E. Echavarria Delft University of Technology, Mekelweg 2, 2628 CD Delft, The Netherlands e-mail: e.echavarriauribe@tudelft.nl B. Hahn Institut für Solare Energieversorgungstechnik (ISET), e.V. Königstor 59, D-34119 Kassel, Germay e-mail: bhahn@iset.uni-kassel.de G. J. W. van Bussel Delft University of Technology, Kluyverweg 1, 2629 HS Delft, The Netherlands T. Tomiyama Delft University of Technology, Mekelweg 2, 2628 CD Delft, The Netherlands Reliability of Wind Turbine Technology Through Time This study attempts to obtain more detailed knowledge of failures of wind turbines (WTs) by using the German “250 MW Wind” test program database. Specific objectives are to show the reliability of some major components and to analyze how their design has advanced through time, what the main failures are, and which technologies have proven to work. Within the program, reports on operation and maintenance are analyzed with respect to WT type, size, and technologies used. This paper presents a comparison of component reliability through time, with respect to their technology. The results show significant differences in reliability for certain subcomponents depending on the size of the WT and especially on the type of power control. For instance, induction generators show half the annual failure rate compared to synchronous generators. The study also includes failures of other components that are affected or added due to the use of the components being analyzed. In general, the results show that failure rates of WTs de- crease with time. Most of the data show a short period of “early failures” and later a long period of “random failures.” However, this is not the case for the megawatt class: As technology is introduced into the market, WTs show a longer early failure behavior, which has not yet become stable. Furthermore, large turbines, included in the database analyzed, show a significantly higher annual failure rate of components, per WT. This may be due to the immature technology of the WTs included in the database. DOI: 10.1115/1.2936235 Keywords: reliability, maintenance, failures, wind turbines 1 Introduction Design of wind turbines WTs has evolved through time, with the aim of becoming less expensive and producing energy more efficiently. Design changes take place at all technological levels. WT manufacturers have tried a broad variety of possibilities. They have investigated different design topologies, such as vertical or horizontal axis of rotation and upwind or downwind placement of the rotor, and also considered the use of different control strate- gies or changes in smaller components such as brakes and blade tips. This evolution of technology for different components is due not only to the experience of what has proven or not proven to work but also to the growth of the wind energy industry. Fossil fuels are becoming less popular, giving opportunities for renew- able energies to develop, and consequently WTs are becoming more commercial, producing energy at competitive prices. For the past 15 years, WTs have grown in size and in rated power, and this has been reflected in their design. However, the current de- velopment slows down as WTs reach rated powers of 3 MW, 5 MW, and more. Manufacturers attempting to make WTs more commercial and less expensive have looked into other fields where technology is proven and where, in many cases, components can be obtained off the shelf. However, in some cases, results show lower reliability than expected, mainly because operational conditions for WTs are not known with enough detail. Unsteady operational conditions differ greatly from conditions in other industries. For instance, gearboxes show high failure rates in spite of their commerciality. The design of the WT seems simple at first sight. The basic structure rarely changes, although there are still different alterna- tives from the technology point of view. For instance, there are vertical or horizontal axis designs; upwind or downwind rotor placement; 1, 2, 3, or even more blades; stall, active stall, or pitch regulation systems; constant or variable speed operation; synchro- nous, induction, doubly fed induction generators, direct-drive train; aerodynamic brakes, etc. Some of these concepts have re- mained, some others have disappeared, or they are just rarely used nowadays. Despite of all these possibilities, the industry has coalesced to a few topologies: WTs are usually horizontal axis, three bladed ma- chines and nowadays they are pitch controlled. Mostly, they are equipped with doubly fed induction generators, and gearboxes convert the rotational speed of the rotor to the speed of the gen- erator. As a second trend, direct driven types are equipped with multipole synchronous generators, which rotate at the same speed as the rotor, in which case a gearbox is not needed. Typically, both concepts use two independent braking systems providing safe op- eration, where at least one is often an aerodynamic brake. In gen- eral, technology has escalated in size with relatively few innova- tive breakthroughs. As the role of wind energy within the public energy supply is now becoming more important, it also becomes more important for operation and maintenance O&M to determine reliability of WTs as systems and failure rates of single components through time in order to identify the best designs or configurations. As of today, data have been analyzed regarding the wind tur- bine as a complete system, without distinguishing between differ- ent component technologies, throughout their operational age. There are historical data on the development of components 1,2, which mention some of the WT topologies included in the data- base. However, this analysis does not consider reliability, espe- cially through time. There are not many sources of information to study reliability of WT technology through time. Only few databases exist with failure information such as the WMEP 3 and LWK 4 in Ger- many, one in Denmark published by Windstats Newsletter 5, and according to Ref. 6, one in Finland published by the VTT, and one in Sweden published by Elforsk. Contributed by the Solar Energy Engineering Division of ASME for publication in the JOURNAL OF SOLAR ENERGY ENGINEERING. Manuscript received July 30, 2007; final manuscript received April 25, 2008; published online June 26, 2008. Review conducted by Spyros Voutsinas. Paper presented at the 2007 EWEC 2007 MILAN: European Wind Energy Conference and Exhibition. Journal of Solar Energy Engineering AUGUST 2008, Vol. 130 / 031005-1 Copyright © 2008 by ASME Downloaded 20 Nov 2008 to 145.94.187.10. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm