Multiple Basin Design Using Innovyze InfoDrainage, Civil 3D, and InfraWorks

BILL NEUHAUSER: Hello, everybody. Glad to see you make it here to New Orleans. We are here the first day of the classes here, and our class here we're talking about is the case study on multiple basin design using Innovyze InfoDrainage, Civil 3D, and InfraWorks along with some other software in the background. But we'll talk about it as we get into it.

But my name is Bill Neuhauser. I have a speaker with me, Ryan Brown here, which he'll show here coming up here. And we get the good old Safe Harbor Statement here. Since I have an Autodesk employee with me, we have to place this up here for everybody to see. So I work for ATG USA, and so I'm not necessarily abiding by this, but we try to live by it, too. But I'll leave it up there for a little bit there. You can read through it if you wish. I'm not going to give you a verbal on this. So I'll go ahead and continue on here.

Again, this is about myself. I'm Bill Neuhauser. I work for ATG USA. We are an Autodesk reseller throughout the whole country here. I've been in the reseller industry here for about 15, 16 years, somewhere in there. Prior to my reseller work, I worked for a consulting firm, and I was also a college instructor at one time.

So software-wise, I go back to the beginning of Civil 3D and prior to that, the beginnings of Land Desktop, and prior to that, the beginning of DCA, and then prior to that, good old-fashioned AutoCAD. So that kind of date me there a little bit. But so yes, I have seen a little bit of everything. And I did work on the boards at one time for just a slight sliver of my career. But Ryan here-- Ryan, I'll go ahead and let you introduce yourself.

RYAN BROWN: Yeah, my name is Ryan Brown. I'm a technical solutions or technical systems engineer with Innovyze and now Autodesk. I've been here for about three years, but prior to that, came from doing consulting work doing a lot of hydraulic modeling of water, wastewater, stormwater systems. And I reside in Raleigh, North Carolina.

BILL NEUHAUSER: And you probably noticed there, too, we are both PEs. So it means we should know something even though we are CAD guys, right? We are rare. We are CAD guys that are engineers who love to play the software. So that's a rarity all by itself. So we'll get into it here.

Now, our case study here is a subdivision that's happening South and East of Colorado Springs, Colorado. And now, as a case study, just to let you guys know, this is not a pics and clicks kind of demo. This is to show you step by step the struggles we've had going through the project here. And we're talking here. This project literally has just started here at the beginning of the year. On the right here, you can see the layout.

The engineering company that hired ATG and myself is out of Burnsville, Minnesota, Core Engineering Group. Now, what happened here is they hired ATG/myself here to come in and help them get-- they actually were using Land Desktop 2006 if you can believe that. We still have a few stragglers out there. And so they made the big jump here up to 2023, which is what is that, 17 years of software changes between one and the other. So let's just say, there's a lot to be shown and taught and so forth.

We first started out with the Fundamentals class with them, and then also, an update in AutoCAD just to get them up to speed. And what they have done, they purchased our ATG Starter Template because we all know rule number one in Civil 3D is always start a drawing with a template file, right? Rule number two is save often. And that's been from day one.

And so during this whole process, after we got through all the training classes here, we've been going through weekly mentoring classes with them. And it's all been based around this project here. And as you'll see here in a minute, I'll give you a little more specifics on the project, but it's a pretty good size. I believe he claims about 7,000 lots. He figures it's going to take between 10 and 15 years from the beginning to end on this project. And so it was a perfect time for them to jump into Civil 3D.

So let me go here a little more information about the project. Now, that picture there is actually hot off the press. It was just taken yesterday. He actually flew out there to have a look at it and some other things. This is a box culvert, which I'm going to show you here in a minute here some things that popped up during our learning process.

But again, the site location info, it's South and East of Colorado Springs. It's about 1,500 acres, about 7,000 units, a whole bunch of different size lots, 50x110, 45x85, et cetera. It's fairly raviny, I guess you want to call it. We've got some hills and valleys and a creek running through it. So slope's 3% to 25%.

The site itself, what we're finding here is that we've got around about 15 larger ponds as of right now and counting. As we get into the design, we'll probably have a couple here or there that gets thrown in. But generally speaking, we've got 15 ponds stretched out throughout the site.

The big thing here as you can see from the picture, we've got this big box culvert, and we have more or less a dry Creek coming through here that is dry one moment, and the next moment, it's got 3,500 CFS coming through it. So pretty wide range here of what we got to design for.

And that road that's going across there, too, the local entity came back and threw at them a 30,000 ADT on it. And so of course, it's going to expand and make it big. Well, what people don't really realize here is just to the East, and I'll show you in a second here, but just to the East and a little bit North is a Space Force Air Force base. And it's got like 8,000 employees on it according to the website. So this is the main route getting out to it. So our thoughts are to eventually the housing out here will probably be a lot of Air Force people.

But now, also to the West and North is, of course, Colorado Springs, which also has another Air Force base, I guess, and also the college there. The Air Force Academy is also there, too. So it's a fairly good-sized town, beautiful area. Of course, he's got it pictured looking the wrong way. We should see the mountains in the background. So this is looking off to the East.

But beyond that, down below here, like I said, 10 to 15 years. You're going to have electrical substation out here, one ground storage water tank, one elevated tank . We got transmission lines going through the site, which you might be able to see on these drawings when they come on up. And then we've got an existing 66-inch raw water line coming through here. Eventually, five schools and one fire station here. But let me continue on.

So one thing we started off in this project here is that when I first got involved with it is, again, being old school, he had an old-school survey that, in actuality, they had they use it on the original site to the South of this, which I'll show in a second here. There's actually a big subdivision to the South of this that they're tying into that was originally done by the same company over the last 10 years or so.

And so when they started working on this one, they used the survey that they had from that. But the problem is they didn't really know how old it was, exactly how-- it was so long ago, they didn't know the source on it. And so again, no idea on the age, not refined enough for the creek and numerous ravines we're finding out. But it was free, right?

So next thing is they started working on doing an actual survey. And one of the drawings I've got, actually, you can see where they started around where that box culvert was if I recall, a little bit south of that, they started doing the old-fashioned survey. Well, one, it was starting to cost a lot. Obviously, it's kind of a big area, and you got the ravines and everything. And I talked him out of it saying, hey, before you start wasting $1,000 on this, let's have a look here on the modern side. Since you're going modern on Civil 3D, why don't you go modern on your surveying?

And so we started looking into drone photogrammetry. So he put out an RFP or a bid or however you want to call it, had a couple companies nearby come back with bids on it. And we looked at them and stuff. And then one of the companies brought up LiDAR. And I talked to them, too. I said, LiDAR's great. It's going to probably cost you more, but keep in mind out here, we have this prairie grass.

And so photogrammetry is going to give you a great aerial, but with all the prairie grass, we're not going to get a very good surface. We might not be any better off with the photogrammetry survey as we were with the original survey. And so we looked into it. He got the price back, found out it really wasn't that much more. And a fast turnaround, the density of the data, though, is-- of course, the raw stuff is too high, and it requires additional post-processing to really get it to where you need it.

So that brings us on to the software side. Now, on the right here, you see the grass that I'm talking about here, right? You get great penetration with LiDAR. Again, photogrammetry, and if it's windy, photogrammetry really hates this stuff. But again, the old survey, not refined enough. The new conventional survey, again, this is going to work, but again, it's still not quite refined enough. It's just when you need to really get around the ditches and the ravines we have through here and stuff, that's a lot of manpower, which equals a lot of money.

The drone photogrammetry, now, this is the path that we originally were looking at taking. We would get the photogrammetry data, process it in Recap Pro, go into InfraWorks, and use the Point Cloud Extraction tool. Time consuming, but it can be done. And from there, we could get into Civil 3D. And actually, I got a link here just FYI. Back in 2017, I actually did another Autodesk University class, and this is the link to it talking about this exact process here, step by step how to do it. And it works fine. Today, it works good.

