Terminology Explained: What is a parametric study?
In this edition of Terminology Explained, we cover the topic of a Parametric Study. Perhaps you and your organization call this by another name (or didn’t realize it even *had* a name!). It is an easy concept to understand, but not always easy to implement!
Imagine you are a typical planner on a typical day: A new project lands on your desk! What could it be this time?
Well, there is a prospective customer to be added into your gas network. Their demand is not certain, but estimated to be between 6 to 10 mmcfd. This customer will be connected into the network by a new 2 mile pipe, which could be either 4”, 6” or 8” in diameter. There are three city gates where set pressures can range from 120 to 160 psig. We will need to attain a minimum pressure of 30 psig for this new customer. (I hope you’ll indulge me with the non-metric measurements; I suppose that reveals where I’m sitting!)
So the question is: What is the best combination of pipe size and regulator set points to meet the minimum pressure for this new customer? Let’s look at this problem in more detail:
Let’s assume a demand of either 6, 8 or 10 mmcfd. And let’s assume set point values of 120, 130, 140, 150 or 160 psig at the city gates. So:
There are 5 set point values at each of the three city gates.
Then there are 3 possible pipe diameters, and 3 possible demand levels.
1,125 scenarios. (5 x 5 x 5 x 3 x 3)
So how could you tackle this problem with Synergi Gas? The steps would look something like this:
- Open the model
- Find the city gate that you want to edit
- Open its editor and enter a new set pressure
- Repeat for city gate 2
- And again for city gate 3
- Set the diameter on the new pipe
- Then set a demand value at the end of that pipe
- Run a steady state analysis
- Look at the resulting pressure at the end of the new pipe
- Store the results somewhere, probably in Excel.
That was run number 1. Now repeat this 1,124 more times! That, my friends, is a parametric study.
Realistically, a planner will not test out every option. Even at only 5 minutes per scenario, that would take over 93 hours, or more than two full working weeks! Likely he or she will try a handful of combinations, using their experience as a guide, and pick the best result from that subset. But what if you could easily run every combination, in a fraction of the time that it would take you to manually run this subset?
So how can I effectively run a parametric study?
For many (many!) years, DNV GL’s Synergi Gas has provided valuable analysis tools that engineers use to make informed decisions about their systems. It is integral in designing for increased capacity, managing outages, enhancing operational efficiency, reducing cost, complying with regulations, and more.
However, using Synergi Gas in this scenario would require varying multiple parameters and running many analyses, which is a manual, time consuming, and error-prone process. Larger and more complex problems are usually not fully analyzed, which could result in incomplete or suboptimal data for decision making.
Our customers have long requested the ability to define these scenarios up front, and “batch up” the analyses to run automatically.
So, coming soon: The Synergi Parametric Study Tool!
DNV GL’s Synergi Parametric Study Tool is a companion to the core Synergi Gas product. It will be a cloud-based, Software as a Service (SaaS) application with a web interface, to define and review the results of studies. And it uses the power of the same Synergi Gas hydraulic engine that you already know and love.
Utilizing the power of Microsoft Azure resources and technologies, the Parametric Study Tool will allow you to run hundreds of analyses in parallel simultaneously.
Build your model in Synergi Gas.
Run a myriad of scenarios with the Parametric Study Tool
Pick the combinations that you want to look at in more detail.
And bring them back into Synergi Gas to analyze further!
We’re hard at work right now to bring the Synergi Parametric Study Tool to life; watch here for more updates!
Many thanks to Paul Cutting, Principal Software Engineer for our Pipeline products, who was an integral contributor to this post!