[ANN] StructuralVibration.jl - v1.0.0

After some months of hard work, I am thrilled to announce the release of the v1 of StructuralVibration.jl.

In this release, I have polished the API to be consistent across the package, but also with SciML conventions. But the major addition to the package is a whole set of functions related to Experimental analysis (EMA). Among those, one can mention:

  • Sdof methods (peak picking, Nyquist circle and least-squares fit)
  • Mdof methods (Least-squares complex exponential, Least-squares complex frequency, Polyreference least-squares complex frequency)
  • Stabilization diagram
  • Quality and analysis indicators
  • FRF reconstruction

For the complete list, you can read the online documentation.

As usual, PRs and issues are welcome!

Finally, the package is referenced on Zenodo for citations :wink:

Congratulations!
great documentation. What did you use to generate it? It doesn’t look like Documenter.jl

Quarto. It is stated at the bottom of the page.

As mentionned by @zdenek_hurak, I use Quarto, because I already use it for publishing my lecture notes. It is very easy to use and very well documented.

v1.1.0 has just been released!

It contains some nice additions related to Operational Modal Analysis. I have implemented Stochastic System Identification methods and extended the EMA algorithms to deal with half-spectrum. Don’t hesitate to read the release notes for a description of the changes.

Hi all,

Just a short message to announce that v1.9.1. has been released.

To remind you what StructuralVibration.jl contains, here is a list of the current functionalities:

Mechanical models

  • Discrete models
    • Spring-mass-damper SDOF system
    • Spring-mass-damper MDOF system
    • FE models of bar, rod, strings and beams
  • Continuous models
    • Longitudinal bars
    • Torsional bars
    • Strings
    • Beams
    • Rectangular plates
    • Rectangular membranes
  • State space model
    • Continuous state-space representation
    • Discrete state-space representation
      • Zero-order hold (ZOH)
      • First-order hold (FOH)
      • Band-limited hold (BLH)
      • RK4

Vibration data generation

  • Excitation signals

    • Rectangular wave
    • Triangular wave
    • Hammer impact
    • Smoothed rectangular wave
    • Sine wave
    • Half-sine pulse
    • Harversine pulse
    • Swept sine wave
    • Gaussian pulse
    • Colored noise
  • Solution for SDOF systems

    • Free response
    • Forced response due to a harmonic force or a base motion
    • Forced response due to any external force or base motion (Duhamel’s integral)
  • Time-domain integration schemes for linear second order systems

    • Central difference scheme
    • RK4
    • Newmark-beta method
    • Linear acceleration method
    • Fox-Goodwin method
    • HHT
    • WBZ
    • Generalized-alpha
    • Mid-Point rule
  • Frequency-domain calculations for linear systems

    • Frequency spectrum
      • Modal summation
      • Direct method
    • Frequency response function (FRF)
      • Modal summation
      • Direct method
  • State-space solvers

    • Time domain
      • RK4 for continuous systems
      • ZOH, FOH, BLH, RK4 for discrete models
    • Frequency spectrum
      • Modal summation
      • Direct method
    • Frequency response function (FRF)
      • Modal summation
      • Direct method
  • Measurement noise

    • Additive Gaussian white noise (agwn) with a prescribed SNR
    • Additive Colored noise (acn)
    • Multiplicative noise
    • Mixed - agwn + multiplicative noise

Signal processing

  • Measurement noise variance estimation algorithms from noisy data

    • Regularization-based estimation
    • D’Errico’s method - Link to the Matlab version
    • SNR estimation from estimated measurement noise variance
  • Denoising algorithms

    • Regularization
    • Kalman filtering
  • Signal preprocessing

    • Detrending data using polynomial fit
    • Gradient calculation using interpolation
  • Signal estimation

    • Transfer functions estimation (H1, H2, H3, Hv)
    • Welch method (PSD, ESD, Autopower, Autopower linear)
    • Signal spectrum estimation

Modal analysis

  • Experimental Modal Analysis - SDOF methods

    • Peak picking method
    • Circle fit method
    • Least-squares fit method
  • Experimental Modal Analysis - MDOF methods

    • Least-squares complex exponential (LSCE)
    • Least-squares complex frequency (LSCF)
    • Polyreference least-squares complex frequency (PLSCF)
    • Stabilization diagram
  • Operational Modal Analysis - EMA-based methods

    • EMA-based approaches (LSCE, LSCF, PLSCF)
    • Frequency-spatial Domain Decomposition (FSDD)
    • Stochastic Subspace Identification (SSI)
      • Covariance-driven SSI (CovSSI)
      • Data-driven SSI (DataSSI)
  • Operational Modal Analysis - ODS-based methods

