In order to better understand the dynamic behaviour of rocket launchers during flight, it is critical to analyse the vibration data recorded during flight. The classical experimental modal analysis techniques are not applicable since it is impossible to control or measure the dynamic loads applied to the launcher.


Client problem

The classical experimental modal analysis techniques are not applicable to flight data since there is no robust and precise manner of controlling and/or measuring the loads that the rocket is subject to.

To analyse the rocket's vibrations without access to the input or vibration source (blind identification), it is necessary to use "operational" modal analysis. Numerous methods have been proposed by researchers but no systematic comparison and benchmark has been conducted.

Solution proposed

After an extensive bibliography review, Cornis identified 4 methods amongst the highest performing ones and most adapted to our client operational context and needs. All four methods were implemented. Special care was taken so our client could easily use them.

To quantify the precision of the modal identification offered by these four methods and the robustness of the identification, a thorough study of their performance was conducted. They were tested on numerous signals simulated on a single degree of freedom system. Numerous modifications and perturbations were then applied to the signals that were feed to the identification methods:

  • noise,
  • harmonic parasites,
  • under-sampling,
  • signal quantification.

Cornis was thus able to develop a multi-method operational modal analysis tool with precise recommendations on the methods to use based on the quality and properties of the recorded signal.

Other missions

  • Industrialisation
  • Data processing