http://www.press.ierek.

com

ISSN (Print: 2537-0731, online: 2537-074X)

Proceedings of Science and Technology

Research Paper
Received: 1 January 2024, Accepted: 14 July 2024, Published online: 30 September 2024
DOI: 10.21625/resourceedings.v4i2.1058

Effect of Nonlinear Magnetic Forces on Transverse Galloping

Dynamics of Square Cylinders
Mostafa R. Rashed1, Mostafa E.A. Elsayed2, M.A. Abdelrahman3, Mahmoud Shaaban4,5
1
Assistant Professor, Mechanical Engineering Department, Faculty of Engineering, Modern University for
Technology and Information, Cairo, Egypt
2
Associate Professor, Automation and Energy Technology Lab, Mechanical Engineering Department, Faculty of
Engineering at Shoubra, Benha University, Cairo, Egypt
3
Associate Professor, Combustion and Energy Technology Lab, Mechanical Engineering Department, Faculty of
Engineering at Shoubra, Benha University, Cairo, Egypt
4
Assistant Professor, Mechanical Engineering Program, School of Engineering and Applied Sciences, Nile
University, Giza, Egypt
5
Assistant Professor, Smart Engineering Systems Center, Nile University, Giza, Egypt

Abstract
Under the influence of cross-fluid flow, a cylinder of a square cross-section may gallop. Galloping is a self-excited
vibration mode that can be utilized for low-power harvesting applications. The harvested power depends on several
factors, including upstream flow velocity and system dynamics. This study explores the potential of magnetically-
induced nonlinear stiffness to improve the power output of galloping-based energy harvesters. In this experimental
study, the vibration response of a square rod with a mass ratio of 10 is investigated at a Reynolds number of 200. The
vibration behavior of two identical coaxial square rods with magnetic monopoles at opposite ends is analyzed. Results
reveal that the magnets’ configuration and strength significantly affect vibration amplitude and the critical flow
velocity necessary for the onset of galloping.

© 2024 The Authors. Published by IEREK Press. This is an open-access article under the CC BY license
(https://creativecommons.org/licenses/by/4.0/).

Keywords
Galloping; Square cylinder; Energy harvesting; Nonlinear stiffness; CFD; Fluid-structure interaction; Vibration

1. Introduction
Flow-induced vibrations, including vortex-induced vibration, galloping, and flutter, can affect slender and flexible
structures exposed to wind or water flows. These instabilities have been observed in a wide range of engineering
structures, such as bridges (Zhang et al., 2020), tall buildings (Kawai, 1995), power transmission lines (Song et al.,
2020), heat exchanger tubes (Paı¨doussis, 2006; Shaaban & Mohany, 2021), and ducts with cavities (Abdelmwgoud
et al., 2021). The detrimental impact of these vibrations on structural fatigue life and safety is widely recognized
(Song et al., 2021). On the positive side, there is an increasing focus on harnessing the energy from flow-induced
vibrations as a sustainable and clean energy source (Lv et al., 2021; Ma & Zhou, 2022; Rostami & Armandei, 2017;
Wang et al., 2020).

Galloping refers to a type of system instability characterized by significant oscillations in a structure when the speed
of the incoming fluid increases. It occurs as a result of aeroelastic instability, wherein small oscillations initiate fluid

pg. 1
 Rashed/ Proceedings of Science and Technology

forces that gradually overpower the inherent damping of the structure. Notably, long and slender structures with non-
circular cross-sections are susceptible to experiencing low-frequency, quasi-periodic motion when the flow velocity
surpasses a certain threshold. If the fluid flows perpendicular to the structure’s axis, it will gallop in the transverse
direction. As the oscillations progress, the amplitude of displacement increases but eventually stabilizes due to a
balance between the fluid damping forces and the reduced system damping, leading to the completion of a cycle.

