Mesoscale modelling of future climate impacts on wind power generation

Graziela Luzia*

*Corresponding author for this work

Research output: Book/ReportPh.D. thesis

26 Downloads (Pure)

Abstract

The need for accurate representations of future wind variability in multiple time scales is emphasized in light of the dependence of wind power generation on weather variability and the potential impact of climate change on weather regimes. This thesis presents a comprehensive study on the creation and validation of wind speed time series for planning efficient future power and energy systems. The current state of the available climate datasets is evaluated, and the limitations in their use for power and energy system applications are identified. A methodology for developing state-of-the-art regional climate simulations is presented. The thesis is presented as a compilation of three parts:

The first part of the thesis, and the related paper, proposes a set of metrics for validating European wind speed time series, specifically aimed at wind energy system integration studies. The paper also presents a sensitivity analysis of a numerical weather model, exploring several configurations and their effects on the resultant wind speed time series. The comparison between the different simulations demonstrate that finer spatial resolution positively impacts the temporal and the spatial correlations. On the other hand, reanalysis data provide time series better correlated with measurements.

The second part validates time series of European wind power generation produced by available regional climate simulations. The challenges of producing time series relevant for power and energy system applications are discussed, and a methodology for extracting, adjusting, and validating the time series is proposed. The validation results, presented in the second paper, show that the time series of wind power generation present accurate results for most of the defined metrics when adjusted by the long-term mean wind speed from a microscale wind atlas. The regional climate ensemble validation, however, is limited to a few models and emission scenarios, and the available temporal frequency is less than optimal.

The final part shows the method developed to create new regional climate simulations tailored for power and energy system applications. The new models aim to address the limitations of the available climate datasets, providing higher spatial and temporal resolution data and a larger ensemble for simulating wind power generation in future scenarios. The validation of three tested models in different configurations showed that the climate simulations can represent the wind temporal dependencies and the spatial correlations approximately as good as widely used reanalyses. The tests with different model configurations demonstrated that the mesoscale model, and not its forcing data, is the one which drives the climate simulations in relation to the analysed metrics. Other relevant contributions and improvements to the models can be added by running more tests.

This thesis provides a significant contribution to the field of power and energy systems, advancing our understanding of the desired qualities of wind speed time series and allowing studies on the potential impacts of climate change on the variability of future wind power generation. The proposed set of metrics provides an objective methodology for validating simulated time series, and the new regional climate simulations, with potential to be extended for other variable renewable energy sources than wind, can help on planning and developing cleaner future power and energy systems.
Original languageEnglish
Place of PublicationRisø, Roskilde, Denmark
PublisherDTU Wind and Energy Systems
Number of pages107
Publication statusPublished - 2023

Fingerprint

Dive into the research topics of 'Mesoscale modelling of future climate impacts on wind power generation'. Together they form a unique fingerprint.

Cite this