The escalating concern over the adverse effects of greenhouse gas emissions on the Earth's climate has intensified the need for sustainable and renewable energy sources. Among the alternatives, wind energy has emerged as a key solution for mitigating the impacts of global warming. The significance of wind energy generation lies in its abundance, environmental benefits, cost-effectiveness and contribution to energy security. Accurate forecasting of wind energy generation is crucial for managing its intermittent nature and ensuring effective integration into the electricity grid. We employed machine learning techniques to predict wind power generation by utilizing historical weather data in conjunction with corresponding wind power generation data. The dataset was sourced from real-time SCADA data obtained from wind turbines, allowing for a comprehensive analysis. We differentiated this research by evaluating not only wind conditions but also meteorological factors and physical measurements of turbine components, thus considering their combined influence on overall wind power production. We utilized Decision Tree, Random Forest, K-Nearest Neighbors (KNN), and XGBoost algorithms to estimate power generation. The performance of these models assessed using evaluation criteria: R2, Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE). The findings indicated XGBoost algorithm outperformed the other models, achieving high accuracy while demonstrating computational efficiency, making it particularly suitable for real-time applications in energy forecasting.
Citation: Asiye Bilgili, Kerem Gül. Forecasting power generation of wind turbine with real-time data using machine learning algorithms[J]. Clean Technologies and Recycling, 2024, 4(2): 108-124. doi: 10.3934/ctr.2024006
The escalating concern over the adverse effects of greenhouse gas emissions on the Earth's climate has intensified the need for sustainable and renewable energy sources. Among the alternatives, wind energy has emerged as a key solution for mitigating the impacts of global warming. The significance of wind energy generation lies in its abundance, environmental benefits, cost-effectiveness and contribution to energy security. Accurate forecasting of wind energy generation is crucial for managing its intermittent nature and ensuring effective integration into the electricity grid. We employed machine learning techniques to predict wind power generation by utilizing historical weather data in conjunction with corresponding wind power generation data. The dataset was sourced from real-time SCADA data obtained from wind turbines, allowing for a comprehensive analysis. We differentiated this research by evaluating not only wind conditions but also meteorological factors and physical measurements of turbine components, thus considering their combined influence on overall wind power production. We utilized Decision Tree, Random Forest, K-Nearest Neighbors (KNN), and XGBoost algorithms to estimate power generation. The performance of these models assessed using evaluation criteria: R2, Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE). The findings indicated XGBoost algorithm outperformed the other models, achieving high accuracy while demonstrating computational efficiency, making it particularly suitable for real-time applications in energy forecasting.
| [1] | Michaelowa A, Dransfeld B, Blodgett C, et al. (2012) IRENA handbook on renewable energy nationally appropriate mitigation actions (NAMAs) for policy makers and project developers. Abu Dhabi, United Arab Emirates: IRENA (International Renewable Energy Agency). |
| [2] |
Breidenich C, Magraw D, Rowley A, et al. (1998) The Kyoto Protocol to the United Nations framework convention on climate change. American J Int Law 92: 315–331. https://doi.org/10.2307/2998044 doi: 10.2307/2998044
|
| [3] | Kumar S, Madlener R (2018) Energy systems and COP21 Paris climate agreement targets in Germany: An integrated modeling approach. 2018 7th International Energy and Sustainability Conference (IESC), 1–6. https://doi.org/10.1109/IESC.2018.8440004 |
| [4] | Crippa M, Guizzardi D, Schaaf E, et al. (2023) GHG emissions of all world countries: 2023. Publications Office of the European Union. European Commission, Joint Research Centre. Available from: https://data.europa.eu/doi/10.2760/953322. |
