Recorded on-site at PSIG 2019 in London, Russel Treat and Giancarlo Milano of Atmos International discuss the factors that go into determining batch arrival time of product that is moved through a pipeline during slack line operations.
This insightful conversation includes a discussion of the variables that affect batch delivery, the importance of communicating accurate ETAs to operators, the models and calculations used to support accurate arrival time, the technology used to enhance accuracy, and much more. Prepare to geek out over the confluence of math and physics!
Determining Batch Arrival Time: Show Notes, Links, and Insider Terms
- Giancarlo Milano is a Principal Simulation Engineer at Atmos International. Connect with Giancarlo on LinkedIn.
- PSIG (Pipeline Simulation Interest Group) facilitates the interchange of information and the advancement in modeling, simulation, optimization, steady-state and transient flows, single and multi-phase flows, and related subjects when applied to fluid pipeline systems.
- The 2019 PSIG Conference was held May 14-17, 2019, in London. Included was a Pipeline Simulation Short Course.
- NEWLY ADDED: Access Giancarlo’s 2019 PSIG whitepaper presentation, “Evaluating Different Approaches for Tracking Batches in a Multi-Product Pipeline During the Presence of Slack.”
- Presentation Preview: This paper compares an incompressible and empirical approach against a full model-based calculation of liquid pipeline behavior under the presence of slack for the purpose of accurately tracking batches in an online system and providing accurate Estimated Time of Arrival alarms to minimize product contamination. During the presentation, side-by-side results of each approach will demonstrate the impact of each method driving to a final conclusion.
- Access Atmos International’s Jason Modisette’s 2018 PSIG whitepaper presentation, “A Slack Flow Model With Moving Regime Boundaries.”
- Slack line is a condition when both liquid and vapor exist in a liquid pipeline at the same time. A similar term is column separation.
- Open-Channel Flow refers to when the liquid and vapor are flowing at the same time.
- Batching is the process of transmitting the same type of product (jet fuels, gasoline, diesel fuel, crude, etc.) through a pipeline to an operator’s delivery location for an extended and planned period of time.
- Batch train refers to sending different batches of product back-to-back (e.g. light crude, then heavy crude) and needing to prevent a mixture of the products.
- Volumetric Rate (volume flow rate) is calculated as the volume of fluid that passes through a certain point per unit of time.
- The Conservation of Mass Theory states that the mass of the product must equal the mass of the reactants in a system.
- The Conservation of Energy Theory states that the total energy in a system remains constant — or conserved — over time. Therefore, energy cannot be destroyed; it can only be transferred from one form to another.
- Flow Meters measure the amount of substance flowing in a pipeline and perform other calculations that are communicated back to the system.
- The Real-Time Transient Model (RTTM) for leak detection simulates the behavior of a pipeline using computational algorithms. The model, which is driven by the field instrumentation, monitors discrepancy between the measured and calculated values potential caused by a leak. RTTM uses flow, pressure, temperature, and density among many other variables.
- Liquid Drop-out is a cross-sectional area of a pipeline that is no longer full with liquid; now it is full of gas that opens and expands as the pressure in the line drops.
- DRAs (Drag Reduction Agents) are additives used in pipelines to reduce friction and increase the flow of the product.
- NTSB (National Transportation Safety Board) releases pipeline safety recommendations. The latest recommendations are part of the 2019-2020 “Most Wanted List.”
- NTSB Recommendation P-11-014 is directed at PHMSA for gas pipeline operators to look at tools and processes for identifying and locating leaks. PHMSA plans to address this recommendation in the “Pipeline Safety: Safety of Gas Transmission Pipelines, MAOP Reconfirmation, Expansion of Assessment Requirements and Other Related Amendments” final rule, which is expected to be published in August 2019.
Determining Batch Arrival Time: Full Episode Transcript
Russel Treat: Welcome to the Pipeliners Podcast, episode 77, sponsored by Gas Certification Institute, providing training and standard operating procedures for gas and custody transfer measurement professionals. Find out more about GCI at gascertification.com.
[music]
Announcer: The Pipeliners Podcast, where professionals, Bubba geeks, and industry insiders share their knowledge and experience about technology, projects, and pipeline operations.
[music]
Announcer: Now your host, Russel Treat.
