Isotrol is a nationally and internationally recognized engineering firm that develops products, projects, solutions and software for supervision and control systems, and also offers specialized outsourcing services. One of the most important business areas of the company is the renewable energy sector, being responsible for the management of wind farms and photovoltaic parks in Spain and other countries.
The main objective of this project is to put into production a system capable of detecting failures in the operation of wind power plants as well as to give a recommendation of the type of failure that occurs. The system must also learn from information provided by external agents so that the recommendation system works with increasing accuracy. This is a very complex problem given the high number of variables monitored in a wind farm as well as the small number of cases from which the system must learn. Different supervised and unsupervised learning techniques will be developed and different approaches to the problem will be proposed to analyze which is the best solution. This is a proof of concept and the learning of the system can be transferred through transfer learning techniques to other wind power plants or even to photovoltaic plants.