So keep in mind, one of the biggest problems you have with photogrammetry that you get from a lot of companies right now is that they have a button on their machine that says Export to CAD. And boom, they click on it. Great. Well, what did it do? Well, it gave a file that's workable in CAD. Yeah, kind of. The problem that I found is that those surfaces are very sparse. They're like a grid pattern more or less, like a 15-foot, 25-foot type grid, which is fine if you're surveying a football field.

But when you're surveying something with a lot of ravines or if you're in town, you've got to look at a lot of curb and gutter and stuff, that's a whole different story. We're missing all those grade breaks. And so doing this method here that I show here, you actually can account for the grade breaks. If you have the current Autodesk Suites, all the Infrastructure Suites, you have that in there.

Now, backing up, though, since we had this flown. We had it processed, Frontier Precision out of the Denver office. I don't remember what they're using. I think it had a yellow scan LiDAR unit mounted on a DGIM300. Data came back. Well, when we first got it from them, guess what?

It was the push the button, give us one-foot contours. Push the button, give us five-foot contours. And it was just like, well, oh, God darn it. Yeah, we got to the ground, but we're not getting what we really need. If we went back to and had the high-resolution data, then Civil 3D can't handle it.

And so that's where it came in here that we got another company involved here. And there's same data. We didn't have to go out and refly anything. But this time, a company out of Fargo that I've dealt with a lot with, I got them involved, and they used some software called TopoDOT and some other software, which I don't remember the name of it, that they will do the post-processing of the data.

Now, the other thing that goes on here is that up here, some of this conventional survey data is not wasted because this data inside of this other software is used as what they call ground truthing. So it not only is still there, it gives you stuff to tie into, and it really nails down your point cloud data.

And then from here, we dropped it into Civil 3D. Now, once in Civil 3D, that's where I get Ryan involved here. From there, we can easily get it over to InfraWorks-- or not InfraWorks, I'm sorry-- InfoDrainage and/or InfraWorks or anything else as it mattered. Now, the file here is actually quite manageable for this 1,500 acres. I know we're well under 100 megs in file size.

And so now, what I'm going to show you here next is just some of the samples from this. Here is the one-foot data not processed. Here is what they call the five-foot data, five-foot contours. And then here is the data being processed after the fact. Now, so if you can see here, we're looking at exactly the same spot. This is where that culvert's at. The five foot, you can't even tell there's a road going through here, now can you? The one foot, yeah, we get a little more resolution. But definitely over here, you can definitely see where we're getting a better resolution.

Now, what's going on here? It's kind of more of a hybrid. As you can see, there is a dense pattering-- pattering-- that's a good word-- a dense pattern going on here where it needs to. But where it's not needed, we get more of the open area. So it's, how to say, automatically doing this for us. Now, I'm sure there's some manual input on the user on their end.

But the end result is kind of more of an oh, my God, I can really take these huge files, process it through it, and now it's usable. And we found, too, that when we got this data into Civil 3D and we got it over into InfoDrainage, InfoDrainage loved the heck out of it, too, because it's much more clean, a lot less overhead.

Now, let me go-- next one here. Now, what we got coming up here, and this is where I got Ryan involved here. But before I jump over to Ryan, though, I am going to jump over to InfraWorks and show you guys. In fact, let me do that right now. I'm going to-- hopefully, we don't lose the screen here.

So just to give me a little more oversight on the whole project here, so what you're seeing here-- now, the outer boundary here, this is from InfraWorks Model Builder. Over here on the left side, this is the airport from Colorado Springs. What you can't see, the Air Force Base is up here in the top right side. Now, what you can see down here is if I can get it to pan up here on us, this is the subdivision that we're tying into. We've got a road actually coming in here. Now, the photograph in here is probably a couple of years old so we're not seeing the full development here.

But now, if I come down in here, now, what Ryan's going to show you here is up here, we have a bunch of ponds here. And I don't show them on this particular drawing, but we've got the ponds kind of scattered out throughout here. Most of these ponds-- not all, but most of these ponds more direct drain into the creek here that you see. Down here in the South, though, what we got going on here, we've got a smaller pond here that's taking all this area. We've got a larger pond here that's going to connect with this pond. And we're going to run a pipe down through here and connect into it. And Ryan's going to show you that.

