 𝝏𝒙 𝝏𝒙
𝝈𝒚𝒚 ; (d) Load paths in the y and x directions

Next, a square plat restrained at top and bottom left corners and subjected to a concentrated load at 45-degree at
!! !!!!
bottom right corner is considered (Fig. 3). The stress components 𝜎!! , 𝜎!" , 𝑎𝑛𝑑 𝜎!! are compared with !, ,
!" !"
!!!!
and  in Fig. 4 (a), (b), and (c), respectively, and the x and y direction load paths are shown in Fig. 4 (d). As can
!"
be seen in Fig. 4 (d), the load paths start from the load application point and end at one of the support points instead
of the entire left edge. It is also noted from Fig. 4 (d), that the hole is deflecting the load paths, compared to those
derived for the plate without hole in Fig. 2 (c). Since the load is applied at 45-degree, the load paths in x and y
directions are very similar to each other except close to the support where the x direction load paths show the
bending close to the bottom corner of plate.

P
Fig. 3 Square plate with a hole

4
Downloaded by EMBRY-RIDDLE AERO UNIV. on January 29, 2016 | http://arc.aiaa.org | DOI: 10.2514/6.2016-0233

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

𝝏𝚿𝒙 𝝏𝚿𝒙 𝝏𝚿𝒚
Fig. 4 (a) Comparison of and 𝝈𝒙𝒙 ; (b) Comparison of - and 𝝈𝒙𝒚 ; (c) Comparison of - and
𝝏𝒚 𝝏𝒙 𝝏𝒙
𝝈𝒚𝒚 ; (d) Load paths in the y and x directions

2. Superposition of load path functions

Since the load path functions and load paths are obtained from the Poisson equation presented in Eq. (11), the
superposition principle can be applied to them. Therefore, if the load path functions for two sets of loadings and
       
boundary conditions are (𝛹!!  
, 𝛹!! ) and (𝛹!! , 𝛹!! ), then the superposition of the two load path functions
       
(𝛹!! + 𝛹!! , 𝛹!! + 𝛹!! ) are the load path functions for the combination of two sets and satisfies the corresponding
Poisson equations:
   
𝜕 ! (𝛹!!    
+ 𝛹!! ) 𝜕 ! (𝛹!!  
+ 𝛹!!  
) 𝜕(𝜎!!!    
+ 𝜎!!! ) 𝜕(𝜎!"! + 𝜎!"! )
!
+ !
= −    
𝜕 𝑦 𝜕 𝑥 𝜕𝑦 𝜕𝑥
(1)
!     !            
𝜕 (𝛹!! + 𝛹!! ) 𝜕 (𝛹!! + 𝛹!! ) 𝜕(𝜎!"! + 𝜎!" ) 𝜕(𝜎!!! + 𝜎!!! )
  + = −
𝜕!𝑦 𝜕!𝑥 𝜕𝑦 𝜕𝑥

An application of the method of superposition is illustrated for the plate with a hole when the concentrated 45-
degree load is decomposed to vertical and horizontal loads (Fig. 5). The load paths in x and y directions for two
loading conditions are shown in Fig. 6 (a) and (b). Then, by applying the superposition principle the load paths for
combined loading are determined. As can be seen in Fig. 6 (a)-(b) and Fig. 4 (d), the superposed load paths are
similar to the load paths determined for the 45-degree load. The ability to use the superposition method for the load
path functions provides the capability to break the loading into simpler components to simplify the load path
calculation of complex problems.
L L L

L L L

P
√
2
2
+ P
√
2
2
=
Fig. 5 superposition of vertical and horizontal loads for the square plate with a hole
P

5
 (a)
+ =
Downloaded by EMBRY-RIDDLE AERO UNIV. on January 29, 2016 | http://arc.aiaa.org | DOI: 10.2514/6.2016-0233

(b)
+ =
Fig. 6 (a) x-direction and (b) y-direction load paths produced by the method of superposition

3. L-bracket under shear load

To further check the accuracy of the present formulations the load paths for an L-shaped cantilever bracket (Fig. 7)
are determined and compared with those obtained by Kelly et al. [7]. As can be seen in Fig. 8 (a) and (b) both load
paths in the x and y directions are in good agreement with Kelly et al.’s results. The load paths in the x direction
(𝛹! ) have wide circulations at the top and bottom of the L-shaped domain. These circulations show the regions that
do not contribute to the load transfers in the x direction, and the load transfer in the x direction is majorly through
the regions above and below the circulations. There are two sets of load paths in y direction (𝛹! ), one set in the right
region that shows how the applied load is transferred from the application point to the middle corner, and another set
in the left region from one support point to another that implies the bending moment in the domain.

