Artificial Intelligence: the beginning of a new reality

Committed to ongoing development, Socomec continually invests in Artificial Intelligence (AI) in order to leverage the benefits of new technologies which can open up new perspectives and extend the life of equipment and installations – and are essential in the development of new solutions.

This development activity encompasses data science and machine learning, as well as working across cloud infrastructure and IoT platforms to enable the deployment of connected products and online applications.  Engaging with the scientific community, Socomec has committed to a five-year sponsorship of the new “Data Sciences and Artificial Intelligence” chair at the Telecom Physique Strasbourg School of Engineering.  Bringing together academic research with real-world applications develops a unique insight into how AI will help to derive the value from data to improve performance, whilst also refining product design and optimising ongoing maintenance.

Leveraging the value in data

Equipment within an electrical infrastructure performs measurements – either to support the operation or for provision to the customer.  Thanks to AI, data can be converted into useful and valuable information to improve performance - to support adaptive, preventive maintenance activities, to detect abnormal changes in the behaviour of electrical installations, and to optimise operations.

By cross-analysing behavioural product data with environmental data and other influencing factors, the installation’s behaviour can be verified. In turn, this means that functionality can be regulated and efficiency can be guaranteed over time.

The detection of atypia can flag the early appearance of faults.  Through learning, AI makes it possible to qualify the atypia and generate alarms to flag concern.  Marc Capot, Applied Research Manager, Socomec, explains; “Our expertise in AI means that we can foresee the development of very valuable new features that could detect abnormal behaviour or the drift of certain parameters of an installation – a game-changer when it comes to monitoring the insulation of an electrical installation or identifying over-consumption as early as possible, for example.  This real-world information means that we will start to create unique insights to support fully informed decision making – decision making with both hindsight and foresight.”

Individually prescribed maintenance

To ensure that products operate at peak performance over several decades, preventive maintenance is vital.  Thanks to AI, and the unsurpassed insight that it can deliver, maintenance activity can be completely adapted to the life of the product – efforts and interventions can be ramped up when conditions are severe and reduced where they are more favourable.  Through the continuous monitoring of connected products – and the use of algorithms to better understand how critical components will age over time – it is now possible to prescribe individual maintenance plans based on the equipment’s actual usage. 

Intelligent design

The advent of IoT and AI allow for truly inventive and intelligent design, by enabling designers to better understand the usage and behaviour of products in real-world settings – with real-world influences. 

Philippe Balleux-Pruvost, IoT Program Director, comments; “The opportunity to refine product designs based on actual performance and conditions has never been greater. By combining human capabilities with Artificial Intelligence, we can leverage AI to spot patterns within limitless data whilst also maintaining an overview of the bigger picture.  This means that we can start to understand past behaviours and events with greater context and insight – and it also means that we can predict what might happen in the future with unprecedented accuracy and speed.

“By bringing together these bodies of knowledge, the future that we have started to reimagine will soon be a reality.  A new era of productivity and efficiency will become our new normal.  Our vision is that 2022 will be the year of realisation – with the first functionalities to integrate AI being available as a piloted maintenance offering for our power conversion equipment, including UPS and storage systems. When an electrical installation is equipped with renewable sources and storage systems it is possible to optimise its operation – focusing on carbon footprint, energy costs or availability.  AI makes this a reality through the analysis of renewable energy consumption and production predictions, using complex analyses to optimise performance.  With AI, a better future is now within reach.”