Applications and benefits of weather, climate and water information to the power sector Laurent Dubus EDF R&D Applied Meteorology & Atmospheric Environment Group WORLD METEOROLOGICAL ORGANIZATION “MEETING OF THE WMO FORUM: SOCIAL AND ECONOMIC APPLICATIONS AND BENEFITS OF WEATHER, CLIMATE AND WATER SERVICES” WMO HEADQUARTERS, GENEVA, SWITZERLAND 8-11 APRIL 2013

WMO Forum, SEB, 8-11 April Overview 1.Preliminary remarks on power systems 2.Important meteorological parameters and data/forecasts used 3.What about economic evaluation of the benefits from weather/water/climate data ?

WMO Forum, SEB, 8-11 April Overview 1.Preliminary remarks on power systems 2.Important meteorological parameters and data/forecasts used 3.What about economic evaluation of the benefits from weather/water/climate data ?

Power systems are more and more complex WMO Forum, SEB, 8-11 April EDF power production, / TWh from renewables

WMO Forum, SEB, 8-11 April Growing importance of Renewables (2010: 12%)

Power Offer/Demand balance: a complex problem WMO Forum, SEB, 8-11 April Production units' program: 58 nuclear reactors 435 hydro power units ~50 thermal (coal, gas, fuel) ~900 Wind farms ~250,000 solar (including households) Huge optimization problem: variables & constraints for day+2 30 minutes forecasts Highly non convex and non linear, discrete and continuous variables Highly demanding on optimality (1% difference  several millions euros/year) and feasibility (all technical constraints must be satisfied) Problems: Production=Demand at each time step Many constraints Financial optimization of production costs 6

WMO Forum, SEB, 8-11 April Overview 1.Preliminary remarks on power systems 2.Important meteorological parameters and data/forecasts used 3.What about economic evaluation of the benefits from weather/water/climate data ?

Power demand depends on temperature WMO Forum, SEB, 8-11 April In France, power demand is highly dependent on temperature.  in winter : -1°C dT  MW of extra production ~ 20 M€ hedging  in summer : +1°C dT  MW of extra production Data from RTE & Météo-France

Increasing renewable energy production increases the dependance on weather variability and climate change WMO Forum, SEB, 8-11 April Daily variations of french hydro power production capacity in 2003 Hydro Wind Solar (PV) Renewables: highly fluctuating resources (especially wind and solar energy)

State of the art NWP models are used WMO Forum, SEB, 8-11 April Example: in-house products built from ECMWF & Météo-France VarEPS/monthly forecasts (temperature) Temperature + Cloud Cover  Demand forecasts  production units scheduling  physical margins calculations  hedging for residual financial risk (mandatory)

Monthly forecasts of river discharge using ECMWF products + in house post-processing methods WMO Forum, SEB, 8-11 April ECMWF Monthly fcst (Z700 & Z1000) Analog Method Hydrological Model Local T2m & Precip over 43 basins Streamflow Prob fcsts Streamflow climatology Hydro Model forced by T2m & Precip climatology ( ) Hydro Model forced by Analog T2m & Precip Observation  New model is now operational, an extension to seasonal forecasts is uner consideration DUBUS, L. In Press. Weather & climate and the power sector: Needs, recent developments and challenges. In: TROCCOLI, A., AUDINET, P.,DUBUS, L. & HAUPT, S. (eds.) Weather matters for energy. Springer

Local forecasts for renewables: a big challenge WMO Forum, SEB, 8-11 April

WMO Forum, SEB, 8-11 April

Many challenges remain WMO Forum, SEB, 8-11 April  Importance of meteorological data for project definition/dimensionning solar rad)  Moderate forecast errors can lead to high production errors  Some important parameters/events still not well known/monitored/forecasted:  solar radiation (global/direct/diffuse)  rapid fluctuations (wind/radiation)  snow, icing events, low level clouds …

WMO Forum, SEB, 8-11 April Overview 1.Preliminary remarks on power systems 2.Important meteorological parameters and data/forecasts used 3.What about economic evaluation of the benefits from weather/water/climate data ?

Economic evaluation of weather/water/climate info benefits WMO Forum, SEB, 8-11 April  Few publications on this subject, most of which come from research institutes, universities, international agencies (NREL, IEA, OECD…)  Not from energy companies, because of the sensitivity of the economic information (competitive markets)  Evaluations are country/technology specific :  depend on national policies (in particular for renewables: incentives vs. pure market rules) BAMS, dec. 2005

EDF’s financial communication (feb. 2011) 17 WMO Forum, SEB, 8-11 April 2013

Importance of observed data for investments WMO Forum, SEB, 8-11 April  An accurate estimation of the resource is essential Example: for an offshore wind turbine in France, an overestimation of 0.5m/s of the average wind speed (6.5 m/s instead of 6.0 m/s) can lead to a difference in production cost of 18% (V. Maupu, EDF R&D, personal communication)  Long time series are needed to assess the wind/solar resource for a given site

Forecasts for wind energy : the NCAR/Xcel Energy experience WMO Forum, SEB, 8-11 April Mahoney, W.P.; Parks, K.; Wiener, G.; Yubao Liu; Myers, W.L.; Juanzhen Sun; Delle Monache, L.; Hopson, T.; Johnson, D.; Haupt, S.E., "A Wind Power Forecasting System to Optimize Grid Integration," Sustainable Energy, IEEE Transactions on, vol.3, no.4, pp.670,682, Oct doi: /TSTE

WMO Forum, SEB, 8-11 April Cost analysis before, during and after the deployment of the new system:  Total decrease of 20% in MAPE between 2008 and 2010  US $6M savings between 2009 and 2010 for wind power forecasts  + US $2M savings due to more efficient commitment and dispatching of fossil fuel ressources From Mahoney et al., 2012 Forecasts for wind energy : the NCAR/Xcel Energy experience From Parks et al., 2011

Wind energy: the ANEMOS.plus & SafeWind Projects WMO Forum, SEB, 8-11 April  Includes 38 references about value of wind forecasting Because in general TSO provide data publicly, free of charge

Results must be taken with care (many parameters to take into account) WMO Forum, SEB, 8-11 April Increased penetration of renewable energies (wind/solar) can a priori decrease energy prices, but have secondary effects like increase in maintenance costs for other energy sources (nuclear, coal fuel, gas) that are used in less optimal conditions  Results should be taken with care, and global assessments are needed

Conclusions (1/2) WMO Forum, SEB, 8-11 April  In the last 10 years, much progress was achieved in NMHSs and research centers (VarEPS & monthly ECMWF for instance)  Only a few energy companies actively collaborate with NMHSs & private weather companies to develop new tailored products (ex.: Xcel Energy & NCAR/NREL, EDF & Météo-France)  Dialog between Providers & Users is essential to translate improvements in science into business improvements  Communication towards & training of end-users is very important  Upstream collaboration and partnerships should be encouraged

Conclusions (2/2) WMO Forum, SEB, 8-11 April  Progress are needed in forecasts at all time scales, but also in observations.  Showing an improvement w.r.t current practise is often enough  An economic evaluation of the benefits of WCW services in the energy sector is :  technically complex  country/technology dependant  generally not publicly available because of strategic aspects  However, even if economic benefit assessment is difficult, the value of WCW services is essential to the power sector

WMO Forum, SEB, 8-11 April Thank you for your attention