Dynamic Mechanical Analysis tests are an established experimental technique that can be used to obtain the material stress-strain relationship by means of forced sinusoidal excitation measurements made on beam specimens. A procedure for the experimental identification of the material standard linear solid (SLS) frequency dependent constitutive equation by means of the DMA estimated material stress-strain relationship is presented. The proposed approach is an extension of an identification technique for the constitutive stress-strain relationship presented by the authors previous works. Uniform and homogeneous beams in clamped-double pendulum boundary conditions, with a sinusoidal flexural force applied at the mobile beam end are considered. Frequency dependent force and displacement test results from reference known specimens are processed in order to estimate the 2-DOFs model frame contribution to the measurement data. The model frame and the material SLS model are both mathematically defined by polynomial rational functions, polynomial order and parameters are to be identified by means of an algebraic numerical technique. The rational function model of the instrument frame is obtained by means of an approach based on the least square error technique and making use of different polynomial bases such as the Forsythe, Legendre, and Chebyshev ones. Unphysical model zeros and poles are eliminated to obtain a minimum order optimal model. A multi-step algorithm is adopted to obtain a n-order SLS material model, where a rational polynomial, n+1-degree numerator and a n-degree denominator polynomial, is assumed. The optimal SLS model order is automatically obtained by eliminating non-physical poles. To minimize computational errors, the influence of the choice of the different polynomial functional bases on the effectiveness of the model results is investigated, and numerical results are discussed. Some application examples, concerning the identification of the SLS material model of viscoelastic standard and non-standard materials, are shown and discussed as well.
Amadori, S. (2021). A procedure for the identification of the SLS material model from dynamical experimental measurements. Elsevier.
A procedure for the identification of the SLS material model from dynamical experimental measurements
Amadori S.
Co-primo
Membro del Collaboration Group
;Catania G.Co-primo
Writing – Original Draft Preparation
2021
Abstract
Dynamic Mechanical Analysis tests are an established experimental technique that can be used to obtain the material stress-strain relationship by means of forced sinusoidal excitation measurements made on beam specimens. A procedure for the experimental identification of the material standard linear solid (SLS) frequency dependent constitutive equation by means of the DMA estimated material stress-strain relationship is presented. The proposed approach is an extension of an identification technique for the constitutive stress-strain relationship presented by the authors previous works. Uniform and homogeneous beams in clamped-double pendulum boundary conditions, with a sinusoidal flexural force applied at the mobile beam end are considered. Frequency dependent force and displacement test results from reference known specimens are processed in order to estimate the 2-DOFs model frame contribution to the measurement data. The model frame and the material SLS model are both mathematically defined by polynomial rational functions, polynomial order and parameters are to be identified by means of an algebraic numerical technique. The rational function model of the instrument frame is obtained by means of an approach based on the least square error technique and making use of different polynomial bases such as the Forsythe, Legendre, and Chebyshev ones. Unphysical model zeros and poles are eliminated to obtain a minimum order optimal model. A multi-step algorithm is adopted to obtain a n-order SLS material model, where a rational polynomial, n+1-degree numerator and a n-degree denominator polynomial, is assumed. The optimal SLS model order is automatically obtained by eliminating non-physical poles. To minimize computational errors, the influence of the choice of the different polynomial functional bases on the effectiveness of the model results is investigated, and numerical results are discussed. Some application examples, concerning the identification of the SLS material model of viscoelastic standard and non-standard materials, are shown and discussed as well.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
