Wind and Power Assessment using High Performance Computing

MARÍA LAURA MAYOL; ALEJANDRO DANIEL OTERO; GONZALO PABLO NAVARRO DIAZ; ANDREA CELESTE SAULO

Buenos Aires

Conferencia; Argentina Wind Power; 2018

Global Wind Energy Council (GWEC)

Worldwide, wind energy occupies a prominent place among the renewable energies, as its resources are abundant, is clean and has lower production costs than other energies. In the last years the global installed capacity has been in an increasing trend. In this context Argentina is no exception, having nowadays expectations of several GW installed in the following years. To follow this growth, the exploitation of this type of energy requires a deep knowledge of the resource, which is a challenge due to its great spatial and temporal variability. That is why, in order to achieve the characterization of the wind in the region, the operation of future wind farms and the management of the electricity market with greater share of this source, all scales of the phenomenon must be studied. In this work different tools for the comprehensive analysis of the wind resource are presented, from wind maps and micrositing to increase the knowledge of the wind potential of a region, to the forecasting of wind and power to reduce uncertainty and allow better planning and integration of this renewable energy into power systems. The wind maps are produced using reanalysis data combined with a mesoscale modeling while micrositing (selection of the best wind turbines location in a specific place) is performed running computational fluid dynamics (CFD) simulations. To add the temporary variation of the resource, wind is forecasted with the Weather Research & Forecasting (WRF) model. This model is also used to forecast power, using wind farm parametrizations to include the wind farm effect on the flow and calculate the power output at each time. Detailed power output forecasts play a fundamental role in the task of balancing supply and demand in a national or utility scale electrical system, that is why different levels of description are required depending on the needs. For a better description of the wind turbine effect on the flow, CFD simulation are performed to capture the details of the wakes and obtain the wind farm efficiency for different inflow velocities and directions. The focus of this work is on the use of open source tools able to run in high-performance computing environments due to the high computing power demand. Examples of all these tools are given for the particular case of a large-scale wind farm located in Patagonia, and the results are contrasted with in situ measurements showing good correlation.