Description du projet
Un système intelligent détecte les défaillances des pompes à vide
Les pompes à vide sont utilisées dans un large éventail d’applications scientifiques et industrielles, notamment dans le traitement des aliments et des semi-conducteurs, ainsi que dans la production de produits pharmaceutiques et chimiques. Elles retirent le gaz d’un volume scellé, laissant un vide partiel. Les dysfonctionnements ou les pannes provoquées par une maintenance ou des réglages inappropriés entraînent d’importantes pertes de revenus. Le projet StarLink, financé par l’UE, s’emploie à mettre au point un système qui tirera parti de l’analyse de données et de l’apprentissage automatique pour identifier les dysfonctionnements des pompes à vide avant qu’ils ne surviennent. Il proposera des mesures de maintenance correctives et ajustera automatiquement les paramètres de fonctionnement. Le système proposé permettra de réduire de 30 % les coûts liés aux arrêts des pompes à vide et d’accroître leur productivité de 50 %.
Objectif
Vacuum pumps are used in thousands of industrial applications, playing a vital role in food processing, semiconductors, chemicals, pharmaceuticals and many other manufacturing and assembly processes. However, today’s pumps are currently unable to provide any type of insights that could help users anticipate a pump malfunction, plan maintenance procedures or setting the adjustments. Pump malfunctions or breakdowns, due to unplanned maintenance or improper settings, cost millions of euros in lost revenues every year as production and logistic lines lie idle waiting for pumps to be fixed, and when they are not optimized their productivity decrease or their energy consumption go up.
But now, DVP, a vacuum pump manufacturer, has developed the solution to these challenges through StarLink, the world’s first intelligent vacuum pump system. StarLink is a patent-pending system that uses data analytics and machine learning to identify pump malfunctions before they happen, propose actions to be taken, and automatically adjust the operation parameters if the problem relates to the setting. This will reduce pump downtime-related costs by 30%, increase their productivity by 50% and make easier the operation manager tasks.
The combination of our deep knowledge of vacuum pumps needs with the machine learning expertise of the university of Ferrara will create the most intelligent device to improve the competitiveness of European companies. Additionally, StarLink will contribute to DVP’s growth in terms of employees and product portfolio since we will be able to offer a wider range of products and services related to vacuum pumps, which will allow us to enter new markets and sell more units. By 2023, it will generate €3M in yearly revenue with net profits of €2M to our company.
Champ scientifique
- natural sciencescomputer and information sciencesinternetinternet of things
- medical and health sciencesbasic medicinepharmacology and pharmacypharmaceutical drugs
- social scienceseconomics and businesseconomicsproduction economicsproductivity
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringsensors
- natural sciencescomputer and information sciencesartificial intelligencemachine learning
Programme(s)
Régime de financement
SME-1 - SME instrument phase 1Coordinateur
40018 SAN PIETRO IN CASALE
Italie
L’entreprise s’est définie comme une PME (petite et moyenne entreprise) au moment de la signature de la convention de subvention.