Sharma Aman, Singh Pradeep Kumar, Makki Emad, Giri Jayant, Sathish T
Department of Mechanical Engineering, GLA University, Mathura, 281406, India.
Department of Mechanical Engineering, College of Engineering and Architecture, Umm Al-Qura University, Makkah 24382, Saudi Arabia.
Heliyon. 2024 Feb 4;10(3):e25800. doi: 10.1016/j.heliyon.2024.e25800. eCollection 2024 Feb 15.
This article explores the use of phase change materials (PCMs) derived from waste, in energy storage systems. It emphasizes the potential of these PCMs in addressing concerns related to fossil fuel usage and environmental impact. This article also highlights the aspects of these PCMs including reduced reliance on renewable resources minimized greenhouse gas emissions and waste reduction. The study also discusses approaches such as integrating nanotechnology to enhance thermal conductivity and utilizing machine learning and deep learning techniques for predicting dynamic behavior. The article provides an overall view of research on biodegradable waste-based PCMs and how they can play a promising role in achieving energy-efficient and sustainable thermal storage systems. However, specific conclusions drawn from the presented results are not explicitly outlined, leaving room, for investigation and exploration in this evolving field. Artificial neural network (ANN) predictive models for thermal energy storage devices perform differently. With a 4% adjusted mean absolute error, the Gaussian radial basis function kernel Support Vector Regression (SVR) model captured heat-related charging and discharging issues. The ANN model predicted finned tube heat and heat flux better than the numerical model. SVM models outperformed ANN and ANFIS in some datasets. Material property predictions favored gradient boosting, but Linear Regression and SVR models performed better, emphasizing application- and dataset-specific model selection. These predictive models provide insights into the complex thermal performance of building structures, aiding in the design and operation of energy-efficient systems. Biodegradable waste-based PCMs' sustainability includes carbon footprint, waste reduction, biodegradability, and circular economy alignment. Nanotechnology, machine learning, and deep learning improve thermal conductivity and prediction. Circular economy principles include waste reduction and carbon footprint reduction. Specific results-based conclusions are not stated. Presenting a comprehensive overview of current research highlights biodegradable waste-based PCMs' potential for energy-efficient and sustainable thermal storage systems.
本文探讨了源自废物的相变材料（PCM）在储能系统中的应用。它强调了这些相变材料在解决与化石燃料使用和环境影响相关问题方面的潜力。本文还突出了这些相变材料的一些方面，包括减少对可再生资源的依赖、将温室气体排放降至最低以及减少废物。该研究还讨论了一些方法，如整合纳米技术以提高热导率，以及利用机器学习和深度学习技术来预测动态行为。本文提供了关于基于可生物降解废物的相变材料的研究概况，以及它们如何能在实现节能和可持续的蓄热系统中发挥重要作用。然而，文中并未明确概述从所呈现的结果中得出的具体结论，为这个不断发展的领域留下了调查和探索的空间。用于热能存储设备的人工神经网络（ANN）预测模型表现各异。高斯径向基函数核支持向量回归（SVR）模型以4%的调整平均绝对误差捕捉到了与热相关的充放电问题。人工神经网络模型在预测翅片管热和热通量方面比数值模型表现更好。在某些数据集中，支持向量机模型的表现优于人工神经网络和自适应神经模糊推理系统（ANFIS）。材料性能预测更倾向于梯度提升，但线性回归和支持向量回归模型表现更好，这强调了要根据应用和数据集进行特定的模型选择。这些预测模型有助于深入了解建筑结构复杂的热性能，辅助节能系统的设计和运行。基于可生物降解废物的相变材料的可持续性包括碳足迹、减少废物、生物可降解性以及与循环经济的契合度。纳米技术、机器学习和深度学习可提高热导率并改进预测。循环经济原则包括减少废物和降低碳足迹。文中未阐述基于具体结果得出的结论。对当前研究进行全面概述突出了基于可生物降解废物的相变材料在节能和可持续蓄热系统方面的潜力。
