# Data Stream Mining Applied to Maximum Wind Forecasting in the Canary Islands

^{1}

^{2}

^{*}

## Abstract

**:**

## 1. Introduction

## 2. State of the Art

## 3. Data Stream Mining

## 4. Data-Set Used and Exploratory Analysis

- Datalogger DLx-MET
- Vaisala HydroMet System MAWS301
- SEAC EMA55

## 5. Methodology

#### 5.1. Linear Regression

#### 5.2. Adaptive Learning Strategy—Data-Stream-Mining-Based

Algorithm 1: Adaptive incremental linear regression gradient descent learner |

#### 5.3. Accumulative Strategy

## 6. Experimental Design and Results

## 7. Results Discussion

## 8. Concluding Remarks and Future Work

#### 8.1. Conclusions

#### 8.2. Future Research

## Author Contributions

## Funding

## Conflicts of Interest

## Abbreviations

EU | European Union |

AEMET | Agencia Estatal de Meteorología |

r.p.m. | Revolutions per minute |

IoE | Internet of Everything |

VMAX10m | Maximum wind speed measured with a 10-min sampling period |

MSE | Mean squared error |

ISTAC | Canary Island Statistics Institute |

## References

- Beven, J. Tropical Cyclone Report: Tropical Storm Delta, 22–28 November 2005; NOAA Technical Notes; Tropical Prediction Center, National Hurricane Center: Miami, FL, USA, 2006. [Google Scholar]
- Seco, A.; González, P.; Ramírez, F.; García, R.; Prieto, E.; Yagüe, C.; Fernández, J. GPS monitoring of the tropical storm delta along the Canary Islands track, 28–29 November 2005. Pure Appl. Geophys.
**2009**, 166, 1519–1531. [Google Scholar] [CrossRef] - Fernández Serdán, J.M.; San Ambrosio Beirán, I.; Martín León, F. La inusual y anómala tormenta tropical “Delta”. Ambienta La revista del Ministerio de Medio Ambiente
**2006**, 52, 60–65. [Google Scholar] - Patricola, C.M.; Wehner, M.F. Anthropogenic influences on major tropical cyclone events. Nature
**2018**, 563, 339. [Google Scholar] [CrossRef] [PubMed] - García-Herrera, R.; Gallego, D.; Hernández, E.; Gimeno, L.; Ribera, P.; Calvo, N. Precipitation trends in the Canary Islands. Int. J. Climatol.
**2003**, 23, 235–241. [Google Scholar] [CrossRef] [Green Version] - Cropper, T.E.; Hanna, E. An analysis of the climate of Macaronesia, 1865–2012. Int. J. Climatol.
**2014**, 34, 604–622. [Google Scholar] [CrossRef] - Wang, Z.; Wang, W.; Liu, C.; Wang, Z.; Hou, Y. Probabilistic forecast for multiple wind farms based on regular vine copulas. IEEE Trans. Power Syst.
**2018**, 33, 578–589. [Google Scholar] [CrossRef] - Lahouar, A.; Slama, J.B.H. Hour-ahead wind power forecast based on random forests. Renew. Energy
**2017**, 109, 529–541. [Google Scholar] [CrossRef] - Shi, Z.; Liang, H.; Dinavahi, V. Direct interval forecast of uncertain wind power based on recurrent neural networks. IEEE Trans. Sustain. Energy
**2018**, 9, 1177–1187. [Google Scholar] [CrossRef] - Krawczyk, B.; Minku, L.L.; Gama, J.; Stefanowski, J.; Woźniak, M. Ensemble learning for data stream analysis: A survey. Inf. Fusion
**2017**, 37, 132–156. [Google Scholar] [CrossRef] - Dawid, A.P. Present position and potential developments: Some personal views statistical theory the prequential approach. J. R. Stat. Soc. Ser. A Gen.
**1984**, 147, 278–290. [Google Scholar] [CrossRef] - European Commission (EC). Europe 2020: A Strategy for Smart, Sustainable and Inclusive Growth; Working Paper {COM (2010) 2020}; European Commission: Brussels, Belgium, 2010. [Google Scholar]
- Bossanyi, E. Short-term wind prediction using Kalman filters. Wind Eng.
**1985**, 9, 1–8. [Google Scholar] - Torres, J.L.; Garcia, A.; De Blas, M.; De Francisco, A. Forecast of hourly average wind speed with ARMA models in Navarre (Spain). Sol. Energy
**2005**, 79, 65–77. [Google Scholar] [CrossRef] - Carpinone, A.; Giorgio, M.; Langella, R.; Testa, A. Markov chain modeling for very-short-term wind power forecasting. Electr. Power Syst. Res.
**2015**, 122, 152–158. [Google Scholar] [CrossRef] [Green Version] - Salcedo-Sanz, S.; Perez-Bellido, A.M.; Ortiz-García, E.G.; Portilla-Figueras, A.; Prieto, L.; Paredes, D. Hybridizing the fifth generation mesoscale model with artificial neural networks for short-term wind speed prediction. Renew. Energy
**2009**, 34, 1451–1457. [Google Scholar] [CrossRef] - Li, G.; Shi, J. On comparing three artificial neural networks for wind speed forecasting. Appl. Energy
**2010**, 87, 2313–2320. [Google Scholar] [CrossRef] - Dalto, M.; Matuško, J.; Vašak, M. Deep neural networks for ultra-short-term wind forecasting. In Proceedings of the 2015 IEEE International Conference on Industrial Technology (ICIT), Seville, Spain, 17–19 March 2015; pp. 1657–1663. [Google Scholar]
- Huang, C.J.; Kuo, P.H. A deep cnn-lstm model for particulate matter (PM2. 5) forecasting in smart cities. Sensors
**2018**, 18, 2220. [Google Scholar] [CrossRef] [PubMed] - Mohandes, M.A.; Halawani, T.O.; Rehman, S.; Hussain, A.A. Support vector machines for wind speed prediction. Renew. Energy
**2004**, 29, 939–947. [Google Scholar] [CrossRef] - Cassola, F.; Burlando, M. Wind speed and wind energy forecast through Kalman filtering of Numerical Weather Prediction model output. Appl. Energy
**2012**, 99, 154–166. [Google Scholar] [CrossRef] - Jiang, Y.; Song, Z.; Kusiak, A. Very short-term wind speed forecasting with Bayesian structural break model. Renew. Energy
**2013**, 50, 637–647. [Google Scholar] [CrossRef] - Cai, L.; Gu, J.; Ma, J.; Jin, Z. Probabilistic Wind Power Forecasting Approach via Instance-Based Transfer Learning Embedded Gradient Boosting Decision Trees. Energies
**2019**, 12, 159. [Google Scholar] [CrossRef] - Salcedo-Sanz, S.; Pastor-Sánchez, A.; Del Ser, J.; Prieto, L.; Geem, Z.W. A coral reefs optimization algorithm with harmony search operators for accurate wind speed prediction. Renew. Energy
**2015**, 75, 93–101. [Google Scholar] [CrossRef] - Marrero, C.; Jorba, O.; Cuevas, E.; Baldasano, J.M. Sensitivity study of surface wind flow of a limited area model simulating the extratropical storm Delta affecting the Canary Islands. Adv. Sci. Res.
**2008**, 2, 151–157. [Google Scholar] [CrossRef] - Qin, Z.; Li, W.; Xiong, X. Estimating wind speed probability distribution using kernel density method. Electr. Power Syst. Res.
**2011**, 81, 2139–2146. [Google Scholar] [CrossRef] - Bradley, J.; Barbier, J.; Handler, D. Embracing the Internet of Everything To Capture Your Share of $14.4 Trillion: More Relevant Valuable Connections Will Improve Innovation Productivity Efficiency & Customer Experience; White Paper; Cisco Systems Inc.: San José, CA, USA, 2013. [Google Scholar]
- Demchenko, Y.; Grosso, P.; De Laat, C.; Membrey, P. Addressing big data issues in scientific data infrastructure. In Proceedings of the 2013 International Conference on Collaboration Technologies and Systems (CTS), San Diego, CA, USA, 20–24 May 2013; pp. 48–55. [Google Scholar]
- Ishwarappa; Anuradha, J. A brief introduction on Big Data 5Vs characteristics and Hadoop technology. Proc. Comput. Sci.
**2015**, 48, 319–324. [Google Scholar] [CrossRef] - Bifet, A.; Kirkby, R.B. Data Stream Mining a Practical Approach. Available online: https://www.cs.waikato.ac.nz/~abifet/MOA/StreamMining.pdf (accessed on 24 May 2019).
- Gama, J.; Gaber, M.M. Learning from Data Streams: Processing Techniques in Sensor Networks; Springer: Berlin, Germany, 2007. [Google Scholar]
- Schlimmer, J.C.; Granger, R.H. Incremental learning from noisy data. Mach. Learn.
**1986**, 1, 317–354. [Google Scholar] [CrossRef] - Widmer, G.; Kubat, M. Learning in the presence of concept drift and hidden contexts. Mach. Learn.
**1996**, 23, 69–101. [Google Scholar] [CrossRef] [Green Version] - Hastie, T.; Tibshirani, R. Generalized Additive Models; CRC Press: Boca Raton, FL, USA, 1990. [Google Scholar]
- Wickham, H. ggplot2: Elegant Graphics for Data Analysis; Springer: New York, NY, USA, 2016. [Google Scholar]
- Bechtel, B. The climate of the Canary Islands by annual cycle parameters. ISPRS
**2016**, XLI-B8, 243–250. [Google Scholar] - Rodríguez, M.; Neris, J.; Tejedor, M.; Jiménez, C. Soil temperature regimes from different latitudes on a subtropical island (Tenerife, Spain). Soil Sci. Soc. Am. J.
**2010**, 74, 1662–1669. [Google Scholar] [CrossRef]

