Improving of the Generation Accuracy Forecasting of Photovoltaic Plants Based on <i>k</i>-Means and <i>k</i>-Nearest Neighbors Algorithms
Renewable energy sources (RES) are seen as a means of the fuel and energy complex carbon footprint reduction but the stochastic nature of generation complicates RES integration with electric power systems. Therefore, it is necessary to develop and improve methods for forecasting of the power plants...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | Russian |
| Published: |
Belarusian National Technical University
2023-08-01
|
| Series: | Известия высших учебных заведений и энергетических объединенний СНГ: Энергетика |
| Subjects: | |
| Online Access: | https://energy.bntu.by/jour/article/view/2287 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832557308878520320 |
|---|---|
| author | P. V. Matrenin A. I. Khalyasmaa V. V. Gamaley S. A. Eroshenko N. A. Papkova D. A. Sekatski Y. V. Potachits |
| author_facet | P. V. Matrenin A. I. Khalyasmaa V. V. Gamaley S. A. Eroshenko N. A. Papkova D. A. Sekatski Y. V. Potachits |
| author_sort | P. V. Matrenin |
| collection | DOAJ |
| description | Renewable energy sources (RES) are seen as a means of the fuel and energy complex carbon footprint reduction but the stochastic nature of generation complicates RES integration with electric power systems. Therefore, it is necessary to develop and improve methods for forecasting of the power plants generation using the energy of the sun, wind and water flows. One of the ways to improve the accuracy of forecast models is a deep analysis of meteorological conditions as the main factor affecting the power generation. In this paper, a method for adapting of forecast models to the meteorological conditions of photovoltaic stations operation based on machine learning algorithms was proposed and studied. In this case, unsupervised learning is first performed using the k-means method to form clusters. For this, it is also proposed to use studied the feature space dimensionality reduction algorithm to visualize and estimate the clustering accuracy. Then, for each cluster, its own machine learning model was trained for generation forecasting and the k-nearest neighbours algorithm was built to attribute the current conditions at the model operation stage to one of the formed clusters. The study was conducted on hourly meteorological data for the period from 1985 to 2021. A feature of the approach is the clustering of weather conditions on hourly rather than daily intervals. As a result, the mean absolute percentage error of forecasting is reduced significantly, depending on the prediction model used. For the best case, the error in forecasting of a photovoltaic plant generation an hour ahead was 9 %. |
| format | Article |
| id | doaj-art-6a4b4bb6e5b844deb8fe1ce5836fbbed |
| institution | Kabale University |
| issn | 1029-7448 2414-0341 |
| language | Russian |
| publishDate | 2023-08-01 |
| publisher | Belarusian National Technical University |
| record_format | Article |
| series | Известия высших учебных заведений и энергетических объединенний СНГ: Энергетика |
| spelling | doaj-art-6a4b4bb6e5b844deb8fe1ce5836fbbed2025-02-03T05:20:03ZrusBelarusian National Technical UniversityИзвестия высших учебных заведений и энергетических объединенний СНГ: Энергетика1029-74482414-03412023-08-0166430532110.21122/1029-7448-2023-66-4-305-3211858Improving of the Generation Accuracy Forecasting of Photovoltaic Plants Based on <i>k</i>-Means and <i>k</i>-Nearest Neighbors AlgorithmsP. V. Matrenin0A. I. Khalyasmaa1V. V. Gamaley2S. A. Eroshenko3N. A. Papkova4D. A. Sekatski5Y. V. Potachits6Novosibirsk State Technical University; Ural Federal University named after the first President of Russia B. N. YeltsinNovosibirsk State Technical University; Ural Federal University named after the first President of Russia B. N. YeltsinNovosibirsk State Technical UniversityNovosibirsk State Technical University; Ural Federal University named after the first President of Russia B. N. YeltsinBelarusian National Technical UniversityBelarusian National Technical UniversityBelarusian National Technical UniversityRenewable energy sources (RES) are seen as a means of the fuel and energy complex carbon footprint reduction but the stochastic nature of generation complicates RES integration with electric power systems. Therefore, it is necessary to develop and improve methods for forecasting of the power plants