Introduction

With a shift from fossil-based systems to renewable energy sources also comes a shift for the transportation. This is also true for pipeline leak detection systems.

Some pipelines transport 100% hydrogen while others a blend with natural gas. CO2 can be transported as gas, liquid or in supercritical state. As a result the leak detection system needs to be able to work with these different mixtures and phases.

For classical leak detection system this is a problem. Negative Pressure Wave leak detection systems are affected due to damping in gas. Classical statistical systems are affected due to compressibility and linepack changes. This leads to long detection times or high thresholds. The systems will react with a loss of sensitivity or in the worst case with false alarms.

This is a typical application for RTTM based systems, which try to consider all the different products and pipeline properties in its calculations. However, if the fluid cannot be accurately modelled by the system this leads to false results.

In the Extended Real-Time Transient Model technology (E-RTTM) these disadvantages are compensated by Signature Analysis. Signature Analysis makes the system less sensitive against inaccuracies in RTTM calculations. Thus challenging pipeline conditions, product parameters or mixtures, e.g. hydrogen blends, leading to inaccuracies become less important. In PipePatrol the RTTM is used for filtering purposes and leaks are detected by pattern recognition techniques.