ENERGY ACCUMULATION FROM SOLAR PANELS AND FORECASTING MODELS FOR ENERGY PRODUCTION USING ARTIFICIAL INTELLIGENCE

Аннотация

This paper explores the process of energy accumulation from solar panels and the possibilities of analyzing energy production through forecasting models using artificial intelligence. The efficiency of photovoltaic systems is influenced by various external factors, such as solar radiation, temperature, dust levels, and seasonal variations. Considering these factors, the feasibility of predicting energy output using artificial neural networks, regression models, and machine learning algorithms is examined. Furthermore, the efficiency of modern energy storage technologies like lithium-ion batteries and supercapacitors is discussed. Research shows that implementing AI-based forecast models helps to properly manage energy reserves and reduce grid load.

International Journal of Political Sciences and Economics
Тип источника: Журналы
Годы охвата с 2023
inLibrary
Google Scholar
 
Выпуск:
Отрасль знаний
  • Tashkent State Technical University
  • National Research Institute of Renewable Energy Sources under the Ministry of Energy
  • National Research Institute of Renewable Energy Sources under the Ministry of Energy
CC BY f
203-205
0

Скачивания

Данные скачивания пока недоступны.
Поделиться
Тоиров O., Жураев E., & Xожиев D. (2025). ENERGY ACCUMULATION FROM SOLAR PANELS AND FORECASTING MODELS FOR ENERGY PRODUCTION USING ARTIFICIAL INTELLIGENCE. Международный журнал политических наук и экономики, 1(4), 203–205. извлечено от https://inlibrary.uz/index.php/ijpse/article/view/125152
0
Цитаты
Crossref
Сrossref
Scopus
Scopus
International Journal of Political Sciences and Economics

Аннотация

This paper explores the process of energy accumulation from solar panels and the possibilities of analyzing energy production through forecasting models using artificial intelligence. The efficiency of photovoltaic systems is influenced by various external factors, such as solar radiation, temperature, dust levels, and seasonal variations. Considering these factors, the feasibility of predicting energy output using artificial neural networks, regression models, and machine learning algorithms is examined. Furthermore, the efficiency of modern energy storage technologies like lithium-ion batteries and supercapacitors is discussed. Research shows that implementing AI-based forecast models helps to properly manage energy reserves and reduce grid load.


background image

Volume 4, issue 6, 2025

203

ENERGY ACCUMULATION FROM SOLAR PANELS AND FORECASTING MODELS

FOR ENERGY PRODUCTION USING ARTIFICIAL INTELLIGENCE

O.Z. Toirov¹, E.T. Juraev², D.O. Hojiev³

¹ ³Tashkent State Technical University, Tashkent, Uzbekistan

²National Research Institute of Renewable Energy Sources under the Ministry of Energy,

Tashkent, Uzbekistan

Abstract:

This paper explores the process of energy accumulation from solar panels and the

possibilities of analyzing energy production through forecasting models using artificial

intelligence. The efficiency of photovoltaic systems is influenced by various external factors,

such as solar radiation, temperature, dust levels, and seasonal variations. Considering these

factors, the feasibility of predicting energy output using artificial neural networks, regression

models, and machine learning algorithms is examined. Furthermore, the efficiency of modern

energy storage technologies like lithium-ion batteries and supercapacitors is discussed. Research

shows that implementing AI-based forecast models helps to properly manage energy reserves

and reduce grid load.

Keywords:

Solar panels, energy accumulation, artificial intelligence, forecast models, energy

efficiency

Introduction

In recent years, the demand for solar energy has significantly increased. Global warming, the

need to reduce carbon emissions, and international commitments to implement renewable energy

sources have considerably enhanced the role of solar panels. Photovoltaic (PV) panels directly

convert solar radiation into electrical energy. However, these systems are intermittent—meaning

they only generate energy during daylight.

Therefore, it is essential to store the generated energy for later use, especially during the night or

on cloudy days. At the same time, predicting energy production, i.e., forecasting the amount of

energy based on future solar radiation, plays a crucial role in balancing the power grid load.

Modern approaches rely on artificial intelligence (AI) technologies, offering advanced methods

for managing solar energy production and storage. This paper is dedicated to analyzing these

technologies.

