Search

Wind power forecasts have been operationally used for over 25 years. Despite this fact, there are still many possibilities to improve and enhance forecasts, both from the weather prediction side and in the use of the forecasts. Until now, most applications have focused on deterministic forecast methods. This is likely to change in the future as penetration levels increase and weather conditions become more unstable due to climate change. Probabilistic methods are therefore receiving more attention from users. The International Energy Agency (IEA) Wind Task 36 on Wind Power Forecasting organises international collaboration, among national weather centres with an interest and/or large projects on wind forecast improvements (NOAA, DWD, UK MetOffice,...), forecast vendors and forecast users to facilitate scientific exchange to be prepared for future challenges. Collaboration is open to IEA Wind member states; 12 countries are already actively collaborating. The Task is divided in three work packages: Work Package (WP) 1 is a collaboration on the improvement of the scientific basis for the wind predictions themselves. This includes numerical weather prediction (NWP) model physics, but also widely distributed information on accessible datasets. This WP will also organise benchmarks for NWP models. The efforts of WP2 resulted in the publication of an international pre-standard (an IEA Recommended Practice) on how to select an optimal wind power forecast solution for a specific application. The focus of WP3 is on the engagement of end users to disseminate the best practice in the use of wind power predictions, especially probabilistic forecasts. The paper presents an overview of the recently completed first phase and the ongoing second phase of IEA Task 36 on Wind Power Forecasting, which provides a forum for international collaboration in this important field for meteorologists, wind power forecasters and end users. For collaboration, please contact the author (grgi@dtu.dk) and see the website at www.ieawindforecasting.dk. [ABSTRACT FROM AUTHOR]

Wind power forecasts have been used operatively for over 20 years. Despite this fact, there are still several possibilities to improve the forecasts, both from the weather prediction side and from the usage of the forecasts. The International Energy Agency (IEA) Task on Wind Power Forecasting organises international collaboration, among national weather centres, forecast vendors and forecast users. The Task looks back on the first 3 years, and just started the second three-year period. Collaboration is open to IEA Wind member states, 12 countries are already therein. The Task is divided in three work packages: Firstly, a collaboration on the improvement of the scientific basis for the wind predictions themselves. This includes numerical weather prediction model physics, but also widely distributed information on accessible datasets and a benchmark. Secondly, we try to improve the derived power forecasts and deal with forecast vendor related matters to bring the entire industry forward. Thirdly, we will be engaging end users aiming at dissemination of the best practice in the usage of wind power predictions. The main result of the first phase is the IEA Recommended Practice for Selecting Renewable Power Forecasting Solutions. This document in three parts (forecast solution selection process, designing and executing forecasting benchmarks and trials and evaluation of forecasts and forecast solutions) takes its outset from the recurrent problem of forecast users of how to choose a forecast solution. The first report describes how to tackle the general situation, while the second report specifically describes how to set up a forecasting trial so that the result is what the client intended. Many of the pitfalls we have seen over the years, are avoided. Other results of the first phase include a comprehensive review paper on the use of uncertainty forecasts in the power industry and an information portal related to forecasting. In the second phase of the Task, we will take up additional topics such as the uncertainty propagation through the modelling chain, the use of distributed measurements for forecasting, and some initial standardisation of data flows and formats. [ABSTRACT FROM AUTHOR]