Chasing the high-resolution mapping of rotational and strain FRFs as receptance processing from different full-field optical measuring technologies

https://doi.org/10.1016/j.ymssp.2021.108428Get rights and content

Abstract

Thanks to the recent advances in digital vision systems, a question might arise about where the full-field optical and contactless methods can bring modern design procedures. An answer lies on the methodologically detailed comparison of the results, processed from different full-field optical techniques, in exploring consistent and high-resolution maps of rotational and strain FRFs that is done through this paper.

Great exertions were first put in thoroughly testing a thin aluminium plate, in its real dynamics as a lightweight structure with broad frequency band dynamics and high modal density, in a unique comparative set-up, to obtain Receptance FRF maps of displacements-over-force by means of 3 different full-field optical techniques (SLDV, DIC, ESPI). There resulted superior quality Receptance maps in a broad and dense frequency domain, with high-resolution and continuity-wise consistent mapping at each frequency line.

This paper exploits the here detailed robust numerical differentiation and signal processing in order to calculate the Impedance-based models of rotational and strain FRFs. Especially for DIC and for ESPI, both rarely used on many frequency lines, the fully populated FRFs for rotations and strains are a clear novelty, with rotational Coherence functions as added quality assessment features.

The systematic comparison of the results obtained in the same location of the sample, by means of spatial and frequency domain metrics, is possible because of the proposed pointwise procedure, permitting the first numerical matching assessment of the 3 optical technologies on these awaited experiment-based quantities, as the full-field rotational and strain FRFs.

Introduction

The accurate sensing of vibration related features in broad frequency band structural dynamics is a demanding task that can affect the reliability of the most advanced physical quantities of interest. Upon the reliability of the latter depend many predictive speculations by means of advanced inferences: simulations about the behaviour of complex parts in a broad frequency range; life cycle estimations; quality assurance of the manufacturing processes, especially when tight dynamic targets need to be reached in working conditions. Beside more traditional quantities like accelerations, velocities and displacements, the rotational dofs are gaining interest (see [1], [2], [3], [4], [5], [6], [7]) for their contribution in refining reliable models of non-conventional structures, despite the difficulties faced in their sensing with traditional transducers. In more recent times, many authors [8], [9], [10], [11], [12] have proved the advantage of using also the rotational dofs in frequency based substructuring, to increase the precision of the hybrid coupling in automotive applications; in [13], [14] only the displacements were available for the hybrid coupling, indeed with lower agreement with the real measurements.

Other authors (see [15], [16], [17], [18]) addressed the rotational dof measurements by adding a T-block-like arrangement of accelerometers, or by putting precisely known distances in between, to use a finite differences approach to evaluate the quantities, but loading the specimen with relevant inertia and damping. In this way, though, the procedure to obtain a many rotational dofs map becomes certainly cumbersome and not practical. The interest for rotational dofs is increased when they are coupled with model updating procedures (see [19], [20], [21], [22]), with the aim to improve the finite element formulation in anisotropic compounds and complex shaped parts. As in [18], the right estimate of rotational Impedances has specific interest for torsional vibrations of shafts and for extended modal analysis of coupled torsional and bending dynamics, although these enquiries were only made with the aid of lumped sensor measurements.

Before the optical full-field techniques appeared in the first viable industrial prototypes, up to recent times, most authors in engineering literature (e.g. [23], [24], [25]) have used lumped sensors for spatially coarse data, like strain gauges or rosettes, coupled with tuned FE models to simulate the missing data on locations where no real data were acquired, also in the spatially dense sensing of strains. This has been the procedure also in structural dynamics for fatigue life estimations, with the appropriate material constitutive parameters. This procedure certainly brings the known drawbacks of finding an effective sensor placement, of spatial averaging, of mass burden, of local distortions, of cross sensitivities to the surface curvature, of the accuracy of the FE model and of its effective updating to have reliable predictions. To answer part of these questions, the author started to work towards enhancing the full-field measurements in broad frequency bands (see [26], [27], [28], [29], [30], [31]) and the predictions on the structural reliability of the components (see [7], [32], [33], [34], [35], [36], [37]). This specific work, instead, explains in details the latest comparative assessment among full-field technologies when the numerical derivatives of raw datasets are taken into account.

