In a previous edition Metas celebrated 10 years in operation and the market introduction of your subsea monitoring system for Oil and Gas the Wide Area Active Monitoring (WAAM) System. What is the purpose is this system?

The WAAM System is continuously monitoring a subsea field for anomalies that might be important for the operator specifically regarding integrity of their equipment, but also any unplanned operations in the area under observation, which is a radius of about 1000 meters around the sensor.

By ‘anomaly’, do you mean leakage?

Yes, that is one type of anomaly we are looking for. The system uses active sonar to gather an enormous amount of data on what is in and around the equipment under observation. The software is designed to provide alarms for specific risks, like leakage. We set the alarm sensitivity levels based on the operator’s parameters to catch any leakage at the earliest signs, to allow action before a minor problem can become a catastrophic one.

How does your system differentiate a minor leak from other noise in the area?

It is important to remember we are using active sonar, so we are only looking at reflections of the signals we send. We are therefore able to ignore frequencies outside of these energy levels. This is very different than passive acoustic systems that have been used before, where other noise could cause false alarms.

But to answer the intent of your question, we do need to differentiate the active responses we get in our energy levels. We do this by tracking the responses over time in the software, then mapping their behavior and categorizing them using machine learning. This allows us to say if a bubble of gas is a fish’s swim bladder or a leakage from the seabed or equipment.

Since the software has mapped the acoustic response of the equipment on the seabed and its location in a 3D map, it can determine if that bubble originated from equipment or from the seabed somewhere else.

What is the minimum leakage level your system can detect, and what impacts or limits that capability?

In field testing using a live leak simulator, where leakage rate could be controlled very accurately, we were able to identify leakage rates down to about 13 liters/day. We were able to identify it in a few minutes, but for comparison purposes it is easier to use liters/day.

For comparison, allowable leakage according to API 6A Annex F (acceptance criteria for subsea connections) is about 0.5 liter/day of gas at ambient temperature, that is in a factory under controlled conditions.

This means any leak we detect is already beyond acceptable rates for subsea equipment and is therefore a serious issue.

Regarding limiting factors on detection, the main challenge with active acoustics is what is known as shadowing, where equipment will block signals and prevent measuring behind them. We have solved this by integrating a pan and tilt function into the WAAM system, allowing us to look over equipment to what is rising behind it. Since we know how far away a response is and we track its progress, we can still determine where it originated.

Considering this risk of leakage and resulting pollution, does your system meet the requirements of National Regulators?

That is a challenging question, because each country has their own regulations and way of specifying requirements on integrity monitoring and pollution prevention. Currently no regulator that we are aware of has specified detection limits a system must meet to be used in their country. They rather expect that Operators will use the methods best able to detect integrity issues in order to prevent pollution and maintain safe operations.

The Petroleum Safety Authority in Norway has been pushing operators to use the latest technologies available to them or develop more sensitive methods. This has resulted in new developments like our WAAM System which was developed with Equinor (formerly Statoil) over the last 6 years.

Likewise, the Offshore Operator’s Committee in the U.S. has been working on new requirements for subsea monitoring and leak detection for U.S. basins like the Gulf of Mexico. Preliminary indications are more advanced, and sensitive sensors and techniques will be required in these regions soon.

Besides your sensitivity, why is your system better than other’s?

We have focused on eliminating false alarms, so the system can be trusted.

Equinor and other operators we have worked with highlighted false alarms as the greatest challenge in using subsea leak detection systems, so we solved it by implementing machine learning. Tracking and characterizing each response so we only alarm on real leaks!