Basem Ali, Opakhai Serikzhan, Elbarbary Zakaria Mohamed Salem, Atamurotov Farruh, Benti Natei Ermias
Air Conditioning Engineering Department, Faculty of Engineering, Warith Al-Anbiyaa University, Karbala, Iraq.
Faculty of Physics and Technical Sciences, L.N. Gumilyov Eurasian National University, 010000, Astana, Kazakhstan.
Sci Rep. 2025 Jan 2;15(1):259. doi: 10.1038/s41598-024-70682-2.
This study presents an in-depth analysis and evaluation of the performance of a standard 200 W solar cell, focusing on the energy and exergy aspects. A significant research gap exists in the comprehensive integration of numerical models with advanced machine-learning approaches, specifically emotional artificial neural networks (EANN), to simulate and optimize the electrical characteristics and efficiency of solar panels. To address this gap, a numerical model alongside a novel EANN was employed to simulate the system's electrical characteristics, including open-circuit voltage, short-circuit current, system resistances, maximum power point characteristics, and characteristic curves. Mathematical equations for calculating efficiency levels under varying operational conditions were developed. The system's operational and electrical parameters, alongside environmental conditions such as solar radiation, wind speed, and ambient temperature, were empirically observed and documented over a day. A comparative analysis was conducted to validate the model by comparing its results with manufacturer data and experimental observations. During the trial from 7:00 to 17:00, energy efficiency varied from 10.34 to 14.00%, averaging 13.6%, while exergy efficiency ranged from 13.57 to 16.41%, with an average of 15.70%. The results from the EANN model indicate that the proposed method for forecasting energy, exergy, and power is feasible, offering a significant reduction in computational expense compared to traditional numerical models. The integration of numerical modeling with EANN enhances simulation accuracy and the developed equations enable real-time efficiency calculations. Empirical validation under varying environmental conditions improves predictive capabilities for solar panel performance. Additionally, operational efficiency assessments aid in better design and deployment of solar energy systems, and computational costs for large-scale solar energy simulations are reduced.
本研究对标准200瓦太阳能电池的性能进行了深入分析和评估，重点关注能量和㶲方面。在将数值模型与先进的机器学习方法（特别是情感人工神经网络（EANN））进行全面整合以模拟和优化太阳能电池板的电气特性及效率方面，存在重大研究空白。为弥补这一空白，采用了一个数值模型以及一种新颖的EANN来模拟系统的电气特性，包括开路电压、短路电流、系统电阻、最大功率点特性和特性曲线。推导了在不同运行条件下计算效率水平的数学方程。在一天的时间里，对系统的运行和电气参数以及诸如太阳辐射、风速和环境温度等环境条件进行了实证观测和记录。通过将模型结果与制造商数据及实验观测结果进行比较，进行了对比分析以验证该模型。在7:00至17:00的试验期间，能量效率在10.34%至14.00%之间变化，平均为13.6%，而㶲效率在13.57%至16.41%之间，平均为15.70%。EANN模型的结果表明，所提出的预测能量、㶲和功率的方法是可行的，与传统数值模型相比，计算成本显著降低。数值建模与EANN的整合提高了模拟精度，所推导的方程能够进行实时效率计算。在不同环境条件下的实证验证提高了对太阳能电池板性能的预测能力。此外，运行效率评估有助于更好地设计和部署太阳能系统，并且降低了大规模太阳能模拟的计算成本。
