”Big Data” Initiative as an IT Solution for Improved Operation and Maintenance of Wind Turbines

    “Big Data” (BD) problems require handling extremely large or complex datasets that would be
    difficult and expensive using traditional relational databases. Software solutions with distributed
    processing, weakened consistency requirements and well-designed data models help overcoming scalability issues.

    Wind energy systems produce extremely large datasets. Today’s wind farm operators either do not collect all available data in a central, easy to access database, or they delete valuable data, because of scalability issues of traditional databases. Emerging “Big Data” tools and algorithms enable collecting all of the most detailed data; moreover, data may not be deleted at all. This is a huge advantage for wind farm operators, because detailed information can be (re)used later for many purposes: e.g., building failure detection and prognosis models, ad-hoc analysis of the past becomes feasible.

    The paper shows how detailed operation data from a large number of wind farms can be collected and stored for further use. A Business Intelligence (BI) reporting prototype system for wind farm analytics is described, with describing typical cases of wind turbine operations where advantages of the “Big Data” initiative can be exploited. Performance tests for collecting and storing of SCADA data from many wind farms prove the advantages and applicability of the proposed method.

    Viharos, Zs. J.; Sidló, Cs. I.; Benczúr, A. A.; Csempesz, J.; Kis, K. B.; Petrás, I.; Garzó, A.
    European Wind Energy Association (EWEA) Conference, “Make your vision reality”, 4-7. February, 2013, Vienna, Austria, S9.3, pp. 184-188.