Machine Learning Based Sensor Selection for Modal Testing

Publication Date

1-1-2024

Document Type

Conference Proceeding

Publication Title

AIAA SciTech Forum and Exposition, 2024

DOI

10.2514/6.2024-0196

Abstract

Modal testing is a common step in the aerostructure design process and is often conducted to verify natural frequencies and mode shapes predicted by computational techniques. Sensor placement in modal testing is crucial to being able to measure the desired natural frequencies. Existing methodologies for sensor placement can be time consuming and require an iterative approach. Using machine learning techniques like those being developed for structural health monitoring may offer a more optimal sensor selection methodology. This work contains a description of three machine learning based sensor selection methodologies and compares their performance with current techniques. Performance of the methodology is evaluated with a beam numerical model.

Department

Aerospace Engineering

Link to Full Text

Share

COinS