<?xml version="1.1" encoding="utf-8"?>
<article xsi:noNamespaceSchemaLocation="http://jats.nlm.nih.gov/publishing/1.1/xsd/JATS-journalpublishing1-mathml3.xsd" dtd-version="1.1" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"><front><journal-meta><journal-id journal-id-type="publisher-id">SSSD</journal-id><journal-title-group><journal-title>Scientific and Social Sustainable Development</journal-title></journal-title-group><issn>3066-8964</issn><eissn>3066-8980</eissn><publisher><publisher-name>Art and Technology</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.61369/SSSD.2026060004</article-id><article-categories><subj-group subj-group-type="heading"><subject>Article</subject></subj-group></article-categories><title>弯曲光伏面板曲率瓶颈破解：基于机器学习的实证优化研究</title><url>https://artdesignp.com/journal/SSSD/2/6/10.61369/SSSD.2026060004</url><author>张吉,王进勇,王琳,钱路宁,王棣威,王彦雄</author><pub-date pub-type="publication-year"><year>2026</year></pub-date><volume>2</volume><issue>6</issue><history><date date-type="pub"><published-time>2026-03-28</published-time></date></history><abstract>本论文提出一种机理化金属弯折太阳能板设计优化的技术路线。不同于目前的技术使用仿真数据本论文首次提出使用受控条件下产生的数据训练模型，使用线性模型并结合改进的进化算法对模型预测神经元网络进行优化，进而用于预测曲面金属太阳能面板能量收集的技术路线。通过太阳能板模拟测试台专门设计的曲率、金属材料、温度、辐射、角度等参数测量实验研究曲率、金属材料、辐射、角度对能量收集效率的影响。测试结果表明曲率影响显著且存在非线性，辐射、角度对能量收集效率的影响随着角度的衰减非常明显，使用优化模型进行预测，其决定系数（R2）相比于传统基线模型提高了约 6%，同时测试均方误差（MSE）大幅降低了约 48%，平均绝对误差（MAE）降低了约 26%。，可视化模型预测结果图形显示了各参数对模型影响的显著度，及其模型预测结果准确度。测试结果对太阳电池的应用技术发展具有直接的指导意义，也可用于对实际应用环境中其他应用的参数显著度和准确度的分析，对可再生能源应用有广泛的实际意义，实验结果显示，采用改进进化模型在存在噪声的测试环境中仍具有高准确度的预测能力。</abstract><keywords>弯曲光伏面板,机器学习优化,实验验证,效率预测</keywords></article-meta></front><body/><back><ref-list><ref id="B1" content-type="article"><label>1</label><element-citation publication-type="journal"><p>[1]Y. Liu, H. Wang, and L. Zhang, &amp;ldquo;Bending strength analysis of ultra-thin glass encapsulated pv modules,&amp;rdquo; Solar Energy Materials and Solar Cells, vol. 250, pp. 112&amp;ndash;125, 2023.[2] 何东坡, 万超, 杨帆, 陈杨, and 彭宇翔, &amp;ldquo;基于机器学习的光伏功率预测模拟方法研究,&amp;rdquo; 中低纬山地气象, vol. 47, no. 6, 2023.[3]X. Chen and M. Li, &amp;ldquo;Machine learning guided material synthesis for flexible pv,&amp;rdquo; Advanced Energy Materials, vol. 14, no. 2, p. 2300123, 2024.[4]A. Manasrah, Y. Jaradat, M. Masoud, M. Alia, K. Suwais, and P. Bevilacqua, &amp;ldquo;Flat vs. curved: Machine learning classification of flexible pv panel geometries,&amp;rdquo; Energies, vol. 18, no. 13, p. 3529, 2025.[5] 郑玉杰, 梁鑫斌, 张起, 孙文博, 施童超, 杜鹃, and 孙宽, &amp;ldquo;基于分子指纹及机器学习回归模型的有机光伏材料效率预测,&amp;rdquo; 材料导报, vol. 35, no. 8, pp. 8207&amp;ndash;8212, 2021.[6]S. Al-Dahidi, M. Alrbai, H. Alahmer, B. Rinchi, and A. Alahmer, &amp;ldquo;En- hancing solar photovoltaic energy production prediction using diverse machine learning models tuned with the chimp optimization algorithm,&amp;rdquo; Scientific Reports, vol. 14, p. 18583, 2024.[7]X. Huang, Q. Li, Y. Tai, Z. Chen, J. Liu, J. Shi, and W. Liu, &amp;ldquo;Time series forecasting for hourly photovoltaic power using conditional generative adversarial network and Bi-LSTM,&amp;rdquo; Energy, vol. 246, p. 123363, 2022.</p><pub-id pub-id-type="doi"/></element-citation></ref></ref-list></back></article>
