The pipeline industry is improving its ability to detect and locate leaks through emerging technologies. There has been a variety of research in recent years aimed at further development of sensor-based technologies for leak detection. A key obstacle to retrofitting existing pipelines with leak detection technologies is the cost and risk of installing hardware, particularly those sensors that require excavation near the pipe. There are many advantages to employing leak detection systems that can leverage existing instrumentation access locations. One such technology may be negative-wave leak detection systems. Negative-wave technologies work by measuring dynamic pressure changes in the pipe. It should be noted that some negative-wave systems require line modifications to accommodate multiple transmitters. While such systems have been on the market for many years, there is insufficient data available about their performance under various pipeline operating conditions for widespread adoption.

In an effort to close many information gaps on the performance envelope of negative-wave technologies, a PRCI-funded field test was performed on a 41-kilometer segment of a 30-inch diameter heavy crude oil pipeline. Products from three suppliers were installed at either end of the test segment. Actual commodity withdrawals were conducted at a remote valve site approximately 21 kilometers into the segment during various operations to test the systems’ abilities to detect the withdrawals without direct user interaction. These test points included withdrawals during steady-state flowing, pump startup, and shutdown conditions.

Data were collected from each system to determine their abilities to detect leaks under various conditions, abilities to locate the leak, false alarm rates, and response times. This test provided significant insight into the performance of such systems over the range of conditions tested. The key focus of this paper is the approach for conducting such multi-vendor commodity withdrawals. This project required some unique considerations for its execution. Such considerations are also documented to provide input to others who are considering such a test.

This content is only available via PDF.
You do not currently have access to this content.