**Figure 2.**Smoothed (using GAM) evolution of VMAX10m (Maximum wind speed in red, upper part) and VV10m (averaged wind speed in blue, lower part) during the sampled period.

**Figure 4.**Learning curve of Gradient Descent applied to station ID C619Y during the first batch, varying $\alpha $.

**Figure 5.**Mean accumulative and adaptive cost ((m/s)${}^{2}$) for weather stations (estimated values vs. observed values for the two strategies) and difference between the two.

**Figure 6.**Cost (mean squared error (MSE), (m/s)${}^{2}$) box-plot comparison between both methods for each weather station.

**Figure 7.**Weather station ID C619Y: Best performance case for the adaptive method. In the lower part, cost (MSE, (m/s)${}^{2}$) obtained with the adaptive strategy (black) and with the accumulative strategy (blue). In the upper part, window size (red) and learning rate (green), both for the adaptive strategy.

**Figure 8.**Weather station ID C457I: Worst performance case for the adaptive method. In the lower part, cost (MSE, (m/s)${}^{2}$) obtained with the adaptive strategy (black) and with the accumulative strategy (blue). In the upper part, window size (red) and learning rate (green), both for the adaptive strategy.

**Figure 10.**Outliers cost (MSE (m/s)${}^{2}$) representation for each weather station, $VMax10m>{Q}_{3}+1.5\times IQR$ (16.3 m/s, 36.5 mph, 58.7 Km/h), and $VMax10m>{Q}_{3}+3.0\times IQR$ (24.1 m/s, 53.91 mph, 86.76 Km/h).

**Figure 11.**Coefficient of determination (${r}^{2}$) box-plot comparison between both methods for each weather station.

