- The intelligent Pipeline Integrity Program (iPIPE) is a collaboration between pipeline operators from several key oil-producing states with fund matching from the North Dakota Industrial Commission and management by experts at the Energy & Environmental Research Center.
This week’s Pipeliners Podcast episode features the latest in a series of several episodes with Jay Almlie of the EERC and iPIPE consortium hosted by Russel Treat.
In this episode, you will learn about future technology in development today to support the Oil & Gas industry, such as augmented reality for a backhoe operator, artificial intelligence, ground-penetrating radar, and more.
Future Technology Today: Show Notes, Links, and Insider Terms
- Jay Almlie is a Principal Engineer at the EERC and a leader of the iPIPE consortium. Connect with Jay on LinkedIn.
- EERC (Energy & Environmental Research Center) is a research, development, demonstration, and commercialization facility for energy and environment technologies development located in Grand Forks, North Dakota. EERC is a leading developer of cleaner, more efficient energy to power the world and environmental technologies to protect and clean our air, water, and soil.
- iPIPE (the intelligent Pipeline Integrity Program) is an industry-led consortium whose focus is to contribute to the advancement of near-commercial, emerging technologies to prevent and detect gathering pipeline leaks as the industry advances toward the goal of zero incidents.
- Find more information on iPIPE members (Gas, NGL, Water, Interstate Crude Oil, and more).
- Backhoe operator drives and manipulates the bucket of a backhoe to dig and move dirt, sand, gravel, or a combination of these materials.
- Augmented Reality is an interactive experience of a real-world environment where the objects that reside in the real world are enhanced by computer-generated perceptual information, sometimes across multiple sensory modalities, including visual, auditory, haptic, somatosensory and olfactory.
- Microsoft HoloLens, also known under development as Project Baraboo, is a pair of mixed-reality smartglasses developed and manufactured by Microsoft.
- Google VR goggles is a head-mounted device that provides virtual reality for the wearer.
- Artificial Intelligence is intelligence demonstrated by machines, in contrast to the natural intelligence displayed by humans and animals.
- GIS (Geographic Information System) is a system designed to capture, store, manipulate, analyze, manage, and present spatial or geographic data.
- Technology Readiness Levels (TRL) are a method for estimating the maturity of technologies during the acquisition phase of a program.
- Ground-penetrating radar (GPR) is a geophysical method that uses radar pulses to image the subsurface.
- Entropy is a thermodynamic quantity representing the unavailability of a system’s thermal energy for conversion into mechanical work, often interpreted as the degree of disorder or randomness in the system.
- LIDAR (Light Detection and Ranging) is a remote sensing method that uses light in the form of a pulsed laser to measure ranges to the earth.
- Federal Aviation Administration (FAA) regulates all aspects of civil aviation in the United States and surrounding international waters. Its powers include the construction and operation of airports, air traffic management, the certification of personnel and aircraft, and the protection of U.S. assets during the launch or re-entry of commercial space vehicles.
- Beyond the Visual Line Of Sight (BVLOS) refers to the available sight of a pilot flying an aircraft.
- Machine Learning is the scientific study of algorithms and statistical models that computer systems use to perform a specific task without using explicit instructions, relying on patterns and inference instead. It is seen as a subset of artificial intelligence.
- Human Intelligence is the intellectual process of humans, which is marked by complex cognitive feats and high levels of motivation and self-awareness.
- OSHA (Occupational Safety and Health Administration) is an agency of the United States Department of Labor.
- Deming Method is a continuous quality improvement model consisting of a logical sequence of four repetitive steps for continuous improvement and learning: Plan, Do, Check (Study), and Act.
Future Technology Today: Full Episode Transcript
Russel Treat: Welcome to the Pipeliners Podcast, episode 119, sponsored by iPIPE, an industry-led consortium advancing leak detection and leak prevention technologies to eliminate spills as pipeliners move toward zero incidents. To learn more about iPIPE or to become an iPIPE partner, please visit ipipepartnership.com.
Announcer: The Pipeliners Podcast, where professionals, Bubba geeks, and industry insiders share their knowledge and experience about technology, projects, and pipeline operations.
Now your host, Russel Treat.
Russel: Thanks for listening to the Pipeliners Podcast. I appreciate you taking the time. To show that appreciation, we give away a customized YETI tumbler to one listener each episode. This week, our winner is Brian Peterson with Southwest Gas. Congrats, Brian, your YETI is on its way. To learn how you can win this signature prize pack, stick around ‘til the end of the episode.
This week, Jay Almlie with iPIPE returns for the last in six episodes about the iPIPE partnership. This time, we’re going to press down on the gas pedal and peg the geek meter and talk about the future of technology. With that, let’s welcome Jay.
Jay, welcome back to the sixth and final episode about iPIPE. I’m looking very much forward to talking to you about future and future tech. I get to be an uber geek with this episode.
Jay Almlie: I’m glad you’re looking forward to it. I’m just getting the hang of this, Russel. I wish there were a few more.
Russel: [laughs] Maybe we could do that in the future.
Jay: [laughs] We’ll look for those opportunities.
Russel: [laughs] We had this conversation a while back. What we talked about is the golden unicorn. What is the golden unicorn that iPIPE is looking for? What do you think is the golden unicorn? Is it out there?
Jay: You and I are understanding this message similarly. There probably isn’t a golden unicorn or a silver bullet. Let’s say there’s a number of really promising future tech pieces that we’ve got our eye on and that we’d like to encourage development in.
That’s not always easy. We’re convincing technology providers to take a basic technology and apply it to this new area, pipeline monitoring or pipeline leak prevention, that they know nothing about. That makes it a tricky space. Let’s dive in. Some golden unicorns, as you say.
Let’s say augmented reality is fascinating, what’s happening with virtual reality and augmented reality right now. If we could, for example, come up with a system where we can take GIS information, latitude and longitude of pipeline segments, in a database available just to pipeline operators and construction people, then we might be able to avoid those third party strikes.
What happens all too frequently is people are out there installing anything. It could be general construction, or it could be even pipeline construction. Sometimes, it’s self-inflicted wounds.
Someone will dig with a backhoe where they’re not supposed to, or they didn’t call 811 first. They don’t know that there’s a pipeline where they’re going to sink their backhoe in. All of a sudden, we have a spill.
If we could avoid that using augmented reality…Let’s imagine now that same backhoe operator wearing a Microsoft HoloLens or Google VR goggles or name your name brand. There’s lots of them out there now. Those goggles are tapped into a central database of all the pipeline files, knowing where their latitude and longitude and depth are.
Now, on those goggles for the backhoe operator, is painted these augmented reality stripes showing where the pipeline is. Now he knows or she knows don’t dig here. That’s a golden unicorn.
Another golden unicorn is really harnessing this emerging technology just broadly called artificial intelligence. It comes in many flavors. It comes in machine vision, machine learning, deep learning, neural nets. All of these are flavors of artificial intelligence. No one has really figured out how to fully harness it yet.
One of your guests on episode four has a good handle on it, but not fully realized yet. It has so much power. That’s why Elon Musk and Bill Gates and others are out there saying they’re scared of this, because it could have so much power once harnessed.
If we can harness it for good, for pipeline leak prevention, that’s potentially a golden unicorn. There are several others. We could spend all day. Those are two that come to mind right away.
Russel: I want to talk a little bit about the idea of augmented reality for a backhoe operator. Maybe a way to think about this conversation, for the listener’s behalf, is how do you come up with a technology or a tool? A lot of people can say, “That makes sense. If I can put on these goggles and I can see where the pipeline is, then I’m going to be able to dig around it.”
Conceptually, that makes a lot of sense, but how do you do that? If I tie it into GIS, not all my records in my GIS system are perfectly accurate. Are they accurate enough? Do they show me the location of the pipeline within a foot, within an inch?
How does that work? Is there some other way of setting it up for the dig that would better and more accurately show the backhoe operator where the pipeline actually exists?
Jay: On our last episode, we talked about this concept of defense in-depth. The Bubba geeks in us translated it to a football analogy. You need upfront linemen. You need the linebackers. You need the secondary.
In this case, yes, today augmented reality may not work perfectly, at least not across all operators. We just heard, during a recent meeting we had with our membership, that sometimes their GIS files aren’t nearly accurate enough. It’s the old as designed and as installed argument. As installed sometimes are way different than as designed.
How do we improve that? Maybe we need a defense in-depth technology where that technology is going to better ascertain where and how deep the pipelines are. Then we follow up with another defense in-depth technology, like augmented reality, that show us now that we have accurate positioning, now don’t dig here, dummy. That’s what augmented reality would tell us.
Russel: The thing that I always think about in this is why aren’t we doing something with ground-penetrating radar.
Jay: Interesting you bring that up.
Russel: [laughs] You’re looking at that, right?
Jay: [laughs] Yeah. You’re leading the question, Russel. That’s perfect. During a recent members forum meeting we had with all of our members, we asked this question. What’s on your horizon? What do you really want us as your tech scouts to keep an eye on? What do you want us to go solicit and actively recruit for technology actually to advance to a proposal stage to iPIPE?
GPR, ground-penetrating radar, came up. The platinum unicorn there, if we can take it one step further, is ground-penetrating radar has fantastic promise. Today’s ground- penetrating radar is heavy, cumbersome, rides on a cart with four wheels, has to be almost in contact with the ground to work.
We’ve got some members who have been exploring, very preliminarily, very fundamentally, very early in the TRL scale, with could they not have that in contact with the ground. Could they make it lighter? Could they pay someone to make it lighter? Could we make that aerial platform based?
Now, if you get into that, that platinum unicorn range, that’s really what iPIPE is about. How do we take some of those Star Trek technologies and operationalize them? Ideally, GPR, loaded on a drone, scanning thousands of square miles every week for an accurate, up to date location not only of the pipeline.
We’ve got some geophysicists here at the EERC who are considering how would we use that GPR signal combined with machine learning to eke out indications of a leak.
The exciting thing about that is we could possibly see a leak before it came anywhere near making a surface expression. The leak plume could be very tiny around the pipeline. Maybe we could see it with something like GPR before it ever hit the surface.
Russel: It’s always interesting to me. This is one of those things, Jay, that drives my wife crazy. I’ll ask a question like this. It’ll be everything’s sweetness and light. It’s all great and gravy.
Then I’ll come right back. I can think of “If you’re going to put a signal out and you’re going to do that, what kind of signal is that? How’s that going to affect vegetation? How’s that affect wildlife? How does that affect people?” [laughs]
I can immediately see the promise. Then I can flip that on its edge and immediately see all the other aspects that would be obstacles or objections or potential problems.
Jay: For sure, yeah. Every opportunity also has a challenge that comes with it. It’s the basic law of thermodynamics, entropy. Everything we possibly gain has some cost that comes with it. We’ve got to navigate that. That’s what operationalizing technology is all about.
Russel: It sounds easy if you do it fast. We talked about augmenting reality. What about drones? What’s going on in the drone space? What would make a drone a golden unicorn?
Jay: Drones, let’s say they are. Drones are a golden unicorn, meaning they’re not here today, but they hold tremendous promise. What if we were able to fly drones beyond visual line of sight and let them run for 6 to 12 to 24 hours at a time?
Imagine the square miles we could cover with a fixed-wing drone with a large enough payload capacity that we can carry a couple of sensors in there. Maybe one is LIDAR. Let’s really think Star Trek. Maybe the other is GPR. I don’t know.
