SVRs and Uncertainty Estimates in Wind Energy Prediction

Prada, J.; Dorronsoro, José R.

doi:10.1007/978-3-319-19222-2_47

Cited by 5 publications

(6 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On one hand, a method to build general noise SVR models using NORMA as the optimization algorithm. On the other hand, an approach to compute error intervals for these regression models avoiding constant width by the use of standard clustering methods, instead of employing ad hoc partitioning approaches as proposed in our previous work [6].…”

Section: Resultsmentioning

confidence: 99%

“…Therefore, a method to compute these intervals is also provided. The computation of these intervals is based on the work we previously carried out in [6], but the problem of constant intervals is solved in a general way in this paper.…”

Section: Introductionmentioning

confidence: 99%

“…1) The problem of constant width in the uncertainty intervals formulations described in [6] is addressed using general techniques, based on standard clustering methods such as k-means and k-prototypes. 2) A framework for general noise SVRs with nonconstant uncertainty intervals is proposed, combining the use of general cost functions, NORMA optimization, clustering techniques and non-fixed error interval estimates for a particular choice of noise distribution assumption.…”

Section: Introductionmentioning

confidence: 99%

“…First, the proposed models give better forecasts than classical SVR when a suitable noise distribution is assumed. Proposed techniques to avoid constant width in the uncertainty intervals described in [6] improve the accuracy of error estimates. Finally, Weibull and Beta distributions seem to capture best the underlying noise distribution in solar radiation prediction tasks.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

General noise support vector regression with non-constant uncertainty intervals for solar radiation prediction

Prada

Dorronsoro

2018

J. Mod. Power Syst. Clean Energy

View full text Add to dashboard Cite

General noise cost functions have been recently proposed for support vector regression (SVR). When applied to tasks whose underlying noise distribution is similar to the one assumed for the cost function, these models should perform better than classical -SVR. On the other hand, uncertainty estimates for SVR have received a somewhat limited attention in the literature until now and still have unaddressed problems. Keeping this in mind, three main goals are addressed here. First, we propose a framework that uses a combination of general noise SVR models with naive online R minimization algorithm (NORMA) as optimization method, and then gives nonconstant error intervals dependent upon input data aided by the use of clustering techniques. We give theoretical details required to implement this framework for Laplace, Gaussian, Beta, Weibull and Marshall-Olkin generalized exponential distributions. Second, we test the proposed framework in two real-world regression problems using data of two public competitions about solar energy. Results show the validity of our models and an improvement over classical -SVR. Finally, in accordance with the principle of reproducible research, we make sure that data and model implementations used for the experiments are easily and publicly accessible.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

General noise support vector regression with non-constant uncertainty intervals for solar radiation prediction

Prada

Dorronsoro

2018

J. Mod. Power Syst. Clean Energy

View full text Add to dashboard Cite

show abstract

“…RT, RF, and ANN are based on empirical risk minimization. Based on general noise functions, a direct approach to building error intervals as functions of residual distributions can be created (Prada & Dorronsoro, 2015). Based on these error intervals, the model's parameters can be selected and further used to fit the data.…”

Section: Literature Reviewmentioning

confidence: 99%

Wind Energy Prediction Using Machine Learning

Buturache¹,

Stancu²

2021

LCE

View full text Add to dashboard Cite

Wind energy prediction represents an important and active field in the renewable energy sector. Since renewable energy sources are integrated into existing grids and combined with traditional sources, knowing the amount of energy that will be produced is key in minimizing the operational cost of the wind farm and safe operation of the power grid. In this context, we propose a comparative and comprehensive study of artificial neural networks, support vector regression, random trees, and random forest, and present the pros and cons of implementing the aforementioned techniques. A step-by-step approach based on the CRISP-DM data mining framework reveals the thought process end-to-end, including feature engineering, metrics selection, model selection, or hyperparameter tuning. Using the selected metrics for model evaluation, we provide a summary highlighting the optimal results and the trade-off between performance and the resources expended to achieve these results. This research is also intended to provide guidance for wind energy professionals, filling the gap between purely academic research and real-world business use cases, providing the exact architectures and selected hyperparameters.

show abstract

General Noise SVRs and Uncertainty Intervals

Prada

Dorronsoro

2017

Advances in Computational Intelligence

View full text Add to dashboard Cite

SVRs and Uncertainty Estimates in Wind Energy Prediction

Cited by 5 publications

References 11 publications

General noise support vector regression with non-constant uncertainty intervals for solar radiation prediction

General noise support vector regression with non-constant uncertainty intervals for solar radiation prediction

Wind Energy Prediction Using Machine Learning

General Noise SVRs and Uncertainty Intervals

Contact Info

Product

Resources

About