Polymer Testing 81 (2020) 106123

Contents lists available at ScienceDirect

Polymer Testing
journal homepage: http://www.elsevier.com/locate/polytest

Analysis Method

A refined numerical model for determining inflation-burst behavior of

composite membrane structures
Taibai Shi a, Jianhui Hu a, b, Wujun Chen a, b, *, Chengjun Gao a
a
Space Structures Research Center, Shanghai Jiao Tong University, Shanghai, 200240, China
b
State Key Laboratory of Ocean Engineering, Shanghai, 200240, China

A R T I C L E I N F O A B S T R A C T

Keywords: The airship structures made of multi-layer composite fabrics or membranes can offer the platform for earth
Airship structures observations, wireless communications and space research due to light weight and good mechanical perfor­
Biaxial response surface mance. The structural safety and serviceability strongly depend on material properties and working conditions.
Composite fabrics
Available studies are limited within service stress limits or are lack of suitable biaxial tensile constitutive models
Constitutive relationship
for understanding structural behavior. This paper thus focuses on a refined numerical model for determining
Failure criterion
FEM inflation-burst behavior of composite airship structures considering new biaxial constitutive equations, novel
Numerical method failure criteria and manufacture factors.
Refined model The differences between ideal and real forms of airship structures, e.g. volume difference, demonstrate the
necessity for incorporating cutting-pattern effects in the initial numerical model. For structural analysis, stress
distributions on real structural forms are different from those on ideal forms because of welding parts that can
enhance local stiffness. The ultimate pressures are 56.7 kPa and 59.5 kPa for ideal and real structural forms.
Structural breaking initiated at the maximum diameter of ideal structural forms propagates fast while welding
parts can prevent breaking propagation for real structural forms. Therefore, the refined numerical model can
reveal basic structural behavior and safety performance of airship structures in the inflation-burst processes.

1. Introduction effects on structural behavior of a high-altitude solar-powered airship

and concluded that solar radiation has a great effect on thermal char­
The high-altitude airship structures made of multi-layer composite acteristics of hybrid airship [8]. In general, these studies mainly focused
fabrics or membranes with light weight and good mechanical perfor­ on ideal structural forms without the consideration of manufacture
mance, can offer the platform for earth observations, wireless commu­ factors and thus possible deviations exist between numerical analysis
nications and space research [1–3]. The basic working principle of these and real working conditions. The manufacture factors play a key role in
structures is that the density difference between filling gas and sur­ affecting final performance of these airship structures as large volume is
rounding air generates the buoyancy to enable stationary states at high a useful way to meet air density reduction at high latitudes [4,9]. Ac­
altitude [4]. Considering complex material properties, manufacture cording to typical manufacture process, two main categories are mate­
process, and operation period, the understanding of structural design rials and structures.
and analysis is indispensable in response to these factors. The materials used for airship structures are multi-layer composite
Stockbridge et al. presented a review of current techniques for fabrics or membranes with high ratio of strength to weight and UV
airship dynamics from concept design, optimization and structural protection. For these composite materials, uniaxial mechanical proper­
analysis [5]. Hygonenc and Soueres analyzed autonomous airship con­ ties are commonly-used because of easy characterization with tensile
trol under low perturbations and proposed a reduced model for different machines [10,11]. Uniaxial tensile/cyclic properties and constitutive
wording phases [6]. Yao et al. developed a heat transient model for models are available for typical fabric materials [12,13]. However, the
thermal behavior prediction of stratospheric airships in launching and biaxially-loaded airship structures indicate that uniaxial mechanical
returning processes, finding that volume, vertical speed and solar radi­ properties are not appropriate for understanding structural behavior
ation are significant factors [7]. Zhang et al. evaluated temperature [14]. In this case, Bridgens and Gosling presented a direct stress–strain

* Corresponding author. Space Structures Research Center, Shanghai Jiao Tong University, Shanghai, 200240, China.
E-mail addresses: [email protected], [email protected] (J. Hu), [email protected] (W. Chen).

https://doi.org/10.1016/j.polymertesting.2019.106123
Received 31 May 2019; Received in revised form 3 August 2019; Accepted 25 September 2019
Available online 1 November 2019
0142-9418/© 2019 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
T. Shi et al. Polymer Testing 81 (2020) 106123

representation for coated-woven fabrics aiming to avoid plane stress the Conclusions.
assumptions [15]. Zhang et al. analyzed tearing behavior of coated
fabrics and evaluated resistance uncertainty and structural reliability of 2. Basic properties and considerations
hyper tensioned membrane structures [16,17]. Chen et al. developed a
multi-functional biaxial machine and suggested that loading ratios in The real applications of airship structures require to consider biaxial
two directions are suitable to obtain proper mechanical properties [18]. strength, welding properties and initial forms. This research topic can be
Apart from mechanical properties of original fabric materials, welding investigated based on a series of our work [22,26,27]. In this section, a
materials and related properties are essential since the width of com­ brief and concise description of these research is given to show the
posite fabric materials is limited regarding production machine, result­ necessary background and work as the basis for following structural
ing in indispensable welding parts in large-volume airship structures. analysis.
Therefore, basic welding properties are essential to analyze reasonable
structural behavior and possible breaking modes. In this case, Jevsnik
et al. reported the effects of ultrasonic welding parameters on bond 2.1. Material levels
strength, seam thickness and seam stiffness. A statistical analysis showed
that ultrasonic welding parameters influence welding properties but 2.1.1. Biaxial strength and failure criteria
obtained values are statistically insignificant [19]. Chen et al. analyzed The composite fabric material used in this study is typical in the
uniaxial and cyclic mechanical properties of welding polytetrafluoro­ airship industry and basic properties are thickness of 0.30 mm, areal
ethylene fabrics [20]. Hu et al. proposed a welding machine for poly­ density of 240 g/m2 and weave count of 48 � 40 ends/5 cm. From the
mers and fabrics in combination with temperature and pressure for viewpoint of symmetric characteristics of airship structures, five warp-
better identification of mechanical properties [21]. Shi et al. obtained to-weft stress ratios in the first quartile are designated to determine a
stress-strain curves of three welding types with the digital image cor­ complete biaxial constitutive relationship listed in Table 1. The biaxial
relation method for characterizing elastic modulus and Poisson’s ratio equipment and DIC technique are activated simultaneously to record
[22]. force and deformation fields. The resulting five stress-strain curves in
The second level of manufacture factors is cutting-pattern effect that warp and weft directions are used to fit biaxial response surfaces. For
is inevitable for membrane structures since ideal airship forms need to more details, please see [27].
be translated to plan forms for arranging fabric materials [23,24]; this Fig. 2 depicts the complete biaxial response surfaces in warp and
process results in possible differences between ideal and real forms. In weft directions by inputting experimental results in three-dimensional
general, structural forms can affect operating performance in a way that stress-stress-strain space. Fitting these results using polynomial equa­
the center deviation (pitching) between gravity and buoyancy, illus­ tions can get expressions for biaxial response surfaces [27].
trated in Fig. 1, can lead to serious results of airship structures. To
evaluate cutting-pattern effects on initial forms of airship structures, the εx ¼ 3:03*10 6 σ 2x þ 1:981*10 6 σ 2y þ 1:832*10 7 σ x σy þ 6:774*10 6 σx
computer-aided method is an appropriate way to address this research 2:424*10 4 σy 1:328*10 4

topic. Wagner investigated patterning effects on airship structures as a (1)

function of elastic-plastic behavior to achieve desired geometry [24].
Tabiei and Jiang employed material nonlinearity in woven-fabric
models for finite element simulation [25]. Qiu et al. further evaluated
initial configuration and nonlinear structural behavior of stratospheric Table 1
airship structures [26]. However, these studies concerning structural Warp-to-weft stress ratio determination for composite fabric materials.
analysis are limited within service stress limits or are lack of suitable Biaxial stress space Ratio Slope Ratio
biaxial tensile constitutive models. sequence angle value
To evaluate real structural behavior in accordance with manufacture ① 15� 0.27:1
factors, material level (biaxial mechanical properties and welding ② 30� 0.58:1
properties) and structural level (initial real form) are needed to be in­ ③ 45� 1:1
60� 1:0.58
tegrated in one refined model to evaluate full-process behavior of airship
④
⑤ 75� 1:0.27
structures. However, this research is not available after a survey of the
literature. This paper thus focuses on the refined numerical model of
composite airship structures and assesses structural behavior in
inflation-burst processes.
In this paper, basic properties and considerations, such as biaxial
strength, failure criteria, welding properties as well as initial forms are
briefly summarized in section 2. The numerical models from structural
forms, structural analysis and bursting process are established with such
factors. The detailed analysis of structural behavior and comparisons
between ideal forms are carried out to emphasize the importance of
refined analysis. The useful values and observations are summarized in

Fig. 1. Normal working and pitching cases for airship structures.

2
T. Shi et al. Polymer Testing 81 (2020) 106123

Fig. 2. Three-dimensional stress-stress-strain space: (a) warp direction, (b) weft direction [27].

εy ¼ 1:034*10 6 σ2x 4:718*10 6 σ 2y þ 1:048*10 7 σ x σ y 1:517*10 4 σ x design, platform reliability and structural stiffness. A typical welding
seam is defined as that two sides of base fabrics align at the interface
4
þ 8:491*10 σ y þ 8:933*10 4
(2)
without overlapping. Then, a knee fabric is welded onto two surfaces
with the heat welding [28]. In this study, welding seams are charac­
where εx and εy are strain values while σx and σ y are stress values in warp
terized under tensile load and noncontact measurement technology, i.e.
and weft directions.
DIC. The strains are given as the original length divided by deformation
The Tsai-Hill, Yeh-Stratton and Norris failure criteria are established
[22]. One thing worth noting is that the stiffness of paint layer is less
on basis of uniaxial properties and are not suitable for biaxially-
than one thousandth of fabric materials and thus the error is negligible.
tensioned membrane materials. Significant deviation exists between
A typical welding seam and corresponding DIC results are given in Fig. 4
these strength criteria for biaxial stress ratios. Therefore, a novel biaxial
[22].
failure criterion is proposed in analog to biaxial criteria for composite
The elastic modulus is defined as the slope of stress-strain curves
plate materials for biaxial failure results [27].
within 20% ultimate strains [12]. For Poisson’s ratio, the
δ2x δ2y δx δy δx δy τ2xy evenly-distributed data points are marked on specimen edges. Average
2
þ 2
þ þ þ þ ¼1 (3) coordinates and displacements for two horizontal lines are determined
ðαXÞ ðβYÞ γXY ϕX ωY S2
and labeled as X1, ΔX1 and X2, ΔX2. Similarly, average coordinates and
where X and Y are tensile strength and α, β, γ, ϕ and ω are parameters to displacements for two vertical lines are given and labeled as Y1, ΔY1 and
describe yarns interactions between warp and weft directions (1.64, Y2, ΔY2. Therefore, the Poisson’s ratio is calculated with following
2.32, 1.02, 1.71 and 1.26 for the fabric material used in this study). equations [12,22]
The five biaxial breaking strengths at different stress ratios can obtain a Δx2 Δx1
failure envelope of composite fabrics in biaxial stress space, see Fig. 3. εx ¼ (4)
x2 x1
The new failure criterion shows that failure envelope for composite
fabrics is a finite part of an ellipse in the first quartile of biaxial stress Δy2 Δy1
space. εy ¼ (5)
y2 y1

2.1.2. Welding properties v¼

εx
(6)
The mechanical behavior of welding seams is essential for airship εy

where εx and εy denote horizontal and vertical strains. v is Poisson’s

ratio. The elastic modulus and Poisson’s ratio are summarized in
Table 2.

2.2. Structural level

Analyzing initial configuration of typical airship models can be

performed with computational cutting-pattern that is composed of cut­
ting subdivision, flattening and pattern assembly integration [26].
For cutting subdivision, numerical airship models can be expressed
as double-axis ellipsoid that is generated by revolving shape profile.
Several geodesic seam lines are assumed to divide airship surface in
accordance with manufacture process. One gore is necessary to flatten as
these gores are identical. A geometrical method with equal-area
approach is employed to minimize distortions resulting from flattening
curved shapes into plan shape. The plan gore is welded from 2D and
transformed to 3D in pattern assembling integration process. The gore
models with nonlinear material properties are utilized to carry out nu­
merical analysis. In addition, the gravity to ensure smooth surface and
Fig. 3. Failure envelope under different failure criteria [27]. avoid wrinkling is incorporated in the analysis. For more details, please

3
T. Shi et al. Polymer Testing 81 (2020) 106123

Fig. 4. Welding specimen and DIC results for determining material properties [22].

presents detailed descriptions, analysis and comparisons on these topics.

Table 2
Elastic modulus and Poisson’s ratio of fabric materials.
Parameters 1 2 3 Average 3.1. Structural forms
Elastic modulus (N/mm) 246.94 265.66 246.57 253.06
Poisson’s ratio 0.7947 0.7762 0.8135 0.7796 In general, the ideal form of airship structures is a streamlined rotor
that can be expressed as the ellipsoid equation. A typical model is given
in Fig. 5 with the long axis of 10 m and maximum diameter of 2.5 m. To
see [26]. consider manufacture process, the surface of airship structure is divided
Generally, it is validated that the airship form is affected by cutting into 12 parts and these parts are assembled to form real structural forms
and patterning processes. The median line is along the given shape while in this initial state.
other surfaces are out of the specified shape. The shape change and The curved surfaces are required to be translated into plan surface in
volume rise are obtained between ideal and real configurations to a accordance with material production. To evaluate cutting-pattern ef­
certain degree. fects, the plan surface is meshed with rectangular elements and then
assembled these 12 parts along the center line. Two adjacent parts are
3. Refined modelling merged to form structural forms. In this process, material properties and
welding properties are taken into consideration in the calculation pro­
The numerical modelling of airship structures consists of three cess. The structural form as well as stress and deformation distributions
interdependent steps, i.e. structural form, structural analysis with me­ on a typical part are exhibited in Fig. 5. The comparisons between ideal
chanical properties, and bursting analysis with new criteria. This section and real forms show that the diameter, structural length and volume

Fig. 5. Cutting-pattern analysis for airship structures.