We’re scanning constantly. Now we have an up to date view of where and how deep the pipelines are. Now we have a view of potential leak spots before they ever make a surface expression, before the plume becomes big enough to impinge on the surface where we could see it with our own eyes.
Maybe that’s the way we’re going to go. All of that, for drones, depends upon one thing. That is the FAA allowing beyond visual line of sight. They’ve got some reasons for not doing that right now. We’ve got to get there. The rest of the world is outpacing us in that area.
Russel: That’s right. This is an interesting situation, drones in particular. The technology is actually way out in front of the regulatory rulemaking for the FAA.
Agriculture has some similar kind of problems. Here’s my theory. Here’s my golden unicorn when it comes to drones. This is a control room kind of answer.
If I could preposition drones that I could fly beyond visual line of sight, if I could launch that drone vertically and fly it horizontally, a combination fixed-wing quad and it could carry an instrument package that would give me methane detection and hydrocarbon detection and beyond visual range imagery and that thing could fly…
If I had some kind of alarm, I could actually dispatch the drone and go look [laughs] and see if the alarm’s for real. To me, that holds a lot of promise. You could probably get these things to a price point where it’s achievable. The challenge is I’ve got to be able to get the FAA to allow me to fly that thing within my pipeline right of way in some way.
Jay: Let me add. Let me one-up you, Russel. I don’t normally do that to friends, but let me one-up you. What if we didn’t have to dispatch that drone? What if it dispatched itself based on machine learning telling us that there might be a possible indication of something anomalous here? It dispatches the drone. The drone flies its route. It comes back.
Russel: Listen, man. I read “I, Robot.”
Russel: I don’t want the drones thinking for themselves.
Jay: We’re joking, but…
Russel: [laughs] I know.
Jay: …it’s not too far. It’s not too far, actually.
Russel: No, it’s not.
Jay: We’re on that precipice right now. Let me take it one step further.
Jay: We’ve talked about drones. We’re working with a company right now, just very preliminary stages. We’re working with a company, in discussion, that is working with a robotics company. They want to harness their knowledge of pipeline risk and tie it to this fairly famous robotics company that is doing fascinating stuff.
What if we thought not only of aerial drones? Now we have ground-based drones to get around the BVLOS question. What if we had robot dogs? I’m using that as a euphemism. What if those robots had sniffers on their nose?
They could roam a pipeline back and forth and not be exposed to those aerial risks. They can put their nose right to the ground, sniffing for leaks. That’s something we’re exploring right now. Could that be a golden unicorn? Possibly. Everyone thinks drones as airplanes, but it doesn’t have to be.
Russel: Absolutely right. The other thing you can think about too is what about putting a drone inside the pipeline…
Jay: Well said.
Russel: …and be able to communicate with the drone. What about miniaturizing those drones so I can fly them around in the flow and look at things other than just leaks? I could look at what’s going on hydraulically and how’s a valve operating and sealing. There’s all kinds of data that you could get that’s just not available to us right now.
Jay: There is a lot of work going on in that space right now, trying to make robots small enough to crawl along the inside of these pipelines, our pipelines specifically, which we typically say…Although we have some transmission partners on board in iPIPE now, we’re still focused on that gathering space not covered by any other program.
That gathering space, gathering pipeline, means roughly three to eight inches. That’s the sweet spot. Let’s fit a drone or a robot inside that eight-inch diameter. Let’s see what that can do. That’s fascinating Star Trek technology.
Russel: It is. Again, there’s issues with all that. The more you miniaturize, the more you have challenges with putting power on it and storing data on it. If it’s inside a pipe, how do I communicate with it? All that kind of thing. It’s fascinating.
What’s fascinating to me is so much of this technology already exists in some form or fashion, but nobody’s actually trying to operationalize that technology, at least not at the level of getting it on pipelines.
