ABSTRACTS OF ARTICLES OF THE JOURNAL "INFORMATION TECHNOLOGIES".
No. 7. Vol. 30. 2024

DOI: 10.17587/it.30.350-356

D. N. Kobzarenko, Dr. of Tech. Sc., Leading Researcher, A. M. Kamilova, Lead Software Engineer,
Institute for Geothermal and Renewable Energy Research — Branch of Joined Institute for High Temperatures of the RAS, Makhachkala, Russian Federation

Investigation Neural Network Models for Wind Speed Prediction Based on Meteorological Observations in Northern Dagestan

The paper presents the experiments' results on the study variants of neural network architectures for predicting wind speed based on meteorological time series for Northern Dagestan. When working with data and models, modern software tools of the Python programming language for Data Science are used, such as Keras — a library for modeling neural networks, Pandas — a library for processing tabular data, AutoKeras — a library for automatically generating a neural network by dataset, TabGan — a library for expansion a dataset with artificial data.
As initial data, regular observations at the Kochubey (Northern Dagestan) meteorological station for the period 2011­2022 were taken with a frequency of generalization of measurements 8 times a day. The original semi-structured data is pre-processed and reduced to a structured CSV dataset format.
The task of predicting wind speed is reduced to the task of classification, in which it is not the wind speed itself that is predicted, but the class number in accordance with the gradation. From the point of view of considering wind speed as a renewable energy resource, three classes with gradations are accepted: class 0: 0—3 m/s (quiet), class 1: 4—7 m/s (average wind sufficient for optimal wind turbine operating), class 2: 8 m/s and higher (strong wind).
When performing experiments, the influence value on the prediction accuracy from several aspects were analyzed, such as: the time data block length, the neural network architecture, the transformation tabular features to normal form or to categorical form, expansion dataset by artificial data, the layout of the verification and test samples, the imbalance of classes, various meteorological parameters as features.
Keywords: wind speed, data analysis, artificial intelligence, neural network, deep machine learning

P. 350-356

References

1. Kobzarenko D. N. Analysis of time series wind speeds and wind directions using neural network models and classification task, Morskie intellektual'nye tekhnologii, 2021, vol. 1, no. 4 (54), pp. 127—133 (in Russian).
2. Kobzarenko D. N., Mustafaev A. G., Shikhsaidov B. I. Efficiency of using one-dimensional convolutional layers in a neural network on the example of classification a meteorological station according to wind speeds time series data, Informacionnye tekhnologii, 2022, vol. 28, no. 9. pp. 497—504 (in Russian).
3. Kobzarenko D. N., Kamilova A. M. Research time series of meteorological observations as features for wind speed forecasting in a neural network, Morskie intellektual'nye tekhnologii, 2022, vol. 1, no. 4 (58), pp. 170—176 (in Russian).
4. Matrenin P. V., Manusov V. Z., Igumnova E. A. Sustai­nable short-term wind speed prediction using adaptive compact neural networks, Problemy regional'noj energetiki, 2020, no. 3 (47), pp. 69—80 (in Russian).
5. Catalao J. P. S., Pousinho H. M. I., Mendes V. M. F. Short-term wind power forecasting in Portugal by neural networks and wavelet transform, Renewable Energy, 2011, no. 36, pp. 1245—1251.
6. Kumar S. P., Lopez D. Forecasting of Wind Speed Using Feature Selection and Neural Networks, International Journal of Renewable Energy Research, 2016, vol. 6, no. 3, pp. 833—837.
7. Doucoure B., Agbossou K., Cardenas A. Time series pre­diction using artificial wavelet neural network and multi-resolution analysis: Application to wind speed data, Renewable Energy, 2016, no. 92, pp. 202—211.
8. AutoKeras [website], available at: https://autokeras.com/ (date of access: 01.11.23).
9. Kiranyaza S., Avcib O., Abdeljaberc O., Inced T., Gabbouje M., Inmanf D. 1D convolutional neural networks and applications: A survey, Mechanical Systems and Signal Processing, 2021, vol. 151.
10. Hochreiter S., Schmidhuber J. Long short-term memory, Neural Computation, 1997, vol. 9, no. 8, pp. 1735—1780.
11. Ashrapov I. Tabular GANs for uneven distribution (2020), arXiv: 2010.00638 [cs.LG].
12. CatBoostClassifier [website], available at: https://catboost.ai/en/docs/concepts/python-reference_catboostclassifier (date of access: 20.12.23).

To the contents