Courses

If you want to participate in the courses then choose OMA course in the registration. If you do not take the OMA course, then after the registration then contact the person directly that is responsible for the course, that is either Michael Döhler for the SSI course, and Ole Dossing for the signal processing course.

Monday 13th May: Course on Operational Modal Analysis

Tuesday 14th May: Advanced Operational Modal Analysis with Stochastic Subspace Identification 

Tuesday 14th: On Data Acquisition and Advanced Signal Processing Function

 

Please find description of all the courses and the contact person below:

Course on Operational Modal Analysis.

Program. Monday 13th May 2019

Morning. Svend Gade

  • Welcome and introduction
  • Signal vs. System Analysis
  • Operational Modal Analysis Theory
  • Frequency Domain Decomposition, FDD
  • Stochastic Subspace Identification, SSI
  • State Space Models
  • Structural Health Monitoring
  • Measurement time needed

Afternoon. Carlos Ventura

  •  Time-frequency Analysis
  •  Applications and use of OMA
  • Test planning, Testing and Data quality control
  • Application examples
  • FEM updating using OMA results
  • Summary & Conclusion

Optional program. Sunday 12th May 2019

You can participate in this part of the course without any further expense. Just write a mail to Line Leth Christiansen Lilch@byg.dtu.dk to notify your interest. This activity will take place at DTU, will cover the OMA testing of different small models, and will extend over both morning and afternoon.

Curriculum Vitae for Svend Gade:

Svend Gade joined Brüel & Kjær, Nærum, Denmark in 1980, where he has been working as an application specialist and as Associate Professor at the Brüel & Kjær University Group in the Sound & Vibration Measurement segment until his retirement in 2017. He is now a consultant with own company, Gade S&V. Gade was heavily involved with the introduction of the world’s first commercially available Sound Intensity system in the 1980es, and he wrote several B&K Technical Reviews covering this topic. He has continued working in this field with array acoustics, i.e. Acoustical Holography and Beamforming techniques. His main focus over the years has been Modal Analysis starting when the modern era of structural measurements began in the 1982 introducing worldwide the concepts of Dual Channel Analysis and input/output measurements techniques. Since 1999 he has also worked with the development of output only modal analysis, i.e. Operational Modal Analysis. His other responsibilities included digital signal analysis, sound power, signature analysis, machine diagnostics, and Sound Quality. He has presented more than 150 papers at conferences such as SAE, JSAE, Inter-noise, ICSV, IMAC, ISMA and IOMAC. Gade has written more than 50 major articles including 20 articles in Sound & Vibration Magazine and 20 Brüel & Kjær Technical Reviews and a chapter in “Handbook of Noise and Vibration Control” (John Wiley & Sons, Inc.). He has lectured extensively throughout the world including Japan, China, India, Australia, Brazil and USA. Gade holds a M.Sc. degree from the Danish Technical University in Copenhagen in electronics/acoustics in 1973 and is educated in 1980 as a lecturer as well. Gade is member of Institute of Noise Control Engineering, INCE, Society for Experimental Mechanics, Inc., SEM and New York Academy of Science.

Curriculum Vitae for Carlos Estuardo Ventura:

Dr. Carlos Ventura is a Civil Engineer with specializations in structural dynamics and earthquake engineering. He has been a faculty member of the Department of Civil Engineering at the University of British Columbia (UBC) in Canada since 1992.  He is currently the Director of the Earthquake Engineering Research Facility (EERF) at UBC, and is the author of more than 480 papers and reports on earthquake engineering, structural dynamics and modal testing.   Dr. Ventura has conducted research about earthquakes and structural dynamics for more than thirtyh years. Three of his most significant contributions in recent years are the development and implementation of performance-based design methods for seismic retrofit of low rise school buildings, a unique seismic structural health monitoring program for bridges in BC, known as the BCSIMS project, and the first network-based earthquake early warning system for schools and public institutions in BC. These projects have contributed in a very significant manner to the seismic risk reduction efforts in BC. In addition to his academic activities, Dr. Ventura is a recognized international consultant on structural vibrations and safety of large Civil Engineering structures. The quality of his research work has been recognized by several national and international awards, as well as being appointed as member of the Canadian Academy of Engineering and of the Engineering Institute of Canada, and Fellow of Engineers Canada.  He is also a member of several national and international professional societies, advisory committees and several building and bridge code committees.

