Southwest Research Institute developing system to pinpoint small hazardous leaks


A team at the Southwest Research Institute has developed a self-teaching leak detection system for hazardous chemical spills.

One day, the system could help you save some money

Using a two camera system, a team at SwRI paired the cameras with an artificial intelligence algorithm they developed.

"This system is able to actually differentiate gasoline from water,” said Maria Araujo, a manager with the High Reliability Systems section. “In other words an almost colorless liquid from water on grass and dirt."

The smart leak detection system can also detect leaks of other hazardous liquids like crude oil and diesel

A slight change in the algorithm and the system could be used for water pipelines too.

San Antonio Water System reported in 2016 nearly 10 billion gallons of water was wasted.

To pump and treat that much water it costs $40 million which was absorbed by ratepayers.

"This camera system can detect very, very small leaks down to a few gallons," said Daniel Davila, a research engineer.

SAWS already uses remote bots for underground leak detection.

With their self-teaching algorithm, researchers say this system can pinpoint tiny leaks.

"That's really what we're trying to solve with this technology is trying to get those leaks before the cause larger ruptures,” said Davila.

The detection system was recently recognized by R&D Technology magazine as one of the 100 most significant innovations of the year.

close video ad
Unmutetoggle ad audio on off