Qatar might be one of the smallest countries in the world, but it is the leading exporter of liquid natural gas, or LNG. In the 1970s Qatar discovered it was sitting on the third largest natural gas reserve of at least 999 trillion cubic feet.
More importantly, Qatar’s gas reserve is a non-associated reserve, meaning it is almost entirely a gas field rather than being associated with an oil field. The massive natural gas field starts on the northern tip of Qatar in Ras Laffan, and extends out eastward into the Arabian Gulf. Qatar’s bountiful oil fields are located on the western side of the country along the border with Saudi Arabia.
At the rate this natural gas is being pulled out of the ground, experts say it will be between 100 to 200 years before the supply is even remotely out of the proven parts. And there is probably more that has yet to be found.
Natural gas is colorless, odorless, tasteless, shapeless and lighter than air. When it is cooled to -162 C (-260 F) it liquefies and decreases in volume by 600 percent. Natural gas contains many elements, including methane, butane, helium and propane.
Its production is considered greener than other types of fossil fuels, and its uses vary from heating homes to powering motor vehicles. Natural gas also is used for producing petrochemical derivative products including some plastics. More recently, natural gas also is being used to produce Diesel products through a chemical Gas To Liquids (GTL) process.
While the supply of natural gas is abundant, getting it out of the ground safely and distributing it from Qatar to the rest of the world can be challenging. It’s not as simple as extracting the gas, putting it in a pipe and sending it off.
“Qatar, in parallel is investing in projects such as Education City, has also invested very heavily in developing new technologies and in building LNG processes,” says Brett Browning, Ph.D., senior systems scientist at Carnegie Mellon.
“You can ship LNG in liquid form but you can’t just sell it in bottles at the other end. You have to turn it back into a gas first, so you need more infrastructure at the receiving end than you do for crude oil. Qatar has invested in that infrastructure and is now reaping the rewards.”
Keeping LNG plants safe is where Browning and his team fit in. Qatargas and RasGas each have many LNG plants that contain multiple trains in which the natural gas is purified and then liquefied. These plants are giant networks of steel pipes, vessels and columns.
Because the raw gas contains noxious and corrosive materials such as hydrogen sulfite, it can corrode these steel pipes. The corrosion can cause a hole or wear the pipe so thin that any type of shock through the plant could cause a rupture.
“Obviously when this happens you lose gas, which is bad. You also have a dangerous situation that could lead to an explosion. It’s pretty rare, but when there is an accident, people can get killed or seriously injured and you can loose a lot of the plant infrastructure,” says Browning.
Because of this, companies spend a lot of time performing inspections to check the thickness of the pipe wall. Currently, inspections are done every few years using external devices equipped with sensors that can see through the pipe. A plant may shut down one of its trains and spend three or four weeks intensely going over it, checking what they need to check and fixing what they need to fix.
This presents challenges when there are four-story high racks of pipes situated in a way that the sensor
is not able to always get in the correct position. Some pipes also may be wrapped in insulation, while others may be surrounded by concrete.
“When this happens, you are not able to look at that section of the pipe. So you have to look at another part of the pipe and make a prediction as to what is happening in the rest,” says Browning. “But there is always an element of uncertainty.”
Browning and his team of roboticists in the Qri8 (pronounced create) lab are working on something different. In the past, people have built robot crawlers that could go through the pipes, which are typically about 400-mm in diameter, and take images. Someone would then look at those images and try to determine if there were any problems, and address them if there were.
Browning, who serves as project lead, and Peter Rander, Ph.D., from Carnegie Mellon’s National Robotics Engineering Center (NREC) in Pittsburgh, are working to automate the process.
The idea is to have a robot go through the pipe and take multiple images that could be registered and stitched together into a 3-D model, much like satellite images that can be stitched together to produce 3-D models of the ground for applications such as Google Earth. This LNG Pipe Vision would create an accurate depiction of each millimeter of pipe throughout the entire plant.
After the robot crawler moved through the pipe and completed the initial mapping, subsequent inspections would take new images, stitch them together and compare them against the initial images to look for any changes.
“We can come up with very good estimates as to how quickly the pipe is degrading, which is what you really care about. With the 3-D vision, we can map out the entire network of pipes and start to automate the process of detecting undesirable changes in the pipe over time,” says Browning.
As far as safety, this model for 3-D mapping could be a big step forward in detecting problems and fixing them before a perilous situation is at hand. LNG plants would always know the exact condition of each millimeter of pipe, and know how rapidly they are changing.
From a robotics standpoint, the core technology of 3-D registration has many applications to other domains. It can be used to detect problems in other types of pipe, including sewage systems, water pipes and electrical conduit. It also can be an effective tool in fields such as medicine where an accurate 3-D depiction can show critical details inaccessible by any other means.
“What makes this project particularly exciting is that we have a great team that is strong in research but also has a proven record of taking research ideas all the way to commercialization,” says Browning. “That means we can have a real impact on a very important industry.”
The Qri8 lab at Carnegie Mellon Qatar is focused on core research that has impact on the industry and community. The Carnegie Mellon NREC is a world leader in commercializing robotics technology.
This project is funded by a generous grant from the Qatar National Research Fund (QNRF), and involves faculty, staff and students on Carnegie Mellon’s campuses in both Qatar and Pittsburgh. The Qatar team includes Brett Browning, Ph.D.; Peter Hansen, Ph.D., postdoctoral fellow; and research engineer Mohamed Mustafa. The Pittsburgh NREC team includes Peter Rander, Ph.D.; Hatem Alismail, robotics masters student and Carnegie Mellon Qatar alumnus; and Joey Gannon, research engineer.
