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

net/publication/338735007

Three Dimensional Dynamic Model Development and Validation for Stranded

Cables

Conference Paper · November 2019

DOI: 10.1115/IMECE2019-11504

CITATIONS READS
2 101

3 authors, including:

Mohammad Hadi Jalali D. Geoff Rideout

Memorial University of Newfoundland Memorial University of Newfoundland
16 PUBLICATIONS   125 CITATIONS    71 PUBLICATIONS   958 CITATIONS   

SEE PROFILE SEE PROFILE

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

Determination of cumulative fatigue life of an oil well drill string based on a combined approach of Bond Graph dynamic simulation and Finite Element Method View project

All content following this page was uploaded by Mohammad Hadi Jalali on 30 January 2020.

The user has requested enhancement of the downloaded file.

 Proceedings of the ASME 2019
International Mechanical Engineering Congress and Exposition
IMECE2019
November 11-14, 2019, Salt Lake City, UT, USA

IMECE2019-11504

THREE DIMENSIONAL DYNAMIC MODEL DEVELOPMENT AND VALIDATION FOR

STRANDED CABLES

Mohammad Hadi Jalali1 Richard McKercher Geoff Rideout

Memorial University Memorial University Memorial University
St. John’s, NL, Canada St. John’s, NL, Canada St. John’s, NL, Canada

ABSTRACT method for utility poles based on modal testing requires

Interactions between cables and structures affect the design numerical modeling of the poles and the attached cables
and non-destructive testing of electricity transmission lines, (conductors) to understand the effect of cables on pole dynamics.
guyed towers, and bridges. In order to understand the effect of A reliable and computationally efficient numerical model for the
conductors on dynamics of utility poles, a three dimensional cable itself is a prerequisite. Models of cables typically treat the
dynamic model for stranded cables is presented based on the cable as a string or a beam or as a series of beam-like segments.
bond graph formalism and validated through experimentation. Models vary in their treatment of bending stiffness and cable sag
The cable is modeled considering the bending stiffness, tension due to self-weight.
and sag due to self-weight. The model consists of three- Ricciardi et. al [1] developed a continuous model for
dimensional rigid segments, connected with translational and vibration analysis of cables with sag considering the bending
rotational springs and dampers. To validate the model, an stiffness. They considered the conductor as a tensioned Euler-
instrumented set up for a stranded cable was tested in the Bernoulli beam under self-weight. They concluded that the
laboratory and the model was validated in both the in-plane and bending stiffness and sag of the cable has significant effect on
out-of-plane directions in the frequency domain by measuring vibration properties of the cable and ignoring them can lead
the cable response to hammer impacts in both directions. Time erroneous results. Sousa et. al. [2] analyzed the effect of
domain response obtained from simulation was also validated considering bending stiffness, shear stiffness and rotational
with experimental time response by performing a free vibration inertia on the natural frequencies of the overhead transmission
test. A set up was designed and built to accurately measure the line conductors. Using Euler-Bernoulli and Timoshenko beam
bending stiffness of a stranded cable in different tensions and the theories, they found the error from neglecting shear and
measured values for bending stiffness were used in the numerical rotational inertia to be very small compared to the error from
model. It was observed that the bending stiffness increased with neglecting the bending and geometric stiffness of the cable.
the increase in tension. The modal properties of the cable were Papailiou [3] proposed a model for bending of stranded cables
obtained numerically and experimentally for different tensions taking into account the slip and friction between the layers. They
and consequently for different amounts of sag. The model is obtained a theoretical bending stiffness for the stranded cables
verified to have sufficient fidelity to predict the dynamics of dependent on curvature and tension and verified the bending
cables undergoing both in-plane and out-of-plane motion, and stiffness model, experimentally. Ni et. al. [4] analyzed the
can be used in a pole-cable system model to investigate the effect vibration of large diameter sagged cables and developed a three-
of cable on dynamics on poles or structures. node finite element for dynamic analysis. A case study compared
Keywords: Stranded Cable, Bond Graph Method, Modal the measured and computed natural frequencies, with and
Testing, Bending Stiffness, Numerical Model without considering cable bending stiffness, of the main cables
of the Tsing Ma Bridge. Ignoring the bending stiffness of sagged
cables led to unacceptable errors in predicting the natural
1. INTRODUCTION frequencies. Jalali et. al. [5] developed a 2D numerical model
Dynamic interaction between cables and the main structure based on the bond graph method for vibration analysis of power
in cabled structures has been a topic of interest for researchers in lines. Their model consisted of lumped segments joined with
recent decades. Developing a nondestructive evaluation (NDE) axial and torsional springs and dampers representing the