Shearography [38] was used in [39] by means of integration along the shear direction to obtain vibration displacement patterns in stroboscopic lighting, while it could have been proficiently used as a direct optical derivative technique to implement the rate of deflection of the tangent plane at each point in the surface map, thus measuring directly the rotations for dynamically deformed flat surfaces. As can be checked instead in [40], the industrial instrumentation manufacturers specialised only in NDT applications of shearography, and no system (to the knowledge of the author) is produced with the mated electronics of dynamic ESPI for complex-valued dynamic rotation distribution. Whereas, also for typical shearography’s applications, the dynamic excitation can be relevant, where faults might be even more easily revealed, because linked to anomalies in higher frequency operative deflection shapes (see [41], [42]). Further derivations might also have led to complex-valued dynamic strain mapping, for the Impedance FRF models, with less expansion of derivative errors, as highlighted below.

The TEFFMA1 fundamental research project aimed at making a comparison between the state-of-the-art of industrial implementations in optical native full-field technologies and the SLDV as reference, to understand at which point of their development these experimental procedures can provide enhanced peculiarities in NVH applications. It was out of the project’s scope, though, to investigate all the specific techniques’ implementations. On the contrary, the TEFFMA project aimed to assess the real potentialities of the data obtainable from commercial instruments of more general use, even if at an immature, but advanced, prototype evolutionary stage, as highlighted in Table 1. The reader interested instead in the basic techniques behind the instrumentations can find the needed survey in scientific literature (e.g. [43], [44] for SLDV, [45] for photogrammetry, [46], [47] for DIC, [48], [49] for ESPI).

In the well equipped laboratory of the TU-Wien, as summarised in Section 2 below, it was possible to build a complex set-up for the comparison of the peculiarities of the different optical technologies there available (SLDV, DIC and ESPI) in acquiring full-field FRFs. No standard sensors were tested to make a cross comparison with the optical technologies, as was done instead for accelerometers in [50] or for strains, with Shearography and fibre Bragg gratings (not SLDV, DIC & ESPI), in [51]. As it will be clearer in Section 2, the set-up would have been otherwise compromised by the adhesion and by the inertial distortions of traditional lumped sensors, like accelerometers or strain gauges and their cabling. It must be remarked that a lot of work was done [52], [53], or is still under refinement, to find the best agreement with optical techniques and strain measurements. For more details the interested reader is invited to look at Refs. [7], [28], [29], [30], [31], [54], [55], [56], [57]. Indeed, the quality achieved in full-field Receptances (FRFs of displacement over force), in terms of field continuity, proved to be a feasible starting point for robust numerical derivative operators to obtain new experiment-based quantities, here in the FRFs’ map shape of complex-valued dynamic rotations and strains over force. All the three tested optical technologies offered an increasing spatial definition with respect to traditional lumped sensors, from a very good to an extreme sampling of the surface, coupled with a detailed frequency domain description, clearly addressing the posed question about the gauge location and the capabilities to improve the reliability of the predictions, beyond the forecast developments highlighted in [58]. Furthermore, most important for lightweight structures, those techniques offer non-contacting and -distorting measurements. The high spatial density solved the sensor location assessment with accurate surface mapping, while the lack of mass burden and tight measured data offered the author the chance to avoid dynamics’ distortions and area averaging respectively, with a clear benefit for the assessment of new quantities, like surface rotations and strains as in [7], and the tuning of numerical models as in [57].

This paper wants to detail the methodology behind the high quality results obtained about rotational FRFs’ estimation and related rotational Coherence function maps, and, further, about strain FRFs, in the framework of a pointwise comparative approach, focused on the derivative results from SLDV, DIC and ESPI measurement technologies, at the same (numerically best obtained) physical location of the test specimen. After Section 2 devoted to briefly describe the testing sessions, Section 3 is therefore dedicated to the methods at the basis of this work: it focuses on the novel and extensive usage of robust numerical differentiations (detailed in Appendix A) on the high quality experimental Receptance maps previously obtained (see [28], [29], [30], [31]), defining the logic to extract refined maps of both dynamic rotation and surface strain FRFs. Section 4 depicts the proposed pointwise comparative framework to have meaningful matchings among the techniques. Section 5 addresses in detail the first pointwise comparison of the evaluated results (rotational FRFs, rotational Coherence functions and strain FRFs), with the systematic usage of the FRAC and MAC metrics on the whole datasets (described in Appendix B), permitting the direct matching of figures coming from competing optical instruments, with related comments, before drawing the final conclusions in Section 6.