Extensive research has been dedicated to understanding galloping in square-shaped structures (Mannini et al., 2014).
Measurements by Nakamura and Mizota (Nakamura & Mizota, 1975) showed sudden changes in frequency and phase
angles when critical flow velocities for vortex excitation and galloping were reached. Bearman and Obasaju (Bearman
& Obasaju, 1982) found notable differences in force coefficients and phase characteristics between stationary square
and circular cylinders. The behavior of the oscillating system is influenced by non-dimensional parameters, including
the Reynolds number and reduced flow velocity. Transverse vibration response in square cylinders can be assessed
using the root mean square displacement 𝑦𝑟𝑚𝑠 = 𝑟𝑚𝑠(𝑌)/𝐿 and the Strouhal number 𝑆𝑡 = 𝑓𝐿/𝑈 (Shaaban &
Mohany, 2022). Galloping amplitude increases almost linearly with reduced flow velocity Ur = U/(fnL) after
surpassing a specific threshold. (Rashed et al., 2024) observed that nonlinear forces do not have a significant impact
on the vibration frequency but can impact the vibration amplitude. Galloping initiation is typically associated with a
minimum displaced mass ratio 𝑚∗ = 𝑚/𝜌𝐿2 , and the onset Reynolds number weakly depends on the mass ratio (Sen
& Mittal, 2011, 2016).

Accurately modeling and predicting the performance of vibrating energy harvesters with nonlinear forces is a
persistent challenge (Ma & Zhou, 2022). Nonlinear devices often face chaos and bifurcation, necessitating an
exploration of the impact of nonlinearity on performance and stability (Huynh & Tjahjowidodo, 2017). (Shaaban &
Mohany, 2021) successfully used coupled modeling to predict forces exerted by airflow due to vortex shedding.
Existing literature suggests that nonlinear forces generally improve power output and stability in energy harvesters,
but the underlying mechanisms and effects require further investigation. Understanding the influence of nonlinear
forces on galloping energy harvesters can facilitate amplitude control and suppression in unfavorable vibration
conditions.

This study aims to examine the influence of a nonlinear magnetic force on the performance of a galloping energy
harvester. To investigate the dynamic behavior, a numerical flow simulation based on the incompressible Navier-
Stokes equation, coupled with a double-structure single-degree-of-freedom solver, is employed. The harvester design
consists of two galloping objects coupled through a magnetic repulsion force. A configuration of the nonlinear
magnetic monopoles is explored. The strength of the magnetic field and repulsion force is varied to assess its impact
on the galloping response and dynamic stability of the system.

2. Methodology

2.1. Numerical Model

Figure 1 displays a schematic of the configuration under investigation. It consists of two identical rods attached to
springs, allowing movement in the transverse direction only. The rods are separated by a distance equal to their edge
length. For the reference case without magnetic forces, the rods have a mass ratio 𝑚∗ = 10 ensuring a typical
galloping response. The flow is laminar at a constant Reynolds number 𝑅𝑒 = 200, eliminating turbulence’s influence
and facilitating comparison with other studies (Joly et al., 2012). The system’s natural frequency is controlled by
varying the structural stiffness, while two magnetic poles at the ends produce repulsive forces as the rods oscillate.

pg. 2
 Rashed/ Proceedings of Science and Technology

(a) (b)

Figure 1: (a) A schematic of the investigated configuration and (b) the parameters of the investigated configuration.

Figure 2 displays the computational domain used in the simulation, with the same extensions used by Li et al. (Li et
al., 2019) in the upstream, downstream, and width directions. Two quadrilateral grid zones, each with 85,000
elements, were employed to calculate the flow around the square cross sections independently. Flow field calculations
are updated in both zones to accurately determine the phase between their motion, which is influenced solely by the
transverse component of the repulsive magnetic force. The grid is orthogonal, and the square walls are resolved by
200 elements. Boundary conditions from Figure 2 are applied to both zones, with varying values for the location and
velocity of the square cross-section wall based on the previous time step. Inlet and side conditions include a fixed
flow velocity and zero pressure gradient. The outlet condition specifies a reference pressure that doesn’t affect the
incompressible solution. The square section walls have a no-slip condition. Pressure-velocity coupling employs a
pressure implicit scheme with operator splitting. Second-order upwind discretization is used for convective terms,
while a least-squares scheme calculates gradients. Time derivatives are resolved using a bounded second-order
scheme.

