Albà Carlos G, Alkhatib Ismail I I, Vega Lourdes F, Llovell Fèlix
Department of Chemical Engineering, ETSEQ, Universitat Rovira i Virgili (URV), Campus Sescelades, Av. Països Catalans 26, 43007 Tarragona, Spain.
Research and Innovation Center on CO2 and Hydrogen (RICH Center) and Department of Chemical and Petroleum Engineering, Khalifa University, PO Box 127788 Abu Dhabi, United Arab Emirates.
ACS Sustain Chem Eng. 2024 Jul 23;12(31):11561-11577. doi: 10.1021/acssuschemeng.4c01961. eCollection 2024 Aug 5.
As the EU's mandates to phase out high-GWP refrigerants come into effect, the refrigeration industry is facing a new, unexpected reality: the introduction of more flammable yet environmentally compliant alternatives. This paradigm shift amplifies the need for a rapid, reliable screening methodology to assess the propensity for flammability of emerging fourth generation blends, offering a pragmatic alternative to laborious and time-intensive traditional experimental assessments. In this study, an artificial neural network (ANN) is meticulously constructed, evaluated, and validated to address this emerging challenge by predicting the normalized flammability index (NFI) for an extensive array of pure, binary, and ternary mixtures, reflecting a substantial diversity of compounds like CO, hydrofluorocarbons (HFCs), hydrofluoroolefins (HFOs), six saturated hydrocarbons (sHCs), hydroolefins (HOs), and others. The optimal configuration ([61 (I) × 14 (HL1) × 24 (HL2) × 1 (O)]) demonstrated a profound fit to the data, with metrics like of 0.999, root-mean-square error (RMSE) of 0.1735, average absolute relative deviation (AARD)% of 0.8091, and SD of ±0.0434. Exhaustive assessments were conducted to ensure the most efficient architecture without compromising the accuracy. Additionally, the analysis of the standardized residuals (SDR) and applicability domain (AD) exhibited fine control and consistency over the data points. External validation using quaternary mixtures further attested to the model's adaptability and predictive capability. The exploration into the relative contribution of descriptors led to the identification of 23 significant sigma descriptors derived from conductor-like screening model (COSMO), responsible for 90.98% of the total contribution, revealing potential avenues for model simplification without a substantial loss in predictive power. Moreover, the model successfully predicted the behavior of prospective industry-relevant mixtures, reinforcing its reliability and opening the door to experimentation with untested blends. The results collectively manifest the developed ANN's efficiency, robustness, and adaptability in modeling flammability, catering to the demands of industry standards, environmental concerns, and safety requirements.
随着欧盟逐步淘汰高全球变暖潜能值制冷剂的指令生效，制冷行业正面临一个全新的、意想不到的现实：引入了更多易燃但符合环境要求的替代品。这种范式转变凸显了对一种快速、可靠的筛选方法的需求，以评估新兴第四代混合物的易燃倾向，为繁琐且耗时的传统实验评估提供了一种实用的替代方法。在本研究中，精心构建、评估并验证了一个人工神经网络（ANN），通过预测大量纯物质、二元混合物和三元混合物的归一化燃烧指数（NFI）来应对这一新兴挑战，这些混合物反映了多种化合物，如一氧化碳、氢氟烃（HFCs）、氢氟烯烃（HFOs）、六种饱和烃（sHCs）、氢烯烃（HOs）等。最优配置（[61（输入层）×14（隐藏层1）×24（隐藏层2）×1（输出层）]）显示出与数据的高度拟合，相关指标如R²为0.999、均方根误差（RMSE）为0.1735、平均绝对相对偏差（AARD）%为0.8091以及标准差为±0.0434。进行了详尽的评估以确保在不影响准确性的前提下获得最有效的架构。此外，对标准化残差（SDR）和适用域（AD）的分析显示出对数据点的良好控制和一致性。使用四元混合物进行的外部验证进一步证明了该模型的适应性和预测能力。对描述符相对贡献的探索导致识别出23个源自导体类筛选模型（COSMO）的重要西格玛描述符，它们占总贡献的90.98%，揭示了在不显著损失预测能力的情况下简化模型的潜在途径。此外，该模型成功预测了与行业相关的预期混合物的行为，增强了其可靠性，并为未经测试的混合物的实验打开了大门。这些结果共同表明所开发的人工神经网络在燃烧建模方面的效率、稳健性和适应性，满足了行业标准、环境关注和安全要求。