Russel: Thanks for listening to the Pipeliners Podcast. I appreciate you taking the time. To show that appreciation, we are giving away a customized YETI tumbler for one listener each episode. This week our winner is Amber Stone with TECO. Congratulations, Amber, your YETI is on its way. To learn how you can win this signature prize pack, stick around to the end of the episode.
This week on the Pipeliners Podcast, I get to sit down with Giancarlo Milano, who’s been with us before to talk about leak detection, and talk about using hydraulic math to determine batch arrivals time in refined products pipelines where there is slack line operations. We had this conversation at the PSIG conference in London, and I hope you enjoy it.
Giancarlo, welcome back to the Pipeliners Podcast.
Giancarlo Milano: Thank you for having me Russel, this is definitely good to be back here, this time on a different note.
Russel: This time we should talk about where “here” is. Actually, Giancarlo and I are in London at my hotel. We’re at the end of the second day for Giancarlo and the end of the fourth day for me at the Pipeline Simulation Interest Group. I’m kind of new to PSIG — this is my second year — but Giancarlo’s a bit of an older-timer, right? How long have you been coming to PSIG?
Giancarlo: I’ve been coming to PSIG for about five or six years now. Compared to some of the guys who have been coming here to the conference, it’s not too much.
Russel: I think most of the board members have probably been doing this for 30 or more years, it’s crazy.
Giancarlo: It is, yes, yes. This year marks the 50-year anniversary of the conference.
Russel: That’s pretty amazing when you think about it.
Giancarlo: It’s been going around for a while.
Russel: For those that don’t know what the PSIG is, why don’t you tell them what the PSIG is.
Giancarlo: PSIG, as Russel mentioned, the Pipeline Simulation Interest Group, it’s an organization that is not funded. It’s a group of individuals that work in the industry doing hydraulic simulations for pipeline, as the name implies.
They pretty much focus on looking at what the technology is doing, what type of problems are out there, and how you can solve them through simulation. Some of the papers that we typically see, some of them are very heavy on the technical levels, while others are a little bit more of an overview, or how to attack the particular problem.
As you’re going through the different dates and the different papers, you encounter problems in liquids, problems in gas, and how you can use hydraulic simulation tools to address them. It’s very interesting.
Russel: Yes. I spent the first two days of this week-long get-together is the Simulation Short Course. I’d actually been looking forward to it. I’ve been wanting to come to it for a while.
I’ve got to say, man, I fancy myself fairly well educated and fairly intelligent, and I felt like I was drinking through a fire hose. It was basically going through your last two years of thermodynamics in two days.
Giancarlo: Yeah, yeah, absolutely.
Russel: It’s intense.
Giancarlo: Yeah. It definitely feels as if you’re going back to university, and you’re going through that thermodynamics, fluid dynamic course, and you’re about to take a final.
[laughter]
Russel: Yeah, exactly. We really didn’t spend a lot of time diving into the details of running calculations. It was just more understanding what are all the different fluid properties, and what are all the calculations and equations of state that are available for calculating that.
Then, looking at single-phase, and then looking at multi-phase, and how do you do that for leak detection? It was pretty comprehensive, really.
Giancarlo: That’s good. It’s definitely a lot of information and material to just cover in two days.
Russel: It is. Your degree is in mechanical engineering, right?
Giancarlo: Electronics Engineering. But, I did focus a little bit more on the instrumentation side at the beginning, so measurements, and on the pipeline, how to take that date into the whole process, and then focused a little bit more on the dynamics and the mechanics, so how the product actually moves through the line.
When I started in the industry, I had a little bit of the background on how to measure it, what to measure, what to look for. It was a nice fit when I started.
Russel: That’s one of the things I’ve found is that my measurement background has been extremely helpful in this context, although it gets in the way sometimes. In measurement, you try all these things that you’re trying to calculate that happen in the pipeline. You try to eliminate them when you’re doing measurement, right?
Giancarlo: Absolutely. You talked about the two days of this course, for me, when I started with Atmos, I pretty much had Jason.
Russel: Yeah, Jason Modisette.
Giancarlo: …Jason Modisette, next door to our office, so to me, it was over a more, longer period that training, just hopping over, knocking on his door. You get in there, you’re asking your five minutes of questions, and the next thing you know, it’s two hours later.
[laughter]
Giancarlo: You’re devolving into these equations and these concepts.
