Evaluation of wind farm efficiency and wind turbine wakes at the Nysted offshore wind farm

Abstract

Here, we quantify relationships between wind farm effi ciency and wind speed, direction, turbulence and atmospheric stability using power output from the large offshore wind farm at Nysted in Denmark. Wake losses are, as expected, most strongly related to wind speed variations through the turbine thrust coeffi cient; with direction, atmospheric stability and turbulence as important second order effects. While the wind farm effi ciency is highly dependent on the distribution of wind speeds and wind direction, it is shown that the impact of turbine spacing on wake losses and turbine effi ciency can be quantifi ed, albeit with relatively large uncertainty due to stochastic effects in the data. There is evidence of the 'deep array effect' in that wake losses in the centre of the wind farm are under-estimated by the wind farm model WAsP, although overall effi ciency of the wind farm is well predicted due to compensating edge effects.When wind turbines are arranged in large wind farms, there will inevitably be a loss of power output due to wind turbine wakes. This means that within the wind farm, wind speeds do not recover to their freestream value after encountering the fi rst turbine (or row of turbines), and thus are lower than the freestream at subsequent turbines because kinetic energy has been extracted. Although this is a well-known phenomenon, power losses due to wakes are diffi cult to predict accurately due to the temporal and spatial variability of wind speed, direction, turbulence and atmospheric stability. Accurate quantifi cation of power losses due to wind turbine wakes in differing wind climates and wind farm layouts is essential to optimal wind farm design. In small and medium size offshore wind farms (less than three rows), wake effects have been quantifi ed experimentally, 1,2 Using the observational data, performance of both state-of-the-art wake and wind-farm models has been demonstrated to be at least satisfactory. 3,4 However, preliminary model results from the fi rst large offshore wind farms suggests that current wind farm models using standard parameter values under-estimate power losses in large wind farms due to wind turbine wakes, leading to over-prediction of power output. 5,6 Better prediction can be obtained by adjusting parameters or using 'added roughness' over the area of the wind farm. 7,8 The implied 'deep array effect' has prompted renewed research into wake behaviour. The focus is to understand and predict power losses from wakes in large wind farms with reduced uncertainty.In the simplest form, the velocity defi cit of an individual wind turbine wake is dictated by the thrust coeffi cient (c t ) (and thereby to wind speed; see Figure 1) and the downstream distance, x [9]. The formula used in WAsP which has a fl at velocity profi le in the wake is: Wake losses at Nysted offshore wind farm R. J. Barthelmie and L. E. Jensen 574 V V c D D kx t 0 0 0 2 Wind Energ. 2010; 13:573-586

Perturbations to the Spatial and Temporal Characteristics of the Diurnally-Varying Atmospheric Boundary Layer Due to an Extensive Wind Farm

“…It is now well understood that when several wind turbines are clustered together forming a wind farm, the net harvested power is less than what theoretically would be extracted by an equal number of isolated turbines (Barthelmie et al 2010). Several works have analyzed the effects of wind-turbine arrangement and wake superposition on the resultant harvested power (Barthelmie et al 2007(Barthelmie et al , 2009; Barthelemie and Jensen 2010;Porté-Agel et al 2013;Stevens et al 2014) as well as wake interactions and the wake-recovery processes (Frandsen 1992;Emeis and Frandsen 1993;Frandsen et al 2006;Cal et al 2010;Markfort et al 2012). Analytical wake models for wind farms developed decades ago (Lissaman 1979;Jensen 1983;Katic et al 1986) continue to serve as the bedrock of various engineering software packages used for wind-farm design such as the Wind Atlas Analysis and Application Program (WAsP).…”

Stokastik sınır analizi kullanarak rüzgâr türbinleri için etkinlik değerlendirmesi

“…[16] Rüzgârın karakteristik özellikleri ve diğer atmosferik veriler de literatürde sıklıkla rüzgâr enerji sistemlerinin performans değerlendirmeleri için kullanılmıştır. Rüzgâr hızının profili [29,30]; rüzgâr hızı, yönü, türbülans ve atmosferik durağanlık ile performans arasındaki ilişki [31]; rüzgâr profili ile ilgili çalışmalara örnek verilebilir. Bunların dışında, Pieralli vd.…”

Synergistic Effects of Turbine Wakes and Atmospheric Stability on Power Production at an Onshore Wind Farm

“…Several modeling studies have looked at simulating downwind velocities within these farms. These include: the Risø WAsP model (Barthelmie and Jensen 2010), the MM5 model (Rooijmans 2004), and the RAMS model (Roy et al 2004). More recently, attempts have been made to include wind turbine parameterizations in the Weather and Research Forecasting (WRF) model (Adams andKeith 2007, Fitch et al 2011).…”

“…Porte-Agel 2009), field experiments (e.g., Neustadter and Spera 1985, Elliott 1991, Conzemius 2010, Schepers et al 2012, numerical simulations (e.g., Ainslie 1988, Dahlberg et al 1991, Schepers 2003, Zahle and Sørensen 2008, Barthelmie et al 2009, Porte-Agel et al 2011, Politis et al 2012, and from studies utilizing both experimental and modeling methods (e.g., Barthelmie and Jensen 2010). It is well documented that downwind turbines in multi-Megawatt wind farms often produce less power than upwind turbine rows (Neustadter and Spera 1985, Elliott 1991, Barthelmie et al 2009, Barthelmie and Jensen 2010.…”

“…It is well documented that downwind turbines in multi-Megawatt wind farms often produce less power than upwind turbine rows (Neustadter and Spera 1985, Elliott 1991, Barthelmie et al 2009, Barthelmie and Jensen 2010. These wake-induced power losses have been estimated from 5% to up to 40% depending on the turbine operating settings (e.g., thrust coefficient), number of turbine rows, turbine size (e.g., rotor diameter and hub-height), wind farm terrain, and atmospheric flow conditions (e.g., ambient wind speed, turbulence, and atmospheric stability) (Sanderse 2009).…”