4
T. Shi et al. Polymer Testing 81 (2020) 106123

increase by 1.2%, 0.78% and 2%; these values quantify the cutting- Based on numerical results, the ultimate pressure for ideal airship
pattern effects on structural behavior. Therefore, this study includes structures is 56.7 kPa. Fig. 6 shows stress distributions in longitudinal
such effects during the refined analysis to reveal proper structural and transverse directions at 50% and 100% ultimate pressure values.
behavior. The stress distributions in longitudinal and transverse directions are
non-uniform and transverse stress is higher than longitudinal stress. The
3.2. Structural analysis material fracture tends to start at the pressure of 56.7 kPa at the
maximum diameter. The corresponding longitudinal and transverse
The main difference between this study and other associated stresses are 37.7 N/mm and 77.7 N/mm, i.e. stress ratio of 0.49:1 (The
research lies in proper mechanical properties and failure criteria. The definition of stress ratio is the ratio of transverse stress to longitudinal
biaxial constitutive equations are obtained as a function of the complete stress). The structural breaking along structural longitudinal direction
biaxial loading ratios for airship structures, see Eqs. (1) and (2) and initiates after the material fracture. The stresses near this area decrease
Fig. 2, which are imported into the software with the VUMAT algorithm. dramatically while the stresses in other parts are still at high stress state
To investigate bursting characteristics, the five-parameter criteria and bursting process goes until the structural breaking.
intended for composite fabrics is utilized and incorporated in the soft­ Unlike ideal airship forms, the manufacture-based model considers
ware to assess structural behavior, given in Eq. (3). initial forms and welding properties (mechanical properties are 212 MPa
After choosing these basic material properties and real structural and 461 MPa in longitudinal and transverse directions [22]). The ulti­
forms, the numerical model is established with ABAQUS that is suitable mate pressure of airship structures is 59.5 kPa. The stress distribution at
for nonlinear analysis for composite fabric structures. The basic steps for 50% and 100% pressure values as well as bursting characteristics are
numerical analysis are composed of pre-processing, simulations and displayed in Fig. 7. The general characteristics of stress distributions are
post-processing where the pre-processing is crucial as material proper­ similar as ideal airship structures. However, the stresses are different in
ties and constitutive equations are defined. In this study, the user- welding parts and in-between welding parts. In detail, the stress in
defined subroutine is employed to develop new programs in terms of welding parts is lower than surrounding area due to larger stiffness. The
material properties, loads and boundary conditions. Such programs can material fracture initiates at the maximum diameter with the maximum
be used by the software and give updated feedback information. Unlike longitudinal and transverse stresses of 26.2 N/mm and 74.3 N/mm, i.e.
UMAT, the VUMAT that can incorporate material fracture and bursting stress ratio of 0.35:1. The following characteristics after material frac­
process is implemented to simulate inflation-burst process of airship ture are structural breaking near welding line edges and dramatical
structures. stress reduction. In this case, the redundant stiffness of welding parts
In this self-developed program, a basic parameter is the STRESSNEW postpones crack propagation in the structures.
that can be calculated using STRESSOLD and STRESSINC as well as
biaxial constitutive model. Then, other conditions are imported to the 4.2. Comparisons
main software for next calculations. The structural behavior is assessed
with the failure criteria to remove breaking elements and adjust struc­ The differences between ideal and real airship forms in inflation-
tural stiffness and geometry. Therefore, this newly-developed algorithm burst process exist and the understanding of basic mechanism and
integrated into the main software can perform the whole inflation-burst reason is useful for future analysis, optimization and design.
process of airship structures. Two main parameters to be analyzed are ultimate pressure and
bursting mechanism. For ultimate pressure, it increases 5% from ideal to
3.3. Modelling information real airship structures. The possible reasons for explaining this obser­
vation are that the 12 welding parts increase structural stiffness which
The boundary condition of airship structures is simply supported to can improve ultimate pressure and that the diameter of real airship
eliminate body displacement. To achieve general structural behavior, structure changes because of cutting-pattern effects, further enhancing
the load taken into account is only the inner pressure since inner and structural stiffness and ultimate pressure.
external pressure difference is the main reason to generate high stress of The busting locations are at the maximum diameter for both ideal
airship structures. The shell element (S4R) is used to perform numerical and real airship structures. A difference is that the cracks propagate
analysis and the total elements are 19392 in the structural model. along welding line edges since stiffness improvement by welding parts
The solution method in the inflation-burst process is the explicit can restrain crack development.
dynamic method that uses the step integral to solve differential equa­
tions; this method is suitable for nonlinear bursting process. The basic 5. Conclusions
principle of this method is given as that old results can be utilized to
calculate new results without iterative calculations. It is conditionally This paper concerns a refined numerical model to determine
stable that a step increment is less than natural frequency. In this case, inflation-burst behavior of composite airship structures considering
the time step of 0.001s is appropriate as the natural frequency of the initial real forms and using new biaxial constitutive equations and novel
airship model is 0.026s [29]. In addition, the linear bulk viscosity failure criteria. Several observations and conclusions are summarized as
parameter is set as 0.6 to simulate damping effects. follows.
On basis of these parameters and preparations, the numerical anal­
ysis is performed to investigate structural behavior and bursting char­ � The differences between ideal and real forms of airship structures
acteristics. Moreover, a comparative analysis with an ideal airship demonstrate the necessity for incorporating cutting-pattern effects in
structure is done to identify differences between these two forms. the initial numerical model.
� Stress distributions for real structural form are different from ideal
4. Results and discussion forms due to the existence of welding parts that can enhance local
stiffness.
4.1. Results analysis � The ultimate pressures are 56.7 kPa and 59.5 kPa for ideal and real
structural forms. Structural breaking propagates fast for ideal
The numerical results of airship structures are composed of stress structural forms while welding parts can postpone breaking propa­
distributions at typical pressures and bursting characteristics of ideal gation for real structural forms.
and real structural forms. This section analyzes typical structural
behavior during the inflation-burst process. In general, the refined numerical model can reveal structural