1
Contact author: [email protected]

1 © 2019 by ASME
compliance and damping of the cable. Pinto et. al. [6, 7]
developed a bond graph model for a pole with a cable attached
to the free end. The cable was modeled as a series of point masses
connected by translational springs and the pole was represented
by a modal expansion based on separation of variables. Modal
parameters of the cable and pole-cable systems were obtained
numerically and experimentally. They concluded that their
model should be expanded and improved from point masses to
rigid bodies with rotational inertia, connected with bending Figure 1: SUCCESSIVE MULTIBODY SEGMENTS
stiffness and damping elements as well as axial ones.
Stiffness values are computed for a segment lengths of 𝑙𝑐 =
The bond graph method is a systematic method that several 𝐿𝑐
simple models can be combined to build a complex model and , where 𝐿𝑐 is the cable length and 𝑛𝑐 is the number of
𝑛𝑐
the dynamic equations of the system are obtained in a segments, [9]:
straightforward manner by investigating energy flow among the 𝐸𝑐 𝐴𝑐 (1)
storage elements [8]. Bond graphs facilitate connection of the 𝐾𝑎𝑥𝑖𝑎𝑙_𝑐𝑎𝑏𝑙𝑒 =
𝑙𝑐
cable model with other subsystem models such as poles.
In this paper, a three-dimensional numerical model is developed 𝐸𝑐 𝐼𝑐 (2)
for a stranded cable similar to the real conductors that are used 𝐾𝑏𝑒𝑛𝑑_𝑐𝑎𝑏𝑙𝑒 =
𝑙𝑐
in electrical transmission lines. The cable is modeled considering
the bending stiffness and sag. The 3D rigid lumped segments are 𝐺𝑐 𝐽𝑐
𝐾𝑡𝑜𝑟_𝑐𝑎𝑏𝑙𝑒 = (3)
connected to each other with translational and rotational springs
𝑙𝑐
and dampers to model the compliance and damping of the cable
and the numerical model is validated with experimental results.
where 𝐸𝑐 is the cable elastic modulus, 𝐴𝑐 is the cross section
A set up was designed for measurement of bending stiffness of
area, 𝐼𝑐 is the cable moment of area, 𝐺𝑐 is modulus of rigidity
the stranded cable under different tensions and measured
and 𝐽𝑐 is the polar moment of area. The damper values are tuned
bending stiffness values were used in the BG model. The tuned
in the model to give a close time response compared to
numerical model can be used to simulate modal testing of power
experimental time responses and 𝐸𝑐 𝐼𝑐 of the cable was measured
lines in in-plane and out-of-plane directions. The following
at various tensions (Section 3.2) and the measured values were
Section describes the bond graph model and theoretical
used in the model.
formulation, Section 3 presents the experimental vibration
testing and cable bending stiffness measurement, and
2.1 Bond Graph of Segments and Joints
verifications and results are presented in Section 4.
In order to create the bond graph model of the system,
dynamic equations of each segment and relative velocity of end
points with respect to the center of gravity must be developed.
2. MULTIBODY BOND GRAPH MODEL DESCRIPTION The following equation relates velocity of end point A of body i
The cable is modeled with 3D rigid lumped segments to the velocity of center of gravity G of body i.
connected with translational and rotational springs and dampers.
In this model, 30 segments are used to model the cable. The 𝑖⃗ ⃗ 𝐺𝑖 + 𝑖 𝑉
⃗ 𝐴𝑖/𝐺𝑖
𝑉𝐴𝑖 = 𝑖 𝑉 (4)
lumped segments are joined with a translational spring/damper
𝑖⃗ 𝑖
to model the axial compliance/damping and three rotational 𝑉𝐴𝑖/𝐺𝑖 = 𝜔 ⃗ 𝑖 × 𝑟𝐴𝑖/𝐺𝑖 = 𝑖𝑟̃𝐴𝑖/𝐺𝑖 𝑖 𝜔
𝑖
⃗𝑖 (5)
springs/dampers to model the torsional behavior and bending
behavior in two directions. Figure 1 shows the segments of the where the left superscript 𝑖 indicates that the vector is presented
cable connected with springs. The dampers are not shown for in a local body-fixed coordinate system, 𝑖 𝑉 ⃗ 𝐴𝑖 is the absolute
clarity. The translational spring corresponds to the axial velocity of point A on body 𝑖 in body-fixed coordinates, 𝑖 𝜔 ⃗ 𝑖 is
compliance, 𝐾𝑎𝑥𝑖𝑎𝑙_𝑐𝑎𝑏𝑙𝑒 . The shear compliance of the cable is the rotational velocity vector, 𝑖 𝑟𝐴𝑖/𝐺𝑖 is the position vector of
not considered in the analyses as it has negligible effect on the point A with respect to G and 𝑖 𝑟̃𝐴𝑖/𝐺𝑖 is the skew-symmetric
dynamics of the cable [2]. Rotary springs 𝐾𝑏𝑒𝑛𝑑_𝑐𝑎𝑏𝑙𝑒 are the matrix containing the relative position vector components. The
bending compliance of the cable about the local body-fixed y and equations below describe Newton’s law in three dimensions for
z directions which are the same because of the axial symmetry of the segment in local coordinates [9]:
the cable and 𝐾𝑡𝑜𝑟_𝑐𝑎𝑏𝑙𝑒 is the torsional compliance of the cable
about the body-fixed x direction. In Figure 1, the bending rotary d
⃗ =
∑ oF ⃗ Gi ) = 𝑚 o𝑉
(𝑚 o𝑉 ̇
⃗ Gi (6)
springs about the z direction are not shown for clarity and the
dt
springs are along the body-fixed coordinates of the body i+1. 𝑑
∑ 𝑖𝜏 = 𝑖𝐻 ⃗ ̇ 𝑖 + 𝑖𝜔
⃗ = 𝐽𝑖 𝜔 ⃗ 𝑖 × 𝐽 𝑖𝜔
⃗𝑖 (7)
𝑑𝑡

