Comparison of pre-construction and operational energy yield assessments for wind farms across the US

Enertis, in partnership with our sister company Barlovento, currently offers world-class wind resource and energy yield assessment (“WR&EYA”) capabilities and technical consulting. Barlovento has been providing WR&EYA services globally for over 20 years and is a member of the MEASNET laboratory network and accredited according to the IEC-17025 standard.

In order to validate Enertis’ and Barlovento’s WR&EYA methodologies and results on US wind power projects specifically, Enertis has partnered with a leading US wind power project owner / operator (the “Partner”) in order to conduct pre-construction WR&EYA validation studies, as well as operational energy production estimate (“OEPE”) benchmarking and deviation analyses, on a selection of the Partner’s US operational wind assets.  

The methodology used by Barlovento for WR&EYA is based primarily on industry-known and widely accepted MEASNET and IEC standards, which standards are further outlined here-in.   

This report outlines the methodologies and results for a selection of the Partner’s US wind projects, which projects are summarized in the table below.  

Table 1. Wind Farm main characteristics 

For Wind Farm 1, given the relatively complex site, our assessment focused on the importance of wind flow model selection by performing a case study to this effect using different wind flow modeling approaches (CFD, linear, mesoscale) in the pre-construction WR&EYA. For Wind Farm 1, deviations ranged from 10.9% (CFD) to 14.9% (mesoscale), depending on the wind flow model used. The largest contributors to these deviations were limitations in modeling wake effects, terrain complexity, and the absence of reasonably estimated avian curtailment impacts at the pre-construction stage.

For Wind Farm 1, based on our analysis, deviations originating from long-term wind speed, wind flow modelling and wake effects have been calculated as 1.0%, 3.1% and 4.7%, for the three different wind flow modelling approaches studied (CFD, linear, mesoscale), respectively. This wind flow modelling case study highlights the importance of correct wind flow model selection in the pre-construction EYA; for the relatively complex site and location of Wind Farm 1, it was expected (and confirmed) by Enertis that a CFD model would produce the best result.

For WR&EYA method validation purposes, Enertis performed a recalculation of its pre-construction EYA for Wind Farm 1, applying a 4.8% avian loss with the corresponding 2.7% uncategorized losses (“other” losses, a significant portion of which likely correspond to avian losses), and has estimated that the total net deviation would decrease to 2.2% for the CFD modelling approach case. Other loss categories such as availability, electrical system regulation, electrical losses, and high temperature derates contributed to deviations between predicted and observed energy production.  

The following figure presents the three different modeling approaches assessed for Wind Farm 1, and their associated deviations, both gross and net of technical and operational (“T&O”) losses. Also shown is the deviation against the original pre-construction assessment by another consultant. The other consultant also used a CFD model but yielded a significantly larger deviation than did Enertis (attributable to a higher long-term wind speed estimation by the other consultant).  

Figure 1. Pre-construction WR&EYA versus OEPE net production P50 with T&O losses – Wind Farm 1

For Wind Farm 1, the total (avian-loss-corrected) net deviation estimate of 2.2% aligns with wind industry standards for intercomparison, as it falls within the deviation range highlighted by industry benchmarking studies (ref. 1).  

For Wind Farm 2, the total net deviation of -0.6% aligns with wind industry standards for intercomparison, as it falls within the deviation range highlighted by industry benchmarking studies (ref. 1).  

For Wind Farm 3, we have found that the main causes of deviation were due to an overestimation in long-term wind conditions (during the pre-construction assessments by other consultants), as well as wind flow and wake modelling processes; however, Enertis has not been able to perform an independent pre-construction WR&EYA or explore this further due to the absence of on-site wind data for Wind Farm 3.   

Overall, the case studies outlined here-in highlight the critical importance of wind flow modeling choices, input data completeness at the pre-construction assessment stage, and post-construction operational feedback, for reducing deviations and uncertainty in energy yield assessments. The results emphasize the need for enhanced project development efforts to support accurate pre-construction operating assumptions, which will meaningfully improve forecast accuracy. Continued benchmarking across diverse sites will be essential to refining industry practices and ensuring more reliable pre-construction energy yield assessments for wind power assets. As an industry, it will be beneficial to continue working to reduce the deviations found between pre-construction WR&EYA compared to actual wind farm energy production, in order to reduce risk and improve financial model revenue build estimations via more accurate pre-construction energy assessments.   

For more information, reach out to our Wind Advisory Services Director: nicholas.capaldo@enertisapplus.com