Method of damage location determination based on a neural network using a single fiber Bragg grating sensor

Area of Science:

• Structural Health Monitoring
• Optical Fiber Sensing Technology
• Artificial Intelligence in Engineering

Background:

• Accurate structural damage identification is crucial for safety and maintenance.
• Traditional methods often require multiple sensors or complex setups.
• Fiber Bragg Grating (FBG) sensors offer a promising alternative for distributed sensing.

Purpose of the Study:

• To propose a structural damage identification algorithm utilizing a single FBG sensor.
• To enhance sensing accuracy through a high-speed FBG demodulation system.
• To validate the algorithm's performance on a realistic aluminum plate model.

Main Methods:

• Signal analysis using wavelet packet decomposition and a backpropagation neural network.
• Development of a high-speed FBG demodulation system (0–4 kHz) employing a tunable Fabry-Perot filter and Mach-Zehnder interferometer.
• Experimental verification using an aluminum plate model simulating structural damage.

Main Results:

• The proposed algorithm successfully identified damage location using only a single FBG sensor.
• The high-speed demodulation system improved sensing accuracy.
• Experimental results demonstrated a damage identification accuracy of up to 90.0%.

Conclusions:

• A single FBG sensor, combined with advanced signal processing and a high-speed demodulation system, is sufficient for accurate structural damage identification.
• The developed algorithm offers a cost-effective and efficient solution for structural health monitoring.
• This approach shows significant potential for real-world applications in damage detection.