And then we've got a man-made channel that we've created. As you can see here, we've actually brought it into InfraWorks. And that flow is going to come over here, and then we eventually will have an output from here. Now, what makes this pond kind of unique is that if you look here, see this drainage here that popped up? This was created from InfraWorks with itself.

You go up here to the Analyze button here, click on this button here, click where you want the analyze to occur. It looks at the surface and comes up with this. Now, the other thing that's cool about this, and this is handy because it helps us double check ourselves with our other software, is that if I click on this, on the right side here, you can see here it's popping up with peak flow numbers.

Now, what's cool here is that you pick on this-- what I found is that you got to go with the regression. You pick on the state. You pick on the zone out here. Now, if you pick on the wrong zone, these will actually go to zero because the software's smart enough to realize, hey, you're in the wrong zone, guy.

So I've got to pick on one of these here. And I'll be honest with you. I'm not positive on what the difference between the two here. I think this one here is for a high elevations. You notice a lower CFS. So Colorado, when you're up high, the rains must not be quite as bad as they are down here. Or maybe it's the snow melting factor or whatever. But anyway, you get the idea. So this is all from within InfraWorks. I just wanted to show you guys this. It's, I guess, a way to maybe easily check yourself.

Now, the other thing we found is that when I picked on I believe it was right down here, that little icon right here, when I picked on this, InfraWorks automatically created this catchment area for me. Now, what's cool is that if I go over here to InfoDrainage, you'll notice here that we have here some catchment areas already drawn in. These actually came from the engineer. This is what they determined what their catchments were. And when you go look at these and you go look at what I got from InfraWorks, they're actually pretty darn close. So I was kind of impressed with that.

But let me go back to my PowerPoint here. I am going to turn this over now to Ryan here in a sec. So a first step of this whole project here is the drainage before we get everything else laid out. And in fact, something here is that that lot layout you see here, I found out here about a week and a half ago from the engineer that all that's changing. The ponds are going to stay pretty much the same location, but all the lots are probably going to be changed and updated. So that's a whole other process.

So maybe next year AU, we'll come back and talk about how we got from this day to the next one because I'm sure we're going to be able to use some other Innovyze software in this. Correct, Ryan? We got all that other water packages you guys have out there. I'm sure we could probably use something else, right?

RYAN BROWN: Oh, yeah, lots for water and wastewater planning and capacity and making sure there's pressure and fire flow all across the development.

BILL NEUHAUSER: Yeah, so-- well, here, let me go through this. I was going to have some music playing here for this. I call it a battle of bands here. So we got-- I'll let Ryan explain what this is. Colorado here, they have by law that the engineers must use this spreadsheet that you'll see here in a minute here to do all of the pond calculations here. So I've got the Excel spreadsheet here. So I've got the Excel spreadsheet.

I had to add your bouncy head coming in there, I guess, Autodesk's own Mr. Ryan Brown, PE, here. So what he's going to do here, he's going to actually jump into the Innovyze site, and he's going to give you a nice detailed study here of what he did comparing these two numbers because Colorado right now, and I'm sure there's a lot of entities around the country that do the same thing, that they have their own programs. In this case, they have an Excel spreadsheet. And they have a lot of training, a lot of training videos just to learn how to run that spreadsheet. And they mandate that you must use this. And that's fine.

What we're trying to show here is that, well, here's another alternate solution. Maybe we could throw this back at the other entities and Colorado and start proving to them that we have a better, faster solution that might save everybody time, which saves everybody money. And that's the bottom line. So here, here's the task here. We've got 15 total interconnected retention ponds. When I say that, really, he's going to look at three of them. All flow into a nearby creek at about 3,500 CFS. So Ryan, I'll go ahead and let you take over. Do you want to share your screen?

RYAN BROWN: Yep. Yeah, thanks, Bill. Yeah, so like Bill was mentioning, basically, we had two different approaches. And we weren't necessarily using InfoDrainage to design anything, but I'll get into that a little bit more once I show the software of some of the components in there.