Figure 2: The utilized computational domain and the boundary conditions.

2.2. Governing Equations

The transverse displacement 𝑌 of a two-dimensional square cylinder, which is supported elastically and exposed to
both flow and magnetic forces, can be described by a straightforward oscillator equation. The equation considers
parameters such as mass 𝑚, damping 𝑐, stiffness 𝑘, and the influence of pressure 𝑃 on displacement.

&+ c Y&+ kY =
m Y& (  PdA + F )  jm (1)

The transient incompressible Navier-Stokes equations are solved to obtain the flow field and calculate the overall
fluid pressure force acting on the cylinders.

pg. 3
 Rashed/ Proceedings of Science and Technology

  ul = 0 , (2)

ul
t
( )
= − ul   ul − p +
1 2
Re
 ul , (3)

where the Lagrangian flow velocity 𝐮𝑙 = 𝐮 − 𝐮𝑚𝑒𝑠ℎ . The absolute velocity vector 𝐮 is normalized by the upstream
flow velocity 𝑈 . The mesh velocity 𝐮𝑚𝑒𝑠ℎ is determined based on the velocity of the solid structure’s boundary,
computed using equation (1). 𝑝 = 𝑃/𝜌𝑈 2 is the normalized pressure. The Reynolds number in a fluid of viscosity 𝜇
is 𝑅𝑒 = 𝜌𝑈∞ 𝐷/𝜇. The formulation for body forces in the flow and structure domains does not incorporate gravity.
The cylinder edge length 𝐿 serves as the reference length scale, while the upstream flow velocity acts as the reference
flow velocity 𝑈.

On the other hand, the magnetic repulsion force is determined by considering a simplified configuration that
represents the ideal scenario of a pair of monopoles with the same charge sign. The repulsion force between two
monopoles of equal strength 𝑞 can be calculated as follows:

q2
Fm = qB = ˆ
r, (4)
40r2

where 𝒓̂ is the direction vector between the two poles, 𝑟 = √𝐿2 − (𝑌2 − 𝑌1 )2 is its magnitude and 𝜇0 is the
permeability. Setting 𝐶𝑏 = 𝑞2 /4𝜋𝜇0 𝐿3 as a constant that represents the relative strength of the pole pair, the cross-
flow force component is dependent on the location of the two squares. Equation 5 shows that the force induced by
the monopoles is a nonlinear stiffness force.

Y 2 −Y1
Fm ,y = Fm  j= C B 32
  Y − Y 2  (5)
1 +  2 1
 
  L  
 

The solver inputs the previous flow field data and cross-section locations/velocities. An iterative flow solver (outlined
in Figure 3) determines the pressure field and provides updated flow data. Using their locations, the structural solver
calculates the transverse magnetic forces on the cross-sections. It then updates their velocity and location. This process
repeats until a steady oscillating behavior is achieved, allowing spectral domain calculations.

Figure 3: The solver algorithm.

pg. 4
 Rashed/ Proceedings of Science and Technology

2.3. Model Validation

In order to ensure the computational model reliability, a grid sensitivity analysis was conducted employing three grid
configurations G1, G2, and G3, listed in Table 1. The study showed that the grid configuration G2 provided
satisfactory results.

Table 1: Grid sensitivity analysis for a stationary square cylinder at 𝑅𝑒 = 200.

Grid Number of Elements 𝐶𝑑

G1 27,500 1.53

G2 85,000 1.47

G3 192,000 1.46

The numerical model’s results are compared and validated against existing data for square cylinders carried out by Li
et al. (Li et al., 2019) at Reynolds number 𝑅𝑒 = 150.

