Optical Multimode Fiber-Based Pipe Leakage Sensor Using Speckle Pattern Analysis.

Area of Science:

• Engineering
• Sensor Technology
• Data Science

Background:

• Water scarcity necessitates improved efficiency in water supply and distribution systems.
• Real-time monitoring tools like leakage sensors are crucial for minimizing losses in water networks.
• Advancements in fiber optics and computing have enabled the development of sophisticated multipurpose sensors.

Purpose of the Study:

• To develop and test a multimode fiber optic sensor for monitoring pipes and detecting leaks.
• To analyze speckle patterns using statistical and machine learning methods for leak detection.
• To evaluate sensor performance under various conditions, including different leak sizes and pipe configurations.

Main Methods:

• Utilized a multimode optic-fiber-based sensor to monitor PVC pipes.
• Captured speckle patterns from the fiber's outlet using a defocused camera.
• Applied statistical and machine learning analyses to the captured speckle patterns.
• Simulated leaks of 2 to 8 mm diameter under varied water flow and pressure conditions.
• Tested sensors placed inside and outside the pipe with covered and exposed fiber core configurations.

Main Results:

• Achieved an overall leak size determination accuracy of 75.8% for a 400 µm covered fiber.
• Recorded an accuracy of 68.3% for a 400 µm exposed fiber.
• Successfully detected pipe bursts, external interventions, and shocks.
• Demonstrated consistent performance for sensors placed both inside and outside the pipe, with both covered and exposed fibers.

Conclusions:

• The developed fiber optic sensor effectively detects water leaks and pipe integrity issues.
• The sensor's performance is robust across different configurations and external factors.
• This technology offers a promising solution for real-time monitoring and loss reduction in water supply networks.