TurbinePredict (Using Machine Learning and Artificial Intelligence to predict component faults and identify turbine underperformance)

Project Description

The project aims to develop and deploy new and innovative predictive processes to improve the performance of existing wind turbines by early identification of underperforming turbines and by early identification of parts that are about to fail. The idea is to get owners and OEMs to change to Proactive rather than Reactive maintenance. This will be done by collecting, collating and analysing data streams from the turbines using Artificial Intelligence (AI), Machine Learning and Mathematical Modelling. Phase 1 will involve consultation with existing Irish windfarm operators to assess all the tools in use by Irish windfarms. Phase 2 will be about streamlining the data capture process and setting up secure cloud-based storage for all the project data and tools. Phases 3 and 4 will be the development of the tools for predictive maintenance and for early identification of poor performance. Phase 5 will be the development of a user interface for applying the tools and reading the results. The final phase, 6, will be dissemination and sales. The process will include rigorous testing throughout and project partners will work hard to keep to abreast of any new developments in the space throughout the project.