Performance-Based Concrete as a Route for CO2 reduction

混凝土是当今建筑业中使用最广泛的材料。据估计，90%的混凝土温室气体排放可以归因于水泥生产。水泥生产约占全球二氧化碳排放量的7%。因此，快速过渡到低水泥混凝土势在必行。提高混凝土生产可持续性的方法之一是使用辅助胶凝材料(SCM)来取代混凝土混合料中的水泥。SCM广泛使用的最大障碍之一是，准确预测所得到的混凝土的性能是具有挑战性的。这项研究项目旨在通过新的测试方法，如电子显微镜结合EDS微量化学分析，更好地了解选定的SCM的物理和化学特性，并建立这些特性与所生产混凝土的性能之间的联系。此外，将开发和训练一个机器学习模型，以预测混合混凝土的强度和耐久性，并确定最大限度地替换水泥的最佳SCM特性。机器学习模型的结果还将被用来创建经验公式，这些经验公式可用于根据SCM特性和配合比组成预测所生产混凝土的性能。这项研究将使混凝土设计者在使用混合混凝土搅拌时更有信心。

Concrete is the most widely used material in today's construction industry. It is estimated that 90% of concrete'sgreenhouse gas emissions can be attributed to cement production. Cement production is the source of approximately 7%of the world's CO2 emissions. Therefore, a rapid transition to lower-cement concrete is imperative. One of the methodsfor improving the sustainability of concrete production is the use of Supplementary Cementitious Materials (SCMs) toreplace cement in concrete mixtures. One of the most significant barriers to the broader use of SCMs is that it ischallenging to accurately predict the resulting concrete's properties. This research project will aim to better understandthe physical and chemical characteristics of selected SCMs through novel testing methods like electron microscopycombined with EDS microchemical analysis and to establish links between those characteristics and the performance ofthe produced concrete. In addition, a machine-learning model will be developed and trained to predict the strength anddurability properties of blended concrete and to identify the optimum SCM characteristics for maximising cementreplacement. The results of the machine-learning model will also be used to create empirical formulas that can be usedto predict properties of the produced concrete based on SCM characteristics as well as mix composition. This researchwill give concrete designers more confidence in using blended concrete mixes.