Wessex will be the world’s first and largest AI-monitored sewer system.

UK-based Wessex Water is a regional water and sewerage business serving 2.8 million customers across the south west of England, owned by by YTL Power International of Kuala Lumpur, Malaysia. The utility has engaged StormHarvester, a provider of solutions for smart drainage, to deploy AI over its sewer network spanning 35,000 km. (New Civil Engineer)

Wessex conducted a three-month trial in the city of Bath to test StormHarvester’s AI with respect to the identification of early-forming sewer blockages, false control room alarms, and for condition-based maintenance. Wessex wanted to glean additional information from its existing network of wastewater sensors.

How they tested the AI

Using existing sensors, StormHarvester’s proprietary AI (machine learning) algorithms and predictive analysis tools comprised within its Intelligent Sewer Suite were used on sensor data along with hyperlocal rainfall forecast data, to predict network levels and detect potential blockage formations in real-time.

They extracted historic sewer level data and historic rainfall levels in a 1.5 square km grid for each of the 98 sensored network assets, and undertook tens of millions of iterative machine learning calculations in order to ‘learn’ sewer asset behaviour in both dry and wet weather periods.

Test results

During the three-month period the AI system detected over 60 early blockage formations in real time – it is estimated that at least two of those had the potential to develop into Category 3 spillages without the AI-flagged early intervention.

The system detected early blockages with an accuracy of 92% and cut control room alarms by 97%.

“We set out to test the ability of machine-learning to predict blockage formations in sewer networks early and intervene before they became service failures, better separate genuine control room alarms from alarms triggered because of high volume rainfall during wet weather and dramatically improve the response time to service failures,” said Wessex Water asset technology manager Jody Knight.

“In short, the technology worked and exceeded expectations.”

Related Story: Verizon Using AI For Optimal Placement Of 5G Transmitters

Image Credit: StormHarvester