Over the lifetime of a wind turbine, the cost for operation and maintenance sums up to a considerable portion of its installation cost. To minimize the total cost resulting from both maintenance activities and production losses due to failures is important to increase the competitiveness of wind energy. Experience has shown a direct impact of service maintenance intervals on wind turbine availability on the one hand, and a correlation between the operating environment (e.g. wind speed, humidity, grid disturbances) and failure frequency of wind turbines on the other hand.
At present, the (preventive-type) service maintenance is carried out according to a time-based strategy in fixed intervals of 6 months. The project intends to develop methods to flexibly adapt the maintenance intervals, taking into account the loads experienced by a turbine as well as information from condition-monitoring systems where this is available. For this purpose, the project involves in-depth investigations of the correlation between different environmental variables and the frequency of failures in wind turbines, based on which quantitative, data-based methods for maintenance management will be developed and implemented.
The project aims at the development and application of data-based, quantitative methods to achieve increased reliability, availability and profitability of wind turbines. By deriving knowledge from data by means of statistical methods and by utilizing advanced models for maintenance planning, the project intends to improve the competitiveness of maintenance service providers on the one hand, and to provide additional value to wind turbine owners by means of increased technical availability on the other hand.
The objective is the development of a generic method for load- and risk-based maintenance planning which is in principle applicable to a variety of wind turbine components and failure types, but which will, in the scope of this project, be implemented for specific wind turbine models and selected components. The focus will be on modern wind turbines with rated capacities of 2MW and above, because for these comprehensive SCADA data (and usually also CMS data) is available and because these are considered to dominate the wind turbine populations in the future.
Specific tasks and expected outcomes:
The expected final outcome of the project is in-depth statistical analysis of wind turbine failure data and SCADA data in order to quantitatively identify the impact of the external influencing factors on failure frequency. A mathematical model based on this correlation will be developed within the project. This project also aims to deepen the understanding of possibilities and constraints of different model based approaches for maintenance optimization. Finally based on the experience gained through this project recommendation will be made for utilization of SCADA and CMS data for maintenance planning, which will lead to economic benefits by saving cost for O&M activities.