2 © 2019 by ASME
where left superscript o indicates that the vector is defined in the system using a pair of MTF’s and then transformed to the body
inertial coordinate system. The first term in the right side of i+1 local coordinate system using another pair of MTF’s. The C
rotational Eq. (7) is the rotational inertia term and the second multiport elements model the stiffness described in Eq’s. (1)-(3)
term is a gyrational term. To be able to add forces such as gravity using a diagonal stiffness matrix. The R elements define the
and equate values in different reference frames, we need to be damping using a damping matrix. The damping values are tuned
able to transform values from the body fixed frame to the inertial to give an accurate time response and natural frequencies
frame. This is accomplished with a rotation matrix. Rotation compared to experimental results.
matrices are made for each of the three rotations 𝜃, 𝜓, 𝜙 about x,
y and z axes, respectively and are multiplied together to create
the transformation matrix.
Vector bond graphs are used to represent the equations. Figure 2
shows the vector bond graph sub model representing the body
segment i. In Figure 2, the 1-junctions are labeled with the
corresponding translational or rotational velocities.
Translational velocity Eq. (4) for defining the velocity of left end
point A of the segment and the similar equation for the velocity
of the right end point B of the segment with respect to center of
mass gravity in the body-fixed coordinate are represented by two
modulated transformer’s (MTF) and two 0-junctions in Figure 2.
⃗ 𝐺𝑖 to 𝑖 𝑟̃𝐴𝑖/𝐺𝑖 𝑖 𝜔 ⃗ 𝐴𝑖 and a FIGURE 3: JOINT i SUBMODEL BOND GRAPH
A 0-junction is used to add 𝑖𝑉 ⃗ 𝑖 to obtain 𝑖 𝑉
𝑖⃗ 𝑖 𝑖 𝑖⃗
0-junction is used to add 𝑉𝐺𝑖 to 𝑟̃𝐵𝑖/𝐺𝑖 𝜔 ⃗ 𝑖 to obtain 𝑉𝐵𝑖 . The two
MTF’s representing relative velocity cross products receive the 3. EXPERIMENTAL METHODS
relative position vector as an input signal. The coordinate
transformation between body-fixed and inertial frames is 3.1 Experimental Modal Testing
represented by an MTF that multiplies 𝑖 𝑉 ⃗ 𝐺𝑖 by the transformation In order to verify the numerical model, experimental
matrix to obtain inertial center of gravity velocity 𝑜𝑉 ⃗ 𝐺𝑖 . vibration tests were carried out. Figure 2 presents the schematic
Translational force Eq. (6) is represented by a mass matrix of the vibration tests. Two rigid frames were secured to the floor
multiport I element and the gravity effort source is connected to on the right and left sides. The right end of the cable was attached
⃗ 𝐺𝑖 1-junction ( 𝑖 𝑉
⃗ 𝐺𝑖 transformed to the inertial frame). The to a threaded rod to adjust the tension of the cable and an in-line
the 𝑜𝑉 load cell at the left end of the cable measured the longitudinal
right hand side of Eq. (7) is represented by a rotational inertia I tension of the cable. Tests were performed with a Bruel & Kjaer
element, a modulated gyrator MGY, external moments from 8205-002 impact hammer, 4507 B 004 70-g accelerometers, and
joints i and i-1 and moments from forces at end points A and B, a National Instruments NI USB-4432 power supply and signal
connected to the 𝑖 𝜔 ⃗ 𝑖 1-junction. The MGY is used to represent conditioner.
the cross product of 𝑖 𝜔 ⃗ 𝑖 × 𝐽 𝑖𝜔⃗ 𝑖 with 𝐽 𝑖𝜔 ⃗ 𝑖 as an input signal.