| [5] | Breitschopf B, Herbst A (2023) Supply chain risks in the EU's energy technologies: Terms of reference. Publications Office of the European Union. European Commission, Directorate-General for Energy. Available from: https://data.europa.eu/doi/10.2833/818557. |
| [6] |
Saini VK, Kumar R, Al-Sumaiti AS, et al. (2023) Learning based short term wind speed forecasting models for smart grid applications: An extensive review and case study. Electric Power Syst Res 222: 109502. https://doi.org/10.1016/j.epsr.2023.109502 doi: 10.1016/j.epsr.2023.109502
|
| [7] | Jørgensen KL, Shaker HR (2020) Wind power forecasting using machine learning: state of the art, trends and challenges. Proceedings of the 2020 the 8th IEEE International Conference on Smart Energy Grid Engineering (SEGE), 44–50. https://doi.org/10.1109/SEGE49949.2020.9181870 |
| [8] |
Ho CY, Cheng KS, Ang CH (2023) Utilizing the random forest method for short-term wind speed forecasting in the coastal area of central Taiwan. Energies 16: 1374. https://doi.org/10.3390/en16031374 doi: 10.3390/en16031374
|
| [9] |
Singh U, Rizwan M, Alaraj M, et al. (2021) A machine learning-based gradient boosting regression approach for wind power production forecasting: A step towards smart grid environments. Energies 14: 5196. https://doi.org/10.3390/en14165196 doi: 10.3390/en14165196
|
| [10] |
Demolli H, Dokuz AS, Ecemis A, et al. (2019) Wind power forecasting based on daily wind speed data using machine learning algorithms. Energy Convers Manage 198: 111823. https://doi.org/10.1016/j.enconman.2019.111823 doi: 10.1016/j.enconman.2019.111823
|
| [11] | Liu T, Fan L (2021) Wind power prediction based on three machine-learning algorithms: Decision tree, k-nearest neighbors and random forest. In: Xu, J., Márquez, F.P.G., Hassan, M.H.A, Duca, G., Hajiyev, A., Altiparmak, F. Author, Proceedings of the Fifteenth International Conference on Management Science and Engineering Management, New York: Springer, Cham. 78: 490–499. https://doi.org/10.1007/978-3-030-79203-9_38 |
| [12] |
Krechowicz A, Krechowicz M, Poczeta K (2022) Machine learning approaches to predict electricity production from renewable energy sources. Energies 15: 9146. https://doi.org/10.3390/en15239146 doi: 10.3390/en15239146
|
| [13] | Demir F, Tasci B (2021) Predicting the power of a wind turbine with machine learning-based approaches from wind direction and speed data. 2021 International Conference on Technology and Policy in Energy and Electric Power (ICT-PEP), 37–40. https://doi.org/10.1109/ICT-PEP53949.2021.9600959 |
| [14] |
Sui A, Qian W (2021) Forecasting the wind power generation in China by seasonal grey forecasting model based on collaborative optimization. RAIRO-Oper Res 55: 3049–3072. https://doi.org/10.1051/ro/2021136 doi: 10.1051/ro/2021136
|
| [15] |
Anushalini T, Sri Revathi B (2023) Role of machine learning algorithms for wind power generation prediction in renewable energy management. IETE J Res 70: 4319–4332. https://doi.org/10.1080/03772063.2023.2205838 doi: 10.1080/03772063.2023.2205838
|
| [16] |
Jin H, Li Y, Wang B, et al. (2022) Adaptive forecasting of wind power based on selective ensemble of offline global and online local learning. Energy Conver Manage 271: 116296. https://doi.org/10.1016/j.enconman.2022.116296 doi: 10.1016/j.enconman.2022.116296
|
| [17] |
Wood DA (2022) Trend decomposition aids short-term countrywide wind capacity factor forecasting with machine and deep learning methods. Energy Conver Manage 253: 115189. https://doi.org/10.1016/j.enconman.2021.115189 doi: 10.1016/j.enconman.2021.115189
|
| [18] | Saini VK, Kumar R, Mathur A, et al. (2020) Short term forecasting based on hourly wind speed data using deep learning algorithms. 3rd International Conference on Emerging Technologies in Computer Engineering: Machine Learning and Internet of Things (ICETCE), Jaipur, India: 1–6. https://doi.org/10.1109/ICETCE48199.2020.9091757 |
| [19] |
Ayene SM, Yibre AM (2024) Wind power prediction based on deep learning models: The case of Adama wind farm. Heliyon 10: e39579. https://doi.org/10.1016/j.heliyon.2024.e39579 doi: 10.1016/j.heliyon.2024.e39579
|
| [20] |