Figure 4 demonstrates that the numerical model employed during the present study effectively captures the oscillation
amplitude and critical flow velocity for all flow-induced vibration mechanisms.

Figure 4: The galloping response of a square rod cylinder at 𝑅𝑒 = 150, compared with the results of (Li et al., 2019).

3. Results and Discussions

In this section, the galloping response of the two bluff bodies with square cross sections is analyzed to determine the
influence of the nonlinear stiffness added by the magnetic repulsion force. As seen in Figure 5, the effect of the
reduced velocity on the galloping response of the bluff bodies is dependent on the strength factor 𝐶𝐵 , correlated to
the relative strength of the nonlinear stiffness to the linear spring stiffness. The impact of reduced flow velocity on
the flow-induced vibration response is shown for the three considered cases: a linear spring system, a system with a
pair of weak magnetic monopoles, and a system with a pair of stronger magnetic monopoles.

For the linear spring case 𝐶𝐵 = 0, vortex-induced vibration results in significant vibration amplitude reaching up to
𝑦𝑟𝑚𝑠 = 0.25, while the galloping response starts after the flow velocity exceeds 𝑈𝑟 = 10. As the flow velocity
increases beyond that limit, the amplitude of vibration increases, reaching twice the vortex-induced value at 𝑈𝑟 = 20.
The influence of non-linear stiffness can be observed in Figure 5. For the vortex-induced vibration, the repulsion
force induced by the magnetic monopoles resulted in a decrease in amplitude at 𝑈𝑟 = 5. It is, however, unclear if the
amplitude decreased as the resonance frequency shifted to a slightly different value, or if the stiffness change resulted
in an overall decrease in the vibration amplitude for vortex-induced vibration. On the other hand, the nonlinear
pg. 5
 Rashed/ Proceedings of Science and Technology

repulsion forces caused by magnetic monopoles resulted in two effects in the galloping flow velocity range. First, an
enhancement of the galloping amplitude is observed such that at 𝑈𝑟 = 20, the amplitude reaches 𝑦𝑟𝑚𝑠 = 0.55, and
𝑦𝑟𝑚𝑠 = 0.8 for 𝐶𝐵 = 0.1 and 𝐶𝐵 = 1.0; respectively, increasing from 0.5 for the linear system. This enhancement can
be attributed to the linear contribution of the stiffness increased by the magnetic force at positions close to the original
mass location.

The second effect of the nonlinear stiffness is a decrease in the onset velocity required for galloping. The reduced
flow velocity at the onset of galloping decreased from 𝑈𝑟 = 10 for the linear system, to about 𝑈𝑟 = 8 for 𝐶𝐵 = 1.0.
This decrease is attributed to the overall decrease of stiffness caused by the nonlinear force, which results in the
galloping range shift to lower reduced velocities, which is equivalent to a reduction in wind speed requirements for
galloping energy harvesting.

Figure 5: Impact of reduced velocity 𝑈𝑟 on the response of flow-induced vibration characteristics under various levels of repulsive force
strength.

Figure 6 compares the vibration-time signal of the two structures under the presence of magnetic monopoles with
different strengths before and after the onset of galloping, as illustrated in Figure 5). Under galloping conditions,
magnetic monopoles with a strength of 𝐶𝐵 = 0.1 result in slightly higher motion amplitude than in situations without
repulsive forces. This effect becomes more noticeable when monopoles of higher strength, 𝐶𝐵 = 1.0, are introduced,
leading to a significant increase in the root mean square of motion, particularly at 𝑈𝑟 = 12. In addition, Figure 6
shows the signal characteristics in the time domain, revealing that while the amplitude of the motion is the same for
the two objects in all cases, the relative phase between them is affected by the strength of the magnetic repulsion
force. As expected, the linear system has the two objects moving at an arbitrary phase as no coupling between them
is present. In contrast, the two cases with a magnetic repulsion force exhibit a phase of 180°, indicating an out-of-
phase behavior. This is to be expected, as this phase results in oscillation between the highest and the least potential
energy states in the magnetic field. As such, this coupling can be used to control the behavior of the bluff bodies and
tie their timing which may be beneficial to their mechanical characteristics in devices designed for energy harvesting
from fluid flow.

pg. 6
 Rashed/ Proceedings of Science and Technology

Figure 6: Vibration-time signal comparison of the two structures with and without nonlinear monopole magnetic effect.

4. Conclusions
The galloping response of a bluff body with a square cross-section depends on the stiffness characteristics of the
system. In this study, the impact of a nonlinear magnetically-induced force on the vibration response of two coaxial
rods with square cross-sections and identical structural parameters is numerically investigated. To calculate the
galloping response of two bluff bodies interacting through a magnetic repulsion force, a fully-coupled finite element
model is used to calculate the forces affecting the two bodies from effects including fluid flow, magnetic repulsion,
and structural forces. The galloping response is investigated at 𝑅𝑒 = 200 and 𝑚∗ = 10 for a linear system and
compared with the case of two levels of magnetic field strength.

Results showed that a nonlinear repulsion force between magnetic monopoles enhances vibration amplitude and
reduces onset flow velocity requirements. Evidently, the increase in vibration amplitude is attributed to a substantial
negative stiffness generated by the magnets' repulsion. On the other hand, the magnetic force strength has a minimal
impact on galloping vibration frequency, which is primarily determined by amplitude. This vibration amplitude
enhancement is significant for powerful magnetic forces and can potentially enhance the performance of galloping-
based flow energy harvesting devices. In addition, the magnetic repulsion force results in a coupling such that the
phase between the motion of the two objects is 180° out of phase.

Acknowledgment
The abstract of this paper was presented at the Environmental Design, Material Science, and Engineering
Technologies (EDMSET) Conference – 1st Edition which was held on the 22nd-24th of April 2024.

Funding declaration
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit
sectors/individuals.

Ethics approval
Not applicable.

Conflict of interest
The authors declare that there is no competing interest.

References
Abdelmwgoud, M., Shaaban, M., & Mohany, A. (2021). Shear layer synchronization of aerodynamically isolated opposite cavities due to acoustic
resonance excitation. Physics of Fluids, 33(5). https://doi.org/10.1063/5.0051226
Bearman, P. W., & Obasaju, E. D. (1982). An experimental study of pressure fluctuations on fixed and oscillating square-section cylinders. Journal
of Fluid Mechanics, 119, 297–321. https://doi.org/10.1017/S0022112082001360
pg. 7
 Rashed/ Proceedings of Science and Technology

Huynh, B. H., & Tjahjowidodo, T. (2017). Experimental chaotic quantification in bistable vortex induced vibration systems. Mechanical Systems
and Signal Processing, 85, 1005–1019. https://doi.org/10.1016/j.ymssp.2016.09.025
Joly, A., Etienne, S., & Pelletier, D. (2012). Galloping of square cylinders in cross-flow at low Reynolds numbers. Journal of Fluids and Structures,
28, 232–243. https://doi.org/10.1016/j.jfluidstructs.2011.12.004
Kawai, H. (1995). Effects of angle of attack on vortex induced vibration and galloping of tall buildings in smooth and turbulent boundary layer
flows. Journal of Wind Engineering and Industrial Aerodynamics, 54–55, 125–132. https://doi.org/10.1016/0167-6105(94)00035-C
Li, X., Lyu, Z., Kou, J., & Zhang, W. (2019). Mode competition in galloping of a square cylinder at low Reynolds number. Journal of Fluid
Mechanics, 867, 516–555. https://doi.org/10.1017/jfm.2019.160
Lv, Y., Sun, L., Bernitsas, M. M., & Sun, H. (2021). A comprehensive review of nonlinear oscillators in hydrokinetic energy harnessing using
flow-induced vibrations. Renewable and Sustainable Energy Reviews, 150, 111388. https://doi.org/10.1016/j.rser.2021.111388
Ma, X., & Zhou, S. (2022). A review of flow-induced vibration energy harvesters. Energy Conversion and Management, 254, 115223.
https://doi.org/10.1016/j.enconman.2022.115223
Mannini, C., Marra, A. M., & Bartoli, G. (2014). VIV–galloping instability of rectangular cylinders: Review and new experiments. Journal of
Wind Engineering and Industrial Aerodynamics, 132, 109–124. https://doi.org/10.1016/j.jweia.2014.06.021
Nakamura, Y., & Mizota, T. (1975). Unsteady Lifts and Wakes of Oscillating Rectangular Prisms. Journal of the Engineering Mechanics Division,
101(6), 855–871. https://doi.org/10.1061/JMCEA3.0002077
Paı¨doussis, M. P. (2006). Real-life experiences with flow-induced vibration. Journal of Fluids and Structures, 22(6–7), 741–755.
https://doi.org/10.1016/j.jfluidstructs.2006.04.002
Rashed, M. R., Elsayed, M. E. A., & Shaaban, M. (2024). Influence of magnetically-induced nonlinear added stiffness on the lift galloping of
square cylinders at low Reynolds number. Journal of Fluids and Structures, 124, 104046. https://doi.org/10.1016/j.jfluidstructs.2023.104046
Rostami, A. B., & Armandei, M. (2017). Renewable energy harvesting by vortex-induced motions: Review and benchmarking of technologies.
Renewable and Sustainable Energy Reviews, 70, 193–214. https://doi.org/10.1016/j.rser.2016.11.202
Sen, S., & Mittal, S. (2011). Free vibration of a square cylinder at low Reynolds numbers. Journal of Fluids and Structures, 27(5–6), 875–884.
https://doi.org/10.1016/j.jfluidstructs.2011.03.006
Sen, S., & Mittal, S. (2016). Free Vibrations of a Square Cylinder of Varying Mass Ratios. Procedia Engineering, 144, 34–42.
https://doi.org/10.1016/j.proeng.2016.05.004
Shaaban, M., & Mohany, A. (2021). Synchronous vortex shedding from aerodynamically isolated side-by-side cylinders imposed by flow-excited
resonant acoustic modes. Experiments in Fluids, 62(10), 205. https://doi.org/10.1007/s00348-021-03301-9
Shaaban, M., & Mohany, A. (2022). Flow–acoustic coupling around rectangular rods of different aspect ratios and incidence angles. Experiments
in Fluids, 63(2), 45. https://doi.org/10.1007/s00348-022-03380-2
Song, Y., Liu, Z., Rxnnquist, A., Navik, P., & Liu, Z. (2020). Contact Wire Irregularity Stochastics and Effect on High-speed Railway Pantograph-
Catenary Interactions. IEEE Transactions on Instrumentation and Measurement, 1–1. https://doi.org/10.1109/TIM.2020.2987457
Song, Y., Wang, Z., Liu, Z., & Wang, R. (2021). A spatial coupling model to study dynamic performance of pantograph-catenary with vehicle-
track excitation. Mechanical Systems and Signal Processing, 151, 107336. https://doi.org/10.1016/j.ymssp.2020.107336
Wang, J., Geng, L., Ding, L., Zhu, H., & Yurchenko, D. (2020). The state-of-the-art review on energy harvesting from flow-induced vibrations.
Applied Energy, 267, 114902. https://doi.org/10.1016/j.apenergy.2020.114902
Zhang, M., Xu, F., & Han, Y. (2020). Assessment of wind-induced nonlinear post-critical performance of bridge decks. Journal of Wind
Engineering and Industrial Aerodynamics, 203, 104251. https://doi.org/10.1016/j.jweia.2020.104251

pg. 8