U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000


Skip to content
Facebook iconYouTube iconTwitter iconFlickr iconLinkedInInstagram

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

 
REPORT
This report is an archived publication and may contain dated technical, contact, and link information
Back to Publication List        
Publication Number:  FHWA-HRT-14-049    Date:  August 2014
Publication Number: FHWA-HRT-14-049
Date: August 2014

 

Mitigation of Wind-Induced Vibration of Stay Cables: Numerical Simulations and Evaluations

CHAPTER 2: THEORETICAL BACKGROUND

VIBRATION OF TAUT STRINGS

The governing equation for the free transverse vibration of a taut string is as follows:(4)

H times second order partial differential of y in respect to x minus rho times A times second order partial differential of y in respect to t equals 0.
Figure 1 . Equation. Equation of motion (EOM) for a taut string.

Where:

H = Axial tension force in a string or cable.
y = Transverse in-plane displacement due to vibration.
ρ = Mass density per unit volume.
A = Cross-sectional area of the string, beam, or cable.
t = Time.
x = Distance.

The equation in figure 1 may be rewritten as follows:

Second order partial differential of y in respect to x equals 1 over c squared times second order partial differential of y in respect to t.
Figure 2 . Equation. One-dimensional wave propagation.

Where c is the phase velocity, which is defined as follows:

c equals the square root of H divided by rho times A.
Figure 3 . Equation. Phase velocity.

Applying the method of separation of variables, a general solution to the equation from figure 1 with a fixed-fixed end condition may readily be derived as follows:

y open parenthesis x, t closed parenthesis equals summation over n from 1 to infinity of C subscript n times cosine of open parenthesis omega subscript n times t minus alpha subscript n closed parenthesis times sine of k subscript n times x.
Figure 4 . Equation. General solution of EOM of a taut string.

Where:
ωn = Natural angular frequency of the nth mode of vibration.
kn = Wave number of the nth mode of vibration.
Cn = Amplitude of in-plane displacement due to vibration.
αn = Phase angle of time-dependent part of transverse in-plane displacement due to vibration.
n = Mode number.

The angular frequencies and wave numbers are not independent of each other but are interrelated as follows:

Omega subscript n equals k subscript n times c equals n times pi times c divided by L
Figure 5 . Equation. Relationship between angular frequency and wave number.

Where:

L = Length of the string.

The equation in figure 4 indicates that the motion of the string is represented by a superposition of standing waves with mode shapes of sin knx and time-varying amplitudes of Cncos(ωnt - αn).

The natural frequencies, ωn, are the eigenvalues representing the discrete frequencies at which the system is capable of undergoing harmonic motion.

The equation in figure 1 is a linearized EOM in which nonlinearities arising from finite sag are ignored. Note that the only significant parameters in figure 1 through figure 5 are L, H, and ρA. Also note that from the equation in figure 5, ωn is proportional to the mode number, n.

From the equations in figure 3 and figure 5, the cable tension H can be determined from the fundamental natural frequency, f, as follows:

H equals 4 times rho times A times L squared times f squared.
Figure 6 . Equation. Cable tension.

f, in Hz is related to the angular frequency ω such that f = ω/2π.

VIBRATION OF CLASSICAL BEAMS

The governing equation for the free transverse vibration of a Bernoulli-Euler beam is given by the following:(4)

E times I times fourth order partial differential of y in respect to x plus rho times A times second order partial differential of y in respect to t equals 0.
Figure 7 . Equation. EOM for a classical beam.

Where:

E = Young's modulus.
I = Moment of inertia.

The equation in figure 7 may be rewritten as follows:

Fourth order partial differential of y in respect to x plus 1 divided by a squared times second order partial differential of y in respect to t equals 0.
Figure 8 . Equation. EOM for a classical beam, rewritten with vibration parameter.

Where a is defined as the vibration parameter for classical beam, which can be solved as follows:

a equals the square root of E times I divided by rho times A.
Figure 9. Equation. Vibration parameter for a classical beam.

Note that the equation in figure 8 is not of the wave equation form and that a does not have the dimension of velocity. Applying the method of separation of variables, a general solution to the equation in figure 8 with a pinned-pinned end condition can be derived and takes the form of the equation in figure 4, with ωn and kn being interrelated as follows:

Omega subscript n equals k subscript n squared times a equals open parenthesis n times pi divided by L closed parenthesis squared times a.
Figure 10 . Equation. Relationship between angular frequency and wave number.

The significant parameters in this formulation are L, EI, and ρA. Note that ωnn2.

VIBRATION OF TAUT STRINGS WITH FLEXURAL STIFFNESS

The governing equation for the free transverse vibration of a taut string with flexural rigidity or, equivalently, a classical beam with axial tension, is given by the following equation:

E times I times fourth order partial differential of y in respect to x minus H times second order partial differential of y in respect to x plus rho times A times second order partial differential of y in respect to t equals 0.
Figure 11 . Equation. EOM for a taut string with flexural stiffness.

A general solution to the equation in figure 11 with a pinned-pinned end condition can be derived and again takes the form of the equation in figure 4, with ωn and kn being interrelated as follows:

Omega subscript n equals k subscript n times c prime equals open parenthesis n times pi divided by L closed parenthesis times the square root of H divided by rho subscript L times the square root of 1 plus n squared times pi squared times E times I divided by H times L squared equals omega subscript n times s times the square root of 1 plus n squared times pi squared divided by xi squared.
Figure 12 . Equation. Relationship between angular frequency and wave number for a string.

Or equivalently as follows:

Omega subscript n equals k subscript n squared times a prime equals open parenthesis n times pi divided by L closed parenthesis squared times the square root of E times I divided by rho subscript L times the square root of 1 plus H times L squared divided by n squared times pi squared times E times I equals omega subscript n times b times the square root of 1 plus xi squared divided by n squared times pi squared.
Figure 13 . Equation. Relationship between angular frequency and wave number for a beam.

Where:

ρL = Cable mass per unit length.
ξ = Flexural stiffness parameter.
ωnb = Natural angular frequency of a classical beam in the nth mode of vibration.
ωns = Natural angular frequency of the taut string in the nth mode of vibration.

The parameter ξ in figure 12 and figure 13 is defined as follows:(5)

Xi is defined as the square root of H times L squared divided by E times I.
Figure 14 . Equation. Flexural stiffness parameter.

For the equation in figure 11, the first term, accounting for the effect of flexural stiffness, is added to the taut-string equation presented in figure 1. The equation in figure 12 indicates that the natural vibration frequencies of a taut string with flexural rigidity can be expressed in terms of those of the simple taut string when appropriate factors are multiplied. The same case may be viewed as a beam with axial tension, and the natural frequencies can be expressed in terms of those of the beam with appropriate factors multiplied as shown in figure 13. The flexural stiffness parameter defined by the equation in figure 14 represents the relative influence of the axial tension over the flexural stiffness in cable vibration.

The solutions presented in figure 12 and figure 13 are for cables with pinned-pinned end conditions. Analysis of cables with fixed-fixed end conditions is more complex and requires the solution of transcendental equations.

VIBRATION OF TAUT STRINGS WITH FLEXURAL STIFFNESS AND SAG-EXTENSIBILITY

The governing equation for the free transverse vibration of a taut string with transverse flexural rigidity and sag-extensibility is as follows:(6,7)

E times I times fourth order partial differential of y in respect to x minus H times second order partial differential of y in respect to x minus h times second order differential of y subscript s in respect to x plus rho times A times second order partial differential of y in respect to t equals 0.
Figure 15 . Equation. EOM for a taut string with flexural stiffness and sag-extensibility.

Where:

h = Horizontal component of tension force due to vibration.
ys = Transverse in-plane displacement due to weight.

No closed-form solution to the equation in figure 15 is available; however, approximate solutions for certain boundary conditions are available. The vibration frequencies of a cable with fixed-fixed end condition can be expressed in terms of those of the taut string as follows:

Omega subscript n divided by omega subscript n times s equals alpha times beta subscript n minus 0.24 times mu divided by xi.
Figure 16 . Equation. Approximate solution to the EOM for a taut string with flexural stiffness and sag-extensibility.

Where:

α = Correction factor for sag-extensibility effects, which is defined in figure 17.
βn = Bending stiffness correction factor for nth mode of vibration.
μ = Mass parameter.

Alpha equals 1 plus 0.039 times mu, beta subscript n equals 1 plus 2 divided by xi plus open parenthesis 4 plus n squared times pi squared divided by 2 closed parenthesis divided by xi squared.
Figure 17 . Equation. Correction factor for sag-extensibility and bending stiffness.

The sag-extensibility parameter, λ2, is defined as follows:

Lambda squared equals open parenthesis rho times A times g times L times cosine of theta divided by H closed parenthesis squared times L all divided by open parenthesis H times L subscript e divided by E times A closed parenthesis.
Figure 18. Equation. Sag-extensibility parameter.

Where:

θ = Inclination angle of the cable.
g = Gravitational constant.
Le = Effective length of the cable, which is defined as follows:

L subscript e equals open bracket 1 plus open parenthesis rho times A times g times L times cosine of theta divided by H closed parenthesis squared divided by 8 closed bracket times L.
Figure 19. Equation. Effective length of cable.

The additional tension h due to cable vibration adds nonlinearity to the formulation and is determined by the equation in figure 20.

h equals integral over x from 0 to L of second order differential of y subscript s in respect to x times y times differential in x divided by integral over x from 0 to L open bracket 1 plus open parenthesis differential in y subscript s in respect to x closed parenthesis squared closed bracket raised to the power of 3 over 2 all divided by E times A times differential in x.
Figure 20 . Equation. Additional tension force due to cable vibration.

The parameter α in figure 16 is given by the following:

Alpha equals 1 plus 0.039 times mu.
Figure 21. Equation. Correction factor for sag-extensibility.

The parameter μ in figure 16 is given by the following:

Mu equals lambda squared for n equals 1 (in-plane) and mu equals 0 for n greater than 1 (in-plane) or for all n (out-of-plane).
Figure 22. Equation. Mass parameter.

The two parameters λ2 and ξ a major role in the formulation in figure 14 and figure 15. The relationship of the equation in figure 16 is known to provide a good approximation when λ2 < 3 and ξ > 50, and many stay cables in cable-stayed bridges fall within this range.