Shakor Zaidoon M, Tayib Yaseen M, AbdulRazak Adnan A, Shnain Zainab Y, Al-Shafei Emad
Chemical Engineering Department, University of Technology, 10066 Baghdad, Iraq.
Fuel and Energy Techniques Engineering Department, Al-Huda University College, 31001 Ramadi, Iraq.
ACS Omega. 2025 Aug 11;10(33):36750-36770. doi: 10.1021/acsomega.5c02250. eCollection 2025 Aug 26.
This review emphasized the role of mathematical models and correlations in thermogravimetric analysis to evaluate the thermal stability of various materials, including biomass, polymers, recycled plastics, and solid fuels of carbon and biomass material. Numerous thermogravimetric analysis kinetic models are driven, and they are broadly divided into model-free and model-based categories. Integral models have proven to be more effective for fitting, particularly for materials with wide decomposition temperature ranges in biomass material and mixed recycled plastic waste. The th order model showed superior predictive accuracy compared with the first-order model, particularly for solid biomass, highlighting the significance of model selection. Traditional thermogravimetric analysis mathematical models are limited in accounting for mass loss as a function of all effective variables. In contrast, artificial neural networks (ANNs) efficiently represent and incorporate these variables, marking a significant advancement in predicting thermogravimetric analysis kinetics. ANNs provide powerful tools for managing complex analysis data, enabling robust predictions and deeper insights into material thermal behavior.
本综述强调了数学模型和相关性在热重分析中对评估各种材料热稳定性的作用，这些材料包括生物质、聚合物、再生塑料以及碳和生物质材料的固体燃料。众多热重分析动力学模型被驱动，它们大致分为无模型和基于模型的类别。积分模型已被证明在拟合方面更有效，特别是对于生物质材料和混合再生塑料废物中具有宽分解温度范围的材料。与一阶模型相比，n阶模型显示出更高的预测准确性，特别是对于固体生物质，突出了模型选择的重要性。传统的热重分析数学模型在考虑质量损失作为所有有效变量的函数方面存在局限性。相比之下，人工神经网络（ANN）能够有效地表示和纳入这些变量，标志着在预测热重分析动力学方面取得了重大进展。人工神经网络为管理复杂的分析数据提供了强大的工具，能够进行可靠的预测并深入了解材料的热行为。
