Mode Decomposition using Principal Component Analysis
This example file shows mode shapes identifcation using principal component analysis (PCA) of 2DOF system subjected to gaussian white noise excitation with added uncertainty (also gaussian white noise) to the response. However it should be taken into acount:
-Eigenvector of PCA are orthognal due to the symmetry property of covariance matrix.
-Modeshapes are only orthognal if the matrix inv(M)*K is symmetric.
-PCA will identify the real mode shapes only if they are orthognal which means inv(M)*K is symmetric.
By changing mass matrix M=[2 0; 0 1]; instead of identity inv(M)*K will not be symmetric even stiffness matrix is symmetric and PCA will fail to identify the real mode shapes.
References:
---------------------
[1] Al Rumaithi, Ayad, "Characterization of Dynamic Structures Using Parametric and Non-parametric System Identification Methods" (2014). Electronic Theses and Dissertations. 1325.
https://stars.library.ucf.edu/etd/1325
[2] Al-Rumaithi, Ayad, Hae-Bum Yun, and Sami F. Masri. "A Comparative Study of Mode Decomposition to Relate Next-ERA, PCA, and ICA Modes." Model Validation and Uncertainty Quantification, Volume 3. Springer, Cham, 2015. 113-133.
Cite As
Ayad Al-Rumaithi (2024). Mode Decomposition using Principal Component Analysis (https://www.mathworks.com/matlabcentral/fileexchange/71878-mode-decomposition-using-principal-component-analysis), MATLAB Central File Exchange. Retrieved .
MATLAB Release Compatibility
Platform Compatibility
Windows macOS LinuxCategories
- AI, Data Science, and Statistics > Statistics and Machine Learning Toolbox > Dimensionality Reduction and Feature Extraction >
Tags
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!Discover Live Editor
Create scripts with code, output, and formatted text in a single executable document.