Validation of Danish wind time series from a new global renewable energy atlas for energy system analysis

TL;DR

This paper introduces a new high-resolution global renewable energy atlas and demonstrates its application in generating and validating 32-year wind power time series for Denmark, highlighting variability and model differences.

Contribution

The paper presents a novel renewable energy atlas capable of producing calibrated, long-term wind power time series for energy system analysis, including validation and comparison with existing models.

Findings

Wind energy production varies by ±10% annually.

Changes in turbine size affect production patterns.

Model differences lead to up to 15% in capacity and 40% in reserve requirements.

Abstract

We present a new high-resolution global renewable energy atlas ({REatlas}) that can be used to calculate customised hourly time series of wind and solar PV power generation. In this paper, the atlas is applied to produce 32-year-long hourly model wind power time series for Denmark for each historical and future year between 1980 and 2035. These are calibrated and validated against real production data from the period 2000 to 2010. The high number of years allows us to discuss how the characteristics of Danish wind power generation varies between individual weather years. As an example, the annual energy production is found to vary by $\pm10\%$ from the average. Furthermore, we show how the production pattern change as small onshore turbines are gradually replaced by large onshore and offshore turbines. Finally, we compare our wind power time series for 2020 to corresponding data from a handful of Danish energy system models. The aim is to illustrate how current differences in model wind may result in significant differences in technical and economical model predictions. These include up to $15\%$ differences in installed capacity and $40\%$ differences in system reserve requirements.