Jay: I try to take off my myopic glasses and blinders. I try to look beyond just pipelines. There are a lot of promising technologies out there. We tend to think, “Why aren’t they applying themselves to pipelines?” It’s just because there are so many other opportunities too.
We just need to educate them a little bit on the needs of pipelines. They’ll dive in. They’re thinking about a thousand different applications. That’s the job of iPIPE, is to harness those and steer them and educate them toward the pipeline needs.
Russel: Here’s another thing that I’ve been noodling on. I’ll just throw it out there for general conversation. As we get these new technologies and as we start creating more and more information that’s coming to us, it seems to me that one of the challenges for us as human beings is how are we going to navigate all that data, how are we going to interact with all that information?
It can very quickly become overwhelming. What would be the management system golden unicorn?
Jay: You’re hitting on a strong point, Russel. In fact, we discussed this during episode one or two of this series. Machine learning, artificial intelligence, deep learning, all of that is playing an increasing role in this. It’s playing an increasing role in the question you just asked. How do we absorb more and more data and do something meaningful with it?
Most of the pipeline operators already have more data than they know what to do with. You’d have to hire a team of 200 analysts to do something with all that data. These pipeline operators can’t do that. It’s not economical. We need to find artificial intelligence, including big data, ways of dealing with all this data so they can extract more.
That goes beyond pipelines. It goes to improved well site performance. It goes to improved reservoir characterization. Certainly, out of the little neck of the woods, pipelines, is a part of that. How do we absorb all that data, do something meaningful with it to prevent leaks? That’s really an emphasis right now. We’re fielding more and more technologies in that space.
Maybe it helps us characterize risk so that we can apply those defense in-depth technologies. It starts with characterizing those segments of pipeline. Maybe one of the only ways to do that economically is let a machine churn through all that data that we physically aren’t capable of. So fascinating.
Russel: You said something too — there’s a mouthful there — which is applying risk. Really, what humans do well is connecting the things that are not easily connected from a…They’re not linear. They’re more abstract. We do really well with abstract ideas and concepts and figure out relationships that are not obvious. That’s what humans do well.
What we don’t do well is a repetitive task or a highly computational task. Machines do that really well. One of the challenges is going to be, in all of this, how do I look at my management systems and my standards, my programs, and start putting risk scoring or risk analysis into this to surface those things that really need to be looked at harder.
Jay: True story. First of all, we’ve fielded an increasing number of proposals that include artificial intelligence of some flavor, to the point where last year it was every single proposal. We’ve been fielding increased interest from technology providers who say, “I can harness that to help you identify the higher risk segments of your pipeline.”
That’s getting to that management piece that you keep bringing up, Russel. How do I manage my fleet of pipelines? They look like spider webs in the gathering space. They’re hugely networked lines for which I’ve got more data than I know what to do with.
How do I manage that fleet? How do I manage the decisions on it? Help me by starting me off in the right direction. Sometimes, that looks like artificial intelligence.
Russel: There’s a reason they call it artificial intelligence.
Jay: True. I think the point you’re trying to make is…
Jay: …it isn’t human intelligence.
Russel: It’s not the same thing.
Jay: We intuit better than any machine can today and hopefully far in the future, but it is a useful tool in that it helps us have to intuit less.
Russel: No doubt. Here’s how I wanted to wrap this conversation up. We’re coming to the end of six episodes. We’ve really talked about many, many aspects of iPIPE and different perspectives and such. What I wanted to ask you is, what do you see as the future of iPIPE? Where are you guys headed? What’s next?
Jay: Predictions are risky. They’re fraught with peril. I’ll make them anyway. I don’t care.
Russel: [laughs] Good man.
Jay: Let’s take a guess. Where are we at the end of 2020? We’ve grown in membership by 25 percent. We’ve grown in resources by perhaps 50 percent — cash and in-kind resources. We will have explored 10 promising technologies. We’ll have four under our belt that are those new tools. I keep saying filling those empty slots in our tool belt. That’s the mark of success.
