Delegates are invited to meet and discuss with the poster presenters during the poster presentation sessions between 10:30-11:30 and 16:00-17:00 on Thursday, 19 November 2015.
Lead Session Chair:
Stephan Barth, ForWind - Center for Wind Energy Research, Germany
Jan Helsen (1) F Gert De Sitter (1) Dirk Van der Linden (1) Christof Devriendt (1) Stefan Milis (2) Pieter Jan Jordaens (2)
(1) OWI-lab/VUB, Brussel, Belgium (2) Sirris/OWI-lab, Leuven, Belgium
Printer friendly version: Print
Presenter's biographyBiographies are supplied directly by presenters at EWEA 2015 and are published here unedited
2007: graduated as Master in Engineering Sciences: Mechatronics
2007-2012: Phd on the dynamics of high power density wind turbine gearboxes funded by IWT and ZF Wind Power. Focus on multibody simulations of wind turbine drivetrains
2012-2014: Post-doc on model based monitoring of wind turbine drivetrains. Funded by ZF Wind Power and Siemens
2014-... Coordinator drivetrain monitoring at OWI-lab
PosterDownload poster (11.75 MB)
Big data wind turbine data intelligence platform in support of (offshore) wind R&D
OWI-Lab, the Belgian research, development and innovation platform for offshore wind energy, is monitoring turbines in all offshore wind farms in the Belgian North Sea. Several research tracks have been running since five years related to design optimization, structural health monitoring, condition monitoring and improved O&M. In order to feed the involved researchers with relevant field data, customized measurement systems have been implemented on several wind turbines. Goal is to gain additional information complementary to traditional SCADA and CMS systems. Moreover, environmental data from met masts in the Belgian North Sea are used to show the links between machine parameters and environmental parameters (extreme waves, wind gusts, temperatures, etc…). The data of these sources needs to be integrated in one database to facilitate this knowledge extraction.
Important to note is that these additional monitoring systems sample on high frequency (e.g. 5kHz) , thus introducing a new challenge with regard to data storage and processing. The benefit of this high frequency data is the ability to have more detailed information for advanced signal processing and detailed study of specific events. However, storing measurement data for historical data analysis over a long time period results in tens of terabytes. Nonetheless, we want to use this integrated high frequency sampled dataset, since it allows much deeper understanding of the degradation mechanisms by investigating the whole picture.
Traditional data approaches are not well suited to handle these types of high volumes combined with high veracity in their datasets. To tackle these challenges OWI-lab is developing a wind intelligence data platform with integrated traditional relational database and big data No-SQL architecture. This platform ensures that advanced preprocessing, data handling and complex correlations between different sources are made much faster, by decreasing post processing time of the obtained data.
This paper discusses the architecture of the OWI-Lab data platform, and illustrates the leverage potential of this integrated monitoring approach by solving several asset monitoring cases for offshore wind turbines.
Main body of abstract
OWI-Lab’s turbine data intelligence platform stores datasets from different sources in a relational database in an automated way. Several steps towards the final design of the data intelligence platform were completed. First a conceptual data model was set-up based on the different data sources. An entity relationship (ER) diagram technique was used. These insights were integrated in dual architecture of the data factory. Lots of attention was paid in this phase to the Extraction, Transformation and Load (ETL) automation of the platform. Data warehouses are typically assembled from a variety of data sources with different formats and purposes. As such, ETL is a key process to bring all the data together in a standard homogeneous environment. The benefit of automated ETL processes is that it supports massive parallel processing for large data volumes and the possibility to schedule data movement jobs on a regular basis in an automated way, making data processing much faster and less complex. The result of the project is a unique ‘automated wind turbine data intelligence platform’ to support research and development projects. Analysis of certain events and data-mining tasks can be supported in much more efficient way.
Furthermore this paper illustrates how this data intelligence platform is of added value for two O&M excellence cases: structural health monitoring and drivetrain monitoring. Turbine fatigue life is assessed based on the data in the platform, whereas failure in rotating components in predicted by progressive alarms based on the integrated dataset.
This paper illustrated an integrated wind turbine intelligence platform for O&M excellence and support of R&D projects. The underlying architecture was discussed. Moreover, real life cases underpinned the added value of this platform for asset monitoring and R&D projects.
Insights in integrated data-intelligence extraction by means of automated data-preprocessing, storage and modeling.
Knowledge about how maximum leverage can be created by combining traditional relational databases with big data approaches.