SHM-EOVLatentVariables

Code to reproduce paper results (or as close as possible, depending on data-availability). The scripts implement both latent variable methods and a conventional fully explicit EOV procedure to address environmental and operational variability in vibration-based SHM.

Code:

• ETH_Blade_visualization.py: Plot all PSDs from the dataset between undamaged and damaged conditions.
• latent_estimation_x.py: Proposed latent variable approach only considering DSFs (X).
• latent_estimation_x_xi.py: Proposed latent variable approach considering both DSFs (X) and EOPs ($\xi$).
• ols_regression.py: Explicit EOV procedure which considers EOPs ($\xi$) to mitigate the effects observed on DSFs (X). This procedure is used to benchmark the results obtained.

Congress paper:

• A latent variable approach for mitigation of environmental and operational variability in vibration-based SHM – A linear approach (EWSHM, 2024)
  • Estimation of latent variables (DSFs and EOPs) of structures to obtain a comprehensive baseline model.
  • Avoid common issues with measured EOPs and DSFs: noise-influence, collinearity and/or acting EOPs not acquired during measurement.
  • Application to a small-scale wind turbine blade.
  • Presentation of the work at EWSHM2024.