Optimising Sensor Placement in Heritage Buildings: A Comparison of Model-Based and Data-Driven Approaches

Area of Science:

• Civil Engineering
• Structural Engineering
• Heritage Conservation

Background:

• Structural Health Monitoring (SHM) is crucial for heritage structure preservation.
• Optimal Sensor Placement (OSP) methods are vital for effective SHM but limited in historical buildings.
• Numerical models for OSP in heritage structures face uncertainties from complex geometries and materials.

Purpose of the Study:

• To investigate the influence of different input data sources on Optimal Sensor Placement (OSP) outcomes for heritage structures.
• To compare OSP results derived from uncalibrated models, calibrated models, and data-driven modal parameters.
• To assess the suitability of data-driven approaches for SHM in heritage buildings.

Main Methods:

• Utilized the Church of Santa Ana in Seville, Spain, as a case study.
• Implemented and compared several OSP methods using three distinct data sources: uncalibrated numerical model, calibrated numerical model, and data-driven modal parameters.
• Systematically evaluated the impact of input data on sensor configuration efficiency.

Main Results:

• The choice of input data significantly impacts the OSP process and outcomes.
• Calibrated models may refine modal parameters but do not guarantee superior sensor configurations.
• Data-driven strategies demonstrated potential for enhanced robustness in OSP for heritage structures.

Conclusions:

• Data-driven OSP approaches offer a promising alternative to model-dependent methods for heritage structures.
• The reliability of OSP in complex, uncertain environments like historical buildings can be improved by prioritizing experimental data.
• This study highlights the need to consider data source quality for effective SHM system design in heritage conservation.