The use of open meteorological information to predict the energy output of a photovoltaic plant for the month ahead. Experimental research

To improve the reliability of power supply in autonomous energy systems, it is necessary to expand the time range for regulation. This is required to create an adequate reserve of resources capable of compensating for peak loads and effectively managing backup capacity. Currently, there are two key methods for solving this problem. The first method is based on mathematical modeling of the expected level of solar insolation by analyzing historical data on actual levels of insolation in the past. The second method uses current insolation forecasts developed based on global and regional climate research and forecasts. In addition, improving climate information collection systems using ground-based and space observation tools increases the accuracy of weather forecasts, making them applicable to practical tasks in the energy sector.

This work is devoted to studying ways to improve the accuracy of medium-term forecasting of energy production by solar power plants using meteorological data and cluster analysis of meteorological conditions. The main goal of the study is to test the effectiveness of the indirect forecasting method, which is based on open meteorological data such as solar insolation and cloud cover.

The article discusses the use of global climate models to predict energy output from solar power plants. A methodology for calculating incoming solar radiation onto a surface inclined at an angle to the horizon is presented.

The results of experiments conducted at the FSNEI-65 solar power plant located on the south facade of the educational building of Ural Federal University in Yekaterinburg are presented. During the experiment, predicted values of solar insolation and cloudiness obtained using the Integrated Forecast System (IFS) climate model were compared with the actual electricity generation by the solar power plant over an average period of time. Graphs of the dependence between real and predicted power were constructed, and forecast deviations were calculated.

The results show high accuracy of climate models in medium-term forecasting and their potential for reducing costs associated with compensating for underproduction of energy by solar power plants.

1. Vissarionov V. I., Deryugina G. V., Kuznetsova V. A., Malinin N. K. Solar energy: a textbook for universities. – M.: MPEI Publishing House, 2008. – 320 p.

2. Denisov K. S. Solving the problem of integrated energy supply of an autonomous consumer in order to reduce economic costs / K. S. Denisov, V. I. Velkin, A. N. Tyrsin // Bulletin of SUSU. Series «Energy». – 2019. – Vol. 19, No. 3. – P. 84-92.

3. Beloborodov I. V., Shcheklein S. E., Nemikhin Yu. E. Study of the operation of a network photovoltaic station // «Energy and resource saving. Energy supply. Alternative and Renewable Energy Sources». – 2015.

4. Bannykh S. M., Shcheklein S. E. Verification of the results of field studies and forecast values for solar energy production // Alternative Energy and Ecology (ISJAEE). – 2024. – No. 04 (421). – Pр. 12-24.

5. Kiseleva S. V., Lisitskaya N. V., Frid S. E. Forecasting the production of solar stations and photovoltaic installations: basic approaches and performance // Alternative Energy and Ecology (ISJAEE). – 2020. – No. 07-18 (330-341). – P. 24-43.

6. Windy.com // [Electronic resource]. Access mode: windy.com / (date of access 11/18/2024).

7. DELTA // [Electronic resource]. Access mode: https://delta-solar.ru/product/solnechnye-moduli/delta-nxt/delta-nxt-500-66-2-m10-hc/ (date of access 10.09.2024).

8. Snegirev D. A. [et al.] Features of forecasting electricity generation by solar power plants / D. A. Snegirev [et al.] // Electric power industry through the eyes of youth – 2017: scientific. t. VIII Int. youth scientific and technical. conf. Samara. – 2017. – Vol. 3. – P. 139-142.

9. Snegirev D. A., Eroshenko S. A., Valiev R. T., Khalyasmaa A. I. Possibilities of algorithmic implementation of short-term forecasting of solar power plant output // International Scientific Conference on Control Problems in Technical Systems. – 2017. – Vol. 1. – P. 236-239.

10. Eroshenko S. A., Khalyasmaa A. I., Snegirev D. A. Approaches to forecasting the solar radiation flux density for SES // Energoexpert. – 2017. – No. 5. – P. 28-31.

11. Tolstykh M. A., Frolov A. V. Some modern problems of numerical weather forecasting // Bulletin of the Russian Academy of Sciences. FAO. – 2005. – Vol. 41. – No. 3. – P. 315-327.

12. Tyunkov D. A., Sapilova A. A., Gritsai A. S., Alekseenko D. A., Khamitov R. N. Methods of short-term forecasting of electric energy generation by solar power plants and their classification // «Electrotechnical systems and complexes». – 2020. – No. 3 (48). – P. 4-10. DOI: 10.18503/2311-8318-2020-3(48)-4-10.

13. Gaak A. V. Forecasting solar energy for consumers of electric energy // Current issues in energy. Proceedings of the All-Russian scientific and practical conference with international participation. – Omsk. – 2015. – P. 138-143.

14. Vorotyntsev D. V., Tyagunov M. G. Forecast of electricity generation by photovoltaic power plants (day ahead) using machine learning // Bulletin of the Moscow Power Engineering Institute. Bulletin of MPEI. – 2018. – No. 4. – P. 53-57.

15. Anufriev O. V., Vorotyntsev D. V., Krapivko P. V., Tyagunov M. G. Forecast of hourly generation of PV power plants for a day ahead using machine learning // In the collection: Fyodorovskie readings – 2017. XLVII International scientific and practical conference with elements of a scientific school. – 2017. – P. 305-311.

16. Zhigalkin A. V., Starchenko D. V. Analysis of problems of forecasting power generation from alternative energy sources // Electric power industry through the eyes of youth-2019. Proceedings of the jubilee X International scientific and technical conference. – V. 1. – 2019. – P. 162-165.

17. Kovalev V. Z., Khamitov R. N., Tyunkov D. A. Spectral analysis of retrospective data on the production of electric energy by solar power plants // Scientific works of KubSTU. – No. 3. – 2019.

18. Gritsay A. S. Hybrid method for short-term forecasting of electric energy consumption for an energy supply company taking into account meteorological factors: diss. Cand. Tech. Sciences. – Omsk. – 2017. – 153 p.

19. Engel N. E., Engel E. A. Comparison of modern intelligent methods for forecasting the production of electric energy by a solar power plant // Engineering technologies: traditions, innovations, development vectors: materials of the VIII All-Russian scientific and practical conference with international participation (Abakan, November 14-16, 2022). – 2022. – P. 61-62.

20. Makambaev T. Kh., Velkin V. I. Forecasting energy production by a power plant based on solar cells using machine learning // WORLD SCIENCE: PROBLEMS AND INNOVATIONS: collection of articles from the XXXI International scientific and practical conference: in 4 parts, Penza, May 30, 2019. Part 2. – Penza: «Science and Education» (IP Gulyaev G. Yu.), 2019. – P. 36-38.