Parameter calibration in wake effect simulation model with stochastic gradient descent and stratified sampling. (English) Zbl 1498.62333

Summary: As the market share of wind energy has been rapidly growing, wake effect analysis is gaining substantial attention in the wind industry. Wake effects represent a wind shade cast by upstream turbines to the downwind direction, resulting in power deficits in downstream turbines. To quantify the aggregated influence of wake effects on the power generation of a wind farm, various simulation models have been developed, including Jensen’s wake model. These models include parameters that need to be calibrated from field data. Existing calibration methods are based on surrogate models that impute the data under the assumption that physical and/or computer trials are computationally expensive, typically at the design stage. This, however, is not the case where large volumes of data can be collected during the operational stage. Motivated by the wind energy application, we develop a new calibration approach for big data settings without the need for statistical emulators. Specifically, we cast the problem into a stochastic optimization framework and employ stochastic gradient descent to iteratively refine calibration parameters using randomly selected subsets of data. We then propose a stratified sampling scheme that enables choosing more samples from noisy and influential sampling regions and thus reducing the variance of the estimated gradient for improved convergence. Through both theoretical and numerical studies on wind farm data, we highlight the benefits of our variance-conscious calibration approach.


62P30 Applications of statistics in engineering and industry; control charts
Full Text: DOI


[1] AINSLIE, J. F. (1988). Calculating the flowfield in the wake of wind turbines. J. Wind Eng. Ind. Aerodyn. 27 213-224.
[2] ÁLVAREZ, M. A. and LAWRENCE, N. D. (2011). Computationally efficient convolved multiple output Gaussian processes. J. Mach. Learn. Res. 12 1459-1500. · Zbl 1280.68153
[3] BARTHELMIE, R. J. and PRYOR, S. (2013). An overview of data for wake model evaluation in the Virtual Wakes Laboratory. Appl. Energy 104 834-844.
[4] BARTHELMIE, R. J., PRYOR, S. C., FRANDSEN, S. T., HANSEN, K. S., SCHEPERS, J. G., RADOS, K., SCHLEZ, W., NEUBERT, A., JENSEN, L. E. et al. (2010). Quantifying the impact of wind turbine wakes on power output at offshore wind farms. Journal of Atmospheric and Oceanic Technology 27 1302-1317.
[5] BOLLAPRAGADA, R., BYRD, R. and NOCEDAL, J. (2018). Adaptive sampling strategies for stochastic optimization. SIAM J. Optim. 28 3312-3343. · Zbl 1461.65131 · doi:10.1137/17M1154679
[6] Breiman, L., Friedman, J. H., Olshen, R. A. and Stone, C. J. (1984). Classification and Regression Trees. Wadsworth Statistics/Probability Series. Wadsworth Advanced Books and Software, Belmont, CA. · Zbl 0541.62042
[7] CHEN, X., LEE, J. D., TONG, X. T. and ZHANG, Y. (2020). Statistical inference for model parameters in stochastic gradient descent. Ann. Statist. 48 251-273. · Zbl 1440.62287 · doi:10.1214/18-AOS1801
[8] Choe, Y., Byon, E. and Chen, N. (2015). Importance sampling for reliability evaluation with stochastic simulation models. Technometrics 57 351-361. · doi:10.1080/00401706.2014.1001523
[9] Choe, Y., Pan, Q. and Byon, E. (2016). Computationally efficient uncertainty minimization in wind turbine extreme load assessments. J. Sol. Energy Eng. 138 041012.
[10] CHURCHFIELD, M. (2013). Review of Wind Turbine Wake Models and Future Directions (Presentation). Technical report, National Renewable Energy Laboratory (NREL), Golden, CO.
[11] DAMIANOU, A. C., TITSIAS, M. K. and LAWRENCE, N. D. (2016). Variational inference for latent variables and uncertain inputs in Gaussian processes. J. Mach. Learn. Res. 17 Paper No. 42. · Zbl 1360.62373
[12] DTU WIND ENERGY (2015). Wind resources for energy production of wind turbines. Available at http://www.wasp.dk/wasp#details__wakeeffectmodel Accessed: 2015-09-18.
[13] FANG, Y., XU, J. and YANG, L. (2018). Online bootstrap confidence intervals for the stochastic gradient descent estimator. J. Mach. Learn. Res. 19 Paper No. 78. · Zbl 1476.62060
[14] FRANDSEN, S. (1992). On the wind speed reduction in the center of large clusters of wind turbines. J. Wind Eng. Ind. Aerodyn. 39 251-265.
[15] GÖÇMEN, T., VAN DER LAAN, P., RÉTHORÉ, P.-E., DIAZ, A. P., LARSEN, G. C. and OTT, S. (2016). Wind turbine wake models developed at the technical university of Denmark: A review. Renew. Sustain. Energy Rev. 60 752-769.
[16] GRAMACY, R. B., BINGHAM, D., HOLLOWAY, J. P., GROSSKOPF, M. J., KURANZ, C. C., RUTTER, E., TRANTHAM, M. and DRAKE, P. R. (2015). Calibrating a large computer experiment simulating radiative shock hydrodynamics. Ann. Appl. Stat. 9 1141-1168. · Zbl 1454.62530 · doi:10.1214/15-AOAS850
[17] HANKIN, R. (2019). calibrator: Bayesian calibration of complex computer codes. R Package, Version 1.2-8.
[18] Higdon, D., Kennedy, M., Cavendish, J. C., Cafeo, J. A. and Ryne, R. D. (2004). Combining field data and computer simulations for calibration and prediction. SIAM J. Sci. Comput. 26 448-466. · Zbl 1072.62018 · doi:10.1137/S1064827503426693
[19] INTERNATIONAL ENERGY AGENCY (2015). Medium-Term Renewable Energy Market Report 2015—Market Analysis and Forecasts to 2020. Technical report.
[20] JENSEN, N. O. (1983). A Note on Wind Generator Interaction.
[21] JOSEPH, V. R. and YAN, H. (2015). Engineering-driven statistical adjustment and calibration. Technometrics 57 257-267. · doi:10.1080/00401706.2014.902773
[22] KATIC, I., HØJSTRUP, J. and JENSEN, N. (1986). A simple model for cluster efficiency. In European Wind Energy Association Conference and Exhibition 407-410.
[23] Kennedy, M. C. and O’Hagan, A. (2001). Bayesian calibration of computer models. J. R. Stat. Soc. Ser. B. Stat. Methodol. 63 425-464. · Zbl 1007.62021 · doi:10.1111/1467-9868.00294
[24] KHEIRABADI, A. C. and NAGAMUNE, R. (2019). A quantitative review of wind farm control with the objective of wind farm power maximization. J. Wind Eng. Ind. Aerodyn. 192 45-73.
[25] LARSEN, G. C. (1988). A simple wake calculation procedure. Technical report, Risø National Laboratory, Denmark.
[26] Lee, G., Byon, E., Ntaimo, L. and Ding, Y. (2013). Bayesian spline method for assessing extreme loads on wind turbines. Ann. Appl. Stat. 7 2034-2061. · Zbl 1283.62241 · doi:10.1214/13-AOAS670
[27] LI, S., KO, Y. M. and BYON, E. (2021). Nonparametric importance sampling for wind turbine reliability analysis with stochastic computer models. Ann. Appl. Stat. 15 1850-1871. · Zbl 1498.62332 · doi:10.1214/21-aoas1490
[28] LIU, B., YUE, X., BYON, E. and KONTAR, R. A (2022). Supplement to “Parameter calibration in wake effect simulation model with stochastic gradient descent and stratified sampling.” https://doi.org/10.1214/21-AOAS1567SUPP
[29] LOHR, S. L. (2010). Sampling: Design and Analysis, 2nd ed. Brooks/Cole, Cengage Learning, Boston, MA. · Zbl 1273.62010
[30] MEASE, D. and BINGHAM, D. (2006). Latin hyperrectangle sampling for computer experiments. Technometrics 48 467-477. · doi:10.1198/004017006000000101
[31] MOLLAND, A. F. and TURNOCK, S. R. (2011). Marine Rudders and Control Surfaces: Principles, Data, Design and Applications. Elsevier, Amsterdam.
[32] NING, S., BYON, E., WU, T. and LI, J. (2017). A sparse partitioned-regression model for nonlinear system-environment interactions. IISE Trans. 49 814-826.
[33] OWEN, A. B. (2013). Monte Carlo theory, methods and examples. Book in progress. Online version available at https://statweb.stanford.edu/ owen/mc/.
[34] PAQUETTE, C. and SCHEINBERG, K. (2020). A stochastic line search method with expected complexity analysis. SIAM J. Optim. 30 349-376. · Zbl 1431.90153 · doi:10.1137/18M1216250
[35] Paulo, R., García-Donato, G. and Palomo, J. (2012). Calibration of computer models with multivariate output. Comput. Statist. Data Anal. 56 3959-3974. · Zbl 1255.62187 · doi:10.1016/j.csda.2012.05.023
[36] POLYAK, B. T. (1990). New stochastic approximation type procedures. Autom. Remote Control 7 937-1008. · Zbl 0737.93080
[37] Polyak, B. T. and Juditsky, A. B. (1992). Acceleration of stochastic approximation by averaging. SIAM J. Control Optim. 30 838-855. · Zbl 0762.62022 · doi:10.1137/0330046
[38] QIAN, P. Z. G. and WU, C. F. J. (2008). Bayesian hierarchical modeling for integrating low-accuracy and high-accuracy experiments. Technometrics 50 192-204. · doi:10.1198/004017008000000082
[39] RASMUSSEN, C. E. (2003). Gaussian processes in machine learning. In Summer School on Machine Learning 63-71. Springer, Berlin. · Zbl 1120.68436
[40] RUPPERT, D. (1988). Efficient estimations from a slowly convergent Robbins-Monro process. Technical report, Cornell Univ. Operations Research and Industrial Engineering.
[41] SHASHAANI, S., HASHEMI, F. S. and PASUPATHY, R. (2018). ASTRO-DF: A class of adaptive sampling trust-region algorithms for derivative-free stochastic optimization. SIAM J. Optim. 28 3145-3176. · Zbl 1403.90541 · doi:10.1137/15M1042425
[42] SNELSON, E. and GHAHRAMANI, Z. (2006). Sparse Gaussian processes using pseudo-inputs. In Advances in Neural Information Processing Systems 1257-1264.
[43] STAID, A. (2015). Statistical modeling to support power system planning. Ph.D. thesis, Johns Hopkis Univ., Washington, DC.
[44] THERNEAU, T. M. and ATKINSON, E. J. (2019). An introduction to recursive partitioning using the RPART routines. R Package.
[45] THERNEAU, T., ATKINSON, B., RIPLEY, B. and RIPLEY, M. B. (2019). rpart: Recursive partitioning and regression trees. R Package, Version 4.1-15.
[46] Tuo, R. and Wu, C. F. J. (2015). Efficient calibration for imperfect computer models. Ann. Statist. 43 2331-2352. · Zbl 1326.62228 · doi:10.1214/15-AOS1314
[47] TUO, R. and WU, C. F. J. (2016). A theoretical framework for calibration in computer models: Parametrization, estimation and convergence properties. SIAM/ASA J. Uncertain. Quantificat. 4 767-795. · Zbl 1383.62354 · doi:10.1137/151005841
[48] YOU, M., BYON, E., JIN, J. J. and LEE, G. (2017). When wind travels through turbines: A new statistical approach for characterizing heterogeneous wake effects in multi-turbine wind farms. IISE Trans. 49 84-95.
[49] YOU, M., LIU, B., BYON, E., HUANG, S. and JIN, J. (2018). Direction-dependent power curve modeling for multiple interacting wind turbines. IEEE Trans. Power Syst. 33 1725-1733.
[50] YUAN, J., NG, S. H. and TSUI, K. L. (2013). Calibration of stochastic computer models using stochastic approximation methods. IEEE Trans. Autom. Sci. Eng. 10 171-186.
[51] ZWAKMAN, J. (2014). Wind turbines and the environment. Available at https://www.wijkplatformsvelsen.nl/ijmuiden-noord/2014/11/23/windturbines-en-het-milieu/ Accessed: 2020-09-11
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.