This week’s Pipeliners Podcast episode features Geoff Roberts and Raj Jayawardena discussing FiberSense, how it works, and what it contributes to the industry. 

In this episode, you will learn about using previously installed fiber optics to conduct readings and how all the various vibrations are able to be differentiated, classified, and acted on. 

Fiber Optic Sensing as a Service Show Notes, Links, and Insider Terms:

• Geoff Roberts is an Advisory Board Member at FiberSense. Connect with Geoff on LinkedIn.
• Raj Jayawardena is the EVP, Business Development & Transformation at FiberSense. Connect with Raj on LinkedIn.
  • FiberSense is delivering a completely new digital platform for digitizing objectsand events in cities over wide area grids, 24/7 – 365 using preexisting fiber optic lines.
• Fiber Optic Cable is telecommunications infrastructure that can also be used to monitor well production, determine producing zones, and measure fracture performance. 
• Fiber Optic Sensing uses the physical properties of light as it travels along a fiber to detect changes in temperature, strain, and other parameters.
• Leak Detection is the process of monitoring, diagnosing, and addressing leaks in a pipeline to mitigate risks.
• SCADA (Supervisory Control and Data Acquisition) though evolving quickly, SCADA is generally a software used to visualize process data, alarms, analytic results, and lately integrating video surveillance, and artificial intelligence.
• R&D stands for research and development.
• HCA (High-Consequence Areas) are defined by PHMSA as a potential impact zone that contains 20 or more structures intended for human occupancy or an identified site. PHMSA identifies how pipeline operators must identify, prioritize, assess, evaluate, repair, and validate the integrity of gas transmission pipelines that could, in the event of a leak or failure, affect HCAs.
• IoT (Internet of Things) is a system of interrelated computing devices, mechanical and digital machines, objects, animals or people that are provided with unique identifiers and the ability to transfer data over a network without requiring human-to-human or human-to-computer interaction.
• Right-of-Way is a strip of land encompassing buried pipelines and other natural gas equipment allowing them to be permanently located on public and/or private land to provide natural gas service.

Fiber Optic Sensing as a Service Full Episode Transcript:

Russel Treat:  Welcome to the “Pipeliners Podcast,” episode 268, sponsored by Gas Certification Institute, providing standard operating procedures, training, and software tools for custody transfer measurement and field operations professionals. Find out more about GCI at GasCertification.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 every episode. This week, our winner is Ernest Binkley with Pacific Gas and Electric. Congratulations Ernest, your YETI is finally on its way.

This week, we are speaking to Geoff Roberts and Raj Jayawardena about sensing as a service. In other words, using fiber optics that are already in place. Geoff and Raj, welcome to the Pipeliner’s Podcast.

Geoff Roberts:  Thank you. Good to be here.

Raj Jayawardena:  Same. Thanks for the opportunity.

Russel:  We’re really glad to have you. I think you guys are doing some interesting stuff, and I’m really excited to get talking about it. Before we dive in, I’d like to ask you guys to do a brief introduction. Tell us a little bit about yourself, what you do, and how you landed there in your career, if you would. Raj, if you don’t mind, why don’t you go first?

Raj:  My name is Raj Jayawardena. I’m an engineer by training. I did a degree in electrical engineering a long time ago, and I got into telecoms. I was into telecoms, fiber operators, submarine cables, that type of thing for more than 20 years.

Then, by a happy coincidence, I got to know about some of the acoustic and vibrational properties of fiber, which I had no idea of in my time in telecoms, which drew me into FiberSense. I’m super excited and glad I took that decision to join the business.

Russel:  Cool. I can tell by your accent. Well, I say I can tell. It sounds like you’re not from North America. Where are you actually on this podcast? Where are you at?

Raj:  I’m based in Sydney, Australia. I’m speaking to you from Sydney right now.

Russel:  I’ve never been there, but that’s on my bucket list. I want to get a sailboat, and take a tour around the Harbour one time.

Raj:  Absolutely. It’s one of those must-do things.

Russel:  Absolutely. Geoff, same question for you.

Geoff:  Sure. Also an engineer, civil engineer. Started working for Florida Power & Light, and now, NextEra. Designing distribution systems for them, both overhead, underground, working with crews. Came out to Houston with Shell Oil, then Enron, then Entergy, and then in the private equity sectors.

I’ve been in the energy and the water space for about 40 years. One of the things that’s always challenged me is, how do we make these networks more safe, more secure, more reliable?