    • Extraction of Operational Deflection Shapes (ODS) from response spectra
    • Extraction of Operational Deflection Shapes (ODS) from cross-spectral density matrices
  • Analysis indicators

    • Modal Overlap Factor (MOF)
    • Mode overcomplexity value (MOV)
    • Mode Phase Collinearity (MPC)
    • Mode Complexity Factor (MCF)
    • Mean Phase Deviation (MPD)
    • Modal Assurance Criterion (MAC)
    • Coordinate Modal Assurance Criterion (COMAC)
    • Enhanced Coordinate Modal Assurance Criterion (ECOMAC)
    • Frequency Response Assurance Criterion (FRAC)
    • Complex Mode Indicator Function (CMIF)
    • Power Spectrum indicator function (PSIF)
  • Modal synthesis

    • Estimation of the FRF/Half-spectrum residuals
    • Reconstruct FRF/Half-spectrum from modal parameters
  • Modal density estimation

    • Finite difference method
    • Global polynomial fitting
    • LOWESS
    • Cubic spline interpolation

Visualization

  • Bode plot
  • 2D and 3D Nyquist plot
  • Waterfall plot
  • General 2D plot
  • Stabilization diagram for Modal Analysis
  • Peaks plot

v1.10.0 is out! This new release adds two extensions for manipulating meshes and animating a structural field (kinematic field, mode shape, ODS, …). These extensions requires Makie.jl and Meshes.jl.

A word of caution though. This update is limited to surface/boundary meshes (i.e. using triangular or quadrangular elements), because these extensions have primarily been developed for experimental modal analysis (EMA and OMA).

v1.11 is out. This minor release adds five Makie apps designed for pedagogical purposes:

  • Free Response of a 1-DOF System: This application simulates the free response of a single degree of freedom system, allowing users to observe how the system behaves when it is displaced from its equilibrium position and then released.

  • Forced Response of a 1-DOF System: This application simulates the forced response of a single degree of freedom system, allowing users to observe how the system responds to an external force applied to it.

  • Free Response of a 2-DOF System: This application simulates the free response of a two degree of freedom system, allowing users to observe how the system behaves when it is displaced from its equilibrium position and then released.

  • Forced Response of a 2-DOF System: This application simulates the forced response of a two degree of freedom system, allowing users to observe how the system responds to an external force applied to it.

  • Newmark Method for a 2-DOF System: This application implements the Newmark method for simulating the response of a two degree of freedom system, allowing users to observe how the system responds to various parameters.

Don’t hesitate to play with it and give me some feedbacks! You can also propose other apps.

Thank you

Congrats for your work!

Do you have any experience with probabilistic models, e.g. 1D beam models with a stochastic distribution of the load or the structure? Thx.

I have some experience in random vibrations. During my Phd, I developed a method for modeling the vibro-acoustic response of structures submitted to random excitations like Turbulent Boundary Layer or Diffuse Acoustic Field.

Thx. I will give it a look.

Great documentation and cool package! One question: I am having a hard time finding some tutorials or examples on how to use this package. I think your docs (which are already very nicely made!) could profit a lot from an easy-to-find beginner tutorial.

Your docs say “This package can be used for educational and research purposes.” That might imply it can’t be used for commercial purposes, but the license doesn’t appear to limit it that way. Is this sentence just a hint about academic origins?

I think that the MIT Licence allows commercial use. What I wanted to highlight was the original purposes of the package. Despite careful review and testing of the code, there is no guarantees that it is bug free. So, if someone wants to use it for commercial purposes, it is at its own risk (like any open source and free software). :grinning_face_with_smiling_eyes:

Could you elaborate please ? I am very biaised as the author of the package and I am not the best to know what users really need.

I built the doc with the idea to provide an example for the most part of the functions, each building block. But if you had difficulties using the package, it means that it is not enough. I would be happy to improve the documentation with additional material.

I can highly recommend this talk by Chris Rackauckas where he explains what I am trying to communicate (from the timestamp up to approx. +3 mins):

Essentialy, imagine you are, like me, a beginner with respect to structural dynamics and especially StructuralVibration.jl: one or more self-contained tutorials or examples can quickly convey how the package is supposed to be used. After that, for specific questions, the broader docs are the main knowledge resource.

I will check that. I think I will have time to prepare some tutorials soon.

v1.12.0 is out!

This release brings with a lot of bug fixes (see here for the complete list).

Apart from that, a collection of Tutorials is now available. These tutorials show how to use the package in various scenarios, and provide more insights on results interpretation.