Ordinal multi-class architecture for predicting wind power ramp events based on reservoir computing

Áreas de investigación:

• Methodology - Machine learning

Año:

2020

Tipo de publicación:

Artículo

Autores:

• Dorado-Moreno, Manuel
• Gutiérrez, Pedro Antonio
• Cornejo-Bueno, Laura
• Prieto, Luis
• Salcedo-Sanz, Sancho
• Hervás-Martínez, César

Journal:

Neural Processing Letters

Volumen:

52

Número:

3

Páginas:

57-74

Mes:

December

ISSN:

1370-4621

BibTex:

@article{1520181, author = "Manuel Dorado-Moreno and Pedro Antonio Guti{\'e}rrez and Laura Cornejo-Bueno and Luis Prieto and Sancho Salcedo-Sanz and C{\'e}sar Herv{\'a}s-Mart{\'i}nez", abstract = "Wind power ramp events (WPREs) are strong increases or decreases of wind speed in a short period of time. Predicting WPREs in wind farms is of vital importance given that they can produce damages in the turbines, and, in any case, they suddenly affect the wind farm production. In contrast to previous binary definitions of the prediction problem (ramp versus non-ramp), a three-class prediction model is used in this paper, proposing a novel discretization function, able to detect the nature of WPREs: negative ramp, non-ramp and positive ramp events. Moreover, the natural order of these labels is exploited to obtain better results in the prediction of these events. The independent variables used for prediction include, in this case, past wind speed values and meteorological data obtained from physical models (reanalysis data). Reanalysis will be also used for recovering missing data from the measuring stations in the wind farm. The proposed prediction methodology is based on Reservoir Computing and an over-sampling process for alleviating the high degree of unbalance in the dataset (non-ramp events are much more frequent than ramps). Three elements are combined in the prediction method: a recurrent neural network layer, a nonlinear kernel mapping and an ordinal logistic regression,to exploit the information provided by the order of the classes). Preprocessing is based on a variation of the standard synthetic minority over-sampling technique (SMOTE), which is applied to the reservoir activations (since the direct application over the input variables would damage its temporal structure). The performance of the method is analysed by comparing it against other state-of-the-art classifiers, such as Support Vector Machines, nominal logistic regression, an autoregressive ordinal neural network, or the use of leaky integrator neurons instead of the standard sigmoidal units. From the results obtained, the benefits of the kernel mapping and the ordinal model are clear, and, in general, the performance obtained with the Reservoir Computing approach is shown to be very robust in the detection of ramps.", awards = "JCR(2020): 2.908 Position: 63/139 (Q2) Category: COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE", comments = "JCR(2020): 2.908 Position: 63/139 (Q2) Category: COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE", doi = "10.1007/s11063-018-9922-5", issn = "1370-4621", journal = "Neural Processing Letters", month = "December", note = "JCR(2020): 2.908 Position: 63/139 (Q2) Category: COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE", number = "3", pages = "57--74", title = "{O}rdinal multi-class architecture for predicting wind power ramp events based on reservoir computing", url = "doi.org/10.1007/s11063-018-9922-5", volume = "52", year = "2020", }

Nota:

JCR(2020): 2.908 Position: 63/139 (Q2) Category: COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Abstract:

Wind power ramp events (WPREs) are strong increases or decreases of wind speed in a short period of time. Predicting WPREs in wind farms is of vital importance given that they can produce damages in the turbines, and, in any case, they suddenly affect the wind farm production. In contrast to previous binary definitions of the prediction problem (ramp versus non-ramp), a three-class prediction model is used in this paper, proposing a novel discretization function, able to detect the nature of WPREs: negative ramp, non-ramp and positive ramp events. Moreover, the natural order of these labels is exploited to obtain better results in the prediction of these events. The independent variables used for prediction include, in this case, past wind speed values and meteorological data obtained from physical models (reanalysis data). Reanalysis will be also used for recovering missing data from the measuring stations in the wind farm. The proposed prediction methodology is based on Reservoir Computing and an over-sampling process for alleviating the high degree of unbalance in the dataset (non-ramp events are much more frequent than ramps). Three elements are combined in the prediction method: a recurrent neural network layer, a nonlinear kernel mapping and an ordinal logistic regression,to exploit the information provided by the order of the classes). Preprocessing is based on a variation of the standard synthetic minority over-sampling technique (SMOTE), which is applied to the reservoir activations (since the direct application over the input variables would damage its temporal structure). The performance of the method is analysed by comparing it against other state-of-the-art classifiers, such as Support Vector Machines, nominal logistic regression, an autoregressive ordinal neural network, or the use of leaky integrator neurons instead of the standard sigmoidal units. From the results obtained, the benefits of the kernel mapping and the ordinal model are clear, and, in general, the performance obtained with the Reservoir Computing approach is shown to be very robust in the detection of ramps.

Comentarios:

JCR(2020): 2.908 Position: 63/139 (Q2) Category: COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE