Skip to main content
Log in

Invariant subspace reduction for linear dynamic analysis of finite-dimensional viscoelastic structures

  • New Trends in Dynamics and Stability
  • Published:
Meccanica Aims and scope Submit manuscript

Abstract

The linear dynamics of finite-dimensional viscoelastic structures is addressed in this paper. The equations of motion of a general, discrete or discretized, dynamical system, made of elements behaving as multiparameter viscoelastic solid models, are formulated in terms of internal variables, whose evolution is ruled by flow laws. The classical elastic-viscoelastic Principle of Correspondence is discussed in conjunction with the Fourier transform, and a new strategy, which leaves the system in the time-domain is proposed. By exploiting the fact that the spectrum of the system is well-separated if damping is small, a Center Manifold-like reduction is performed, which eliminates the internal variables, lowering the dimensions to those of the corresponding elastic system, thus filtering the fast dynamics. The order of magnitude of the error related to the reduction is investigated. A comparison with the popular Kelvin–Voigt model is performed for homogeneous structures. Examples on sample structures are worked out, namely the one degree-of-freedom viscoelastic system and the discretized elastic beam on viscoelastic soil. In all the examples the (3-Parameters) Standard Model is adopted.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10

Similar content being viewed by others

Notes

  1. For example, for the 5-Parameter Model, the internal power reads (apex j omitted):

    $$\begin{aligned} \begin{aligned}\mathcal {P}_{int}:= \,&\sigma \left( \frac{\dot{\sigma }}{E_{0}}+\dot{\varepsilon }_{v1}+\dot{\varepsilon }_{v2}\right) \\ =&\frac{d}{dt}\left( \frac{\sigma ^{2}}{2E_{0}}+\frac{1}{2}E_{v1}\varepsilon _{v1}^{2}+\frac{1}{2}E_{v2}\varepsilon _{v2}^{2}\right) +\eta _{1}\dot{\varepsilon }_{v1}^{2}+\eta _{2}\dot{\varepsilon }_{v2}^{2} \end{aligned} \end{aligned}$$

    in which use has been made of Eq. (2-a) differentiated with respect to the time, and of the flow laws (2-b), each solved with respect to \(\sigma \). It is, therefore, \(\mathcal {P}_{int}=\dot{\psi }+\delta \), with \(\psi \) the elastic energy of the three springs and \(\delta \) the dissipation (see, e.g., [5, 6, 40]). Positive viscous coefficients \(\eta _{k}\) assure that \(\delta \ge 0\), and, therefore, dissipation of the mechanical work.

References

  1. Flugge W (1975) Viscoelasticity. Springer, London

    Book  MATH  Google Scholar 

  2. Moreau JJ (1970) Sur les lois de frottement, de viscosité et de plasticité. CR Acad Sci Paris 271:608–611

    Google Scholar 

  3. Rice JR (1971) Inelastic constitutive relations for solids: an internal-variable theory and its application to metal plasticity. J Mech Phys Solids 19(6):433–455

    Article  ADS  MATH  Google Scholar 

  4. Germain P, Suquet P, Nguyen QS (1983) Continuum thermodynamics. ASME Trans Ser E J Appl Mech 50:1010–1020

    Article  ADS  MATH  Google Scholar 

  5. Maugin GA, Muschik W (1994) Thermodynamics with internal variables. Part I. General concepts. J Non Equilib Thermodyn 19:217–249. doi:10.1515/jnet.1994.19.3.217

    ADS  MATH  Google Scholar 

  6. Lemaitre J, Chaboche J-L (1994) Mechanics of solid materials. Cambridge University Press, Cambridge

    MATH  Google Scholar 

  7. Di Paola M, Pirrotta A, Valenza A (2011) Visco-elastic behavior through fractional calculus: an easier method for best fitting experimental results. Mech Mater 43(12):799–806

    Article  Google Scholar 

  8. Di Paola M, Failla G, Pirrotta A (2012) Stationary and non-stationary stochastic response of linear fractional viscoelastic systems. Probab Eng Mech 28:85–90

    Article  Google Scholar 

  9. Jia-ju S, Ke-hwa J (1981) Dynamic response of viscoelastic beam. Appl Math Mech 2(2):255–264

    Article  MathSciNet  Google Scholar 

  10. Cabański J (2002) An exact method for the free vibration analysis of Timoshenko–Kelvin beams with oscillators. J Sound Vib 253(3):669–685

    Article  ADS  MATH  Google Scholar 

  11. Kirillov ON, Seyranian AP (2005) The effect of small internal and external damping on the stability of distributed non-conservative systems. J Appl Math Mech 69(4):529–552

    Article  MathSciNet  Google Scholar 

  12. Mahmoodi SN, Khadem SE, Kokabi M (2007) Non-linear free vibrations of Kelvin–Voigt visco-elastic beams. Int J Mech Sci 49(6):722–732

    Article  Google Scholar 

  13. Marynowski K (2002) Non-linear dynamic analysis of an axialy moving viscoelastic beam. J Theor Appl Mech 40(2):465–482

    MATH  Google Scholar 

  14. Ghayesh MH, Alijani F, Darabi MA (2011) An analytical solution for nonlinear dynamics of a viscoelastic beam-heavy mass system. J Mech Sci Technol 25(8):1915–1923

    Article  Google Scholar 

  15. Ghayesh MH (2012) Nonlinear dynamic response of a simply-supported Kelvin–Voigt viscoelastic beam, additionally supported by a nonlinear spring. Nonlinear Anal: Real World Appl 13(3):1319–1333

    Article  MathSciNet  MATH  Google Scholar 

  16. Ghayesh MH (2008) Nonlinear transversal vibration and stability of an axially moving viscoelastic string supported by a partial viscoelastic guide. J Sound Vib 314(3):757–774

    Article  ADS  MathSciNet  Google Scholar 

  17. Ghayesh MH, Moradian N (2011) Nonlinear dynamic response of axially moving, stretched viscoelastic strings. Arch Appl Mech 81(6):781–799

    Article  MATH  Google Scholar 

  18. Ghayesh MH, Amabili M, Farokhi H (2013) Coupled global dynamics of an axially moving viscoelastic beam. Int J Non-Linear Mech 51:54–74

    Article  MATH  Google Scholar 

  19. Ghayesh MH, Amabili M (2012) Nonlinear dynamics of axially moving viscoelastic beams over the buckled state. Comput Struct 112:406–421

    Article  Google Scholar 

  20. Ghayesh MH, Amabili M, Farokhi H (2013) Two-dimensional nonlinear dynamics of an axially moving viscoelastic beam with time-dependent axial speed. Chaos Solitons Fractals 52:8–29

    Article  ADS  MathSciNet  MATH  Google Scholar 

  21. Luongo A, D’Annibale F (2013) Double zero bifurcation of non-linear viscoelastic beams under conservative and non-conservative loads. Int J Non-Linear Mech 55:128–139

    Article  Google Scholar 

  22. Luongo A, D’Annibale F (2014) On the destabilizing effect of damping on discrete and continuous circulatory systems. J Sound Vib 333(24):6723–6741

    Article  ADS  Google Scholar 

  23. Esmailzadeh E, Jalali MA (1999) Nonlinear oscillations of viscoelastic rectangular plates. Nonlinear Dyn 18(4):311–319

    Article  MATH  Google Scholar 

  24. Marynowski K (2004) Non-linear vibrations of an axially moving viscoelastic web with time-dependent tension. Chaos Solitons Fractals 21(2):481–490

    Article  ADS  MATH  Google Scholar 

  25. Wang ZM, Zhou YF, Wang Y (2007) Dynamic stability of a non-conservative viscoelastic rectangular plate. J Sound Vib 307(1–2):250–264

    Article  ADS  Google Scholar 

  26. De Haan YM, Sluimers GM (2001) Standard linear solid model for dynamic and time dependent behaviour of building materials. HERON 46(1):2001

    Google Scholar 

  27. Sturla FA, Argento A (1996) Free and forced vibrations of a spinning viscoelastic beam. J Vib Acoust 118(3):463–468

    Article  Google Scholar 

  28. Marynowski K, Kapitaniak T (2007) Zener internal damping in modelling of axially moving viscoelastic beam with time-dependent tension. Int J Non-Linear Mech 42(1):118–131

    Article  MATH  Google Scholar 

  29. Hatami S, Ronagh HR, Azhari M (2008) Exact free vibration analysis of axially moving viscoelastic plates. Comput Struct 86(17–18):1738–1746

    Article  Google Scholar 

  30. Marynowski K (2010) Free vibration analysis of the axially moving Levy-type viscoelastic plate. Eur J Mech—A/Solids 29(5):879–886

    Article  Google Scholar 

  31. Muscolino G, Palmeri A (2007) Response of beams resting on viscoelastically damped foundation to moving oscillators. Int J Solids Struct 44(5):1317–1336

    Article  MATH  Google Scholar 

  32. Hatada T, Kobori T, Ishida M, Niwa N (2000) Dynamic analysis of structures with Maxwell model. Earthq Eng Struct Dyn 29(2):159–176

    Article  Google Scholar 

  33. Lewandowski R, Bartkowiak A, Maciejewski H (2012) Dynamic analysis of frames with viscoelastic dampers: a comparison of damper models. Struct Eng Mech 41(1):113–137

    Article  Google Scholar 

  34. Buhariwala KJ, Hansen JS (1988) Dynamics of viscoelastic structures. AIAA J 26(2):220–227

    Article  ADS  MathSciNet  MATH  Google Scholar 

  35. Banks HT, Inman DJ (1991) On damping mechanisms in beams. J Appl Mech 58(3):716–723

    Article  MATH  Google Scholar 

  36. Gaul L (1999) The influence of damping on waves and vibrations. Mech Syst Signal Process 13(1):1–30

    Article  ADS  MathSciNet  Google Scholar 

  37. Vasques CMA, Moreira RAS, Rodrigues JD (2010) Viscoelastic damping technologies—Part I: modeling and finite element implementation. Mech Eng 1(2):96–110

    Google Scholar 

  38. Fabrizio M, Giorgi C, Naso MG (2004) Viscoelastic solids of exponential type I. Minimal representations and controllability. Meccanica 39(6):531–546

    Article  MathSciNet  MATH  Google Scholar 

  39. Fabrizio M, Giorgi C, Naso MG (2004) Viscoelastic solids of exponential type. II. Free energies, stability and attractors. Meccanica 39(6):547–561

    Article  MathSciNet  Google Scholar 

  40. Fabrizio M, Giorgi C, Morro A (1994) Free energies and dissipation properties for systems with memory. Arch Ration Mech Anal 125(4):341–373

    Article  MathSciNet  MATH  Google Scholar 

  41. Guckenheimer J, Holmes PJ (2013) Nonlinear oscillations, dynamical systems, and bifurcations of vector fields, vol 42. Springer, New York

    MATH  Google Scholar 

  42. Wiggins S (2003) Introduction to applied nonlinear dynamical systems and chaos, vol 2. Springer, New York

    MATH  Google Scholar 

  43. Troger H, Steindl A (2012) Nonlinear stability and bifurcation theory: an introduction for engineers and applied scientists. Springer, New York

    MATH  Google Scholar 

  44. Nayfeh AH (2008) Perturbation methods. Wiley, New York

    Google Scholar 

  45. Luongo A (1993) Eigensolutions sensitivity for nonsymmetric matrices with repeated eigenvalues. AIAA J 31(7):1321–1328

    Article  ADS  MATH  Google Scholar 

  46. Luongo A (1995) Free vibrations and sensitivity analysis of a defective two degree-of-freedom system. AIAA J 33(1):120–127

    Article  ADS  MATH  Google Scholar 

  47. Luongo A, Zulli D (2012) Dynamic analysis of externally excited NES-controlled systems via a mixed multiple scale/harmonic balance algorithm. Nonlinear Dyn 70(3):2049–2061

    Article  MathSciNet  Google Scholar 

  48. Luongo A, Zulli D (2012) A paradigmatic system to study the transition from zero/hopf to double-zero/hopf bifurcation. Nonlinear Dyn 70(1):111–124

    Article  MathSciNet  MATH  Google Scholar 

  49. Luongo A, Zulli D (2014) Aeroelastic instability analysis of NES-controlled systems via a mixed multiple scale/harmonic balance method. J Vib Control 20(13):1985–1998

    Article  MATH  Google Scholar 

  50. Nayfeh AH, Mook DT (2008) Nonlinear oscillations. Wiley, New York

    MATH  Google Scholar 

  51. Bolotin VV (1964) The dynamic stability of elastic systems, volume 3 of Holden-Day series in mathematical physics. Holden-Day

  52. Hagedorn P (1970) On the destabilizing effect of non-linear damping in non-conservative systems with follower forces. Int J Non-Linear Mech 5(2):341–358

    Article  MATH  Google Scholar 

  53. Luongo A, D’Annibale F (2016) Nonlinear hysteretic damping effects on the post-critical behaviour of the visco-elastic Beck’s beam. Math Mech Solids 22(6):1347–1365. doi:10.1177/1081286516632381

    Article  MathSciNet  MATH  Google Scholar 

  54. Luongo A, D’Annibale F, Ferretti M (2016) Hard loss of stability of Ziegler’s column with nonlinear damping. Meccanica 51(11):2647–2663

    Article  MathSciNet  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Angelo Luongo.

Appendix: Free dynamics of the Standard Model viscoelastic oscillator for medium or large viscosity coefficients

Appendix: Free dynamics of the Standard Model viscoelastic oscillator for medium or large viscosity coefficients

This "Appendix" is devoted to furnish some detail on how the free dynamics of the SO changes when medium or large viscosity coefficient are considered.

Fig. 11
figure 11

Exact versus asymptotic moduli of the eigenvalues of the SO when b varies: a real parts; b imaginary part. \(c_{0}=15, c_{v}=5\)

The discussion is developed by referring, first, to the graphs displayed in Fig. 11, where the moduli of both the exact roots of Eq. (44) (gray curves), and the asymptotic roots (black curves), furnished by Eq. (45), are plotted vs b, when \(c_{0}=15, c_{v}=5\); in particular, Fig. 11a is relevant to the moduli of the real part of the eigenvalues \(\left| \mathrm {Re}\left( \lambda _{1}\right) \right| , \left| \mathrm {Re}\left( \lambda _{2}\right) \right| \), respectively, while in Fig. 11b the modulus of the imaginary part \(\left| \mathrm {Im}\left( \lambda _{2}\right) \right| \) is shown. It is found that, when b increases from zero, while \(\left| \mathrm {Re}\left( \lambda _{1}\right) \right| \) monotonically decreases (Fig. 11a), i.e. the real negative root moves towards the imaginary axis, \(\left| \mathrm {Re}\left( \lambda _{2}\right) \right| \) first grows with b, reaching a maximum at \(b=4.13=:b_{opt}\) (exact solution), and then it decreases, that is the complex eigenvalues first move away from the imaginary axis and then come back towards it. Since the damping coefficient of the oscillatory decaying part of the response is proportional to \(\left| \mathrm {Re}\left( \lambda _{2}\right) \right| \), then \(b_{opt}\) represents an optimum value of the viscosity coefficient, which maximizes the damping. Moreover, the two curves \(\left| \mathrm {Re}\left( \lambda _{1}\right) \right| , \left| \mathrm {Re}\left( \lambda _{2}\right) \right| \) cross each other at \(b=4.86=:b_{cr}\) (exact solution), thus entailing that, increasing the viscosity coefficient produces a change of the character of the decaying motion, namely: if \(b<b_{cr}\) the exponential decaying part of the motion is faster than the oscillatory one, otherwise, if \(b>b_{cr}\) it is slower than the oscillatory one. In addition, with reference to (Fig. 11b), as b overcomes a threshold, a not negligible dependence of the (exact) frequency \(\left| \mathrm {Im}\left( \lambda _{2}\right) \right| \) on the viscosity coefficient is recognized. As a final comment, the asymptotic solutions of Eq. (45), as it is expected, lose their validity when b is not small and becomes larger than a certain value, i.e. the black curves of Fig. 11a, b leave the gray ones.

Fig. 12
figure 12

SO free dynamics when \(b=4.13\) in sub-figures (a), (b), and when \(b=200\) in sub-figures (c), (d): a, c eigenvalues; b, d orbits. \(c_{0}=15, c_{v}=5, u(0)=-0.1, v(0)=0.03, w(0)=0.2\)

The above mentioned qualitative responses of the SO, relevant to medium or large viscosity coefficients b, are sketched in Fig. 12 where, for the same values of \(c_{0}, c_{v}\) and for the same initial conditions adopted in Sect. 6.1, the spectra of the eigenvalues and the free motion-orbits in the state-space, are shown when \(b=b_{opt}\) (Fig. 12a, b), and when \(b=200\) (Fig. 12c, d).

When \(b=b_{opt}\) (Fig. 12a, b), it is seen that: (1) the spectrum of eigenvalues (gray circles in Fig. 12a) is not well-separated, i.e. \(\left| \mathrm {Re}\left( \lambda _{1}\right) \right| \) is of the same order of \(\left| \mathrm {Re}\left( \lambda _{2}\right) \right| \); (2) since \(\left| \mathrm {Re}\left( \lambda _{1}\right) \right| >\left| \mathrm {Re}\left( \lambda _{2}\right) \right| \) the exponential character of the motion has a faster in-time decaying with respect to the oscillatory one; (3) since \(\left| \mathrm {Re}\left( \lambda _{2}\right) \right| \) is maximized by \(b_{opt}\) the in-time decaying of the oscillations is faster with respect to the case of small viscosity coefficient. These latter considerations are corroborated by the plot of Fig. 12b, where the orbit in the state-space is shown: the free motion develops itself along a spiral trajectory, wrapped around the one-dimensional subspace \(r:=\mathrm {Span}\left[ \mathbf {u}_{1}\right] \); differently from the small viscosity-case, the two phases of the motion cannot be distinguished, namely a part of the trajectory is not more parallel to r (remember that the spectrum of the eigenvalues is not well-separated) and, therefore, there is not a recognizable fast transient motion tending to \(\pi :=\mathrm {Span}\left[ \mathrm {Re}\left( \mathbf {u}_{2}\right) ,\mathrm {Im}\left( \mathbf {u}_{2}\right) \right] \). Moreover, as it is expected, when the viscosity coefficient is not small, the effectiveness of the RO in reproducing the dynamical behavior of the SO decreases; e.g., in Fig. 12a, it is apparent that the complex eigenvalues of the RO are quite far from those of the SO.

When b is large (Fig. 12c, d), it is found that: (1) the spectrum of eigenvalues (Fig. 12c) is not well-separated; (2) since \(\left| \mathrm {Re}\left( \lambda _{1}\right) \right| <\left| \mathrm {Re}\left( \lambda _{2}\right) \right| \) the exponential character of the motion has a slower in-time decaying with respect to the oscillatory one; (3) \(\left| \mathrm {Re}\left( \lambda _{2}\right) \right| \) is about an order of magnitude smaller with respect to that of the small viscosity-case, thus implying that the in-time decaying of the oscillations is very slow; (4) since \(\left| \mathrm {Re}\left( \lambda _{1}\right) \right| \) is about two orders of magnitude smaller with respect to that of the small viscosity-case, also the in-time decaying of the exponential component of the motion is very slow. The orbit in the state-space, shown in Fig. 12d, confirms these considerations: indeed, the spiral trajectory, is wrapped around the one-dimensional subspace r, as it occurs in the previous cases, but it is able to reach the plane \(\pi \) only after several elliptical orbits, which are almost parallel to \(\pi \) are experienced; remarkably, this kind of dynamics reflects the slow in-time decaying of both exponential and oscillatory components of motion.

In conclusion, when b is not small, the SO does not more behave as a one d.o.f. system, the displacements u and w are not more in-phase, and, depending on the magnitude of the viscous coefficient, an highly damped response (Fig. 12b), as well as a weakly damped motion, in which the mass experiences larger in-time oscillations with respect to the Kelvin–Voigt element (Fig. 12d), are possible.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Luongo, A., D’Annibale, F. Invariant subspace reduction for linear dynamic analysis of finite-dimensional viscoelastic structures. Meccanica 52, 3061–3085 (2017). https://doi.org/10.1007/s11012-017-0741-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11012-017-0741-y

Keywords

Navigation