Monitoring Flexible Gas Pipeline with a Microphone and Artificial Neural Networks

Abstract

Pipeline networks are complex systems of ducts transporting gas and chemical products through long distances. With the purpose to track these leaks a technique, based on the analysis of sound noises captured by a microphone and on pressure transients generated by leak occurrence, was developed. Neural Artificial Networks were applied to determine leak magnitude and leak location. The experimental results showed that it is possible to detect leaks in pipelines. The dynamics of these noises in time were used as input to the neural model to determine the location and magnitude of the leaks.