Power Forecasting and Dynamic Line Rating

Transcription

1 Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen October 2016 PRESENTED BY: DR. ARNE GRAVDAHL Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

2 Content Background A brief history from Norway Why power forecasting? Bridging scales, Strategies Accuracy Validations Way Forward Power Forecasting Technology available on the Portal Using the Portal Demo Dynamic Line Rating Video Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

3 Norway: Geography and meteorology Extension north-south is approximately km Coastline km Fjords, valleys and mountains Long traditions within meteorology (Fishing industry) From 58 to 71 deg. north Fjords, valleys and mountains Just after World War I, a group of meteorologists in Norway led by Vilhelm Bjerknes developed the Norwegian cyclone model that explains the generation, intensification and ultimate decay (the life cycle) of mid-latitude cyclones, and introduced the idea of fronts, that is, sharply defined boundaries between air masses. The group included Carl-Gustaf Rossby (who was the first to explain the large scale atmospheric flow in terms of fluid dynamics), Tor Bergeron (who first determined how rain forms) and Jacob Bjerknes. en.wikipedia.org/wiki/meteorology Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

4 Norway: Energy situation Electricity Norway has the world's largest per capita hydropower production, and is the sixth largest hydropower producer in the world In a year with normal precipitation, hydropower generation is around 121 TWh, corresponding to approximately 99 percent of Norway's total power production Normal production capability and consumption in Norway [TWh] Electricity is cheap light is always on in Norway Difficult to develop wind due to low electricity prices strong focus on profitability strong focus on the wind resources and forecasting Source: NOU 2012: 9 Energiutredningen verdiskaping, forsyningssikkerhet og miljø Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

5 Content Background A brief history from Norway Why power forecasting? Bridging scales, Strategies Accuracy Validations Way Forward Power Forecasting Technology available on the Portal Using the Portal Demo Dynamic Line Rating Video Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

6 Power forecasting in a renewable energy mix Merit order effect How do you get your electricity? Most likely it is from a mixture of coal, gas, oil, nuclear and hydro, standing from more than 90% of the worlds electricity production. Times are changing - wind has become part of the mix, and the trend is clear, its contribution will become even more significant in the coming years. What iare the market consequences? The introduction of wind as an additional source in the energy mixture has a profound market impact, explained by the so called merit order effect. The merit order effect describes the mechanism by which the market price of electricity is set. Wind energy has no fuel costs the wind is free and the operational costs are low too. Hence, wind energy is first introduced in the energy mix followed by the more expensive sources until the power demand is met, and thereby the price of electricity is set. Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

7 Bridging scales Global Models ECMWF, GFS ( km) Downscale order 10:1 Meso/Regional Models WRF (9-1 km) Downscale order 100:1 Micro Model WindSim ( m) Accurate description of the local flow field and the wake effects, Model size 2x2 km Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

8 Power Forecasting Strategies Depending on available data and requested accuracy level, we will configure a power forecasting service in accordance with your needs: Day-ahead or intraday Set-up Period Mesoscale Hindcast Historical Wind Measurements ANN Training CFD Look-up Tables Forecasting Mode Global Forecast Mesoscale Forecast ANN Climate Correction CFD Downscaling Online Data Measurements Configurable with respect to weather input data Flexible time resolution Historical Wind Measurements Historical Power Production Historical Power Production Historical CFD Forecasted Power Empirical Power Curve ANN Training Wake Modeling Power Production Forecast ANN Power Correction Power Production Forecast Online Data Production WindSim software/data Integration of online data Multiple strategies, combinations of statistical and physical models, Hybrid solutions Input data The Power Forecasting is divided in two, set-up and forecasting mode, for accuracy and computational speed reasons Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

9 Power Forecasting Strategies The power forecasting system couples Numerical Weather Prediction (NWP) data, Artificial Neural Network (ANN) and Computational Fluid Dynamics (CFD) The power forecasting can be setup in various ways, using so-called strategies, depending on available data and required accuracy level «Wind to Power» «ANN Wind to Power» «ANN Wind to Wind + CFD» Power Forecasting strategies using various combinations of Artificial Neural Networks and Computational Fluid Dynamics Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

10 Content Background A brief history from Norway Why power forecasting? Bridging scales, Strategies Accuracy Validations Way Forward Power Forecasting Technology available on the Portal Using the Portal Demo Dynamic Line Rating Video Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

11 Average MAE Benchmark 1000 MW WindSim takes part in a power forecasting benchmark over 6 months Benchmark includes 10 wind farms in Europe with approximately1000 MW capacity Benchmark includes 16 participants Our meso scale provider for all wind farms is UK Met Office Below is the intermediate results after 3 months: Evaluation start: Evaluation end: Num. forecasts: MAE averaged over all forecasted sites and horizons Benchmark Ranking Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

12 Validation ANN and CFD have different strengths ANN good for the mean values CFD good for the extreme values A hybrid solution combining the two strategies will work best Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

13 Content Background A brief history from Norway Why power forecasting? Bridging scales, Strategies Accuracy Validations Way Forward Power Forecasting Technology available on the Portal Using the Portal Demo Dynamic Line Rating Video Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

14 Development of WindSim Power Forecasting Power Forecasting has been developed through R&D and Joint Industry Projects (JIP) It started with extensions in WindSim, developed into a separate software, further into services and now we plan to gather both software and services in a web portal 2008 JIP; China Electric Power Research Institute; Power time series JIPS; KIER, TUBITAK, ZAMG 2011 JIP; StormGeo, Leviathan, WindSim; Chinese market 2012 Launch of WindSim Power Forecasting 2012 JIP; Aquiloz; Common software/service offer 2013 R&D, NFR Energix; Artificial Neural Networks, Meso-micro coupling 2014 JIP; Shenyang University of Technology; common software/service offer 2014 Launch of service with UK Met Office 2015 R&D, NFR Energix; Nowcasting 2015 JIP; Pattern Energy; Launch of WindSim Portal 2015 Partner outlets; Breeze, Baze technology, KOS Pre-qualified for major trial (power forecasting of 1500 MW wind energy) 2016 R&D,NFR Energix; Nowcasting (Met.no) Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

15 Flexible Power Forecasting Deliverables Depending on the operational setup the power forecasting output is delivered in various ways: As a service, operated by WindSim, with power forecasting results delivered by or ftp Through the WindSim Portal, configurable by the customer Through a Wind Farm Management System, operated by third parties WindSim Portal configurable by the customer Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

16 WindSim Portal Demo user WindSim Portal: Request your own user on the Portal and test out the various forecasting strategies Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

17 Content Background A brief history from Norway Why power forecasting? Bridging scales, Strategies Accuracy Validations Way Forward Power Forecasting Technology available on the Portal Using the Portal Demo Dynamic Line Rating Video Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

18 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

19 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

20 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

21 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

22 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

23 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

24 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

25 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

26 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

27 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

28 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

29 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

30 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

31 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

32 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

33 Web Portal Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

34 Content Background A brief history from Norway Why power forecasting? Bridging scales, Strategies Accuracy Validations Way Forward Power Forecasting Technology available on the Portal Using the Portal Demo Dynamic Line Rating Video Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

35 Dynamic Line Rating (DLR) Increase amount of current by 10-40% An increase in wind speed blowing at a right angle to a high-voltage line can cool the line enough to safely increase the amount of current it can carry by 10 to 40 percent Two operational modes: Nowcasting (Weather stations) Forecasting (Numerical weather prediction model) Winner EWEA Poster Award 2015 Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

36 Idaho National Laboratory Cooperation since INL buys a WindSim license, training and guidance towards their special usage within DLR 2014 First JIP were power line object with wind speed and direction in the mid section of each power line segment was established. The Power line module was first launched in WindSim Second JIP were we have developed a «Wind Atlas» method for wide area usage of the «Power line» module. INL has merged 4 models with 50 million cells each 2016 Third JIP is ongoing, were we will create wind statistics for power line routing optimization Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October

37 Thank you WindSim AS Fjordgaten Tønsberg, Norway Tel: WindSim Americas 2945 Townsgate Road Westlake Village California 91361, USA Tel: WindSim Brasil Market Place II, Av. Doutor Chucri Zaidan, ºandar, Vila Gertrudes São Paulo SP , Brasil Tel: WindSim China No. 101 Shaoyang Beili Chaoyang District Beijing, China Tel: WindSim France 3 Rue du Fin F Saint Vaast de Longmont, France Tel: WindSim India Suite # 617 Regus Milenia Business Park Phase 2, Level - 6, Campus 4B, No , Dr.M.G.R Road Kandanchavady, Perungudi Chennai , India Tel: WindSim Sub-Saharan Africa 16th Floor Norton Rose House Riebeeck Street Cape Town, 8000, South Africa Tel: Power Forecasting and Dynamic Line Rating WindSim User Meeting, Xiamen, October