Domain-Adapted Power Curve for Cross-Farm Applications

2026-07-22

AI Summary

This study addresses the challenge of wind power forecasting in site selection for new wind farms, where historical operational data are unavailable. To overcome this limitation, the authors propose a domain-adaptive approach for transferring power curves across wind farms. Departing from conventional assumptions based on geographical proximity or layout similarity, the method integrates temporal environmental variables with spatial topographic features to construct a semantic similarity metric between the source domain (operational wind farms) and the target domain (prospective sites). This enables effective transfer of power curve knowledge to data-scarce locations. Experimental results demonstrate that the proposed method significantly outperforms existing transfer strategies across multiple real-world site selection scenarios, substantially improving power prediction accuracy for sites lacking historical operational data.

Abstract

The wind energy industry relies on accurate power curve models to make power forecast, evaluate turbine performance, quantify upgrade, or support site-planning decisions. In this paper, we focus on site-planning power curves, i.e., we investigate how power curve models trained using turbine data on an operating wind farm can be transferred to a new, undeveloped farm. The traditional wisdom in the wind energy literature relies on distance, layout, or terrain characteristics for making cross-farm power curve transfer. Through the lens of domain adaptation, we propose a more reliable transfer learning approach for cross-farm power curve modeling. In the cross-farm applications, a domain is specified by the temporal environmental variates and spatial terrain variables. Domain adaptation is to find a capable similarity metric to adapt the domain on the new farm to that on the existing farm. Empirical results show that our domain adapted power curve consistently outperforms competing approaches by an appreciable margin for site-planning power predictions.

Problem

• power curve
• cross-farm transfer
• domain adaptation
• wind energy
• site planning

Innovation

• domain adaptation
• power curve
• transfer learning
• wind farm
• site planning