Section snippets

The testing activity in brief

In Fig. 1 an overview of the test session arrangement in the common set-up at TU-Wien is available. The three measurement technologies of Table 1 were focused on the same portion of the specimen surface to sense the same dynamic behaviour of the restrained plate, as the latter was chosen to express a high modal density inside the frequency range of interest, with a complex mixing of different shape patterns.

The specimen was an aluminium thin rectangular plate (250 × 236 × 1.5 mm, slightly

Experimental Impedances for rotation & strain distributions

As the vibration data are sampled on a finer grid by optical techniques, a clearer map of the deforming pattern becomes available, with much more freedom in choosing better sensor locations (see [67]), keeping in mind that full-field technologies do not need any FE model. The scanning vibrometry (like stepped grid SLDV and its continuous scanning variants, see [43], [44], [68], [69], [70], [71]) can challenge any accelerometric measurement chain in the reachable sight, especially for light

Details on the pointwise comparative approach proposed

When the modal density is high and near or repeated poles are present, the dynamic behaviour of a structure may markedly change along the sampled locations with the blending of shapes, as was clearly faced during the tests at the basis of the TEFFMA project [7], [28], [29], [31], [54], making the comparison among different tools quite prohibitive, if not meaningless. A precise common reference for all the three techniques was therefore needed, to be sure that the data in the comparisons were

Results of the comparisons in the pointwise matching

The details on the Receptance FRF maps acquisition can be found in Section 2 and extensively in [29], [30], [31]. The basic features of the datasets here retained are the restricted geometry grid (57 × 51 dofs) of the reference based on SLDV, and the common range of [20.312–1023.438] Hz, with the SLDV native spectral even spacing of 0.78125 Hz among the 1285 frequency lines. By means of these choices, the SLDV dataset was only restricted (by dropping the outlier grid points), but not

Conclusions

This paper has highlighted the relevance that native full-field measurements are gaining in deriving novel advanced physical features, here rotational- and strain-FRFs from experiment-based Receptances, which are becoming strategical for complex design procedures, in particular showing the relevant advantage of consistent and almost continuous data fields when numerically processed. The chasing of accurate mapping of dynamic rotational and strains FRFs faced the first thorough, methodologically

CRediT authorship contribution statement

Alessandro Zanarini: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Software and computing tools for the paper, Data Curation, Writing – Original Draft, Writing – Review & Editing, Visualization, Supervision, Funding acquisition.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This activity is part of the Project TEFFMA - Towards Experimental Full Field Modal Analysis, funded by the European Commission at the Technische Universitaet Wien , through the Marie Curie FP7-PEOPLE-IEF-2011 PIEF-GA-2011-298543 grant, for which the Research Executive Agency is greatly acknowledged. TU-Wien, in the person of Prof. Wassermann and his staff, are kindly acknowledged for hosting the TEFFMA project of the author at the Schwingungs- und Strukturanalyse/Optical Vibration

References (97)

  • ZanariniA.

    Competing optical instruments for the estimation of full field FRFs

    Measurement

    (2019)
  • YangL. et al.

    Vibration analysis by means of digital shearography

    Opt. Lasers Eng.

    (1998)
  • LopesH. et al.

    Evaluation of noise in measurements with speckle shearography

    Mech. Syst. Signal Process.

    (2019)
  • StanbridgeA.B. et al.

    Modal testing using a scanning laser doppler vibrometer

    Mech. Syst. Signal Process.

    (1999)
  • StanbridgeA.B. et al.

    Measuring area vibration mode shapes with a continuous-scan LDV

    Measurement

    (2004)
  • WarrenC. et al.

    Comparison of FRF measurements and mode shapes determined using optically image based, laser, and accelerometer measurements

    Mech. Syst. Signal Process.

    (2011)
  • ZanariniA.

    Full field optical measurements in experimental modal analysis and model updating

    J. Sound Vib.

    (2019)
  • MitchellL. et al.

    An emerging trend in experimental dynamics: merging of laser-based three-dimensional structural imaging and modal analysis

    J. Sound Vib.

    (1998)
  • LecompteD. et al.

    Quality assessment of speckle patterns for digital image correlation

    Opt. Lasers Eng.

    (2006)
  • CrammondG. et al.

    Speckle pattern quality assessment for digital image correlation

    Opt. Lasers Eng.

    (2013)
  • PoozeshP. et al.

    Feasibility of extracting operating shapes using phase-based motion magnification technique and stereo-photogrammetry

    J. Sound Vib.

    (2017)
  • AllenM.S. et al.

    A new method for processing impact excited continuous-scan laser Doppler vibrometer measurements

    Mech. Syst. Signal Process.

    (2010)
  • Di MaioD. et al.

    Continuous Scan, a method for performing modal testing using meaningful measurement parameters; Part I

    Mech. Syst. Signal Process.

    (2011)
  • GiulianiP. et al.

    Six degrees of freedom measurement with continuous scanning laser doppler vibrometer

    Mech. Syst. Signal Process.

    (2013)
  • Di MaioD. et al.

    Continuous Scanning Laser Vibrometry: A raison d’être and applications to vibration measurements

    Mech. Syst. Signal Process.

    (2021)
  • BenasciuttiD. et al.

    Spectral methods for lifetime prediction under wide-band stationary random processes

    Int. J. Fatigue

    (2005)
  • BraccesiC. et al.

    An equivalent uniaxial stress process for fatigue life estimation of mechanical components under multiaxial stress conditions

    Int. J. Fatigue

    (2008)
  • Mrs̆nikM. et al.

    Frequency-domain methods for a vibration-fatigue-life estimation – Application to real data

    Int. J. Fatigue

    (2013)
  • KangJ. et al.

    Digital image correlation studies for microscopic strain distribution and damage in dual phase steels

    Scr. Mater.

    (2007)
  • De la TorreI.M. et al.

    Laser speckle based digital optical methods in structural mechanics: A review

    Opt. Lasers Eng.

    (2016)
  • AvenhausW. et al.

    Gastric wall elasticity assessed by dynamic holographic endoscopy: Ex vivo investigations in the porcine stomach

    Gastrointest. Endosc.

    (2001)
  • EhrhardtD.A. et al.

    Full-field linear and nonlinear measurements using Continuous-Scan Laser Doppler Vibrometry and high speed three-dimensional Digital Image Correlation

    Mech. Syst. Signal Process.

    (2017)
  • LiuW. et al.

    The importance assessment of RDOF in FRF coupling analysis

  • QUATTRO Brite-Euram project no: BE 97-4184Tech. rep.

    (1998)
  • HelderweirtS. et al.

    Application of accelerometer-based rotational degree of freedom measurements for engine subframe modelling

  • de KlerkD. et al.

    Solving the RDoF Problem in Experimental Dynamic Substructuring

  • HaeusslerM. et al.

    Component TPA: benefit of including rotational degrees of freedom and over-determination

  • MuellerT. et al.

    Airborne transfer path analysis for an e-compressor

  • WagnerP. et al.

    Application of dynamic substructuring in NVH design of electric drivetrains

  • ZanariniA. et al.

    Using FBS for road noise NVH predictions: first outlines

  • ZanariniA. et al.

    Direct and indirect vibro-acoustic measurements for road noise NVH predictions

  • FriswellM. et al.

    Finite Element Model Updating in Structural Dynamics

    Solid Mechanics and Its Applications

    (1995)
  • LinkM.

    Updating of analytical models - Review of numerical procedures and application aspects

  • BaqersadJ. et al.

    A Noncontacting Approach for Full-Field Strain Monitoring of Rotating Structures

    J. Vib. Acoust.

    (2016)
  • PingleP. et al.

    Full Field Dynamic Stress/Strain from limited Sets of Measured Data

  • PingleP. et al.

    Limited experimental displacement data used for obtaining full-field dynamic stress strain information

  • AvitabileP. et al.

    Full Field Dynamic Stress/Strain from limited Sets of Measured Data

    Shock Vib.

    (2012)
  • ZanariniA.

    Dynamic behaviour carachterization of a brake disc by means of electronic speckle pattern interferometry measurements

  • Cited by (0)

    View full text