5
T. Shi et al. Polymer Testing 81 (2020) 106123

Fig. 6. Stress distributions in two directions at 50%, 100% Pcr and breaking condition (ideal airship structure).

Fig. 7. Stress distributions in two directions at 50%, 100% Pcr and breaking condition (real airship structure).

behavior and safety performance of airship structures in the inflation- Declaration of competing interest
burst processes.
The authors declare that there are no conflict of interests.
Data statement
Acknowledgements
Some or all data, models, or code generated or used during the study
cannot be shared at present due to the ongoing projects. The work was supported by the National Natural Science Foundation
of China (Nos. 51778362 and 51608320) and the Key research and

6
T. Shi et al. Polymer Testing 81 (2020) 106123

development program of the ministry of science and technology (No. [15] B. Bridgens, P. Gosling, Direct stress–strain representation for coated woven
fabrics, Comput. Struct. 82 (2004) 1913–1927.
2016YFB1200200). The authors are grateful to the editors and anony­
[16] Y. Zhang, J. Xu, Y. Zhou, Q. Zhang, F. Wu, Central tearing behaviors of PVC coated
mous reviewers for professional comments and suggestions in improving fabrics with initial notch, Compos. Struct. 208 (2019) 618–633.
the quality of the paper. [17] Y. Zhang, Y. Lu, Y. Zhou, Q. Zhang, Resistance uncertainty and structural reliability
of hyper tensioned membrane structures with PVC coated polyesters, Thin-Walled
Struct. 124 (2018) 392–401.
References [18] W. Chen, C. Gao, Z. Qiu, T. Shi, Recent development of bi-axial shearing test
method for fabrics, in: Proceedings of IASS Annual Symposia, International
[1] W. Chen, Design of Membrane Structure Engineering, China Building Industry Association for Shell and Spatial Structures (IASS), 2016, pp. 1–10.
Press, 2005. [19] S. Jev�snik, S.H. Eryürük, F. Kalao�glu, B.K. Kayao�
glu, P. Komarkova,
[2] L. Liao, I. Pasternak, A review of airship structural research and development, Prog. V. Golombikova, et al., Seam properties of ultrasonic welded multilayered textile
Aerosp. Sci. 45 (2009) 83–96. materials, J. Ind. Text. 46 (2017) 1193–1211.
[3] S.D. Ilcev, Stratospheric communication platforms as an alternative for space [20] W. Chen, Y. Tang, D. Zhao, X. Ren, G. Fu, S. Dong, Experiments on the mechanical
program, Aircraft Eng. Aero. Technol. 83 (2011) 105–111. properties and the welded seam properties of architectural membranes, Spat.
[4] G.A. Khoury, Airship Technology, Cambridge university press, 2012. Struct. 13 (2007) 006.
[5] C. Stockbridge, A. Ceruti, P. Marzocca, Airship research and development in the [21] J. Hu, W. Chen, S. Zhang, Q. Cai, D. Yang, A welding machine for thermoplastic
areas of design, structures, dynamics and energy systems, Int. J. Aeronaut. Space polyimide (TPI) films: novel apparatus and experimental investigations, Polym.
