Come meet the poster presenters to ask them questions and discuss their work
Check the programme for our poster viewing moments. For more details on each poster, click on the poster titles to read the abstract. On Wednesday, 6 April at 15:30-16:15, join us on Level 3 of the Conference area for the Poster Awards!
PO227: An innovative edge computing solution with a universal machine learning algorithm for condition monitoring in electrical y mechanical components of the wind turbine
Iñigo Zumaran Gil, Business Development Manager, Elliot Cloud
While the entire sector is positioning itself in complex cloud solutions for condition monitoring using Biddata and machine learning, we have developed a standard algorithm for the detection of mechanical and electrical failures, both slow and fast evolution. The algorithm can be configured to detect different failure modes in a very simple way. This algorithm has been integrated into an edge computing solution that allows our module to be directly connected to the control PLC of the wind turbine or one of its components (such as the generator, etc.) and record operating data. From these data, the algorithm makes a model of the normal behavior of any equipment and later, by means of a complex algorithm, it evaluates the degradation of the equipment, generating alarms that are returned to the PLC for its use. It the congress it will be seen how the algorithm has proven its effectiveness detecting failures in bearings and generator windings, transformers, gearbox oil and gears and other critical equipment. In addition, the maintenance technician can connect to the device through his mobile phone or a tablet and can access the history of operation data registered in the equipment, being able to make a simple analysis of what has happened in the equipment, which can help to carry out the diagnosis, identify the origin of the malfunction and evaluate if the maintenance actions have been carried out correctly. Whatever equipment manufacturer could increase the value of their equipment including this low cost module.