Bolt Preload Identification Method Based on Multi-Frequency Guided Wave Reconstruction and Spectral Centroid Fusion

Area of Science:

• Mechanical Engineering
• Materials Science
• Structural Health Monitoring

Background:

• Bolted joints are vital for structural integrity in critical applications.
• Preload loss in bolted joints can lead to stiffness degradation, fatigue, and safety issues.
• Existing methods for monitoring preload loss using ultrasonic guided waves face challenges like frequency dependence and environmental interference.

Purpose of the Study:

• To develop an early-warning system for detecting preload reduction in bolted joints.
• To overcome limitations of current ultrasonic guided wave techniques for preload monitoring.
• To provide a robust and interpretable method for structural health monitoring of bolted connections.

Main Methods:

• Utilized broadband excitation (20-250 kHz Chirp) and multi-frequency narrowband reconstruction (50-180 kHz).
• Employed spectral centroid extraction and optimized weighted fusion for preload identification.
• Developed a fixed SC-domain linear calibration framework using a reference loading group.

Main Results:

• The fused spectral centroid index demonstrated a stable, monotonic, and linear correlation with preload.
• High Pearson correlation coefficients (>0.99) were achieved when calibration was transferred to new loading groups.
• The spectral centroid method significantly improved robustness against temperature variations compared to energy-based features.

Conclusions:

• The proposed method offers a reliable and interpretable approach for early detection of preload loss.
• The developed spectral descriptor and calibration transfer framework enable repeatable early warnings.
• This technique enhances the safety and longevity of structures utilizing bolted joints.