European Pattern Recognition (EPR) is a research and development project funded by the EU and the Swedish Energy Agency, with Metrum leading the project. One part of the project is based on the development of software applications for trend analysis, where AI (artificial intelligence) and pattern recognition are used to predict what will happen in energy grids and to detect discrepancies and errors at an early stage. And there has now been a breakthrough.
“We’re very excited and pleased to be able to say that we have achieved a breakthrough in our project,” says Metrum CTO Robert Olofsson. “This means that we can now develop practical solutions and begin to apply them in electrical grids.”
By using existing measurement data from many different measurement instruments and performing continuous automated analyses of expected future values for both power quality and variations, predictions can be made of voltage changes as well as power demands, something that is of enormous value to both utilities and major industries. Better assessments of the anticipated power demand that determine tariffs can be made, contributing to direct cost savings.
“The breakthrough is a major technical advance where a combination of AI and pattern recognition can save huge amounts of money by giving companies the capability to avoid costly disruptions, to perform condition based maintenance at the right place, and not the least, to plan for appropriate power demands,” says Magnus Andersson, CEO at Metrum.
Metrum’s new instrument platform PQX3 and the PQX3-FR instrument is designed and capable to include big data and similar analyses, often referred to as third-generation measurements, in the respective instruments and provide new capabilities for classification and identification of disruptions.
For more information: www.europeanpatternrecognition.eu