Main Content

1. Operation Principle of Solar Panels

Solar panels consist of photovoltaic cells that generate electric current through the photoelectric

effect when exposed to sunlight. The amount of energy produced depends on solar irradiance,

ambient temperature, panel orientation, dust accumulation, and other external factors.

Main components of a solar panel system include:

Photovoltaic cells (usually silicon-based)

Glass protective layer

Inverter (converts DC to AC)

Monitoring system

Due to the intermittency of solar power, energy storage becomes a necessity.

2. Energy Accumulation Systems

Energy accumulation refers to storing produced energy for later use. The most widely used

storage technologies for solar energy include:


background image

Volume 4, issue 6, 2025

204

Lithium-ion

batteries

Known for their high energy density, long service life, and rechargeability. Widely used in small

to medium-scale solar systems. However, they are temperature-sensitive and expensive.

Supercapacitors

Capable of storing and discharging large amounts of energy quickly. Suitable for voltage

stabilization and frequency control but not long-term storage.

Pumped

hydro

storage

systems

Used in large-scale power plants. Excess energy pumps water to a higher elevation, which is

later released to generate electricity through turbines.

If not properly managed, energy storage systems can result in losses or failure. Hence, predictive

models are critical.

3. Forecasting Solar Energy Production Using AI

AI, particularly machine learning methods, is widely used to forecast solar energy production.

These models rely on data such as:

Historical solar irradiance

Temperature and humidity

Cloud cover

Panel tilt angle

Panel performance metrics

Commonly used algorithms include:

Artificial Neural Networks (ANN) – Predict energy generation based on radiation and

other variables

Random Forest – Tree-based ensemble algorithm evaluating multiple variables

Support Vector Machine (SVM) – A high-accuracy classification model

Long Short-Term Memory (LSTM) – Neural networks suitable for time-series data

Studies (Yang et al., 2021) show that ANN models can forecast daily solar energy production

with over 90% accuracy.

4. Model Outputs and Practical Applications

AI-based forecast models enable:

Prevention of battery overcharging

Timely connection to the energy grid

Accurate estimation of stored energy

Efficient system control with minimal energy loss

For example, SmartGrid-integrated solar energy systems with AI support enable real-time

monitoring and optimization. As a result, overall energy efficiency improves by 15–20%.

Conclusion

Efficient storage and utilization of energy from solar panels are critical for the stable operation of

energy systems. With energy storage technologies, solar power can be used even during cloudy

days or nighttime. However, effective operation of these systems requires accurate forecasting of

energy production.

AI-based forecast models allow real-time monitoring and control of production, storage, and grid

load, enhancing both energy efficiency and system reliability.

Future research should focus on developing new AI models, integrating with IoT devices, and

evaluating economic efficiency.

References:

1.

Yang, L., Zhao, X., & Liu, C. (2021). Solar Power Forecasting Using Artificial Neural

Networks with Weather Data. Energy Reports, 7, 3147–3155.


background image

Volume 4, issue 6, 2025

205

2.

Chen, J., Wang, H., & Li, Z. (2020). Energy Storage Systems for Solar Applications: A

Review. Renewable and Sustainable Energy Reviews, 134, 110275.

3.

Kumar, R., & Singh, M. (2019). AI-based Solar Energy Management in Smart Grids.

Journal of Clean Energy Technologies, 7(4), 204–210.

4.

IEA (2022). Global Photovoltaic Forecasting Report. International Energy Agency.

5.

Zhang, Y., & Lee, D. (2022). Smart Energy Systems with Solar Panels and AI-based

Control. IEEE Access, 10, 33621–33630.

Библиографические ссылки

Yang, L., Zhao, X., & Liu, C. (2021). Solar Power Forecasting Using Artificial Neural Networks with Weather Data. Energy Reports, 7, 3147–3155.

Chen, J., Wang, H., & Li, Z. (2020). Energy Storage Systems for Solar Applications: A Review. Renewable and Sustainable Energy Reviews, 134, 110275.

Kumar, R., & Singh, M. (2019). AI-based Solar Energy Management in Smart Grids. Journal of Clean Energy Technologies, 7(4), 204–210.

IEA (2022). Global Photovoltaic Forecasting Report. International Energy Agency.

Zhang, Y., & Lee, D. (2022). Smart Energy Systems with Solar Panels and AI-based Control. IEEE Access, 10, 33621–33630.