SafeWind

Due to the variable nature of the wind resource, the large-scale integration of wind power causes several difficulties in the operation and management of a power system. Short term forecasts of wind generation from a few hours to a few days ahead are necessary for carrying out various management tasks related to the integration of wind generation in power systems (i.e. quantification of reserves, optimal scheduling, congestion management, or the design of optimal trading strategies).

In the past 15 years, considerable research has been carried out in the field of wind power forecasting leading to several operational tools. However, the current wind forecasting technology presents several bottlenecks. Depending on the sensitivity of end-users, the quality of the forecasts is not adequate for about 10-20% of times. In situations such as fronts or wind speeds near cut-off speed the forecast error may reach high levels. Large forecast errors significantly reduce the value and acceptability of wind power due to their impact on the grid, the penalties they involve when trading and other consequences.

The aim of SafeWind is to substantially improve wind power predictability in these challenging or extreme situations. Going beyond this, wind predictability is considered as a system parameter linked to the resource assessment phase, where the aim is to take optimal decisions for the installation of a new wind farm.

Breakthrough improvement of wind predictability is far from trivial. It necessitates research back to the basics, bringing together excellence from different disciplines like meteorology, statistics, applied mathematics, physics, power engineering, wind power forecasting, information & communication technologies. SafeWind aims at developing this synergy through a work program inspired by the needs of the industry,for solutions of an operational nature to the problem of extremes. For this reason, the project associates industrials such as Transmission Systems Operators (TSOs), utilities and wind farm developers.

The project aims to improve predictability and especially to treat the issue of extremes at various temporal scales:

• very short -term horizons (i.e. 5 minutes to next hour) that become of high importance especially in systems with high penetration.
• short-term horizons (i.e. hourly predictions for the next 2-3 days). This is the main focus of the actual technology. SafeWind extends this by considering predictability of intra-hourly variations and extremes which existing models are unable to handle.
• and also to longer-term horizons (i.e. up to 10 days or more).

The project aims also to improve wind predictability and the impact of extremes at different spatial scales:

• local scale: Extreme gusts or shears affect wind turbines or wind farms with loss of energy or structural damage.
• regional scale: Extreme events (like extreme winds due to thunderstorms, etc) can cause the loss of significant amounts of wind energy with potential impact on the grid management.
• European scale: Extreme weather situations can propagate causing impacts in different countries, e.g. if a storm takes a wrongly predicted track leading to large prediction errors; if these situations with potential risk at EU level are not well managed a big scale failure or blackout can happen.

Given the deployment of wind energy in Europe, the project aims at developing research towards a European vision for wind power forecasting. This is expected to be necessary in the coming years in the frame of an enhanced coordination among Transmission System Operators (TSOs) to manage large wind capacities. The project aims at bringing together existing but not yet coherently collected meteorological data as well as wind power and nacelle anemometer measurements distributed over Europe to detect extreme events in an early stage, to understand the spatio-temporal characteristics of such weather situations and related wind generation, and finally use them for better prediction of wind power.

Although current forecasting technology mainly encompasses deterministic models for the power output the project develops the concept of complementary tools that can be used jointly to traditional forecasts to assess wind predictability in the above temporal and spatial scales. The project will develop:

• new forecasting methods for wind generation focusing on uncertainty and challenging situations/extremes;
• models for "alarming": providing information for the level of predictability in the (very) short-term. They use near-real time online observations for alerts on potential extreme prediction errors and for immediate updates of short-term (0-6h) wind power predictions on regional and local scale;
• models for "warning": providing information for the level of predictability in the medium-term (next day(s)). Such tools, based on ensemble weather forecasts and weather pattern identification, can be used to moderate risks in decision making procedures related to market participation, reserves estimation etc.

A validation phase, in synergy with the ANEMOS.plus project, is foreseen to evaluate operationally the most promising models developed here.

At the early stages of wind energy, the focus was in resource assessment where the aim is to take optimal decisions where to install new wind farms. As penetration increases predictability and forecasting tools become of paramount importance. Nowadays, with the development of very large scale wind farms (i.e. offshore) and also with the development of electricity markets, wind predictability may become an issue at the level of site selection and design of a new wind farm. The project develops research on the new concept of "resource assessment versus predictability". The predictability of a site and especially the issue of extreme events can be considered when taking decisions for the installation of a new wind farm.

Another synergy between the resource assessment and the forecasting communities is to use historical runs of meteorological models for long periods, as a basis to calculate the probability of extreme events in a given area. New methods to estimate Vref using historical NWP will be developed, helping to improve the assessment of the IEC wind turbine class for new wind farm projects. These improved methods will contribute to reduce the risk of wind turbine damage under extreme conditions, due to wrong estimations of Vref based on very limited periods of measured data.

Finally, the project aims at analysing how new measurement technologies like Lidars can be beneficial for better evaluation of external conditions, resource assessment and forecasting purposes. Using new measuring devices for a more detailed knowledge of the wind speed and the wind energy available at the local level, permits to produce more accurate wind (and wind power) statistics at local geographical scale by removing some of the uncertainty into the measurement, or alternatively resulting into better tuning of the prediction models especially for the very-short-term.