Russel: Yeah, I could very much see how that would happen, very much see how that would happen.
Giancarlo: It’s fantastic. It’s really, really amazing.
Russel: I think the thing that amazes me the most, Giancarlo, here the most is that just how many different ways people can take these equations and apply them to do something useful.
Giancarlo: Yes.
Russel: We’re going to talk about your paper here in a second. I saw papers on calculating a fluid drop out and recovery in piping streams, and trying to optimize a pipeline system, looking at that. We talked about doing modeling to try and figure out how best to run smart pigs in a gas system. That one was really interesting.
The most interesting of that was, here is this nice, smooth line, where they predicted the speed of the pig, and then there was this [laughs] the very much, not smooth at all line showing what the actual speed was going through the pipeline.
They had a nice little sidebar with the lady that presented that paper, and a number of others. It’s kind of unlimited.
Let’s talk about your paper. I am looking at your paper. It’s evaluating different approaches for tracking batches in a multi-product pipeline during the presence of slack. That’s a whole lot of words. Why don’t you put that title into layman’s terms?
Giancarlo: That is a whole lot of words. This is something that we’ve been working on over the last few years. The thing is — when we are working on being able to track how different products come through the pipeline, you want to make sure that you know when those batches are arriving at a specific location.
Pipelines, they can be multi-product, so you can send a batch of a heavy crude followed by a batch of light crude, followed by a batch of gasoline or diesel. It’s something that’s called a batch train.
Russel: Yeah, you used that term in your presentation. It’s very descriptive.
Giancarlo: I think that’s the best way that describes it, because you pretty much sending one behind the other one. You have these two products, and they come together. They’re moving through the line. There’s a little bit of mixing in there and contamination.
You want to make sure that when that pure batch arrives at its delivery location — and by pure, I mean the part where it’s actually that product and not a mixture of it — you want to make sure that you have estimated time of arrival to provide to the operator.
The idea is that when he gets an alarm — he gets an event — to let him know this batch is within 30 minutes of arrival so that he can get ready. What you want to do — what the operator wants to do — is be ready to switch that valve to deliver to the corresponding tank.
You don’t minimize. You don’t want to put crude on a refined products or jet fuel tank, because that’s going to affect that product. If you do it the other way around, maybe the jet fuel into the crude, that’s not so bad.
Getting to know when this batch arrives is highly critical.
Russel: Yeah. I think people that know particularly pipeline, would understand that. I think what people probably don’t know is just how complicated knowing when a batch is going to arrive actually is. All the things that could impact that timing and make a difference by a number of minutes, that number of minutes is material, right?
If you’ve got a long haul pipeline, and you’ve got a terminal sitting off of it, and I’ve got to take 10,000 barrels of jet fuel off the line, I need to know when to open the valves and when to close the valves to make that happen correctly, and 5 or 10 minutes could ruin the whole batch.
Giancarlo: Yes, absolutely. The other thing is tracking these batches. When you think about a pipeline, all a pipeline really is, it’s a big storage facility. It has a fixed area. It has a length. It has a fixed volume where you can put product in it.
Then, you’re pushing it through, so you’re moving that product consistently. If you have a pipeline that is nice and level, there’s no elevation changes, kind of like that drive from San Antonio to Houston, you put a pipeline there, everything is great. There’s no changes. You’re just pushing through a straw, right?
Russel: Right, exactly, so if the pipeline is straight, level, consistent internal diameter, consistent internal shape — meaning there’s no ovality or anything — and consistent smoothness, then it’s pretty simple.
Giancarlo: Yes.
[laughter]
Russel: The problem is this is never the case.
Giancarlo: That’s never the case, yeah. That’s when you need to bring theory into actual practical reality here, or the other way around, and that’s what we did.
We had this pipeline. It’s going through a very drastic elevation profile, and then what happens is when you have this product going over the top of a hill, as soon as that batch reaches the other end of that hill, you have that column separation, that slack, that’s actually changing the product from liquid into some sort of gas, the gas phase. It’s no longer liquid. That section of the pipeline is no longer a liquid full, which is one of the main assumptions that real-time transit models do.
