Back to the programme printer.gif Print



Thursday, 13 March 2014
09:00 - 10:30 The model chain: First steps towards tomorrow's technology
Resource Assessment  


Room: Tramuntana
Session description

Recent advances in the software and computational resources available to the wind industry have opened a new frontier; the ‘model chain’. A single approach to such a concept has yet to emerge as the industry standard, although a general idea of a dynamic process across progressively smaller scales is emerging. This session intends to give delegates a broad view of how research institutes and private companies are dealing with this challenge, what the most promising approaches are and which range of applications is foreseen for the coming years.

Learning objectives

  • Understand some of the challenges of multi-scale modelling
  • Get a first glimpse of current approaches in this topic
  • Talk directly to the main players in this research field
Lead Session Chair:
Pep Moreno, Vortex, Spain

Co-chair(s):
Hans E. Joergensen, DTU Wind Energy, Denmark
Claude Abiven Natural Power, France
Co-authors:
Claude Abiven (1) F P
(1) Natural Power, Strasbourg, France

Printer friendly version: printer.gif Print

Presenter's biography

Biographies are supplied directly by presenters at EWEA 2014 and are published here unedited

Claude has been working for Natural Power, a wind energy consultancy, for over six years. He is specialized in wind flow modelling for energy yield assessment. His research interests include CFD modelling of complex or forested terrain, as well as thermally driven flows.

Abstract

Are mesoscale-coupled CFD models the next generation models?

Introduction

Re-analyses combine observations and global flow models to provide wind flow variables on a global scale. Mesoscale models are used to downscale these variables to meso scales. We present a CFD model coupled to a mesoscale model which completes the model chain to provide flow variables at micro scales, which is required for energy yield assessment.
Such models are greatly anticipated by the industry, as they are theoretically able to provide flow variables without the need for onsite measurements, take into account micro scales, and model thermally-driven flows, features not fully addressed by models currently available in the industry.


Approach

The coupled model is first described.
Its performance is then compared to the one of linear, standard CFD, and mesoscale models. All four models were run on a selection of sites on which data from 2 masts or more were available. On each site data from one of the masts was assumed to be known, while data at the other masts was compared to model predictions.
We conclude by discussing advantages and drawbacks of the coupled model compared to existing tools.


Main body of abstract

The coupled CFD model solves the time-dependant Reynolds Averaged Navier-Stokes equations including the transport equation for potential temperature. It uses the k-epsilon model of turbulence. The model uses time-series obtained from the WRF mesoscale model in order to compute initial and boundary conditions at each time-step.
The model typically requires the use of 1 second time-steps, which implies extensive computational time. A methodology was therefore developed to select a limited number of days representative of one full year, in order to minimize computational cost.
The model was then run on a selection of sites* located worldwide and on which two masts or more were available. Values of wind speed, turbulence and wind shear predicted by the model were then compared to measurements and to predictions obtained from linear, standard CFD and mesoscale models.
The coupled model was found to be able to provide absolute values of wind speed and turbulence with a level of accuracy comparable to the ones of CFD and linear models which both require on-site measurements. When fed with measurements, the coupled model was found to provide the most accurate values of wind speed out of all models (3.9% absolute error on average, against 5.4% for linear model).
The coupled model is able to provide mast-like time series. In particular it is shown to reproduce observed diurnal variations of wind shear and turbulence, as well as variations of turbulence with wind speed to a great level of accuracy.
*Ten at the time this abstract is being written.

Conclusion

Based on the selection of sites used for the computations presented here it appears that coupled CFD models could help reduce uncertainties on wind power production and enable precise turbine suitability assessment on sites associated with complex, thermally-driven flows, and on sites with no or lack of on-site measurements.
The main drawback of such models appears to be their expensive computational cost, which prevents extensive testing and can limit model resolution at the current level of computer performance.


Learning objectives
State of the art of modelling for wind resource assessment.
CFD and mesoscale coupling.
Modelling of atmospheric stability.
Modelling error and uncertainty of models commonly used in the industry.
Modelling without or with limited on-site measurements.