Reduct live – Borehole water extraction.

Reduct is Verco’s newly launched AI enabled technology to automate energy management and optimise your factory.  Reduct sits alongside Carbon Desktop (or other corporate energy management platforms) to optimise the systems in your operation.

Our Early Adopter’s programme is well underway and already showing exciting real-world results.  This is the first in a series of bulletins showing live examples of Reduct’s ability to reduce cost, workload and carbon.

Reduct optimises borehole water extraction.

For this project we set up an automatic data stream into Reduct, capturing energy consumption for three (of three) borehole extraction pumps.  These were already setup on VSDs from a previous energy reduction initiative.  These VSDs are set to provide water at the desired pressure but without excess to maximise energy efficiency.

Pattern recognition immediately identifies and reports abnormal consumption trend.
When one of the pumps started consuming much more power than normal, Reduct’s pattern recognition immediately identified an abnormal trend in consumption and reported the issue to our trials team.  Reduct identified that a site technician had left one of the pumps set to “hand” while carrying out some maintenance work, resulting in excessive energy consumption.

Otherwise undetectable wastage prevented within the hour.
Within an hour the excess had been detected by Reduct and resolved by the site maintenance team, avoiding potentially thousands of pounds of additional electricity consumption.  With no immediate impact on the site water supply, it’s likely this issue would have gone unnoticed.

Optimal performance understood and maintained.
Shortly after this correction had been made, a second scheduled maintenance task resulted in this same pump being turned off and left off.  Reduct immediately identified an abnormally low consumption and a corresponding net increased consumption was recorded across the entire pump set for this system.  The system identified that operating on just two of the three pumps created a sub-optimal condition which resulted in an increase in overall electricity consumption.  This system alert directed the site technician to reinstate the third pump.  It also highlighted the performance difference between running two pumps versus three to meet the same site water demand.

Intelligent optimisation with machine learning.
Reduct uses pattern recognition to constantly assess running conditions.  In the real world, what is considered to be normal is constantly changing due to a range of factors including seasonality, production variance and time of day.  Setting up a static alert can be difficult to filter out true anomalies from normal operational variance.  Reduct uses machine learning to create dynamic thresholds to generate high-value alerts within constantly variable environments.

Become a Reduct Early Adopter.
Want to know more?  Our Early Adopters Programme is now full but due to high demand, we’re creating a second wave of Reduct early adopters in Q2 of 2021 and we would love you to join us:

Our latest news & insights