generation using the energy of the sun, wind and water flows. One of the ways to improve the accuracy of forecast models is a deep analysis of meteorological conditions as the main factor affecting the power generation. In this paper, a method for adapting of forecast models to the meteorological conditions of photovoltaic stations operation based on machine learning algorithms was proposed and studied. In this case, unsupervised learning is first performed using the k-means method to form clusters. For this, it is also proposed to use studied the feature space dimensionality reduction algorithm to visualize and estimate the clustering accuracy. Then, for each cluster, its own machine learning model was trained for generation forecasting and the k-nearest neighbours algorithm was built to attribute the current conditions at the model operation stage to one of the formed clusters. The study was conducted on hourly meteorological data for the period from 1985 to 2021. A feature of the approach is the clustering of weather conditions on hourly rather than daily intervals. As a result, the mean absolute percentage error of forecasting is reduced significantly, depending on the prediction model used. For the best case, the error in forecasting of a photovoltaic plant generation an hour ahead was 9 %.https://energy.bntu.by/jour/article/view/2287short-term forecastingelectricity generationphotovoltaic plantrenewable energy sourcesmeteorological factorsinsolationsolar radiationneural networksdata clusteringpredictive modeldata preprocessingmachine learningprincipal component analysisadaptive boostinglinear regression |
| spellingShingle | P. V. Matrenin A. I. Khalyasmaa V. V. Gamaley S. A. Eroshenko N. A. Papkova D. A. Sekatski Y. V. Potachits Improving of the Generation Accuracy Forecasting of Photovoltaic Plants Based on <i>k</i>-Means and <i>k</i>-Nearest Neighbors Algorithms Известия высших учебных заведений и энергетических объединенний СНГ: Энергетика short-term forecasting electricity generation photovoltaic plant renewable energy sources meteorological factors insolation solar radiation neural networks data clustering predictive model data preprocessing machine learning principal component analysis adaptive boosting linear regression |
| title | Improving of the Generation Accuracy Forecasting of Photovoltaic Plants Based on <i>k</i>-Means and <i>k</i>-Nearest Neighbors Algorithms |
| title_full | Improving of the Generation Accuracy Forecasting of Photovoltaic Plants Based on <i>k</i>-Means and <i>k</i>-Nearest Neighbors Algorithms |
| title_fullStr | Improving of the Generation Accuracy Forecasting of Photovoltaic Plants Based on <i>k</i>-Means and <i>k</i>-Nearest Neighbors Algorithms |
| title_full_unstemmed | Improving of the Generation Accuracy Forecasting of Photovoltaic Plants Based on <i>k</i>-Means and <i>k</i>-Nearest Neighbors Algorithms |
| title_short | Improving of the Generation Accuracy Forecasting of Photovoltaic Plants Based on <i>k</i>-Means and <i>k</i>-Nearest Neighbors Algorithms |
| title_sort | improving of the generation accuracy forecasting of photovoltaic plants based on i k i means and i k i nearest neighbors algorithms |
| topic | short-term forecasting electricity generation photovoltaic plant renewable energy sources meteorological factors insolation solar radiation neural networks data clustering predictive model data preprocessing machine learning principal component analysis adaptive boosting linear regression |
| url | https://energy.bntu.by/jour/article/view/2287 |
| work_keys_str_mv | AT pvmatrenin improvingofthegenerationaccuracyforecastingofphotovoltaicplantsbasedonikimeansandikinearestneighborsalgorithms AT aikhalyasmaa improvingofthegenerationaccuracyforecastingofphotovoltaicplantsbasedonikimeansandikinearestneighborsalgorithms AT vvgamaley improvingofthegenerationaccuracyforecastingofphotovoltaicplantsbasedonikimeansandikinearestneighborsalgorithms AT saeroshenko improvingofthegenerationaccuracyforecastingofphotovoltaicplantsbasedonikimeansandikinearestneighborsalgorithms AT napapkova improvingofthegenerationaccuracyforecastingofphotovoltaicplantsbasedonikimeansandikinearestneighborsalgorithms AT dasekatski improvingofthegenerationaccuracyforecastingofphotovoltaicplantsbasedonikimeansandikinearestneighborsalgorithms AT yvpotachits improvingofthegenerationaccuracyforecastingofphotovoltaicplantsbasedonikimeansandikinearestneighborsalgorithms |