Course on Advanced Operational Modal Analysis with Stochastic Subspace Identification

Szymon Gres, Structural Vibration Solutions A/S, Aalborg, Denmark

Michael Döhler, Inria, Rennes, France

The 4h short course will go through the theory and practice of:

  1. Stochastic state space modeling The state space model with its system matrices and noise terms is developed step by step, and explained why it is so useful for Operational Modal Analysis.
  2. Subspace identification of system matricesWe will develop the identification of the system matrices from measurements with subspace identification in detail, and subsequently the computation of the modal parameters. The theory behind the subspace methods is made accessible, and we will discuss their appealing theoretical and practical properties.
  3. Estimation of modal parameter uncertaintiesModal parameters computed from measurements are always afflicted with uncertainties. We will discuss the sources of uncertainties and describe a simple way to obtain the modal parameter estimates together with their standard deviations.
  4. How to obtain global models from a stabilization diagramOnce the modal parameters are estimated at different model orders together with their uncertainties in a stabilization diagram, the question is how to obtain one global estimate for each mode. Since the uncertainties indicate the quality of the estimates at each model order, they can be taken into account when computing global modes.
  5. Damage detection using the SSI frameworkThe properties of subspace identification can be used to detect changes in the modal parameters directly from the measurements, without actually computing the modal parameters. We will discuss the background of the related subspace-based damage indicator and illustrate how it works on some case studies.

Contact person: Michael Döhler , michael.doehler@inria.fr

Course on Data Acquisition and Advanced Signal Processing Functions

1.   About vibration sensors and sensor application

In experimental modal analysis, we most often think of vibration, and hence, vibration sensors is of main interests. Here we shall discuss the properties and application of displacement-, velocity- and acceleration sensors

2.   About the concept of Signals as information carriers

  • What is a signal?
    • any observable pattern. Time, spatial or multi-
      dimensional domains
    • The sought information may be deeply buried in noise or
      other, but irrelevant information
  • Why interested in signals?
    • signals are carriers of information (on structural dynamics)
  • What is signal analysis?
    • the art and science of extracting relevant information
    • data reduction (parameter representation)
    • transform to domain for easy extraction of information
      and interpretation
    • model signal to predict future behaviour from history

3.   The Fourier Domain for the analysis, understanding and signal processing tasks

The Fourier Transform is probably the single most important function in solving differential equations and signal analysis, and we shall study, in details the interpretation and application in many aspects. Such as:

  • The Concept of Negative Frequencies

and

  • Full Spectrum

 

4.   Signal Processing Functions

 

For CM systems, many signal processing functions are required. The importance of understanding possibilities and limitations for each of these functions of very high importance. Following is a list (possibly not conclusive):

  • Auto Spectrum (Power Spectrum)
  • Cross Spectrum and Frequency Response Function
  • 1/N Octave Spectra
  • Hilbert Transform
  • Modulation Spectrum (AM and FM demodulation)
  • Order Spectrum
  • Synchronous time domain
  • Wind turbine order resampling
  • Digital Filter Functions

Contact person: Ole Dossing ole@dossing.eu

Curriculum Vitae for Ole Døssing

Ole Døssing has forty+ years’ experience in structural dynamics, modal analysis, signal processing, machine heals monitoring and diagnostics and the development of software applications in these fields.

Worked 12 years with A P Moller Maersk with experimental dynamics of ships and modal analysis, another 12 years with Bruel & Kjaer, with signal processing, modal analysis and lecturing/teaching in 54 countries. 20 years with Rovsing Dynamics and G P Strategies with systems for machine health monitoring, diagnostics and prediction, and lastly with Exensor Technologies in the field of unmanned sensor network systems for detection and classification of field activities for protection of military camps and plant assets.