IDEMA | UBI | ISLAND | IDEMA | UBI | ISLAND |
---|---|---|---|---|---|

C018J | TIAS-LAS VEGAS | LANZAROTE | C457I | VICTORIA-DEPÓSITO MARRERO | TENERIFE |

C019V | YAIZA-PLAYA BLANCA | LANZAROTE | C458A | TACORONTE-A S.E.A. | TENERIFE |

C029O | LANZAROTE/AEROPUERTO | LANZAROTE | C459Z | PUERTO DE LA CRUZ | TENERIFE |

C038N | HARÍA-CEMENTERIO | LANZAROTE | C469N | SILOS-DEPURADORA | TENERIFE |

C048W | TINAJO-LOS DOLORES | LANZAROTE | C611E | SAN MATEO (CORRAL DE LOS JUNCOS) | GRAN CANARIA |

C117A | PUNTAGORDA | LA PALMA | C612F | TEJEDA-CRUZ DE TEJEDA | GRAN CANARIA |

C117Z | TIJARAFE-MIRADOR TIME | LA PALMA | C614H | TEJEDA CASCO | GRAN CANARIA |

C126A | EL PASO-C.F. | LA PALMA | C619X | AGAETE-CASCO | GRAN CANARIA |

C129V | FUENCALIENTE-SALINAS | LA PALMA | C619Y | LA ALDEA DE SAN NICOLAS | GRAN CANARIA |

C129Z | TAZACORTE | LA PALMA | C623I | SAN BARTOLOME TIRAJANA (CUEVAS DEL PINAR) | GRAN CANARIA |

C139E | LA PALMA/AEROPUERTO | LA PALMA | C625O | SAN BARTOLOME TIRAJANA-LOMO PEDRO ALFONSO | GRAN CANARIA |

C148F | SAUCES-S.ANDRÉS-BALSA ADEYAHAME | LA PALMA | C628B | SAN NICOLAS T.-TASARTE/COPARLITA | GRAN CANARIA |

C229J | PÁJARA-PUERTO MORRO JABLE | FUERTEVENTURA | C629Q | MOGAN (PUERTO RICO) | GRAN CANARIA |

C239N | TUINEJE-PUERTO GRAN TARAJAL | FUERTEVENTURA | C629X | PUERTO DE MOGÁN | GRAN CANARIA |

C248E | ANTIGUA-EL CARBÓN | FUERTEVENTURA | C635B | SAN BARTOLOME TIRAJANA-H.LAS TIRAJANAS | GRAN CANARIA |

C249I | FUERTEVENTURA/AEROPUERTO | FUERTEVENTURA | C639M | SAN BARTOLOME TIRAJANA-C.INSULAR TURISMO | GRAN CANARIA |

C258K | LA OLIVA (CARRETERA DEL COTILLO) | FUERTEVENTURA | C639U | SAN BARTOLOME TIRAJANA (EL MATORRAL) | GRAN CANARIA |

C259X | LA OLIVA-PUERTO DE CORRALEJO | FUERTEVENTURA | C648C | AGÜIMES-EL MILANO | GRAN CANARIA |

C314Z | VALLEHERMOSO-ALTO IGUALERO | LA GOMERA | C648N | TELDE-CENTRO FORESTAL DORAMAS | GRAN CANARIA |

C317B | AGULO-JUEGO BOLAS | LA GOMERA | C649I | LAS PALMAS DE GRAN CANARIA/GANDO | GRAN CANARIA |

C319W | VALLEHERMOSO-DAMA | LA GOMERA | C649R | TELDE-MELENARA | GRAN CANARIA |

C328W | HERMIGUA-DEPÓSITO AYUNTAMIENTO | LA GOMERA | C656V | TEROR-OSORIO | GRAN CANARIA |

C329Z | SAN SEBASTIÁN DE LA GOMERA | LA GOMERA | C658X | LAS PALMAS G.C.-TAFIRA/ZURBARÁN | GRAN CANARIA |

C406G | CAÑADAS PARADOR | TENERIFE | C659H | LAS PALMAS G.C. SAN CRISTÓBAL | GRAN CANARIA |

C419X | ADEJE-CALDERA B | TENERIFE | C659M | LAS PALMAS DE GC. PLAZA DE LA FERIA | GRAN CANARIA |

C428T | ARICO-DEPURADORA LA DEGOLLADA | TENERIFE | C665T | VALLESECO | GRAN CANARIA |

C429I | TENERIFE/SUR | TENERIFE | C668V | AGAETE - SUERTE ALTA | GRAN CANARIA |

C430E | IZAÑA | TENERIFE | C669B | ARUCAS-BAÑADEROS | GRAN CANARIA |

C438N | CANDELARIA-DEPOSITO CUEVECITAS | TENERIFE | C689E | MASPALOMAS | GRAN CANARIA |

C439J | TENERIFE-GÜIMAR | TENERIFE | C839X | TEGUISE LA GRACIOSA-HELIPUERTO | LA GRACIOSA |

C446G | LAS MERCEDES-LLANO LOS LOROS | TENERIFE | C916Q | PINAR-DEPÓSITO | EL HIERRO |

C447A | TENERIFE/LOS RODEOS | TENERIFE | C925F | SAN ANDRÉS-DEPÓSITO CABILDO | EL HIERRO |

C449C | SANTA CRUZ DE TENERIFE | TENERIFE | C929I | EL HIERRO/AEROPUERTO | EL HIERRO |

C449F | ANAGA-COL. REP. ARGENTINA | TENERIFE | C939T | SABINOSA-BALNEARIO | EL HIERRO |

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Sánchez-Medina, J.J.; Guerra-Montenegro, J.A.; Sánchez-Rodríguez, D.; Alonso-González, I.G.; Navarro-Mesa, J.L.
Data Stream Mining Applied to Maximum Wind Forecasting in the Canary Islands. *Sensors* **2019**, *19*, 2388.
https://doi.org/10.3390/s19102388

**AMA Style**

Sánchez-Medina JJ, Guerra-Montenegro JA, Sánchez-Rodríguez D, Alonso-González IG, Navarro-Mesa JL.
Data Stream Mining Applied to Maximum Wind Forecasting in the Canary Islands. *Sensors*. 2019; 19(10):2388.
https://doi.org/10.3390/s19102388

**Chicago/Turabian Style**

Sánchez-Medina, Javier J., Juan Antonio Guerra-Montenegro, David Sánchez-Rodríguez, Itziar G. Alonso-González, and Juan L. Navarro-Mesa.
2019. "Data Stream Mining Applied to Maximum Wind Forecasting in the Canary Islands" *Sensors* 19, no. 10: 2388.
https://doi.org/10.3390/s19102388