On The Benefit Of A Multivariate Description Of Wind For A Better Long .
On the benefit of a multivariate description of wind for a better long-term extrapolation Abel Tortosa-Andreu¹, Pau Casso¹, Patrícia Puig¹ Javier Viscarret², Inma Murillo³, Ricardo Martínez⁴ Nuria Sal de Rellán⁵, Luís Prieto⁶ and Günter Vahlkampf⁷ ! (1) Vortex, S.L., Barcelona, Spain (2) Acciona Energía, S.A., Sarriguren, Spain (3) Alstom Renovables España, S.L., Barcelona, Spain (4) Barlovento Recursos Naturales, S.L., Logroño, Spain (5) E.ON Renovables S.L.U., Madrid, Spain (6) Iberdrola Renovables Energía, S.A., Madrid, Spain (7) Sowitec development GmbH, Sonnenbühl, Germany !
Outline MCPs/LTCs MCPs shortcomings: illustrative problems of the industry Remodeling: description Remodeling success: illustrative cases Validation results Users feedback Summary
MCPs: Long-term correction method based on a (linear) regression analysis of a wind reference SERIES with on-site measurements. MEASURE on-site measures CORRELATE long-term reference data (NWP, stations) PREDICT long-term extrapolation
Industry Problem 1: Wind-Rose MEASURES SERIES A Wind-Rose - Site Data (5 years) MCP does not properly correct the Wind-Rose INDUSTRY - MCP B B Wind-Rose - Vortex Series (same period) Wind-Rose - MCP (same period) Direction R² RMSE MCP 0,58 20,9 SERIES 0,58 22,0
Industry Problem 2: IN & OUT-of-training sensitivity INDUSTRY - MCP IN-training OUT-of-training R 2 IN OUT MCP 0,96 0,56 SERIES 0,71 0,71 on-site measures long-term reference data long-term extrapolation MCP performance degrades considerably from the IN to the OUT-of-training sample.
Industry Problem 3: Metric degradation MEASURES INDUSTRY - MCP MEASURES R² SERIES MCP 0,54 0,34 U 7,69 6,26 (18,6%) 7,75 (0,78%) A 8,68 6,93 (20,2%) 8,73 (0,58%) k 2,26 2,59 (14,6%) 2,18 (3,54%) MCP produce a exceptional good fit for some metrics while significantly degrades others.
Remodeling: yet another MCP/LTC? YES The principe is the same: fit campaign measurements with reference data to extrapolate the long-term wind resource. It employs a variety of modeled variables at different levels/heights (multivariate approach), not just wind reference series It is not linear NO The long-term correction keeps the representativeness of the series in all its attributes (correlation, frequency distribution, Rose,.) Reduces out-of-training degradation . Thus, it does not preclude the use of traditional MCPs as second layer (if needed)
Multivariate approach A multivariate analysis allows to identify (and fit) regime dependent winds (i.e. associated to a particular climate pattern or weather situation). ! ! U component R 0.53 V component R 0.65 Temperature R 0.42 Wspeed R 0.78 ! ! Model variables@80m vs. Measures . Scatter plot loess smoothers pair wise correlation Wspeed Measures The highly dynamic and non-linear fashion in which the atmospheric variables are linked suggests the need for a more sophisticated statistical tools for MCPs/ LTCs than current linear approaches based solely on wind.
Remodeling: Data reduction Reduced variables (Components) NWP output Linear transformation
Remodeling: non-linear ensemble Linear models can account for non-linear trends/relationships if the non-linear nature of the data is known.The non-linearity of wind is unknown a priori and it is site and weather regime dependent. The alternative: inherently non-linear models. Remodeling follows an ENSEMBLE approach of non-linear models: Palliate inherent drawbacks of single nonlinear models (local solutions, over-fit, instabilities, etc.) ! Stable solutions ! Characterize uncertainty BoxPlot: non-linear models Outliers: in red Color scale: P90-P10 band width Measures: black solid line ! !
Remodeling Success 1: Wind-Rose MEASURES REMODELING A Wind-Rose - Site Data (5 years) INDUSTRY - MCP SERIES A Wind-Rose - Remodeling (same period) Remodeling corrects the Wind-Rose: increases the correlation and decreases de RMSE/Bias B B Wind-Rose - Vortex (same period) Wind-Rose - MCP (same period) Direction R² RMSE REMODELING 0,64 19,5 MCP 0,58 20,9 SERIES 0,58 22,0
Remodeling Success 2: IN & OUT-of-training INDUSTRY - MCP IN-training REMODELING OUT-of-training Remodeling consistently extrapolates the lessons learned in the period coincident with measures, to the rest of the SERIES. The sensitivity/dependence of results to the training period is reduced. OUT-of-training R IN OUT REMODELING 0,81 0,77 MCP 0,96 0,56 SERIES 0,71 0,71
Remodeling Success 3: Metric degradation REMODELING INDUSTRY - MCP MEASURES R² REMODELING MCP SERIES 0,61 0,34 0,54 U 7,69 7,68 (0,13%) 7,75 (0,78%) 6,26 (18,6%) A 8,68 8,67 (0,11%) 8,73 (0,58%) 6,93 (20,2%) k 2,26 2,29 (1,33%) 2,18 (3,54%) 2,59 (14,6%) Remodeling does not degrades the attributes of the SERIES in order to improve a particular metric, it improves overall all the attributes.
Validation at 100 sites worldwide
Validation at 100 sites worldwide IMPROVEMENTS R² (hourly) R² (daily) R² (monthly) R² dir (hourly) RMSE BIAS PROD k A Events 100% 100% 100% 95% 100% 100% 97% 82% 97% Average Val. 0,06 0,06 0,05 0,12 0,4m/s 10,6% 26,3% 7,4% 10,6% Std. Deviation 0,04 0,05 0,06 0,10 0,3m/s 7,20% 19,1% 6,8% 7,1%
Users feedback POSITIVE FEEDBACK “Remodeling provides good potential to improve long-term estimates”. POINTS TO BE IMPROVED ! “In complex terrain there is a significant * “Remodeling combines very good results improvement in some parameters while and low effort into an astonishing in others the original series reproduces effective standardized procedure”. better the data”. “Remodeled series significantly improves The results for the wind direction are the original series in simple terrain sites”. satisfactory overall, although there is still room for improvement. “Remodeling improves in a high ** percentage of cases the R² respect to the Although the Weibull fit is good, the Vortex Series and other Industry MCPs”. wind histogram for the remodeled series should be improved. “Remodeling provides good results for the Wind Bias/RMSE and the Weibull parameters”. (*) Check Poster Id. 088: "Benefits of multiscale modelling to reduce long term wind resource uncertainty: the Gouda case” (**) in the process of testing the re-modeling
Summary SERIES MCP-A-LINEAR MCP-B-LINEAR MCP-A-MATRIX MCP-B-MATRIX REMODELING Full colored petal identifies the BEST average performance of that metric. ! ! Statistical sample includes sites worldwide with different terrain complexities. ! ! OUT-of-training: 4 yrs 1N-training: 1 yr
Industry Problem 1: Wind-Rose Wind-Rose - Site Data (5 years) Wind-Rose - Vortex Series (same period) Wind-Rose - MCP (same period) A B B Direction R" RMSE MCP 0,58 20,9 SERIES 0,58 22,0. IN-training INDUSTRY - MCP OUT-of-training MCP performance degrades considerably from the IN to the OUT-of-training sample.
May 02, 2018 · D. Program Evaluation ͟The organization has provided a description of the framework for how each program will be evaluated. The framework should include all the elements below: ͟The evaluation methods are cost-effective for the organization ͟Quantitative and qualitative data is being collected (at Basics tier, data collection must have begun)
Silat is a combative art of self-defense and survival rooted from Matay archipelago. It was traced at thé early of Langkasuka Kingdom (2nd century CE) till thé reign of Melaka (Malaysia) Sultanate era (13th century). Silat has now evolved to become part of social culture and tradition with thé appearance of a fine physical and spiritual .
On an exceptional basis, Member States may request UNESCO to provide thé candidates with access to thé platform so they can complète thé form by themselves. Thèse requests must be addressed to esd rize unesco. or by 15 A ril 2021 UNESCO will provide thé nomineewith accessto thé platform via their émail address.
̶The leading indicator of employee engagement is based on the quality of the relationship between employee and supervisor Empower your managers! ̶Help them understand the impact on the organization ̶Share important changes, plan options, tasks, and deadlines ̶Provide key messages and talking points ̶Prepare them to answer employee questions
Dr. Sunita Bharatwal** Dr. Pawan Garga*** Abstract Customer satisfaction is derived from thè functionalities and values, a product or Service can provide. The current study aims to segregate thè dimensions of ordine Service quality and gather insights on its impact on web shopping. The trends of purchases have
Chính Văn.- Còn đức Thế tôn thì tuệ giác cực kỳ trong sạch 8: hiện hành bất nhị 9, đạt đến vô tướng 10, đứng vào chỗ đứng của các đức Thế tôn 11, thể hiện tính bình đẳng của các Ngài, đến chỗ không còn chướng ngại 12, giáo pháp không thể khuynh đảo, tâm thức không bị cản trở, cái được
Le genou de Lucy. Odile Jacob. 1999. Coppens Y. Pré-textes. L’homme préhistorique en morceaux. Eds Odile Jacob. 2011. Costentin J., Delaveau P. Café, thé, chocolat, les bons effets sur le cerveau et pour le corps. Editions Odile Jacob. 2010. Crawford M., Marsh D. The driving force : food in human evolution and the future.
Le genou de Lucy. Odile Jacob. 1999. Coppens Y. Pré-textes. L’homme préhistorique en morceaux. Eds Odile Jacob. 2011. Costentin J., Delaveau P. Café, thé, chocolat, les bons effets sur le cerveau et pour le corps. Editions Odile Jacob. 2010. 3 Crawford M., Marsh D. The driving force : food in human evolution and the future.
