EN
Back to the archive

The encyclopedia · Engineering & Operations · Operational decision · 2016–2017

Aqualia's wastewater plant cut power 13.5% by policing its own setpoints

An IBM cognitive optimizer ran a one-year pilot at Aqualia's Lleida plant, cutting electricity 13.5% and phosphorus chemicals 14%.

Aqualia · IBM

the move

Aqualia's wastewater plant in Lleida, Spain, ran on conservative setpoints that over-used electricity and chemicals to stay safely within effluent limits.

IBM built a cognitive optimizer that models the plant as a constrained Markov-decision-process and continuously recommends better setpoints.

In a one-year pilot at the plant, the system held water quality within regulation while cutting energy and chemical use.

why it works

  • It targets the conservative operating setpoints, not the equipment.
  • Optimizing control loops keeps effluent within legal limits.
  • It reduced electricity consumption by 13.5% in the pilot.
  • It cut phosphorus-removal chemicals by 14%.
the payoffLet the process tune its own operating pointsclever

what transfers

For environmental processes, there is headroom between 'safe' and 'optimal'; a real-time optimizer can find it without crossing the compliance line.

what came after

The one-year pilot at Aqualia's Lleida plant showed a 13.5% reduction in electricity consumption and a 14% reduction in the chemicals used to remove phosphorus, demonstrating an optimization-based way to save resources in wastewater treatment.

references

spotted an error? The archive wants to know.

same kind of clever