FIGURE 2: BODY i BOND GRAPH

FIGURE 4: SCHEMATIC FOR EXPERIMENTAL VIBRATION
TESTING a) MODAL IMPACT TEST b) FREE VIBRATION TEST
Figure 3 shows the bond graph sub model of joint i between
successive segments. The 0-junctions represent the translational Two types of vibration tests were carried out. In Figure 4-a,
and angular velocity constraints that are caused by parallel the modal impact test is shown and Figure 4-b shows the free
spring/dampers between the left end point A of body i+1 and vibration test. The modal tests were carried out with five
right end point B of body i (Figure 1). The velocity of B on body different tensions 25N, 50N, 100N, 200N and 300N in the cable.
⃗ 𝐵𝑖 and 𝑖 𝜔
i, 𝑖 𝑉 ⃗ 𝑖 are first transformed to the inertial coordinate

3 © 2019 by ASME
The amount of sag of the cable changes depending on the tension TABLE 1: MATERIAL PROPERTIES OF CABLE
in the cable. In place of the accelerometer in Figure 4, another
accelerometer was used in the out-of-plane direction 𝑘𝑔
𝐸𝑐 (𝐺𝑃𝑎) 𝐴𝑐 (𝑚𝑚2 ) 𝑚𝑐 ( ⁄𝑚)
(perpendicular to the plane of Figure 4), to measure the out-of-
plane vibration due to hammer impact load. The hammer hits
were made at the quarter span length of the cable and the 190 31.66 0.166
accelerometer was placed at mid span. Each test was performed
five times and the average result was considered for the analysis.
The RMS averaging method with exponential weighting was
used to average the results and an exponential window was 3.2 Bending Stiffness Measurement
applied with duty cycle of 10 percent. In the free vibration test The bending stiffness of stranded cables is bounded by two
(Figure 4-b), a 1 kg weight was hung from the quarter span of extreme flexural stiffnesses 𝐸𝐼𝑚𝑎𝑥 and 𝐸𝐼𝑚𝑖𝑛 [10]. The
the cable and by cutting the string attached to the weight, an maximum bending stiffness 𝐸𝐼𝑚𝑎𝑥 corresponds to a solid cross
initial displacement was created to cause free vibration of the section with no slippage of the strands: [11]
cable. The in-plane vibration response at mid-span was measured
by the accelerometer. Figure 5 shows the setup used and the 𝑁𝑠
𝜋𝑑𝑠𝑖 4 𝜋𝑑𝑠𝑖 2 2 𝜋𝑑𝑐 4
material properties of the cable are presented in Table 1. The 𝐸𝐼𝑚𝑎𝑥 = 𝐸 ∑ ( + 𝑦𝑠𝑖 ) ≈ 𝐸 (8)
cable (7x19 stranded cable) was supported with a pin connection 64 4 64
𝑖=1
at both ends and the span length between the rigid frames was
2.4 m. where 𝑑𝑠𝑖 is the diameter of the ith strand, 𝑦𝑠𝑖 is the distance from
the center of gravity of the ith strand to the center of gravity of
the conductor, 𝑑𝑐 is the diameter of the cable and 𝑁𝑠 is the
number of strands in the cable.
The minimum bending stiffness 𝐸𝐼𝑚𝑖𝑛 corresponds to all the
strands of the cable slipping past each other as static friction
between the strands is overcome. Therefore, the bending
stiffness in this condition is the sum of bending stiffness of all
the strands:
𝑁𝑠
𝜋𝑑𝑠𝑖 4
𝐸𝐼𝑚𝑖𝑛 = 𝐸 ∑ (9)
64
𝑖=1