Wang X, Hao Y, Yang W (2024) Novel wind power ensemble forecasting system based on mixed-frequency modeling and interpretable base model selection strategy. Energy 297: 131142. https://doi.org/10.1016/j.energy.2024.131142 doi: 10.1016/j.energy.2024.131142
|
| [21] |
Shinde SK, Tirlangi S, Kumar NK, et al. (2024) Enhancing wind power generation forecasting with advanced deep learning technique using wavelet-enhanced recurrent neural network and gated linear units. International J Renewable Energy Res 14: 324–338. https://doi.org/10.20508/ijrer.v14i2.14577.g8893 doi: 10.20508/ijrer.v14i2.14577.g8893
|
| [22] |
Wang Y, Zhao K, Hao Y, et al. (2024) Short-term wind power prediction using a novel model based on butterfly optimization algorithm-variational mode decomposition-long short-term memory. Appl Energy 366: 123313. https://doi.org/10.1016/j.apenergy.2024.123313 doi: 10.1016/j.apenergy.2024.123313
|
| [23] |
Piotrowski P, Rutyna I, Baczyński D, et al. (2022) Evaluation metrics for wind power forecasts: A comprehensive review and statistical analysis of errors. Energies 15: 9657. https://doi.org/10.3390/en15249657 doi: 10.3390/en15249657
|
| [24] |
Farrar NO, Ali MH, Dasgupta D (2023) Artificial Intelligence and machine learning in grid connected wind turbine control systems: A comprehensive review. Energies 16: 1530. https://doi.org/10.3390/en16031530 doi: 10.3390/en16031530
|
| [25] |
Xie Y, Li C, Li M, et al. (2023) An overview of deterministic and probabilistic forecasting methods of wind energy. IScience 26: 105804. https://doi.org/10.1016/j.isci.2022.105804 doi: 10.1016/j.isci.2022.105804
|
| [26] |
Wood DA (2023) Feature averaging of historical meteorological data with machine and deep learning assist wind farm power performance analysis and forecasts. Energy Syst 14: 1023–1049 https://doi.org/10.1007/s12667-022-00502-x doi: 10.1007/s12667-022-00502-x
|
| [27] |
Benti NE, Chaka MD, Semie AG (2023) Forecasting renewable energy generation with machine learning and deep learning: Current advances and future prospects. Sustainability 15: 7087. https://doi.org/10.3390/su15097087 doi: 10.3390/su15097087
|
| [28] |
Alkhayat G, Mehmood R (2021) A review and taxonomy of wind and solar energy forecasting methods based on deep learning. Energy AI 4: 100060. https://doi.org/10.1016/j.egyai.2021.100060 doi: 10.1016/j.egyai.2021.100060
|
| [29] |
Garg S, Krishnamurthi R (2023) A survey of long short term memory and its associated models in sustainable wind energy predictive analytics. Artif Intell Rev 56: 1149–1198 https://doi.org/10.1007/s10462-023-10554-9 doi: 10.1007/s10462-023-10554-9
|
| [30] | Nazir MS, Wang Y, Bilal M, et al. (2022) Wind energy, its application, challenges, and potential environmental impact. Handbook of Climate Change Mitigation and Adaptation New York: Springer, 1–38. https://doi.org/10.1007/978-1-4614-6431-0_108-2 |
| [31] |
Malakouti SM (2023) Improving the prediction of wind speed and power production of SCADA system with ensemble method and 10-fold cross-validation. Case Stud Chem Environ Eng 8: 100351. https://doi.org/10.1016/j.cscee.2023.100351 doi: 10.1016/j.cscee.2023.100351
|
| [32] | Wirth R, Hipp J (2000) Crisp-dm: Towards a standard process modell for data mining. Computer Sci |
| [33] |
Mansoury I, El Bourakadi D, Yahyaouy A, et al. (2023) A novel wind power prediction model using graph attention networks and bi-directional deep learning long and short term memory. Int J Electr Comput Eng, 6847–6854. http://doi.org/10.11591/ijece.v13i6.pp6847-6854 doi: 10.11591/ijece.v13i6.pp6847-6854
|
| [34] | Kaggle, Wind Turbine Power (kW) Generation data, 2023. Available from: https://www.kaggle.com/datasets/psycon/wind-turbine-energy-kw-generation-data. |
| [35] | Saini VK, Mathur F, Gupta V, et al. (2020) Predictive analysis of traditional, deep learning and ensemble learning approach for short-term wind speed forecasting. International Conference on Computing, Power and Communication Technologies (GUCON), Greater Noida: IEEE, 783–788. https://doi.org/10.1109/GUCON48875.2020.9231081 |
Figures(7) / Tables(3)
Asiye Bilgili, Kerem Gül. Forecasting power generation of wind turbine with real-time data using machine learning algorithms[J]. Clean Technologies and Recycling, 2024, 4(2): 108-124. doi: 10.3934/ctr.2024006
DownLoad: