当前位置: X-MOL 学术Build. Environ. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A review on cooling performance enhancement for phase change materials integrated systems—flexible design and smart control with machine learning applications
Building and Environment ( IF 7.1 ) Pub Date : 2020-05-01 , DOI: 10.1016/j.buildenv.2020.106786
Yuekuan Zhou , Siqian Zheng , Guoqiang Zhang

Abstract Climate-adaptive design, smart control, latent thermal storages, multi-dimensional uncertainty analysis, and multi-objective optimisations are effective solutions for cooling performance enhancement of buildings through integrated techniques, such as hybrid ventilations, nocturnal sky radiation, radiative cooling and active PV cooling for the self-consumption. However, there is no systematic and in-depth analysis on this topic in the academia. In this study, a state-of-the-art review on novel PCMs based strategies to reduce cooling load of buildings has been presented. The investigated strategies include the structural configuration, systematic control and the multi-criteria for assessment. The roles of ventilations, radiative cooling and the underlying heat transfer mechanism have been characterized for the in-depth understanding. In order to realise the multivariable optimal design and robust operations under multi-level scenario uncertainties, parametric and uncertainty analysis, single- and multi-objective optimisations have been comprehensively reviewed, together with technical challenges for each solution. Research results show that, integrated passive and active systems with flexible transitions on operating modes are full of prospects for the multi-criteria performance improvement. Trade-off solutions along the multi-objective Pareto frontier are multi-diversified, dependent on the adopted approach and the studied scenario. Furthermore, machine learning methods are promising for the thermal and energy performances improvement, through the surrogate model development, the model predictive control and the optimisation function. Future studies and prospects have been demonstrated as avenues for future research. This study presents a systematic overview on novel PCMs based strategies, together with the application of machine-learning methods for cooling performance enhancement, which are critical for the promotion of novel PCMs based cooling strategies in buildings.

中文翻译:

相变材料集成系统冷却性能增强综述——机器学习应用的灵活设计和智能控制

摘要 气候适应性设计、智能控制、潜热蓄热、多维不确定性分析和多目标优化是通过混合通风、夜间天空辐射、辐射冷却和主动冷却等综合技术提高建筑物冷却性能的有效解决方案。光伏冷却自用。然而,学术界并没有对这个话题进行系统深入的分析。在这项研究中,对基于新型 PCM 的降低建筑物冷负荷的策略进行了最先进的审查。研究的策略包括结构配置、系统控制和多标准评估。通风、辐射冷却和潜在的传热机制的作用已被表征,以便深入了解。为了实现多级场景不确定性下的多变量优化设计和稳健运行,对参数和不确定性分析、单目标和多目标优化以及每种解决方案的技术挑战进行了全面审查。研究结果表明,具有灵活转换工作模式的被动和主动集成系统充满了多准则性能提升的前景。多目标帕累托边界上的权衡解决方案是多样化的,取决于采用的方法和研究的场景。此外,通过代理模型开发、模型预测控制和优化功能,机器学习方法有望改善热和能源性能。未来的研究和前景已被证明是未来研究的途径。本研究系统概述了基于新型 PCM 的策略,以及机器学习方法在冷却性能增强方面的应用,这对于在建筑物中推广基于新型 PCM 的冷却策略至关重要。
更新日期:2020-05-01
down
wechat
bug