An old contact of mine that was involved with FiberSense at the time reached out to me, and we started discussing. He’s like, “Wow, this could solve a whole lot of issues,” so it got me excited. From the electric side to the gas side to the water side, a confluence of solutions for those markets. It got me excited about it.

Russel:  Why don’t you talk to us a little bit about what FiberSense is doing? I think it’s unique. That’s why I asked you guys to come on and talk.

Raj:  What we’re doing, which is a bit different from others that are out in the marketplace using fiber optics (as a) sensing device, is that we’ve turned that capability into a service.

We provide those, the fiber optic sensing, to our clients as a subscription-type model. We tune it to the client’s specific requirements, and support that service on a continuous basis. It’s a 24/7 support model that allows us to achieve performance levels that are probably not being achieved elsewhere.

Russel:  That’s interesting. Before we dive into the details of how you guys do this as a services model, it’d be helpful for the listeners to understand what’s the current state of the art, the current understood practice for fiber optic sensing. Maybe you guys could walk us through that.

Raj:  Sure. What we’re fundamentally using is a technology or a technique called distributed acoustic sensing. What that essentially is, it taps into the acoustic and vibrational properties that are inherent in fiber optic cables, cables that are deployed in city environments across countries, even in marine environments, subsea.

The fiber itself is very sensitive to strain. Any kind of strain, minute that it may be, it’s detectable through the fiber if you have the right equipment, so we have that equipment that can connect to the fiber.

Basically, anything that’s in motion that’s in proximity to that fiber creates a strain on that fiber. The laser that we’re firing down that particular fiber strand is adjusted or the light from that laser is backscattered by that movement or vibration.

Russel:  Yes, you’re able to basically send a light beam down the fiber and measure that light reflect refraction and correlate that to movement or other things that are happening to the fibers.

Raj:  Yes.

Geoff:  As well as the location where that movement is occurring.

Russel:  Right, right. Most people in pipelining are familiar with fiber, at least notionally. The question always comes up, “Well, why don’t we just drop fiber in the trench when we build a pipeline?” That’s the currently understood practice.

You open a trench. You put fiber in the trench when you open the trench. What is the current kind of practice versus what do you guys do, and how is that different? Is that making sense to you when I’m driving out?

Geoff:  It is, and I’ll take a stab at it, and Raj jump in. There are two things. One of the things that we realized in looking at the potential opportunity was that it made sense in particular to focus on urban environments. They’re hard. When we are in the middle of nowhere, it’s easier to sense a sound, because there are less sounds competing with it.

We focus on the urban environments, which again, are challenging. The second thing we did was we really focus on using existing fiber to the extent that we possibly can, rather than going and having to build new, or it’s different if you’re building a new pipeline, that may be the right solution, put fiber in the trench.

What we really saw was there’s a way that we can leverage the existing fiber optic network, that’s a far more cost-effective way to get comprehensive coverage than you can if you have to build it out inch by inch, foot by foot. That’s part of the process.

The other interesting thing is, talking with some pipelines very recently, we were talking about this very issue. One of their comments was, “I don’t want it in the same trench, because I don’t want to have to dig it up and fix it every time we have to work on the pipeline.”

We’re going to create another issue for them. Is there another way that you can solve it, and that really feeds into part of our approach. Raj, do you want to continue?

Raj:  No, you’ve hit the nail on the head, Geoff. One of the things that we discovered, especially in cities and urban areas, is the ubiquity of fiber. Typically, there’s fiber running down every street, in a lot of cases, both sides of the street. The ability to plug into that existing fiber is the quickest, easiest and cheapest solution to create a sensing product.

Russel:  That makes a huge amount of sense, but it raises some pretty big questions, because to my mind, given what I know about fiber, there’s a correlation to how close you are to the pipe versus your ability to detect. We start talking about using existing fiber then you’re going to have some kind of distance. How do you deal with that particular challenge?

Raj:  One of the things that should be noted is there’s a variety of different fiber optic sensing solutions. We’ve developed some products that, for example, specifically look at external threats to pipelines and other linear infrastructure, such as excavation and boring, those types of machinery.

The offset of the fiber to the infrastructure and to the specific activity that’s occurring, that can actually be quite a big offset. Because of the nature of the vibrations that have been created, we can detect that tens of yards or 50 yards off of the fiber itself.

If you’re looking at things of the nature of leaks in pipelines and that type of thing, those particular events present with lower energy, and so in those situations, you would need to be closer to the physical asset, a couple of yards or five yards, offsets of that nature. It really depends on what you’re trying to achieve in terms of the fiber optic sensing itself.

Geoff:  That energy gradient that you just talked about on that pressure differential, I’ll use a contrast and natural contrast of water in a water pipeline versus a natural gas pipeline that might be at 700 to 1,000 PSI. That’s a completely different issue than it is on the water side. It creates a much higher energy gradient. It sends off basically a different signature.

I’ll back up. One of the things that I think about and describing this as I asked them the logical question of, has anyone seen the movie “The Hunt for Red October?”

Most people have. They realized it was a sub hunting each other, and there’s that whole scene where you have the person over the scope, looking for that signature of the other ship, the sound, the vibration that that other submarine makes.

Take us fast forward, where we now have the cloud. We have algorithms and machine learning, and we have the ability to separate out all that other noise and focus on the vibration that’s important.

With that comes the ability, as Raj said, to be able to look further than you might have otherwise previously. You can do more on that offset, and then it depends upon the particular asset that you’re trying to protect and the types of situations that you have there.

Russel:  That makes sense to me. Being a guy with a background leak detection and understanding how leaks are detected with SCADA systems and such and other kinds of direct detection technology, certainly, I could see where a big leak on a high pressure line could be picked up pretty quickly, because it’s going to have a very distinct signature.

Where I think the challenge becomes, like with all other kinds of leak detection, is how small of a leak can I detect and how reliably can I detect it? That’s going to be different on a six-inch line and 50 pounds of pressure versus a 30-inch line at 2,100 pounds of pressure.

The whole idea of being able to offset the fiber and still be able to detect, that’s pretty compelling, because it opens up a whole lot of possibilities particularly from an operations and maintenance perspective on the pipeline itself.

Geoff:  Correct. The other thing that you have…I’ll use an example from the waterside, and let Raj jump in, but we’re working with a client on the water side that was very concerned about mainline leaks.

For a water company, typically, they don’t know that they have a mainline leak until a crater appears in the middle of the street or a business floods out in the central business district or something like that. They were asking us to focus on what I would consider to be a sizable leak, which was four liters a second and asking us if we can help identify something that can see that, and we started working with them.

This is where the magic of really using modern technology, the cloud, etc. In very short order, we obviously identified the leak, but then we took it a step further and continued to let the system learn from itself not just here, but on every other system that we’re looking at.

Very quickly, we got down to 0.1 liters per second. That learning process and the ability to scale that globally really gives you this opportunity to get better and better at finding the things that initially you might need to be right next to, but you don’t on a longer term basis.

Russel:  That’s actually a great point, Geoff, because that’s one of the things I want to talk to you about, because when you start doing things like this in the cloud and you start applying artificial intelligence, or machine intelligence is probably a better way to say that, there’s nothing really artificial about it. 

Anyways, when you start applying machine intelligence, the bigger the data set you have, the better your algorithms can be. By doing this as a service and by running it up to the cloud and creating huge data lakes that you can use to support your analysis, there’s a level of tuning that you can do that a regular pipeline operator couldn’t get to.

Raj:  Correct. That’s exactly right. That’s a very important and a unique advantage of this type of approach to fiber optic sensing. Not only are we continuously monitoring each one of these systems that were deployed across, we’re also a very distributed system.

It’s a very distributed, continuous fiber optic sensing scheme that we put forward into the field, as opposed to pockets of point sensors that you might attach to pipelines at certain locations. We can actually see the entire breadth of the pipeline continuously.

Russel:  Yeah, that’s just the nature of fiber and the way that this technology works. You basically take that fiber, and you turn it into a long series of sensing points, and read those out six feet, one foot, whatever, but you get a long series of sensing points.

Geoff:  The other the other aspect to it, which is also unique, and it’s about our approach that we’ve chosen is it really takes a deep level of technical expertise in photonics and sonar, and sonar, and telecoms, and it’s kind of weaving that all together.

You quickly find that it’s very challenging to find the people with the right skills. It would be very hard for, if I was on utility out there, “Oh, I want to go and build this.” I’d have to think long and hard, because I’m not going to be able to attract the right people to help me think through some of the problems and the challenges that naturally come up when you’re in a dynamic environment.

Russel:  Like you, Geoff, I’ve been in the business a long time, and I remember back when all of the big pipeline companies had their own internal R&D groups, and they had the best and the brightest in every domain. When new tech came, they hired somebody to do that tech, right?

Geoff:  Yep.

Russel:  They don’t do that anymore.

Geoff:  No.

Russel:  They’re way too lean for that, and then the other thing is, there’s so much more technology, and so many more vertical disciplines that you have to have expertise in, it’s virtually impossible.

Geoff:  Correct.

