Chem Plants Dream of Electric Bleeps

Article by Simon Jagers and Gerben Gooijers

Simon Jagers and Gerben Gooijers show how you can raise plant reliability using current and voltage data

EVERY year since 2017, Nouryon (formerly Akzo-Nobel Specialty Chemicals) runs an innovation challenge called Imagine Chemistry, which recognises novel technologies with the potential to drive growth in the chemical industry. In 2018, our company was one of the winners. 

At Semiotic Labs, we develop machine learning technology that can accurately predict over 90% of developing failures in rotating equipment such as blowers, pumps, and compressors, up to five months in advance. In late 2018 Nouryon trialled our system – named SAM4 – at its chlorine plant in Ibbenbüren, Germany to monitor the health of several blowers and pumps. Based on the success of that project, it extended its SAM4 pilot in mid-2019 to include two dozen pumps, blowers, and compressors at the company’s salt plant in Delfzijl, the Netherlands. Now, in 2020, Nouryon is rolling out SAM4 to additional assets at both trial plants and five other sites in Europe.

So what is this novel technology that has Nouryon so enthused?

Motor current signature + voltage

SAM4’s anomaly detection algorithms are based on a technique called motor current signature analysis (MCSA). Where other techniques analyse vibrations or oil or temperature to monitor equipment health, MCSA analyses alternating current signals.

The technique has a long track record. It was first suggested in the 1970s to reliably monitor motors in nuclear power plants from a safe distance. Since then a vast body of research has proven MCSA’s value for a wide range of industrial equipment, from agitators to yaw systems (used in wind turbines). 

SAM4 adds voltage measurements to the basic technique. That raises the system’s sensitivity and accuracy. It also confers the ability to automatically handle varying loads and speeds and calculate energy efficiency, which allows it to monitor assets operating in flexible production systems.

For the rest of this article, we’ll use the acronym MCSA to refer to systems that capture both current and voltage.

Making maintenance 100% predicatable

Article by Simon Jagers and Gerben Gooijers

Simon Jagers and Gerben Gooijers are Founders at Semiotic Labs

Recent Editions

Catch up on the latest news, views and jobs from The Chemical Engineer. Below are the four latest issues. View a wider selection of the archive from within the Magazine section of this site.