I think that’s where we’ll be, I don’t think success ends there. Until we actually get to zero leaks, iPIPE will continue. It’s like a safety program. Where OSHA might impose a safety program with a goal of zero safety incidents, you never get to zero. You keep improving your program.
That’s actually the Deming cycle. Plan, execute, improve, and then analyze your results. You complete the circle again and again and again. We’re just going to keep getting better at our job by pushing ourselves to include more and more technology so we get better and better, so we learn more, so we don’t have to watch our pipelines as intensely because our technology is helping us.
Russel: There you go. What you’re saying is you’re going to make pipelining easy?
Jay: [laughs] No. I don’t think I ever said that.
Jay: You can say easier, and I can say we’re going to succeed. It’ll never be easy.
Russel: One of the things that this approach does do and what the future could hold is it becomes less stressful. Really, one of the things about operating pipelines that makes it tough is there’s so much you don’t know.
We have all these systems and processes and approaches to manage that reality. These tools and technologies actually are beginning to show up. They actually promise the opportunity to actually know or at least have confidence that your system knows and the system will elevate those things you need to be addressing.
That’s a radically different way of thinking about how a pipeline’s run than what the current state of the art is, at least in my opinion.
Jay: Let me try and expand on that point and just reinforce you, Russel. What we want to do is we want to apply technology so that the human brain can apply itself to the higher-level, higher-order thought processes.
We can be looking over bigger and more important things on the pipeline and let the machines do the low-level stuff so we don’t have to. In that, we’ll keep getting better as we apply that human brain to what it’s really good at.
Russel: Well said. We’re coming to the close of this episode and also the close of the six-episode series. Jay, one of things I wanted to give you the opportunity to do is, if you’d like, make some closing remarks. What would you like the listeners to hear or to be thinking about as we wrap this up?
Jay: The pipeline operators that are part of iPIPE are intensely interested in doing their job better and better and better, continuous Improvement. Why? Remember that all those pipeline operators are also landowners and hunters and fishermen and nature lovers, just like everyone else.
There aren’t people in the pipeline operations industry who are out to do harm to the environment. They’re all interested in doing better. That’s why iPIPE took root. I guess I would do myself a disservice if I also didn’t point out that you can learn more about iPIPE at www.ipipepartnership.com. Check us out.
I would encourage all the Bubba geeks out there who love this technology stuff. If you know of a technology that you think, “Hey, this could be promising for my field. I’m a pipeline operator. This could be promising,” point them our way. Let us explore that for you so that we can bring that to the forefront.
You can point them to the website. You can have them call me. You can call Russel. Any of the above. Just keep those technologies flowing to us so that we can keep this exciting conversation alive.
Russel: Jay, with that, you can expect to hear me knocking on the door and telling you what I think would help and see if I can’t come and play as well. This sounds like a fun game. I’d like to come play.
Jay: Bring it.
Russel: [laughs] Jay, thanks so much for all this information. It’s certainly my hope that the listeners can connect some dots and that we as a community can actually make meaningful progress together toward zero incidents.
Jay: This has been fun, Russel. Thank you.
Russel: I hope you enjoyed this week’s episode of the Pipeliners Podcast and our conversation with Jay Almlie. Just a reminder before you go. You should register to win our customized Pipeliners Podcast YETI tumbler. Simply visit pipelinerspodcast.com/win to enter yourself in the drawing.
If you would like to support the podcast, the best way to do that is to leave us a review. You can do so on Apple Podcast, Google Play, or wherever you listen to the podcast. You can find instructions at pipelinerspodcast.com.
Russel: If you have ideas, questions, or topics you’d be interested in, please let me know on the Contact Us page at pipelinerspodcast.com or reach out to me on LinkedIn. Thanks for listening. I’ll talk to you next week.
Transcription by CastingWords