When that happens, if you are thinking that this section of the pipeline over this hill is full or product, but it’s not, then what that’s going to happen is your model is going to assume that your batch is way further upstream than where it actually is, and…
Russel: Yeah, so let me try and paint a picture. If you think about this — it’s rising up the hill, it’s kind of like you’re sucking on a straw. There’s a boundary of the fluid, and it’s coming up to the top of the hill.
Giancarlo: Correct, yes.
Russel: Then, if you think about when you get a straw, and you put your thumb over it, and then you pick it out of the liquid, and you let your thumb off, and it just drains out of the straw…
Giancarlo: Automatically drains.
Russel: …that’s exactly what’s happening, except that when you do that with a straw, you’re mixing with air. In this case, what’s happening is the property of the fluid’s actually changing.
Giancarlo: Correct.
Russel: It’s going from a pure liquid to a combination of a gas…It’s actually a combination of a gas and a vapor and a liquid.
Giancarlo: And a vapor, yeah. It’s about the product finding that critical point when it changes phase from liquid to gas, and obviously, that gas or different phase product needs to come from somewhere.
The product is actually changing its properties into a different phase of it, but for the purpose of the batch, the batch, it’s speeding through that hill, coming down really fast. The head of that batch is going to be more much further downstream, and if you don’t take that into account, your ETA’s for that head is going to be way off.
Russel: That’s actually making me think of something. I want to ask this question.
Giancarlo: Okay.
Russel: You’re actually moving the same amount of mass?
Giancarlo: Yes.
Russel: The mass is moving at the same velocity, but the nature of the mass is changing. Is that right?
Giancarlo: When it goes over the hill, it actually will speed up.
Russel: Yeah, so the velocity’s going to change.
Giancarlo: The velocity changes…
Russel: The nature of the product, the mass velocity stays the same.
Giancarlo: It stays the same, but the mass velocity… the volumetric rate and the velocity of it does change.
Russel: I just had an engineering nerd moment. When I was in college and had got to this understanding, it wasn’t as big a deal. Now it’s like there’s something gratifying when you get to that understanding, is that the mass velocity’s staying the same, but the pure velocity is accelerating because the fluid state is changing.
Giancarlo: Another analogy that, when I started looking into slack and vapor pockets and bubbles, there’s many terminologies to define this phenomenon, let’s call it that way.
If you think about open-channel flow, you pretty much have an open-channel flow, not a pipeline, like you take a pipeline, and you cut it in half.
Russel: Like a ditch.
Giancarlo: Like a ditch, yeah. You send some water through it, think about that, and then think about that coming to the end of the waterfall. As soon as that water hits that peak of the waterfall, and it starts to drop, the velocity of that fluid is just slamming down, based on gravity and all the other effects.
If you think about that analogy, you can think a little bit about slack, but with a differentiation that it’s open, so it’s using the air that’s already out. Here, there’s no air within the pipeline.
It’s a magnificent phenomenon, and it actually is very dangerous for pipelines. There’s a lot going on physically there.
Russel: There’s a lot of complexities with pipelines…
Giancarlo: There’s a lot of complexity, a lot of complexity.
Russel: It’s not nearly as simplistic as the open channel of water analogy.
Giancarlo: No.
Russel: It works, because you can visualize the water accelerating.
Giancarlo: It’s a way to visualize.
Russel: But what’s really happening is the fluid is accelerating, but in water, to make that acceleration, because it just has to displace air. In a closed pipeline, there’s no air to displace.
Giancarlo: There’s no air to displace, yes.
Russel: Something else has to happen.
Giancarlo: Anyhow, we can talk about this forever. It’s fascinating, but for the purpose of this paper, and for the purpose of the end user for these type of pipelines, they want to know when that batch actually arrives. Whether it’s going through a flat pipeline or through a drastic elevation profile, they need to know when it gets there. There’s sort of two ways of doing this.
The way that we originally did it is by doing a calculation based on how much volume comes in and out of the pipeline, or a particular region. Using the conservation of mass theory, everything that comes in should come out of the line.
If you’re flat and level, nice and steady, everything is great, but as soon as you go through these elevation changes, you might have more product coming in than what’s coming out at the very end. During those particular situations, you have a packing situation, which, although liquids are, in theory, incompressible, there’s still some level of compressibility there, especially when you’re going through a slack.
That is actually shifting the heads or the heads of the batch, which is the most downstream portion of a batch, and obviously, affecting your ETA. We’re using what we call an empirical approach in which we are measuring how much volume comes in and out of the region.