But just a quick summary, like Bill was saying, it's required to be used within the Mile High Flood District. That's generally the Denver kind of area of Colorado. It does use a Modified Puls Method, which is basically a continuity equation very similar to what InfoDrainage does. A big key thing, and I'll really hone in on it later, but that the ponds are analyzed independently. So there's no interaction between the two. And so you'll see that becomes a problem when we analyze some of these interconnected ones.

The outlet structures are sized, and there's macros and things like that to help guide those workflows and things like that. But at the end of the day, it's still a spreadsheet like Bill was mentioning. And I can show you the spreadsheet in a second here. There's still a lot of black box things going on with it, and it's not very clear what numbers mean what and that kind of thing.

For the InfoDrainage, though, it is a comprehensive program where everything from the stormwater controls to the pipes to the hydrology is all done within one package. Everything's interacting with each other. It does use an industry standard EPA SWMM5 engine. It does use the Dynamic Wave Equations Method for solving the shallow water equations.

The routing is very, very similar to what the spreadsheet's essentially doing is just subject to the continuity of flow in, flow out. And then there is lots of calculators. Just like there are macros and things like that in that spreadsheet, there are lots of calculators to help guide the users to be able to size the [? BMP, ?] size the pipes, size the outlet structures. And I'll show that in a little bit when I get into the software.

So as far as the methodology that we used, the ponds were already designed in the Mile High Flood District. And like I said, the point was just to compare the different methodologies for sizing it in the Mile High Flood District and what does it look like doing the same kind of thing within InfoDrainage. Is it going to be similar? Or are they going to be different? What's going to go on?

The Colorado Unit Hydrographic Procedure is a specific method that's used in Colorado and used within that Mile High Flood District spreadsheet. And because of that, I wasn't trying to model the hydrology even though you could within InfoDrainage. I just put in basic inflows, so basic hydrographs straight from that spreadsheet. So we're doing more of an apples to apples kind of comparison and just picked the 25-year storm as my example that I wanted to use as part of this study.

Then I recreated the entire network using the pond size, the pipe networks that Bill has described to me, the outlet structures. All of that was taken from that Mile High Flood District spreadsheet and pretty much dumped into InfoDrainage to see how things compared. So this was my initial look at comparing one of the ponds in InfoDrainage to what the output hydrograph from the spreadsheet was.

And as you can tell, there's a lot of differences going on here. So one, there's a ton more volume. Two, we've got lots higher flows. We've got some strange peaks. We've got this pretty significant decrease in flow for some reason, and of course, a longer drain time for everything. And so I'm concerned. I'm frustrated. I'm like, what's going on exactly? Why are the results so different?

And then I started to think about, oh, we've got a fully dynamic system. Everything's connected up to each other. So D1, Pond 1D is that pond that I was looking at before there. And you can see I've got the pond right there, and then I've got a bunch of other connections. I've got multiple ponds entering this.

There's even one that's up here that's entering into pond E that then empties into pond D1. And then there's some tail water effects from the pipe network here where water might be backing up. So clearly, there's some effects going on from all of these things being impacted together, which really just highlights the importance of analyzing these systems together rather than independently to really get an idea of how well things are going to perform.

So what I did is-- well, let's go to the software first. So just taking a look at the software, actually, I'll start in the spreadsheet first. So this is a look at the Mile High Flood District's spreadsheet. It is very, very comprehensive, put together by very, very intelligent people. It's a tried and true method for sure with lots of macros and things like that and ways to do things.

However, like Bill was mentioning, there's lots of resources and things like that of trying to train and understand what you're doing in this spreadsheet, trying to understand what each one of these different tabs are and what you need to do, understanding what these different orifice types are without a lot of visual guidance or anything like that. And so it can just be-- really, I've heard it described as just being a black box, not really understanding what exactly is going on or what's happening. You just type the numbers in, make sure it works, and then call it good at the end of the day.

Switching over to InfoDrainage, this is looking at one example here. We did put together some initial plans and things like that. So like I said, I just took the hydrographs from the spreadsheet, the inflow hydrographs from each one of those and dumped them into these customized hydrographs so I could get an analysis of the hydrology itself.

