Wind distribution and capacity factor estimation for wind turbines in the coastal region of South Africa T.R. Ayodele ⇑ , A.A. Jimoh, J.L. Munda, J.T. Agee Department of Electrical Engineering, Tshwane University of Technology, Private Bag X680, Pretoria 0001, Staatsartillerie Road, Pretoria West, South Africa article info Article history: Available online 24 August 2012 Keywords: Wind distribution Weibull distribution Capacity factor Wind turbine South Africa abstract The operating curve parameters of a wind turbine should match the local wind regime optimally to ensure maximum exploitation of available energy in a mass of moving air. This paper provides estimates of the capacity factor of 20 commercially available wind turbines, based on the local wind characteristics of ten different sites located in the Western Cape region of South Africa. Ten-min average time series wind-speed data for a period of 1 year are used for the study. First, the wind distribution that best models the local wind regime of the sites is determined. This is based on root mean square error (RMSE) and coef- ficient of determination (R 2 ) which are used to test goodness of fit. First, annual, seasonal, diurnal and peak period-capacity factor are estimated analytically. Then, the influence of turbine power curve param- eters on the capacity factor is investigated. Some of the key results show that the wind distribution of the entire site can best be modelled statistically using the Weibull distribution. Site WM05 (Napier) presents the highest capacity factor for all the turbines. This indicates that this site has the highest wind power potential of all the available sites. Site WM02 (Calvinia) has the lowest capacity factor i.e. lowest wind power potential. This paper can assist in the planning and development of large-scale wind power- generating sites in South Africa. Ó 2012 Elsevier Ltd. All rights reserved. 1. Introduction In recent years, efforts have been made around the world to generate electricity from renewable energy sources. This is due to the infinite availability of their prime movers and in an effort to reduced harmful emissions into the environment. One of the ways in which electricity can be generated from these sources is to use wind turbines that convert the kinetic energy in a mass of moving air into electricity. At present, the wind power growth rate stands at over 20% annually. At the end of 2010, global cumulative wind power capacity reached 194.4 GW [1] and it is predicted that 12% of the world electricity may come from wind power by 2020 [2]. In South Africa, the interest in wind as a resource for electricity generation is receiving considerable support from stake holders. This was evident at the last wind power summit (Wind Power Afri- ca 2011) held in Cape Town in May 2011. Currently, the major indigenous energy resource for electricity generation in the coun- try is coal, which constitutes 85% of the primary energy mix. This contributes significantly to environmental pollution and leads to high emission of green house gas (GHG) in the country. South Africa is the 14th highest emitter of GHG in the world [3]. However, as a signatory to the UN Framework Convention on Cli- mate Change, the country has committed itself to the international community to reduce GHG emission. Part of the plan to honour this commitment is contained in the ‘‘Integrated Resource Plan (IRP)’’ which has 56% of wind power in the first phase of its renewable en- ergy feed-in tariff programme. In view of this, it is essential to have a reliable knowledge of the wind distribution and the appropriate turbine selection based on the analysis of local wind regimes at dif- ferent sites in the country. An understanding of the performance of a wind turbine, in re- sponse to different wind speeds at a proposed site, is a prerequisite for the successful planning and implementation of a wind power project. Both the wind speed and its distribution have an influence on the performance of a wind turbine. Therefore, the operating parameters of wind turbines which are characterised by four parameters, the cut-in (v in ), rated (v r ), the cut-out (v co ) wind speed and the turbine nominal power should be a good match with the prevailing wind characteristics and distribution of the local wind regimes [4]. However, selecting wind turbines to match a specific site has traditionally been done by designing a new turbine based on the wind characteristics of a given site. However, this method is time consuming and uneconomical. A more practical approach could be to select the turbine that best matches the wind charac- teristics of a specific site from the commercially available ones [5,6]. The turbine power curve operating parameters can be combined with the statistical wind distribution parameters of a 0196-8904/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.enconman.2012.06.007 ⇑ Corresponding author. Tel.: +27 735605380. E-mail addresses: tayodele2001@yahoo.com, Ayodeletr@tut.ac.za (T.R. Ayodele). Energy Conversion and Management 64 (2012) 614–625 Contents lists available at SciVerse ScienceDirect Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman