Chang Yi-Hsun, Zhang You-Lun, Cheng Cheng-Hao, Wu Shu-Han, Li Cheng-Han, Liao Su-Yu, Tseng Zi-Chun, Lin Ming-Yi, Huang Chun-Ying
Department of Applied Materials and Optoelectronic Engineering, National Chi Nan University, Nantou 54561, Taiwan.
Department of Electrical Engineering, National United University, Miaoli 360302, Taiwan.
Nanomaterials (Basel). 2025 Jul 17;15(14):1112. doi: 10.3390/nano15141112.
Accurate identification of active-layer compositions in organic photovoltaic (OPV) devices often relies on invasive techniques such as electrical measurements or material extraction, which risk damaging the device. In this study, we propose a non-invasive classification approach based on simulated full-device absorption spectra. To account for fabrication-related variability, the active-layer thickness varied by over ±15% around the optimal value, creating a realistic and diverse training dataset. A multilayer perceptron (MLP) neural network was applied with various activation functions, optimization algorithms, and data split ratios. The optimized model achieved classification accuracies exceeding 99% on both training and testing sets, with minimal sensitivity to random initialization or data partitioning. These results demonstrate the potential of applying deep learning to spectral data for reliable, non-destructive OPV composition classification, paving the way for integration into automated manufacturing diagnostics and quality control workflows.
准确识别有机光伏(OPV)器件中的活性层成分通常依赖于侵入性技术,如电学测量或材料提取,而这些技术存在损坏器件的风险。在本研究中,我们提出了一种基于模拟全器件吸收光谱的非侵入性分类方法。为了考虑与制造相关的变异性,活性层厚度在最佳值周围变化超过±15%,从而创建了一个真实且多样的训练数据集。应用了具有各种激活函数、优化算法和数据分割比例的多层感知器(MLP)神经网络。优化后的模型在训练集和测试集上均实现了超过99%的分类准确率,对随机初始化或数据分区的敏感性极小。这些结果证明了将深度学习应用于光谱数据以进行可靠的、非破坏性的OPV成分分类的潜力,为集成到自动化制造诊断和质量控制工作流程铺平了道路。