We compare them side by side, and if there’s a difference between them, then we say, “Hey, you know, we’re packing, so let’s calculate a difference between these two volumes and see how much made-up air we have.”
This will determine what we call “Empty Volume” in our software. That gives you an idea of the amount of slack or the size of the bubble that you have in order to maintain an accurate ETA for this particular batch you send. That seems to work fairly well.
Over a 1,000 kilometer pipeline, we’ve been able to track ETAs anywhere between 5-12 minutes, which after that length and those drastic elevation profiles, and then different products throughout the line, it’s magnificent.
It really gives the operator a sense of knowing when those batches arrive and being able to react. We’re doing this through a very simplistic approach, just taking these measurements – volume totalizers/accumulators – in those particular regions, and doing a calculation that will give you a very good result for the operator.
Russel: Yeah, but just to be clear, though, you’re not actually measuring the volume, you’re using the hydraulic equation to calculate.
Giancarlo: No, no. On this approach, we are actually measuring the volume that is going into that pipeline…
Russel: You’re measuring the entry volume, and you’re measuring the exit volume.
Giancarlo: Exit volume on the other side, on the station, but not on that particular region. We are identifying a region which is flow meter bounded.
Let’s say, for instance, it could be that between flow meter and flow meter, there’s 200 kilometers, but the slack, it’s only a 50 kilometer section of those 200…
Russel: Yeah, that’s what I was getting at.
Giancarlo: I’m defining a region based on those flow meters, and I’m calculating this empty volume for the whole 200 kilometers, not just the small section.
The idea is to get an estimate of what’s happening in that particular region for the purpose of calculating the position of these batches and knowing when they’re going to get to the other side.
Russel: I’m just thinking. Basically what you have is a volume plus an empty volume that gives you a total volume.
Giancarlo: Yeah. You have the pipeline internal volume based on the internal diameter, which calculates a cross-sectional area of the pipe, and then you have the length of the pipeline.
Russel: Are you just making a simplifying assumption that the difference between your volume in and your volume out is your slack?
Giancarlo: Pretty much, yes. Very much, yeah.
If you think about it, if you go back to the analogy of the straw, what you’re doing is that you’re squeezing that straw in to maintain the assumption inside the model that the pipeline is always full. You’re making the internal diameter of the pipeline smaller if you think about it.
Russel: Yeah.
Giancarlo: That way, you hold less volume, and then the model can continue to run, making the assumption that this new pipeline that you just squeezed in is full of liquid.
That assumption is maintained in the model, but your ETA gets, actually, pushed forward, or the older batch gets pushed forward to maintain the accurate ETAs, estimated time of arrivals, of these batches.
Russel: Having sat through your paper and listen to you run through the math about how you’re actually doing this versus kind of the simplifying explanation of, basically what you’re doing is you’re kind of modifying the shape of the pipe to take away the empty volume, and using that to calculate where the head of the batch is, that’s… [laughs] I’m sorry, that’s just overly simplistic. I mean, it makes sense, and I get it, it works that way, but it’s overly simplistic compared to what you’re actually doing to get there.
Giancarlo: Yes, absolutely. You think about it, there’s obviously, we’re taking all these volumes every sample. We’re making the calculations, but yeah, you think about it as simplistic that at the time of the problem, you’re like, “How can we get around this issue without having to implement a full hydraulic calculation model to know what this slack volume is.”
Russel: Or do something even more complicated where you’re trying to do a multi-phase calculation based on how the fluid is actually changing.
Giancarlo: Correct.
Russel: You could do that, but complexity always comes with a cost.
Giancarlo: Yes, and that’s where the other approach comes in. We have this implemented. It’s been working very well, and we’re like, “Well, let’s take a look at what it would have been if we would have configured this full real-time transit model [RTTM] to calculate this out of the slack.”
When we started doing that, we developed a two-fluid model to calculate the size of the slack in these particular regions. As soon as we started doing this, we started to run a full model, take the instrumentation, tune the model for each individual pipeline segment in order to get the right roughness; to get the right pressure drop across it so the pressures make sense throughout that particular region.
Russel: Not to mention you’ve got to have all the different fluids kind of profiled, and you’ve got to put those fluid properties in to get all that to work, as well.
Giancarlo: Correct, because the fluid directory takes the temperature that they’re coming in. We are taking all of these theories to solve this conservation of mass equation — momentum, energy, and mass — and those equations are going through partial differential equations to solve a big problem. Just think about doing that equation every five seconds. That’s how often the calculation is coming in.
The tuning of the pipeline’s roughness; making sure that you have the calculated correct flow rates through each individual segment based on the instrumentation — there’s a lot of work that needed to be done in order to get the model to a point where we say, “Okay, now we can go ahead and turn on this two-fluid model in order to calculate the size of the slack.
It was a lot of work to get the model to run and to converge a lot of data validation on the instrumentation that’s coming in. Because, that’s another thing, if a meter jumps to zero all of the sudden, the model’s not going to like that. The equation is not going to solve, and it’s going to say, “Eh, sorry.”
There’s going to be a lot of validation, so we ensure that the data that we’re feeding these equations are as good as they actually are, and they’re within a valid range where the model won’t fall over.
That’s a whole different story for RTTM systems, whether they’re doing batch tracking or leak detection, or any type of other approach, that the data that you’re reading is fundamental.
Once we got all of that work done, and we had a model, then it’s “Okay, let’s turn to this two-fluid model that Jason’s been developing.” We were able then to calculate the slack in these particular regions. Not just by the simplistic approach, but actually through physics.
We did the test. You drop the pressure on the upstream side, the dynamic head goes below the elevation, which is the ambient pressure conditions, and your slack starts to form. The liquid drop-outs is what the term is called for the cross-sectional area of the pipe that is no longer full with liquid, but now full with gas that starts to open and expand as your pressure starts to drop.
Whether you’re going through normal operating conditions — and your flow changes a little bit — or you’re going through a full draining or filling condition where your pressure is dropping, you could encounter these slacks at the peaks or at the changing of elevation.
It did give me a very good understanding of where the slack was forming. It obviously is not uniform throughout the whole segment. You actually get it every time there’s a drop in elevation — you get slack — as we were talking on the waterfall effect.
It was interesting to see that, but after so much work to get this model working and identifying where these slack regions are, we realized that for the purpose of tracking where this batch head is, we don’t really care much, because we want to know when that head’s going to get to the very end of the line.
It did give a very good calculation of the slack within that region, and we know how the head now moves when we have this liquid drop out throughout the pipeline segments.
At the end of the day, if we calculate in the simplistic form 1,000 cubes of empty volume, and the slack gives us 1,000 cubes of slack, I’m still going to be able to calculate the exact same ETA at the very end of that region.
Russel: To me, Giancarlo, that is really the value of PSIG, because some of these guys here, they are crazy smart. They think in differential equations — “which differential equations” is what separates the mathematicians from the engineers.
Giancarlo: You’re seeing it on these papers, and all you’re seeing are formulas and numbers and equations being thrown at you, like, “Woo,” you know?
Russel: Yeah, exactly, but when you get to this, the ability to understand the full formula and all the variables that go into the formula, and then be able to make reasonable, constraining, and simplifying assumptions to apply it to a particular purpose, that’s really what this conference is about. That, to me, is what’s so fascinating.
Giancarlo: At the end of the day, what we’re trying to do is solve a real problem through simulation.
Russel: And do it in a way that it’s easy to understand, apply, and maintain.
Giancarlo: Absolutely, and that’s the key.
Russel: That’s the tricky bit.
Giancarlo: Yeah, that is the tricky bit.
Russel: You can throw the math at it, and you can get it working in a lab, but “let’s put it out there and make it work in production” is a different kind of thing.
Giancarlo: Yeah. At the end of the day, you’re just bringing it to reality.
Russel: Other than your paper, obviously, what was your favorite paper that you’ve seen so far at the conference?
Giancarlo: So far, [laughs] because I’m very involved on the backtracking side of things, the one before me — actually I wasn’t aware — was about mixing and tracking the interface growth and using an equation to calculate that mixing and the combination.
I’m seeing already the potential of another paper in the future by taking their research and seeing how that can be applied to improve our tracking application.
There’s also been a lot of talks about DRA [drag reduction agents] on the liquid side.
Russel: Yeah, I wasn’t able to get to any of those. They had several conversations about DRA, and how it changes fluid properties, and what impact that has on simulation. I’m sure that was some interesting stuff, as well.
Giancarlo: That was very interesting stuff.
The first day at this conference this year, the first day was split into one liquid track and one gas track. The DRA stuff was fascinating. Today, there were some good presentations on leak detection as well. It’s been very fascinating to hear about these topics.
Russel: I spent yesterday mostly in the gas track. Ed Nicholas does the leak detection part of the short course. I was listening to him, and he was talking about RTTMs for gas.
There’s an NTSB recommendation related to support, requiring gas pipeline operators to look at tools and processes for identifying and locating leaks. The way that that’s written from a regulatory standpoint is pretty broad, but we were talking about leak detection, and from a gas perspective, and how a lot of it’s done on a pure volume basis.
I was asking him the question, “What about doing leak detection on an energy basis?” because if you do it on an energy basis, you get rid of some of the complexities of just doing simple volume.
If I’m looking at MMBtu [1 million BTUs], I’ve got conservation of energy built in. I’m not just doing conservation of volume. To some degree, it’s simplifying, so I had this thought.
The reason I bring this up is you mentioned that you had an idea for a paper, and I sat there and listened to this, and I thought, “Oh, that would be a really awesome paper to do is to show it being done both ways,” but it won’t be Russel writing that paper.
[laughter]
Russel: Not unless I retire and something strange happens and that’s how I wanted to spend my summer for some reason.
Giancarlo: It’s amazing. You’re doing your day-to-day job. You are getting through a day-to-day with all these tasks and the work that you have on your day-to-day. Then you have to spend the time on writing this paper on the stuff that you’ve done. You’re looking at your notes from a year or two ago and say, “What did we do here? How can I put this together?”
Then you’re writing this paper. You’re putting these ideas together. You’re traveling. We do a little bit of traveling. Travel, come back. Through those last weeks, you’re suffering to meet the deadline. Then you say, “Oh, I’m not doing this again.” Then you’re here. I was like, “Oh, that’d be an interesting idea. Let me put myself through that again.”
Russel: [laughs] It’s what makes this fun. Some people will think you and I are a little weird because we’re a little geeky about the math and about this kind of stuff. To me, it’s fascinating. I always come to something like this and always get more ideas than I have the resources to execute. That’s part of what makes this fun. There’s always something new to be looking at and thinking about.
Giancarlo: Yes. When you’re sitting there listening and it’s a topic of your interest or close to your area, you start thinking about, “Wow, that’s magnificent. Why didn’t I think of that? How can I use this to tackle this problem that I have back at home?” It’s very fascinating.
Russel: Yes. The PSIG, it’s interesting. It’s a very small conference, probably, what do you think, 200 people here or less?
Giancarlo: No. I think they put the numbers between 130 and 150 this year. Their number always have been between 130, 140 and 170, 180. It varies between those numbers. It’s not very large.
Russel: There’s people here from literally all over the…
Giancarlo: All over, yes. They had a chart today. There were people from about 18 or 19 different countries.
Russel: Some really talented physicists and engineers. It’s probably becoming one of my more favorite conferences that I go to. I go to a lot. I think one of the reasons I like it so much is it’s hyper geeky. [laughs]
Giancarlo: Yes. It could be very technical very fast.
Russel: No doubt. Look, man, thanks for taking a few moments out of your busy schedule here at the conference and talking to us about your paper.
Giancarlo: My pleasure.
Russel: Certainly, again, as always, we’ll link it up in the show notes. You’re a regular guest. If you guys want to find Giancarlo and talk about batch tracking, you guys can geek out, talk about batch tracking. You’ll find Giancarlo’s contact information on the pipelinepodcastnetwork.com website. Thanks again.
Giancarlo: Thank you, Russel. Thank you for having me again. Enjoy London.
Russel: Hope you enjoyed this week’s episode of the Pipeliners Podcast and our conversation with Giancarlo Milano. Just a reminder, before you go, you should register to win our customized Pipeliners Podcast YETI tumbler.
Simply visit pipelinepodcastnetwork.com/win to enter yourself in the drawing. If you would like to support the podcast, please leave us a review on Apple Podcast, Google Play, Stitcher, SoundCloud, or whatever application you use to listen to podcasts. You can find instructions at pipelinepodcastnetwork.com.
[background music]
Russel: If you have ideas, questions, or topics you’d be interested in, please let us know 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