Desalination Plant Performance Monitoring demo
Clean each membrane train when its data says it is due. 1:16
What the demo shows
A system solution for desalination operators that predicts membrane fouling and cleaning needs.
Six agents normalize every reverse-osmosis train's flow, pressure drop and salt passage overnight, forecast the day each train reaches its cleaning trigger, name the kind of fouling from the feed water and the clean history, and stagger the cleans against the demand forecast so the plan keeps output above forecast demand on each day. The plant is drawn as a schematic, each train colored by its drift with its forecast underneath; the process engineer books the clean and the water production manager signs the plan.
Transcript
- Desalination Plant Performance Monitoring tells operators when each membrane train needs cleaning.
- Today, trains are cleaned by the calendar, while fouling burns energy.
- Here are Corrin Point's eight trains, colored by pressure drop drift.
- Overnight, train three turns amber, pressure drop up almost 12%.
- Its clean falls due in nine days on the 22nd, train five's day.
- Two trains out in a heat wave, leaves the city short, so train three moves to the 20th.
- The diagnosis cites the algal bloom and August's acid-only clean, alkaline first.
- Farrah books the clean, Daniel signs the plan, and output stays above demand on every day.
- Clean membranes when they need it, an estimated $900,000 a year.
- Built on Botminds: agents, data, screens, and governance on one platform.
Transcribed from the film’s narration. Captions are on the player.