Figure 6 shows the test apparatus for measuring the bending

stiffness of the cable. A hydraulic cylinder is secured to the
frame at the top of the setup to apply the load at the middle of
the cable, and an LDVT (Linear Variable Differential
Transformer) is attached to the cylinder to measure the deflection
of the cable at midspan at the same time. The inline load cell
was used to measure cable tension, which was adjusted with a
threaded rod. An S-type load cell was also attached to the
cylinder to measure the applied load. The cable was secured with
clamped-clamped end conditions and the theoretical procedure
used to measure the bending stiffness follows [12]. The bending
stiffness of the stranded cables increased with tension in [10, 12]
and the results obtained in the present study confirm this trend.
The bending tests were performed at the same five tensions as in
the modal testing. The measured bending stiffness values are
presented in Table 2 and the theoretical limits of bending
stiffness are 𝐸𝐼𝑚𝑖𝑛 = 0.0306 and 𝐸𝐼𝑚𝑎𝑥 = 15.5632,
respectively.
FIGURE 5: CABLE VIBRATION TEST a) TEST SET UP

b) INLINE LOAD CELL END c) THREADED ROD

4 © 2019 by ASME
 were the modes that could not be captured numerically or
experimentally.

TABLE 3: IN-PLANE NATURAL FREQUENCY COMPARISON OF

THE CABLE

Modal Test Error

Tension (N) Simulation [Hz]
[Hz] (%)
5.26 6.23 18.44
25 8.03 9.20 14.57
13.19 13.43 1.82
7.63 7.85 2.88
11.73 11.41 2.73
50
16.17 16.50 2.04
23.17 21.15 8.72
6.33 - -
100 11.29 11.02 2.36
16.94 13.71 19.04
7.81 7.72 1.21
-- 9.37 --
200
15.53 15.75 1.44
23.06 23.96 3.94
9.08 9.68 6.61
300 17.60 16.99 3.46
FIGURE 6: EXPERIMENTAL SET UP FOR CABLE BENDING 25.91 26.00 0.35
STIFFNESS MEASUREMENT

TABLE 2: MEASURED BENDING STIFFNESS OF CABLE WITH

DIFFERENT TENSIONS TABLE 4: OUT-OF-PLANE NATURAL FREQUENCY
COMPARISON OF THE CABLE
Tension (N) 25 50 100 200 300
Modal Test Error
𝐸𝐼 (𝑁𝑚2 ) 0.2549 0.3420 0.4948 0.5972 0.8283 Tension (N) Simulation [Hz]
[Hz] (%)
2.94 3.06 4.08
25 5.97 6.28 5.19
4. RESULTS AND DISCUSSION 9.19 9.75 6.09
Two vibration tests (Figure 4-a and Figure 4-b) were simulated 3.87 3.88 0.26
in 20sim® software using bond graph method. Modal impact
7.92 7.90 0.25
testing (Figure 4-a) was simulated and the frequency domain 50
11.84 12.14 2.53
toolbox in the software was used to generate transfer functions
18.57 16.65 10.34
from which the eigenvalues can be extracted. In the simulation,
an impulse force is applied at the quarter-span cable location and 4.51 5.27 16.84
the time response is obtained at mid-span of the cable similar to 100 9.77 11.20 14.60
the experimental test procedure (Figure 4). ModalView [13] 13.90 11.22 19.27
software was used to analyze the experimental data and a 6.82 6.86 0.53
stability diagram based on the frequency response functions was 200 8.11 9.50 17.10
used to extract the natural frequencies with a sampling rate of 14.22 14.06 1.14
1024 Hz. Figure 7 shows an experimental FRF which was 8.64 8.32 3.71
obtained from in-plane vibration of the cable with 300 N tension. 300 17.15 16.98 0.99
Table 3 presents the comparison of in-plane natural frequencies 24.62 24.85 1.34
of the cable with different tensions and Table 4 compares the out-
of-plane natural frequencies of the cable. In order to excite and
capture the in-plane modes, the hammer hit was performed in the
plane of cable in Figure 4 and for out-of-plane modes, the cable
was excited in the plane perpendicular to the plane of the cable.
Some of the natural frequencies are missing in the tables and they

