Description
GAMESA has a Predictive Maintenance System (PMS) for wind turbines in operation. This system collects measurements of some performance variables in real time and, after being compared with the predefined reference values, sends alarms to the operator of the wind energy park. Depending on the alarms received, the operator decides the operation strategy or maintenance work to carry out. In this project, GAMESA’s aim is to improve its current Predictive Maintenance System (PMS) in order to develop an advanced monitoring and prognosis system. Taking into account that GAMESA has been maintaining wind farms from the last 20 years, there is a huge amount of historical performance data available. This data will be compare with real time data to improve alarms generation system. To do this, GAMESA has to develop what an ordinary performance pattern is, and then the monitoring and comparison system will be created. After that, this new system will be able to predict wrong performance, display an alarm, and recommend maintenance operations, and through that avoid possible future breakdown in the wind turbine generator.
Summary of project results
Within this project Gamesa has worked on the development of a new advanced Health Monitoring System (HMS) that, based on a normal pattern of behavior of the main components, generates recommendations for operating and maintaining the wind turbine. This has implied the need to carry out the following tasks: Identification and selection of variables to be monitored in the main components of the wind turbine, development of the monitoring systems to enable better diagnosis of the state of the components - focusing the work primarily on the Drive-train and the tower and the foundation as a whole - and development and validation of two complementary firmware solutions for the definition of the normal pattern of behaviour. The result of the project is considered very positive for Gamesa, having developed a new solution for advanced monitoring of wind turbines based on the normal pattern of behavior of its components.
Summary of bilateral results