Great progress has been made in the recent past in the technical handling of large amounts of operating data from wind turbines (WTGs). However, there is still a lack of suitable analysis procedures especially for high-frequency operating data of the wind energy application for implementation in control room software and maintenance practice. On the other hand, there is a large number of new methods and findings in data and system analysis which were developed in basic research but are not or hardly ever applied in the wind energy sector. Non-linear physical processes are often considered, which typically require adjustments of the analytical methods. From the company's point of view, there is an urgent need for action in the wind energy sector to make use of the existing potential of these already available data and developed analytical methods for optimisation and cost reduction.

The aim of the "WiSA big data" project is to contribute to early fault detection and diagnosis on wind turbines by analysing operating data with high temporal resolution, thus supporting decisions in maintenance planning and implementation. For this purpose, methods are developed and tested which have proven themselves on the basis of operating data averaged every 10 minutes for application to high-resolution data. On the other hand, novel methods for early fault detection are transferred to wind energy applications. The developed and tested methods will be subjected to a practice-oriented quantitative comparative evaluation.

• Data Stream Analysis
• Relational and non-relational databases
• Scalable aggregation and transformation methods