5 © 2019 by ASME
 accurately measure the bending stiffness of the stranded cable
with different tensions and the measured stiffness values were
used in the dynamic model. The bending stiffness was seen to
increase with the increase in tension. The developed dynamic
model is concluded to be a reliable and accurate model to be used
for cable dynamic analysis in in-plane and out-of-plane
directions.

REFERENCES
1. Ricciardi, G. and F. Saitta, A continuous vibration
analysis model for cables with sag and bending
stiffness. Engineering Structures, 2008. 30(5): p. 1459-
1472.
FIGURE 7: EXPERIMENTAL FRF: IN-PLANE EXCITATION 2. Sousa, R.A., et al., The influence of bending and shear
WITH 300 N TESNION stiffness and rotational inertia in vibrations of cables:
An analytical approach. Engineering Structures, 2011.
As can be seen from Table 3 and Table 4, there is typically good 33(3): p. 689-695.
agreement between the numerical bond graph frequency domain 3. Papailiou, K.O., On the bending stiffness of
results (in-plane and out-of-plane) and the experimental results. transmission line conductors. IEEE Transactions on
The free vibration test (Figure 4-b) was also simulated in the Power Delivery, 1997. 12(4): p. 1576-1588.
numerical model and the acceleration in-plane time response at 4. Ni, Y.Q., J.M. Ko, and G. Zheng, Dynamic analysis of
the mid-span obtained from experiment and simulation are large-diameter sagged cables taking into account
compared in Figure 8. As can be seen, there is a good agreement flexural rigidity Journal of Sound and Vibration, 2002.
between the numerical and experimental time responses and the 257(2): p. 301-319.
numerical model is able to replicate the decay rate, vibration 5. Jalali, M.H. and G. Rideout. Development and
amplitude and frequency content of the experimental time Validation of a Numerical Model for Vibration of Power
response. Lines. in Proceedings of The Canadian Society for
Mechanical Engineering International Congress 2018.
2018. Toronto.
6. Ferreira Pinto, P. and D. Rideout. Development and
Validation of an In-Situ Utility Pole Simulation Model
for Virtual Modal Testing. 2016. ASME 2016
International Mechanical Engineering Congress and
Exposition.
7. Pinto, P., Experimental and numerical investigation of
the influence of cables on utility pole modal property
extraction. 2017, Memorial University.
8. Uddin, N. and J.T. Gravdahl, Bond graph modeling of
centrifugal compression systems. SIMULATION,
FIGURE 8: TIME RESPONSE COMPARISON OF SIMULATION 2015. 91(11): p. 998-1013.
AND EXPERIMENT 9. Karnopp, D.C., D.L. Margolis, and R.C. Rosenberg,
System dynamics: modeling, simulation, and control of
The time domain and frequency domain verification of the mechatronic systems. 2012: John Wiley & Sons.
numerical model shows that the developed model is an accurate 10. Filiatrault, A. and C. Stearns, Flexural Properties of
and reliable model to capture dynamics of stranded cable and can Flexible Conductors Interconnecting Electrical
be used in connection with the pole(s) model for cable-beam Substation Equipment. Journal of Structural
system dynamic investigations. Engineering, 2005. 131(1): p. 151-159.
11. Jalali, M.H. and G. Rideout, Analytical and
5. CONCLUSION experimental investigation of cable–beam system
A three-dimensional dynamic model for a stranded cable was dynamics. Journal of Vibration and Control: DOI:
developed and validated with experiments. The lumped segment 10.1177/1077546319867171.
method was used to model the system and axial, bending and 12. McConnell, K.G. and W.P. Zemke, The measurement of
torsional behavior of cable was considered. In comparison to flexural stiffness of multistranded electrical conductors
analytical methods or finite element analysis, the bond graph while under tension. Experimental Mechanics, 1980.
formalism is a very computationally efficient and 20(6): p. 198-204.
straightforward method for connecting different sub models of 13. ABSignal, ModalView. 2008, National Instrument.
cable and pole to each other. A set up was designed and used to

6 © 2019 by ASME

View publication stats