Skip to content
Episode 39: ILI Data Integration with Marc Lamontagne episode artwork

Sep 4, 2018 32 min

Episode 39: ILI Data Integration with Marc Lamontagne

from Pipeliners Podcast Hosted by Pipeliners Podcast

The Pipeliners Podcast is continuing a series on inline inspection (ILI) with Marc Lamontagne of the Lamontagne Pipeline Assessment Corporation. In this episode, host Russel Treat and Mr. Lamontagne discuss data. Specifically, what should operators do with all of the data they're collecting? Mr. Lamontagne provides his perspective on data integration, what to look for in the data as it pertains to the condition of pipes in the field, the importance of using visualization and mapping to enhance your view of data, and other tools that can help you process the data to make informed decisions and reduce risk.

Episode details come from the podcast's RSS feed.

Embed this episode

Ready to play

Episode 39: ILI Data Integration with Marc Lamontagne

0:00 32:39
of matches

Transcript Auto-generated, may contain errors

Welcome to the Pipe Liners Podcast, episode 39. The Pipe Liners Podcast, where professionals, public eeks, and industry insiders share their knowledge and experience about technology, projects, and pipeline operations. And now your host, Russell Treet. Thanks for listening to the Pipe Liners Podcast.

We appreciate you taking the time and to show that appreciation, we're giving away a customized Yeti Tumblr to one listener each episode. This week, our winner is Michael Knight with Dominion East Ohio. To learn how you can win this signature Yeti Tumblr, stick around to the end of the episode. Mark, welcome back to the Pipe Liners Podcast.

Thank you, Russell. It's always great to be here. So I don't know if you know this, but a week ago today, we had an episode come out on the de-magnetizing pipe, and that came up in our conversation when we were talking about magnetic flux leakage. I don't know if you had a chance to listen to that, but you might want to.

It was very interesting. And I certainly learned some things that I didn't even know I needed to know about until you and I had talked about. Well, I'll have a listen. Sorry, I haven't yet.

So anyways, I asked you to come back because I wanted to talk, we've been talking about magnetic flux leakage and ultrasonic and geometric tools and the various kinds of inline inspection. What we really haven't talked about is the data other than it creates a lot. So I'd like to just kind of tee this up and ask the question, you know, as a pipeline operator, how do you deal with all this data that you collect? Well, and that has been a question and a mountain to climb for many years.

So the data from each and every run can be, you know, fairly substantial. Of course, what we're looking for is the needle in the haystack of anomalies. And you know, there are issues such as growth of crotant cracks, interacting anomalies and the like. So data integration is very important to inline inspection examinations.

And so the data that's provided by the vendors are typically in excel spreadsheet format, providing locations and sizing and comments of features within the pipelines, including not only the anomalies that you're looking for that to create the pipe, but also locations of valves and various wall thickness changes, et cetera, that may help in the location of these anomalies. So typically, this information used to be handled one at a time and anomalies gone after based on singular inspections. But more and more, the anomalies are actually being integrated with multiple runs to examine, as I mentioned, growth and interaction with other anomalies. Yeah, so I guess I have a hard time visualizing, you know, what kind of information would you have in these spreadsheets?

What would that even look like? So, you know, as everybody, I'm sure that's listening to those podcast is familiar with excel spreadsheets. There are multiple columns containing various information. And for instance, if you take a metal loss inspection, it would give you a distance from beginning of run location of the upstream growth weld, the distance from the upstream growth weld to the anomaly plus the orientation of the anomaly on the pipe, sizing information of the anomaly, and then any commentary that might be provided by the vendor based on the anomaly plus, plus all the hardwired features to the pipeline.

So there could be teas and bends and valves and anchors. Anything that might be able to be detected by the tools that are going through the pipe and including above ground markers that define specific locations for each and every inspection. Goodness. You know, it always, when we talk about these things, my mind starts kind of racing and I'm thinking about all this and I'm just compelled by how much information this is.

Mm-hmm. Yeah, and that's, you know, has been overwhelming for operators to get a handle on all this information coming in. And, you know, more and more as all computer products go, the software to integrate these runs has become more and more prevalent. Right.

Right. Certainly, you know, it's one thing, you need the numbers, but at some level, but it's hard to work exclusively in the numbers. You need a way to visualize that. Is there any kind of standard approach to visualization of this kind of information?

Well, each and every vendor has a typical visualization project product and to examine the raw data that is provided. It gets more and more difficult once you get away from the specific vendor. When you're combining information from various runs, typically these days there are either in-house packages that are more prevalent, created by the operator, as opposed to packages that can be bought off the shelf. And visualization, of course, if you can look at a joint-by-joint basis containing anomalies defined by various inspections, it's a very powerful tool when available.

So I have seen some commercial software that does this. And I'm also, I don't know if you're familiar with it, but I'm not familiar with it because I know about it. I'm not intimate with it because I know the details, but there is a data standard, the pods standard, which is the pipeline open data standard. I know that's used in the GIS world for, you know, kind of as an inventory of what's been constructed.

And that standard, do you know, is that standard used in the ILI world as a way to normalize the output data from these various inspection technologies? Yes, yes it is. As you say, it's across, at least North America. I'm not entirely familiar with the whole pod's process, but I know that it is used for, along with inline inspection information.

Yeah, there's actually a conference coming up in Houston here in a couple of months. It's recently been renamed. I'll look at the details to get it linked up in the show notes, but I know it's the conference where pods gets talked about. And I'm hopefully going to be able to get to that and walk the floor and hopefully I can find some people that may be coming to podcast and talk about those kind of details.

We shall see that that is yet to be determined. But anyways, I guess the reason I'm bringing this up, if I can get to a normal way to organize the data, it starts getting a lot easier to visualize the data. And I know from our previous conversations that kind of the best practice is to track GIRTH welds, which is the place where the different segments of pipe are put together, and then to track the anomalies, not only in terms of mile marker, but also in terms of the relationship to the GIRTH weld. Why is the GIRTH weld so critical in all of this?

Well, that is an easily definable location with the inline inspection tools, and it's a unique signature to each and every joint and a very good relative starting point to define, as you say, the locations of those anomalies. And it can be baseline from previous inspections and well-defined, not only location, but also pipe properties, manufacturer, well-signals, et cetera. And so it is, I guess, the most minute point, if you want to put it that way, that is fixed within the pipeline. And therefore, the best reference.

Yeah, so I guess it's a way, if I'm running a tool 20 miles down a segment of pipe, I can locate features more accurately because I can cross-reference what the mile marker reading I'm getting is to the GIRTH weld and find things in relation to the GIRTH weld if I'm going to have to dig and mitigate. And after baselining inspections, operators can define a specific GIRTH weld number to each GIRTH weld and therefore, each and every inspection thereafter can have the same GIRTH weld number and be more easily relatable to the anomalies. Yeah, I guess another way to think of this is that in constructing, operating, and maintaining a pipeline, really what I'm looking at is my asset is each segment of pipe is a unique asset. If I can track it from manufacture through its life, then pulling all that information together can be very helpful in terms of maintaining that and ensuring it's being operated reliably and so forth.

So I'm trying to get a picture in my mind about how would I try and represent data from these different tool runs and from these different tool technologies in a standard way? Is there, for the off-the-shelf or in-house tools, is there a standard way that people approach that? Well, generally the inline inspections themselves denote boxes where anomalies are located and is a relative position downstream from the GIRTH weld as a start point and then a length and width is provided for each and every anomaly as well as depth. But these boxes, length, width and relative position of the GIRTH weld define its location on and orientation to find its location on a single joint.

So in portraying multiple inspections, a visual tool can show a joint of pipe that is basically opened up to make it 2D from say the top of the pipe around to the top of the pipe and it will display these boxes of anomalies from the various tool runs within that joint whether they overlap or not. So, kind of like if I took a pair of scissors and I took a paper tube and I cut it along the top and then laid it flat and drew a box around every place where there was a problem or something I wanted to look at or know about on the paper tube. That's right. That's a fairly typical view of joints with anomalies.

Of course, with more computing horsepower and there are visualizations tools that are actually three-dimensional and have heat maps of showing locations of anomalies on the joints as well. So the flat note piece of pipe with the boxes on it of course is pretty rudimentary but it works. Well, I guess it's also a way to look at a lot of data quickly and then if I wanted to look at information in more detail presenting it in 3D, it would be helpful. Yes.

Being a geek, I'm sitting here thinking about VR goggles and the way to put on the VR goggles and CM3 dimensions is I run down the tool. Run with the tool down the pipeline. Well, and that is definitely being worked on and there's software for that as well presently. Well, there you go.

That's pretty cool. Is anybody using anything like that in production yet that you know of? Or is that all R&D at this point? I believe it's just, well, there are operators that are using VR type software but of course it's just getting a leg hold and just in its initial stages.

Interesting. Let's kind of move on from talking about how you visualize all this stuff and let's talk about what you do when you start finding these features that overlap. If I have multiple features that are on the same part of the pipe, I've got some metal loss, if I've got some cracking, if I've got some denting and those things are working together, what do I do with that? What do I do about and with that reality?

There are typical descriptive regulations about certain interacting anomalies and requirements to get after them in the field. So there are means by which you can actually determine the remaining strength and criticality of anomalies either through finite element analysis or other techniques and methods that have been developed over the years. And actually with this data integration, there's more than just the overlapping or interacting anomalies. There's also growth of specific anomalies that can be captured as well and taken into account.

So what would I guess this is probably one of those questions where the answer is it depends, but what would be more important, the severity of the issue or how quickly it's moving? Yeah, well, that would be, it depends, but they would potentially be hand in hand. The growth would increase the criticality of the combined anomaly even that much more. So the consideration has to be given to both and typically you look at a period of time in the future where as this progresses, what is the issue and what is the remaining strength?

Right. So how do I go through a, if I look at a new pipeline run and I've got that data's been analyzed, it's been integrated, and now I'm going to take a look at how do I work through the process of determining where I'm at in terms of the integrity of that run of pipe and what I need to do about it? Right. Well, so typically you start off with the singular anomalies and determine their criticality, whether it be cracking, corrosion or geometric anomalies.

Then you take into consideration what you found from the integration process and whether there's some combined anomalies as well, then you take that into consideration and then beyond that you look at some trending issues such as particular orientations based on product and the other influences such as elevation changes. Do you have a higher density of anomalies at say low points within the pipeline? Do you have a high voltage line running close to a segment of pipe? So you know what other interactions external to the pipeline could cause issues, which you want to get to the root cause of when you're repairing a pipeline.

Yeah, and I guess that also raises another question. I know that it's really only been recently in my experience within the last couple of years that I began to understand how these kinds of things are identified and repaired. So maybe you could talk a little bit about when you find, when you make a decision and you look at something and you say, well, I've got an issue going on at this particular joint pipe at this particular location. And based on my calculations, I'm beginning to lose my ability to hold pressure.

What do you do to mitigate that? What's the process for fixing the issue once you identify? Well, of course you have to get together with operations and contractors as well to be able to ex-gulate the pipe. And you want to do so safely.

So the consideration is given to reductions in pipeline pressure while the excavation is taking place. And based on the type of anomaly, the criticality of the anomaly, there are different reparations that can happen. If the anomaly has an appropriate remaining strength, the pipe can be cleaned, sandblasted, and recoded. If there is a potential for growth, whether it be an internal anomaly or an anomaly that is more critical, then a sleeve can be put on the pipeline to reinforce in that particular location.

If it's a widespread issue where there is a number of frozen features over a greater area, then pipeline recoding can take place over a overly required distance. So there are a number of ways to treat anomalies. Liquid pipelines typically recode or sleeve depending on criticality. Whereas gas pipelines, you have more of an option to actually replace a joint pipe if it's substantial type anomalies.

And again, that depends on operations and location, et cetera. I didn't even think about that, but it's easier to, I mean, if I have to empty a oil pipeline in order to replace a joint, that's a much different kind of problem than I need to empty a gas pipeline. Yes. Because a gas pipeline, I just vin it.

So the whole pipeline, I've got to figure out a way to pump out the oil. Whatever looking, even if I first get the oil out and put water in, I still got to get the water out before I can replace the joint pipe. Yeah, that's very interesting. Very interesting.

So this kind of brings another question to mind as we talk about that. And that is, how do you prioritize all this work? I've got to, I'm sitting here and listening to this, I've got to assemble the data, I've got to analyze each run, I've got to analyze interactions and I'm finding because of multi-runs, I've got to perform calculations and determine severity, I've got to figure out what the appropriate mitigating action is. And I would suspect that the pipeline of any size at all, and particularly if there's any age involved in the pipeline, that how do you prioritize what you're doing?

Well, for each of the inspections that are received, you have to look at the immediate concern. So you look at depth of anomaly and you're given a specific depth over which it becomes an immediate concern. Similarly, the remaining strength at each of the locations of the anomalies is taken into account to look for immediate concerns. So for instance, if there is a anomaly that has a remaining strength of less than 110% of the maximum operating pressure, then you would consider that as an immediate excavation or anomaly that might be greater than 80% in depth would be considered an immediate excavation.

So there are those defined immediate requirements, then you have to consider growth, either like corrosion growth or fatigue for cracking, and look at it as well as looking at its depth and remaining strength, remaining life as the growth continues based on your operations or the location of the pipe based on for corrosion growth, whether it be a heating influence or strictly a soil influence or even the product causing the growth. So and then those are prioritized based on the calculated remaining life and or depth that achieves within the next few years before another reinspection occurs. How far apart are these inspections typically? Well, there are rules by code depending on what industry you are in that inspections have to take place within a certain period of time.