Fig. 7 L-shaped domain under transvers load

6
Downloaded by EMBRY-RIDDLE AERO UNIV. on January 29, 2016 | http://arc.aiaa.org | DOI: 10.2514/6.2016-0233

(a)

(b)
Fig. 8 Load path in an L-bracket (a) using load paths functions (b) presented by Kelly et al. [7]

4. Plate with a pin loaded hole

Since the material properties have not been used in the load path function derivation, the proposed algorithm should
be able to show the load paths in anisotropic structures. In order to validate the applicability of the load path
function formulations for the load path identification of these structures, a pin-loaded plate (Fig. 9 (a)) with material
properties of E1= 181000MPa, E2=10300MPa, 𝜈!" = 0.28, G12= 7170, and staking sequence of [0/45/90]s is
considered. The x-direction load paths (𝛹𝒙 ) in the mid-layer and its comparison with those obtained by Tosh and
Kelly [16] are shown in Fig. 10 (a) and Fig. 9 (b), respectively. The load path starts from the distributed applied load
at the hole with angle zero and stays zero until it gets farther from the hole toward the right. It is also seen that the
load paths have zero angle for the entire left region of structural domain. This means the load transfer in the left side
as well as close to the hole is pure axial loading due to 𝜎!! and the shear stress does not contribute to the load
transfer. When the load path gets to the right end of the plate the shear stress becomes larger and axial stress
becomes zero, then the axial stress becomes significant until the load path gets to the left edge of the plate. The x-
direction load paths (𝛹𝒙 ) for an isotropic pin-loaded plate is shown in Fig. 10 (b), and as can be seen, the load paths
turning points are closer to the hole than those of composite plate. Also, the load flow is distributed in a wider region
in the composite structure than the isotropic structure which results in a better use of materials and provides the
capability to design a more efficient structure.

7
 (a) (b)
Fig. 9 (a) Pinned plate under axial load and (b) its x-direction load paths [16]
Downloaded by EMBRY-RIDDLE AERO UNIV. on January 29, 2016 | http://arc.aiaa.org | DOI: 10.2514/6.2016-0233

(a) (b)
Fig. 10 X-direction load paths for (a) composite and (b) isotropic plate with a pin loaded hole

5. Wing plane subjected to aeroelastic load

Because the loading and boundary conditions have not been used in any stages of the theoretical or computational
development, the proposed load path algorithm should also be able to provide the load paths for any type of loading
and boundary conditions, such as mechanical, thermal, and aerodynamic loads. As long as the numerical solver can
calculate the stress components in the structure, the proposed load path algorithm is able to determine the load flow
and load paths. We plan to validate this through an example of a wing under Aeroelastic loading. The aeroelastic
analysis is conducted using Doublet-Lattice Methods and MSC Nastran Aeroelastic Trim analysis. The aeroelastic
pressure distribution on the aerodynamic mesh is presented in Fig. 11 (a). The load paths in the x and y directions
(i.e. 𝛹! 𝑥, 𝑦  𝑎𝑛𝑑  𝛹! 𝑥, 𝑦 ) are determined by using Eq. (4) and are shown in Fig. 11 (b) and (c). The load path in x
direction (𝛹! ) starts from the leading edge on the left side of the wing with small angle in the region close to the
wing root. The angle starts increasing closer to the wing tip, which means close to the root, 𝜎!! has larger
contribution to the load flow in the x direction, whereas, close to the tip, the load flow is the resultant of both 𝜎!!
and 𝜎!" . The load path in the y direction (𝛹! ) starts from the leading edge with approximately 45-degree angle,
which means the 𝜎!" and 𝜎!! have the same contribution in the y directional load transfer. It is also quite interesting
to note that the pattern in the x and y directions load-paths resembles the spars and ribs patterns in a wing box [17].

(a) (b) (c)

Fig. 11 (a) Aeroelastic pressure distribution on the Aerodynamic mesh; Load paths in the (b) x direction and
(c) y direction for a wing plane

8
 IV. Conclusion
In this paper, definition and formulations for load flow and load path functions based on the equilibrium equations
are introduced. By taking advantage of the load path functions, the equilibrium equations are converted to the Poison
equations. The loading, boundary conditions, and material properties have not been used in any stages of the
theoretical or computational development. Therefore, the load paths for various plane stress structures with different
boundary conditions, loadings and materials were determined with the new method, and the results were compared
with those available in the literature. The four main advantages of proposed load path method are its computational
efficiency by requiring only a single computational analysis; computational simplicity which makes it possible to be
easily integrated in a computational software, elimination of post-processing to show the load path contours, and its
capability of obtaining both load paths and load flow.

Acknowledgment
The work presented here was partially supported by the Air Force Summer Faculty Fellowship Program (SFFP). We
Downloaded by EMBRY-RIDDLE AERO UNIV. on January 29, 2016 | http://arc.aiaa.org | DOI: 10.2514/6.2016-0233

appreciate the support and are thankful to the experts at Air Force Research Laboratory's Multidisciplinary Science
and Technology Center for their helpful comments and suggestions.

References

1. A. G. M. Michell, “The limits of economy of material in frame structure,” Philosophical Magazine, Vol. 8, No.
6, pp. 589-597, 1904.
2. G. N. Vanderplaats and E. Salajegheh, “New approximation method for stress constraints in structural
synthesis,” AIAA journal, Vol. 27, No .3, pp. 352-358, 1989.
3. H. L. Thomas and G. N. Vanderplaats, “An improved approximation for stress constraints in plate structures,”
Structural optimization, Vol. 6, No. 1, pp. 1-6, 1993.
4. D. Kelly and M. Elsley, "A procedure for determining load paths in elastic continua," Engineering
Computations, vol. 12, pp. 415-424, 1995.
5. D. Kelly and M. Tosh, "Interpreting load paths and stress trajectories in elasticity," Engineering Computations,
vol. 17, pp. 117-135, 2000.
6. W. Waldman, M. Heller, R. Kaye, and F. Rose, "Advances in two-dimensional structural loadflow
visualisation," Engineering Computations, vol. 19, pp. 305-326, 2002.
7. D. Kelly, C. Reidsema, A. Bassandeh, G. Pearce, and M. Lee, "On interpreting load paths and identifying a load
bearing topology from finite element analysis," Finite Elements in Analysis and Design, vol. 47, pp. 867-876,
2011.
8. H. Harasaki and J. S. Arora, "New concepts of transferred and potential transferred forces in structures,"
Computer methods in applied mechanics and engineering, vol. 191, pp. 385-406, 2001.
9. T. Sakurai, K. Takahashi, H. Kawakami, and M. Abe, "Reduction of calculation time for load path U* analysis
of structures," Journal of Solid Mechanics and Materials Engineering, vol. 1, pp. 1322-1330, 2007.
10. H. Hoshino, T. Sakurai, and K. Takahashi, "Vibration reduction in the cabins of heavy-duty trucks using the
theory of load transfer paths," JSAE Review, vol. 24, pp. 165-171, 2003.
11. K. Marhadi and S. Venkataraman, "Comparison of quantitative and qualitative information provided by
different structural load path definitions," International Journal for Simulation and Multidisciplinary Design
Optimization, vol. 3, pp. 384-400, 2009.
12. M. Santer and S. Pellegrino, "Topological optimization of compliant adaptive wing structure," AIAA journal,
vol. 47, pp. 523-534, 2009.
13. K.-J. Lu and S. Kota, "An effective method of synthesizing compliant adaptive structures using load path
representation," Journal of Intelligent Material Systems and Structures, vol. 16, pp. 307-317, 2005.
14. S. Venkataraman, S., K. S. Marhadi, and M. A. Haney, “Investigating alternate load paths and damage tolerance
of structures optimized for multiple load cases,” 50th AIAA/ASME/ASCE/AHS/ASC Structures, Structural
Dynamics, and Materials Conference, Palm Springs, CA, 2009.
15. S. P. Timoshenko and J. N. Goodier, Theory of Elasticity, McGraw-Hill, 1951
16. M. Tosh and D. Kelly, "On the design, manufacture and testing of trajectorial fibre steering for carbon fibre
composite laminates," Composites Part A: Applied Science and Manufacturing, vol. 31, pp. 1047-1060, 2000.
17. D. Locatelli, Q. Liu, A. Y. Tamijani, S. B. Mulani, and R. K. Kapania, "Multidisciplinary Optimization of
Supersonic Wing Structures Using Curvilinear Spars and Ribs (SpaRibs)," in 54th AIAA/ASME/ASCE/AHS/ASC
Structures, Structural Dynamics, and Materials Conference, 2013.
9

View publication stats