Direct Use of the Savitzky-Golay Filter to Develop an Output-Only Trend Line-Based Damage Detection Method.

Area of Science:

• Structural Health Monitoring
• Signal Processing
• Mechanical Engineering

Background:

• Bridges are critical infrastructure requiring reliable damage assessment.
• Existing methods for damage detection can be sensitive to noise and operational variations.
• The Savitzky-Golay filter (SGF) is a signal processing technique for smoothing data and identifying trends.

Purpose of the Study:

• To investigate the direct application of SGF for damage localization and quantification in bridges.
• To propose a single-stage, trend line-based damage detection method using SGF.
• To evaluate the method's performance under various conditions, including noise and changing velocities.

Main Methods:

• Numerical simulation of a simply supported beam under a moving sprung mass.
• Acquisition and noise-polluted acceleration data along the beam.
• Application of SGF to determine signal trend lines for damage analysis.
• Development of a baseline-free approach using normalization factors and Gaussian curve fitting.

Main Results:

• The proposed SGF-based method accurately located and quantified simulated bridge damage.
• The method demonstrated insensitivity to noise and variations in load velocity.
• The baseline-free approach enhances the method's practical applicability.

Conclusions:

• The Savitzky-Golay filter offers a robust and effective tool for structural damage detection in bridges.
• The proposed trend line-based method is reliable under dynamic loading and noisy conditions.
• This baseline-free technique provides a promising advancement for real-world structural health monitoring.