Wednesday, March 9, 2016

Model-ing Citizens

At the end of this month I'll be heading off to Europe for the latest wastewater process modeling conference - wwtmod2016 - which I'm sure will be as interesting as previous seminars in the series.  It's a pretty cool gathering of process engineers and/or modelers to dig into the process models that are a big part of what I do as a poop engineer.

In thinking about the seminar, I started to reflect on the fact that there are a lot of good wastewater process simulators on the market right now, 4 of which I'll talk about in this blog.  Before I get to them though, I have to comment on the fact that there are so many simulators in what is a pretty niche market.  If I think back to spreadsheets in the 1980s' there were several on the market - Lotus 123, Excel and Quattro Pro (the last being my fave for some time), but today the market is dominated by Excel alone (OK, there are a few diehards using OpenOffice, Numbers on the Mac or Google Sheets for simple collaborative tasks).  So how come there are so many good wastewater process simulators on the market right now? I'd like to suggest a few possible reasons:
  1. Wastewater treatment is complex (and fascinating!) and so we need models to help us figure it out.
  2. Wastewater modeling has strong "champions" who have really driven the ideas and industry in a positive direction.  In another blog I might list out several of these individuals, but from my own personal experience and because he's turning 60 soon I'll just mention Imre Takacs here.  Super nice fella with a real passion for modeling, including his latest venture with SUMO.
  3. Wastewater modeling is cool.  Let's face it, producing diurnal graphs and "playing" with a virtual plant is way cool.  That's why I like it, right?

Simulators

OK, so here is a list of my personal favorite simulators.  They each have their particular strengths but I resist the temptation to say which is "best" despite being asked many times.  They're all good and useful tools.  Some do some things better than others and it's always a moving target as they each add new features. There are other simulators on the market too, but my exposure to them is limited, so the fact that I've not listed them is in no way a slight on their capabilities.  So here are, in order of my exposure to them and not in order of preference, my fave four!

GPS-X

For me, it all started with GPS-X 2.0 running on Unix on a HP computer.  That's real modeling!  They and everyone else shifted over to Windows which was more convenient, for sure, but our models slowed considerably until recent years as a result.  I still wonder if GPS-X running on modern machine with Linux wouldn't be the way to go!

The big plus for GPS-X is the user interface. Maybe it's because I started with this simulator, but I still love the ability to use sliders for control, set up graphs and have scenarios all in one interface.  It has it's quirks you need to learn like all the simulators, but in terms of being easy to run and adjust models, it's great.


BioWin

So, I find GPS-X one of the easiest simulators to run, but BioWin has deservedly won the reputation amongst design engineers for being the easiest to set up.  There are many engineer-friendly features in the model that make it the goto for many process engineers.  Another strength of the model is the biokinetic model (ASDM), which has been termed a "super model", because it carries all variables and all rate equations around all process units which makes it easier to ensure the mass balance holds.  They also pride themselves on having decent defaults for most parameters under most conditions. You should never use any model "out of the box" without knowing what you're doing but with BioWin you maybe don't have to move too far out of the box!


SUMO 

Here's the new kid on the block.  Having done his time with Hydromantis (producers of GPS-X) and Envirosim (BioWin folks), Imre struck out on his own to develop a whole new modeling platform from the bottom up.  It's pretty exciting to see the development.  What I've seen so far there are two main tenets guiding it's development: (1) modern user interface (pretty cool); and (2) ease of access to the biokinetic models.  The latter is the most exciting piece for me as it's allowing us to do a lot of investigations for WERF projects and other applications.  For those at WWTmod2016, you'll get to see some of this when my colleague Patrick Dunlap presents some initial modeling for one of those projects.


SIMBA#

Lastly but only in my own chronology as it's been around for years in Europe, is SIMBA#.  This simulator is very strong for anyone wanting to look at control.  It also has some nice energy features which they're continuing to develop and refine.  I'm thinking this "old kid on the block" may start to get some traction in North America in the next year or so.



So... what's your favorite wastewater process simulator (assuming you have one), and why?

(Copyright disclaimer... all graphics on this blog were taken from the software supplier's websites.  Please check out their sites for the original graphics and further info)

3 comments:

  1. I am flattered Andy (talk about a super nice fella), but it is undeserved - without prof Marais and the original UCT group we would be still counting BODs and MLSS. And when it comes to doers, a lot of things happened really because of Mark Fairlamb and our own Robert Kovacs - who are not so much in the limelight but they deserve to be.

    ReplyDelete
  2. Cheers mate, and see you in the South of France soon!

    ReplyDelete