Russel:  That’s not just in fiber, but that applies to a lot of things that we do these days.

Geoff:  Absolutely.

Russel:  What are the major challenges you guys are finding as you’re working with pipeliners and trying to help them apply, utilize fiber that’s already there for their operations? What are some of the challenges you’re finding?

Geoff:  I think there are probably a couple. One is, with any industry when you’re introducing a new technology, there’s generally a reluctance. It’s the, “OK, that sounds amazing. How do I know?” Part of that process, I’ll call it the education and adoption process, is one of, “We believe firmly that this works. Let’s go and test it on your system, in the real world.”

What we found is that taking that approach, it goes from, “OK, that’s interesting,” to, “Wow, it really works. OK, how do we think about building this out for my system?” If I’m a pipeline operator and I’ve got a transmission, maybe some high-pressure distribution, maybe some low-pressure distribution, where am I going to focus?

I’m going to focus on the HCA areas first. I want to focus on the transmission system, then the high pressure distribution, where those risks are, and work my way down the food chain. Maybe what I start with…

The important thing for me is I’m trying to avoid strikes. Then, I happen to be in an area with a lot of subsidence, or an area subject to seismic. How do I better understand what’s happening with my underground network in relation to those issues? Obviously, leaks always come up as an interest and an area as well. Raj, thoughts there?

Raj:  That’s right. Especially in urban environments, one of the biggest challenges right up front is that cities are very noisy.

To be able to discriminate between the things that are important to defending assets or detecting leaks, and that kind of thing, it was very challenging. To be able to filter out traffic noise, and other things that are in the background. To isolate the things that were really important.

We probably picked the toughest environment to try and cut our teeth, and develop a solution. In hindsight, the pain we went through in developing those algorithms and the filtering, and those types of things, have allowed us to now operate very confidently with high levels of performance in those types of environments.

Geoff:  That reinforces an experience I’ve had from several different discussions where many times, an individual or a team has had experience with trying to implement IoT out in the field, and discovering it doesn’t scale well.

They’ve had issues where they’ve implemented a solution, but they have a very high level of false positives or, even worse, the false negative issue.

I can’t tell you the number of pipeliners – water as well as gas and liquids – that had basically said, “We looked at the system a while ago, but there was so much noise that we ultimately had to turn it off, because it wasn’t giving us actionable intelligence.”

It goes right to the heart of what Raj has mentioned. We focus on the hard spot because we figured that if we could solve it there, then the rest is a whole lot easier. We focus a lot of attention and resources on trying to get that right.

Then, we go through a very rigorous process on new capabilities. That’s a multi-level similar to what NASA goes through in trying to adapt the new technology, etc. They go through a very rigorous process, and we do the same thing. Our experience is that’s proven to be absolutely the right approach.

Russel:  For peoples that develop and commercialize technology, they’re going to understand that conversation. For people that are users of technology, not so much.

I remember, my first exposure to fiber optic sensing was in the military when they were using it for intrusion detection. It’s rather difficult to distinguish the footsteps of a man from the footsteps of a deer, as an example.

You guys are talking about the same thing. Trying to implement this in a cityscape where you’re going to have to be able to distinguish a trash truck rolling down the neighborhood versus a backhoe stopping and operating its bucket.

When you start framing it that way, all of a sudden, you realize that these distinctions are really subtle, very, very, very subtle, but they are distinct. You got to figure out how to identify them. That’s why a big data set is important. That’s why deep knowledge in this is important.

Raj:  Yes, that’s absolutely right. One of the important things that we realized fairly early on is not to focus on one thing, not just to focus on the acoustics. There are other things that are out there in the environment.

We can look at the activity temporarily. We can look at it over time, is it changing? Is there motion? Is it in motion? When you start to bring all these things together, then you build a picture about what you’re looking at, what you’re observing.

Russel:  Being able to do things like understand ground moisture, understand ground movement, and be able to see that over time and correlate it to weather, that has huge potential for helping people more proactively manage their pipelines.

Having an understanding of, “Well, this much rain at this location is going to cause these kinds of problems,” is way different than, “Hey, we had a lot of rain. Let’s go look and see if we have problems.” It’s very different.

Geoff:  Right. A good example is we were helping solve a problem for a client in the Northeast US.

They were looking to protect the fiber optic network, going between some important locations in the Northeast. One of the things that they also asked of us is what we could tell them further about what’s going on underground. “There’s a lot of construction in the area. How do we deal with subsidence?” etc.

The team developed a great solution to be able to provide Vs30 – a 30 meters down, call it roughly a 100 feet down – and to continually assess shear waves, so you could start to see subsidence. You could start to see the impact of seismic events, etc.

To make something that was, prior to that point in time, completely invisible, but to now see it on a real time continuous basis provided tremendous insight to them. They allowed them to think about their system, and how they protected it differently.

Russel:  Wrapping this conversation up, guys, what should pipeliners – oil and gas pipeliners, in particular – take away from this conversation?

Raj:  I might kick off, Geoff. Fundamentally, it’s an offset detection capability that we brought to bear in different utility pipeline environments.

It’s a proven technology in terms of performance. Our false positive and false negative rates are extremely low when compared to other kinds of technologies, or technologies that are similar to the ones that we’re using.

We’ve developed the model that allows the pipeliners to be the pipeliners, and basically outsource the sensing and the asset monitoring, and protect protection to someone like FiberSense.

Geoff:  With that, doing it in a way that they can avoid a big capital expenditure, because it is sensing as a service. Not only do we basically deliver that capability, but we’re delivering a capability that’s constantly learning from a global foundation and just will help to accelerate their insight on their networks in a very positive way.

Russel:  That raises one more question for me and that is, if you think about doing this in a city, it’s pretty easy to visualize there’s fiber everywhere, but when you start getting into rural areas and you start talking about long-haul pipeline, I might see fiber in a right-of-way where I have other things like overhead power and so forth.

What do you all do when you start getting into these gaps of coverage where there is no existing fiber? How do you address that with your customers?

Raj:  You’re absolutely right. There are situations, especially in regional rural settings, that the fiber is not as ubiquitous as in cities. What we’ve done is we’ve actually developed some techniques to deploy fiber in a very low-cost manner.

We’ve done that in a number of situations for clients that haven’t had fiber in the vicinity or in the right place to do the sensing. We would suggest adopting those techniques, as part of the sensing solution to get the fiber infrastructure deployed in the first place.

Geoff:  I described it with a couple clients that, obviously, they had some networks coming into urban environments, and there was good coverage there. Then, there were some gaps in the coverage.

Really, we’ve been developing a plan to stitch together the points between areas where there was a lot of fiber to areas where there were none and how do we bring that all together to provide the type of service and the capability that that particular client was looking for.

Russel:  Yeah, that answers the question. I would assume that there’s an opportunity to work with your customers and help them get fiber into the areas where they don’t have it. Then, there’s other value propositions that you guys might be able to bring to bear to that to help underwrite the costs, because you have the need for backhaul telecom, too.

Geoff:  There’s also a benefit to a customer and looking at things like, “OK, right now, I’m doing SCADA whether it be microwave or cellular or whatever. Well, wait a minute, if I’ve got fiber going right by there now, I can solve that myself. Maybe, I have a more secure, more reliable, more resilient solution along with it.” There are lots of different ways to look at the opportunity.

Russel:  Look, guys, this is really interesting. My takeaway, and what I would want pipeliners to take away, is you don’t have to install fiber to get the benefit of fiber.

The whole idea of being able to put fiber in and do sensing from an offset versus having to install it directly underneath the pipe. Certainly, there is some loss in fidelity for leak detection. There’s a whole lot of other benefits to be gained.

What I would want for pipeliners takeaway is be aware of that. You gotta keep watching what people are doing with fiber because we’re just really starting to understand how to deploy this tech.

Geoff:  Agree.

Raj:  Yeah, that’s absolutely right. There are innovations happening virtually daily. We’re seeing new things, and we’re developing new algorithms and software to pull out different pieces of information. It’s an ever evolving space that we’re only starting to really get into.

Russel:  Yeah, I couldn’t agree more. Gentlemen, thanks so much for your time. I appreciate it. I’ll be very interested to hear from you in another year or two and find out what’s new or different.

Geoff:  Sounds great. Russel, thank you.

Raj:  Thank you, Russel. Great to talk to you.

Russel:  I hope you enjoyed this week’s episode of The Pipeliners Podcast, our conversation with Geoff and Raj. Just a reminder before you go, you should register to win our customized Pipeliners Podcast YETI tumbler. Simply visit PipelinePodcastNetwork.com/Win and enter yourself in the drawing.

If you’d like to support the podcast, please leave us a review. You can do that wherever you happen to listen. You can find instructions at PipelinePodcastNetwork.com.

[background music]

Russel:  If you have ideas, questions, or topics you’d be interested in, please let me know either on the Contact Us page at PipelinePodcastNetwork.com, or reach out to me on LinkedIn. Thanks for listening. I’ll talk to you next week.

Transcription by CastingWords