Pilegis Martins, Gardner Diane, Lark Robert
Cardiff School of Engineering, Cardiff University, Queens Buildings, Cardiff CF24 3AA, UK.
Materials (Basel). 2016 Jun 2;9(6):440. doi: 10.3390/ma9060440.
Manufactured sand differs from natural sea and river dredged sand in its physical and mineralogical properties. These can be both beneficial and detrimental to the fresh and hardened properties of concrete. This paper presents the results of a laboratory study in which manufactured sand produced in an industry sized crushing plant was characterised with respect to its physical and mineralogical properties. The influence of these characteristics on concrete workability and strength, when manufactured sand completely replaced natural sand in concrete, was investigated and modelled using artificial neural networks (ANN). The results show that the manufactured sand concrete made in this study generally requires a higher water/cement (w/c) ratio for workability equal to that of natural sand concrete due to the higher angularity of the manufactured sand particles. Water reducing admixtures can be used to compensate for this if the manufactured sand does not contain clay particles. At the same w/c ratio, the compressive and flexural strength of manufactured sand concrete exceeds that of natural sand concrete. ANN proved a valuable and reliable method of predicting concrete strength and workability based on the properties of the fine aggregate (FA) and the concrete mix composition.
机制砂在物理和矿物学性质上与天然海砂和河砂不同。这些性质对混凝土的新拌性能和硬化性能可能既有益处也有不利影响。本文介绍了一项实验室研究的结果,该研究对一家工业规模破碎厂生产的机制砂的物理和矿物学性质进行了表征。当机制砂完全替代混凝土中的天然砂时,研究了这些特性对混凝土工作性和强度的影响,并使用人工神经网络(ANN)进行建模。结果表明,由于机制砂颗粒的棱角性较高,本研究中制备的机制砂混凝土通常需要更高的水灰比(w/c)才能达到与天然砂混凝土相同的工作性。如果机制砂不含粘土颗粒,则可使用减水剂来弥补这一不足。在相同的水灰比下,机制砂混凝土的抗压强度和抗折强度超过天然砂混凝土。人工神经网络证明是一种基于细集料(FA)性能和混凝土配合比组成预测混凝土强度和工作性的有价值且可靠的方法。