But then as you go through your data integration and you determine the remaining life of anomalies, whether it be depth or by depth or by remaining strength, then you can base your reinspection interval based on those calculations as well. And it's often a practice that the reinspection interval is less than that mandated within regulations and based on calculations of the anomalies found within the in-line inspections. Wow. I have a whole new appreciation for what people do in integrity management.

There's a lot of information. I mean, ILI is just one portion of it. The other data integration can be that can be brought in, of course, is cathodic protection and all those measurements along the pipe. Even maybe flow characteristics of certain products have some sort of issue.

The temperatures of the products, there's a whole myriad. Yeah, I could see where soil condition and soil movement and even weather cycling, I mean, all of those things are going to have impact on the serviceability of the integrity of the body. That's right. So it just, you just start thinking about it and it's like, oh, well, yeah, there's this and there's this and there's this and there's this.

It's really, I guess for somebody who's kind of a novice in this domain, frankly, it's a bit overwhelming. It can be and to try and get a handle on all this information and put it together so that it is aligned correctly. It's quite a chore. Yeah, yeah, no doubt.

And of course, it's very critical to the safe operating of pipe. If you look at the major incidents that we've had in the last five or ten years, almost all of them have an aspect of a mechanical failure in them. And it could be for all kinds of reasons, but there seems to always be an aspect of that. So there's always an aspect of the inspection or the integrity management program that needs to be looked at following one of these incidents.

That's right. That's right. There's, you know, unfortunately, there are incidents, but learnings do come from incidents and and and the processes hopefully improve based on those. Yeah.

Yeah, well, I definitely think you see that. I mean, you can in fact, if you follow the FIMSA statistics, you can actually see it in the statistics for the industry that there certainly seems to be improvement. I'm always a little, I don't like patenting the industry on the back or, you know, claiming victory because you never know when the next problems around the corner. It's one of those things that you're always striving to be perfect and yet, you know, perfection is not really possible in the big scheme.

So, but yeah, it's it's very interesting. I think of the of all the things that we've talked about, the one I think that maybe you and I need to have some conversation and figure out a way to dive into this a little bit more incrementally because this data integration conversation, it seems like the biggest of the conversations we've had so far. Yeah, definitely. I mean, there's there's there's tool technology is of course interesting and and ever progressing, but the data integration is certainly a huge portion of integrity management and and very important.

Yeah, I'm kind of surprised that that you know, as we're talking about visualization and being a software guy myself, I'm kind of surprised that there's not, you know, some off the shelf tools for this might be an opportunity. Who knows? Well, you know, there are a couple couple tools that can be can be purchased by operators, but it is it is a version and converging area. Yeah.

Well, hey, Mark, again, as always, thank you for participating. I'm going to do the normal thing I try to do where I try to get, you know, sum it up in three key takeaways. I guess the first takeaway is as I've said a number of times in our conversation, it's a lot of information and pulling it together and, you know, combining it is really critical to an effective program. I think the second part that I take away is that there really needs to be people who have knowledge of the particular program and the particular pipe and that that is really critical.

You kind of have to have a, I guess there's a fair amount of art in this just in terms of understanding the reality of a particular situation for a particular run of pipe in terms of, you know, where it's operated, what fluids it's moving, how it was manufactured, how it's being maintained, all that. And then lastly, ideally there's a risk management approach to all this because you can't keep it all perfect. You have to make decisions about how to deploy your resources and, you know, where to be looking and how to be looking. So that gets us my three takeaways.

Pull the data together, you know, understand what's important about your specific operation and, you know, have a program to prioritize. So you get to the most important stuff first. Sounds easy if you say it fast. Yeah, very good summary.

That's excellent. Yes. Right. Well, again, thanks, Mark.

And I'm certain we'll be having you back. Oh, well, thank you Russell and enjoyed it as usual and have a great day. All right. Thanks, Mark.

You too. I hope you enjoyed this week's episode of the pipeline is podcast and our ongoing conversation with Mark LaMontaine. Just a reminder before you go, you should register to win our customized High Pliners podcast Yeti Tumblr. Simply visit pipelineerspodcast.com slash win to enter yourself in the drawing.

If you have ideas, questions, or topics you'd be interested in, please let us know on the contact us page at pipelineerspodcast.com or reach out to me on LinkedIn. Thanks again for listening. I'll talk to you next week.

Frequently asked questions

How long is this episode of Pipeliners Podcast?

This episode is 32 minutes long.

When was this Pipeliners Podcast episode published?

This episode was published on September 4, 2018.

Is there a transcript available for this episode?

Yes, a full transcript is available for this episode. You can read the complete transcript on the episode page.

Can I download this Pipeliners Podcast episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.