A GPR-based framework for assessing corrosivity of concrete structures using frequency domain approach.

Area of Science:

• Civil Engineering
• Materials Science
• Non-Destructive Testing

Background:

• Ground-penetrating radar (GPR) is crucial for evaluating concrete structure integrity.
• Traditional GPR interpretation relies on rebar reflection intensity, which is susceptible to complex environmental factors.
• Accurate assessment of rebar corrosion is vital for structural safety and maintenance.

Purpose of the Study:

• To develop a more robust GPR data interpretation method for concrete corrosivity evaluation.
• To enhance the accuracy of detecting and quantifying corrosion-induced deterioration in reinforced concrete.
• To overcome limitations of traditional amplitude-based GPR analysis.

Main Methods:

• Implemented efficient GPR data collection and pre-processing techniques.
• Utilized deep learning for rebar recognition and Short-Time Fourier Transform (STFT) for time-frequency analysis.
• Normalized center frequency of rebar responses and applied depth correction for standardized analysis.
• Optimized and validated GPR condition mapping thresholds using ground truth data (hammer tapping, reinforcement exposure).

Main Results:

• The developed GPR method demonstrated superior performance over traditional amplitude-based approaches.
• Achieved an average accuracy of 0.80 for detecting active corrosion.
• Reached an average accuracy of 0.84 for detecting active corrosion with delamination.

Conclusions:

• The proposed time and time-frequency domain GPR analysis offers a more reliable method for assessing concrete corrosivity.
• This comprehensive approach significantly improves the detection and quantification of corrosion-induced structural damage.
• The findings support the adoption of advanced GPR interpretation techniques for enhanced infrastructure monitoring.