Sci. 13 (2012) 170–187. Test. 66 (2018) 327–333.
[6] E. Hygounenc, P. Soueres, Automatic airship control involving backstepping [22] T. Shi, W. Chen, C. Gao, J. Hu, B. Zhao, X. Wang, et al., Investigation of mechanical
techniques, in: IEEE International Conference on Systems, Man and Cybernetics, behavior of weld seams of composite envelopes in airship structures, Compos.
IEEE, 2002, p. 6. Struct. 201 (2018) 1–12.
[7] W. Yao, X. Lu, C. Wang, R. Ma, A heat transient model for the thermal behavior [23] D. Str€
obel, L. Gründig, P. Singer, Selected examples for the optimization of cutting
prediction of stratospheric airships, Appl. Therm. Eng. 70 (2014) 380–387. patterns for textile membranes, in: VI International Conference on Textile
[8] L. Zhang, J. Li, J. Meng, H. Du, M. Lv, W. Zhu, Thermal performance analysis of a Composites and Inflatable Structures. Munich, Germany, 2013, pp. 258–267.
high-altitude solar-powered hybrid airship, Renew. Energy 125 (2018) 890–906. [24] R. Wagner, Basics in tension structures, Int. J. Space Struct. 24 (2009) 223–231.
[9] W. Chen, S. Dong, Research and development of airship and high altitude long [25] A. Tabiei, Y. Jiang, Woven fabric composite material model with material
endurance platform in Germany (Europe), Spat. Struct. 4 (2006) 3–7. nonlinearity for nonlinear finite element simulation, Int. J. Solids Struct. 36 (1999)
[10] MSAJ, Testing Method for Elastic Constants of Membrane Materials, 1995. 2757–2771.
[11] ISO, Rubber-or Plastics-Coated Fabrics - Determination of Tensile Strength and [26] Z. Qiu, W. Chen, C. Gao, B. Zhao, J. Hu, X. Wang, et al., Initial configuration and
Elongation at Break, International Organization for Standardization, 2000. nonlinear mechanical analysis of stratospheric nonrigid airship envelope,
[12] J. Hu, C. Gao, S. He, W. Chen, Y. Li, B. Zhao, et al., Effects of on-axis and off-axis J. Aerosp. Eng. 32 (2018), 04018155.
tension on uniaxial mechanical properties of plain woven fabrics for inflated [27] T. Shi, W. Chen, C. Gao, J. Hu, B. Zhao, Pa Wang, et al., Biaxial constitutive
structures, Compos. Struct. 171 (2017) 92–99. relationship and strength criterion of composite fabric for airship structures,
[13] Y. Zhang, S. Xu, J. Xue, Q. Zhang, Anisotropic mechanical properties and Compos. Struct. 214 (2019) 379–389.
constitutive relations of PTFE coated glass fibers, Compos. Struct. 179 (2017) [28] V.K. Stokes, Joining methods for plastics and plastic composites: an overview,
601–616. Polym. Eng. Sci. 29 (1989) 1310–1324.
[14] Z. Qiu, W. Chen, C. Gao, Y. Hu, Experimental and numerical study on nonlinear [29] T. Shi. Experimental Study and Strength Model for Envelope Materials and
mechanical properties of laminated woven fabrics, Constr. Build. Mater. 164 Structures, Shanghai Jiao Tong University, 2019.
(2018) 672–681.