Training to work at a water treatment plant is a time-consuming and complicated process. There are often times when the engineers who run these plants make decisions based on their experience rather than a strict manual. Some of the work is done by feel, with engineers who deeply understand the various flows, pressures and pumps inside a plant working on an instinct honed by many years wrangling the water. This is part of the reason bringing in and training young engineers has been so difficult. The job takes a long time to get comfortable in, and many other engineering jobs pay more.
“The major problem is the difficulty to maintain drinking-water infrastructure with an aging workforce …,” said Pingbo Tang, an associate professor of civil and environmental engineering at Carnegie Mellon University. “What I observe is that here in the U.S., at least in Pennsylvania, we see in a lot of those water treatment plants that the average age is above 60.”
Researchers at CMU are trying to solve this problem using the U.S. National Science Foundation’s ACCESS program. Pengkun Liu, a CMU graduate student who works in Pingbo Tang’s lab, water resources expert Damon Weiss of Ethos Collaborative and Christopher Miller from the University of Akron are creating a virtual water treatment plant using Pittsburgh Supercomputing Center’s Bridges-2. The idea is that in a virtual plant, you can run all kinds of simulations that can’t be tested in a real plant. Anything from an emergency to a simple part replacement can be simulated and trained on without worrying about harming the treatment plant or the supply of clean water.
“… If we want to build the virtual reality environment, we need a lot of strong computing resources … We get a lot of human data, and if we want to train a large-language model, we [also] need strong computing resources … from PSC. [We need powerful] GPUs for rendering the virtual reality environment and also for training and testing the models.”
—Pengkun Liu, CMU [pull-quote]
By tapping into the decades of expertise that water treatment plant managers have now, researchers will preserve this knowledge even after the senior staff retires. This is just the first step in a long process of making sure the water continues to flow cleanly into all the homes that need it.If you’re interested in how the researchers are teaching a supercomputer to create a water plant simulation, you can find more details in the original story here: Human-AI Knowledge Sharing for Water Treatment Plants Operation
Project Details
Resource Provider Institution(s): Pittsburgh Supercomputing Center (PSC)
Affiliations: Carnegie Mellon University, University of Akron, Ethos Collaborative
Funding Agency: NSF
Grant or Allocation Number(s): CIV220006
The science story featured here was enabled by the U.S. National Science Foundation’s ACCESS program, which is supported by National Science Foundation grants #2138259, #2138286, #2138307, #2137603, and #2138296.