But as you can see, I've got a background CAD layer in here. Like Bill was mentioning, well, we can start to bring in CAD files for what are called inflow areas. So these are basically the subcatchment areas if we wanted to do the hydrology within here. So we can utilize InfraWorks to be able to delineate those watersheds for us. We don't want to have to spend a lot of time doing it.

Likewise, with surfaces and things like that, surfaces can be brought in here. And you see we have one here. And really, that does, I guess, to highlight some of the things that we can do a deluge analysis, which is basically a 2D analysis. We can also start to pick up cover levels and things like that over pipes.

But generally speaking, InfoDrainage is supposed to be fairly intuitive, easy to use, drag and drop. Everything is spatially accurate. You're just walking through the top ribbon here and going through. You can validate models to make sure that everything makes sense. You can audit them to make sure you have enough cover, and then, of course, looking at the results and the profiles and things like that to be able to get a good, nice picture of what exactly is going on during your simulations.

I'm just going to flip back to the presentation here. So what I did in order to compare it to the Mile High Flood District a little bit better was to simply just disconnect all those ponds, just make it so that there's no connections, there's no interaction between them. So it's more of an apples to apples type of comparison.

So looking at that same pond, this is that same pond, D1. We can see these hydrographs match much, much better than what they did before. It's still not perfect. The peak's a little different. The peak is shifted a little bit. It's down a little bit. But for the most part, these are very, very similar. And it's always nice to see just an overlay and see how things compare. It's hardly ever that things are going to be 100% exactly the same. But to get an idea of how well things are compared, it can be nice to just match them up on a graph like this.

However, I personally think-- and this is a little bit of the modeling background, but being able to use just the cool types of measures to be able to understand really how well things match up. So the RSR, or the Route Squared Ratio, is a ratio between the Root Mean Squared Error and the standard deviation of the observations.

And so basically, these two equations up here where the Root Mean Squared Error is essentially a look at the differences between the observed and predicted values. So if we assume that the spreadsheet is giving us the observed values and the InfoDrainage is giving us the modeled, we can subtract those two and then run these two through these equations, sum them all up, and then run the standard deviation over the observed values through a similar kind of calculation.

Once we do that, we're able to calculate the Root Squared Ratio. And if you can see over on the right side here, we have a listing of ranges. So these are just typical ranges of what you would consider a good model to be or a good place for a model to be in terms of calculating these types of numbers. So between 0 and 0.5, we've got a very good performance, a very good relationship between the observed and the prepared values. And so what I did was basically to calculate this for each one of the outputs for the differences.

And so looking at all the ponds across it instead of boringly just going through it and just looking at the pictures, this is just a summary of those results and being able to see that they're all below that 0.5 value really means that all of these ponds showed very good correlation between the observed and the predicted or the spreadsheet and the InfoDrainage values.

So really encouraging to see that the methodology used within InfoDrainage is fairly similar, if not almost identical to what you can produce with that Mile High Flood District. But you do get that added benefit where you're really analyzing the entire system rather than analyzing things just independently. So I think with that, I'll hand it back to Bill.

BILL NEUHAUSER: OK, well, thank you, Ryan. I learned a lot. I mean, I'm an engineer, and I remember going through a lot of the stuff back in college days. And how much you forget in 30 years, right? But being civil engineers, we got-- what are we like, jack of all trades, master of none? Well, there's a few people that are masters at their field. And I can definitely see that from what you were going through there. I appreciate your time on this. I really, really do. This helped out a lot.

Again, maybe next year, we'll be back here again talking about maybe something else, maybe another colleague of yours over on the water side on the pressure pipes or something like that because eventually, we're going to have to figure out sanitary on this. We've going to have to figure out the water distribution on this, the whole gamut.

And I was never so happy when I found out that Autodesk went out and bought Innovyze and brought them in. We've been waiting for decades, decades for some of these solutions, and not just the InfoDrainage side, but everything else they did bring along with it. So that was a pleasant surprise to say the least. But otherwise, if anybody else-- we'll have a question and answer session when this is all done, and we can talk some more then. But thank you very much.