Wind farm layout optimization under uncertainty with landowners' financial and noise concerns

Modeling noise and lease soft costs improves wind farm design and cost-of-energy predictions

“…This coefficient is included in many layout optimization studies (DuPont and Cagan, 2012;Grady et al, 2005;Marmidis, et al, 2008;Mosetti et al, 1994), but underestimates costs. The optimal COE in the authors' previous work (Chen, 2013;Chen andMacDonald, 2015, 2013) ranged from USD 42 -48 per MWh, about 10 USD lower than the real industry data. To investigate the impact of the cost reduction coefficient on optimal COE, the authors conducted an optimization analysis for a portion of the Story County Wind Farm (Wikipedia, 2013), varying the number of turbines and the inclusion of the cost reduction coefficient, as shown in Figure 3.…”

“…This 21% calculation comes from the application of the enhanced cost model developed by the authors (Chen and MacDonald, 2014) which is based on the Wind Turbine Design Cost and Scaling Model (Fingersh et al, 2006) by National Renewable Energy Laboratory (NREL) and the Turbine System Cost Report from Lawrence Berkeley National Laboratory (Bolinger and Wise, 2011). This paper enhances this existing work with a detailed investigation and probabilistic modeling of two major soft costs in wind farm projects: the costs of landowner acquisition (land lease costs) as discussed in the authors' previous work (Chen, 2013;Chen andMacDonald, 2015, 2013), and noise disturbance compensation-an important contribution of this work, explained in detail in Section 2.4. Noise not only annoys landowners, but may also induce health issues (Dai et al, 2015), which makes the consideration of noise disturbance compensation a necessity.…”

“…(Fingersh et al, 2006) by NREL and the Turbine System Cost Report from Lawrence Berkeley National Laboratory (Bolinger and Wise, 2011), a wind shear model (Ray et al, 2006), Jensen's (1983) wake loss model, and a power model for a GE 1.5sle turbine (Archer and Jacobson, 2007). The authors' previous work (Chen, 2013;Chen andMacDonald, 2015, 2013) discusses the latter four models. The NREL model includes all costs associated with the construction of a wind farm, including the costs of cabling, roads, and transportation, although these costs do not change based on the locations of the turbines.…”

“…Wind condition is modeled as aleatory uncertainty using a wind rose, shown in detail in section 2.5.2. Three epistemic sources of uncertainty are modeled in this paper: wind shear, same as in previous work (Chen, 2013;Chen andMacDonald, 2015, 2013); landowner participation, with more detail given in previous work (Chen, 2013;Chen andMacDonald, 2015, 2013) and Section 2.4.3; and monetary compensation required for a given noise disturbance, as represented by a Willingness-to-Accept-Noise utility model, explained below. There are multiple sources of uncertainty within the noise propagation model, but they are self-contained and not detailed here.…”

“…This paper uses an optimization-under-uncertainty framework developed previously by the authors (Chen, 2013;Chen andMacDonald, 2015, 2013), to specifically address two major soft-cost challenges: (a) determine land lease terms for all participating landowners and (b) determine noise disturbance compensation. The robust design optimization problem has two objectives for the farm's COE: minimize the normalized mean value and standard deviation.…”