Size distribution of pores and their geometric analysis in red mud-based autoclaved aerated concrete (AAC) using regression neural network and elastic mechanics. (12th December 2022)

• Record Type: Journal Article
• Authors: Dong, Minghao; Ma, Rui; Sun, Guangcheng; Pan, Chenyu; Zhan, Shulin; Qian, Xiaoqian; Chen, Ruohong; Ruan, Shaoqin
• Is Part Of: Construction & building materials. Volume 359(2022)
• Publication Date: 2022-12-12
• Subjects: Autoclaved aerated concrete -- Thermal conductivity -- Pore structure analysis -- Neural network regression analysis -- Red mud; Building materials -- Periodicals; 624.18
• DOI: 10.1016/j.conbuildmat.2022.129420
• Languages: English
• ISSNs: 0950-0618

Abstract

Highlights: Neural network regression analysis is performed when studying red mud-AAC with different w/s ratios. Relationship between pore structures and performances of red mud-based AAC is clarified. A model reflecting pore structures and compressive strength of AAC is proposed with better regression and applicability. A modified series–parallel model of thermal conductivity is provided in red mud-based AAC. Neural network regression analysis could be effective based on a mass of measured data. Abstract: A large amount of red mud needs to be disposed of in China, which causes environmental pollution. The feasibility of using red mud in autoclaved aerated concrete (AAC) has been confirmed, and its strength and thermal properties are highly related to its pore structures. To further clarify the relationship between the pore diameter (shape) and performances of red mud-based AAC, this study aims to establish a quantitative relationship between the compressive strength and the pore structure of AAC with red mud as well as its average shape factor of pores with a diameter above 0.5 mm, which was relied on the principle of elasticity. The experimental results show the strength formula of red mud-based AAC considering its porosity and its pore shape is R = R 0 0.03 1 - 1.21 V p 2 3 + 0.95 S H π 2 - 4 π + 8 + S H 2 π 4 + 16 S H π 2 - 8 S H π 3 16 + S H π 2 - 4 π + 8 + S H 2 π 4 + 16 S H π 2 - 8 S H π 3 . At the same time, to explore the constitutive relationship between the pore structure and thermal properties of red mud-based AAC, the series–parallel model of thermal conductivity is analyzed and modified, indicating that the series model has greater contribution to the effective thermal conductivity of AAC, where the expression is λ eff = 0.9 1 v 1 / λ 1 + v 2 / λ 2 + 0.19 ( v 1 λ 1 + v 2 λ 2 ). Finally, through the neural network regression analysis, it is proved that the established model reveals good accuracy.