4.7 Article

Prediction of compressive strength and feature importance analysis of solid waste alkali-activated cementitious materials based on machine learning

Related references

Note: Only part of the references are listed.
Article Mathematics, Applied

Prediction of Mechanical Properties of Rubberized Concrete Incorporating Fly Ash and Nano Silica by Artificial Neural Network Technique

Musa Adamu et al.

Summary: The use of waste products and by-products, such as silica fume, fly ash, and crumb rubber, in concrete production as partial substitutes has become an important approach for environmental sustainability. To counteract the negative effects of crumb rubber, nanomaterials like nano silica are added to enhance the mechanical properties of concrete. This study successfully predicted the mechanical characteristics of concrete with artificial neural network (ANN) technique, and the ANN model showed high accuracy in predicting the strengths.

AXIOMS (2023)

Article Chemistry, Physical

Investigation of Alkali-Activated Slag-Based Composite Incorporating Dehydrated Cement Powder and Red Mud

Aref A. Abadel et al.

Summary: This study investigated the production process of eco-efficient alkali-activated slag-based samples using recycled construction cementitious materials (RCCM) and red mud (RM). The findings showed that when a high quantity of dehydrated cement powder (DCP) and red mud were used, the strength and durability of the specimens improved significantly. It was concluded that DCP and RM could be potential sustainable binder substitutes, allowing waste valorization and reducing negative environmental impact.

MATERIALS (2023)

Review Chemistry, Physical

Carbonation of Alkali-Activated Materials: A Review

Ghandy Lamaa et al.

Summary: This paper presents a literature review on the effects of accelerated carbonation on alkali-activated materials. It provides a better understanding of the influence of CO2 curing on the chemical and physical properties of alkali-activated binders. The changes in chemistry, mineralogy, and physical properties caused by induced carbonation are discussed, along with the influence of this curing method on the strength development of alkali-activated materials.

MATERIALS (2023)

Article Construction & Building Technology

Prediction & optimization of alkali-activated concrete based on the random forest machine learning algorithm

Yubo Sun et al.

Summary: This study correlates the mix design parameters of alkali-activated concrete (AAC) to their performances using an artificial intelligence approach. Five random forest models were established to predict the fresh and hardened properties of AAC with equivalent accuracy reported in the literature. In addition, an inverse optimization was performed to reduce the sodium silicate dosages and mitigate the environmental impact of producing AAC. This study provides practical information for AAC mix design cases.

CONSTRUCTION AND BUILDING MATERIALS (2023)

Review Computer Science, Interdisciplinary Applications

Predicting the shear strength of concrete beam through ANFIS-GA-PSO hybrid modeling

Jie Li et al.

Summary: This study developed an ANFIS-GA-PSO hybrid model to predict the shear strength of concrete beams. Predicting the shear strength before construction is crucial for assessing the structure's ability to withstand external forces. The hybrid model showed higher accuracy in assessing shear behavior compared to the ELM model, and the ELM model exhibited fast training performance. The results identified key factors determining shear strength in reinforced concrete beams.

ADVANCES IN ENGINEERING SOFTWARE (2023)

Article Construction & Building Technology

Development of autogenous shrinkage prediction model of alkali-activated slag-fly ash geopolymer based on machine learning

Jiale Shen et al.

Summary: This paper developed a highly accurate autogenous shrinkage prediction tool for alkali-activated slag-fly ash geopolymer using machine learning. The influencing factors of autogenous shrinkage for geopolymer paste and mortar were analyzed. The results showed that the Extreme Gradient Boosting (XGB) algorithm provided the best prediction performance with an R2 value above 0.90 and strong generalization ability for predicting the autogenous shrinkage of alkali-activated slag-fly ash geopolymer paste and mortar.

JOURNAL OF BUILDING ENGINEERING (2023)

Article Engineering, Environmental

Resource utilization of solid waste from steel industries in cement-based cementitious materials: Mechanical properties, hydration behaviors, and environmental impact

Jie Liu et al.

Summary: This study investigated the synergistic effect of granulated blast furnace slag (GBFS), steel slag (SS) and steel slag mud (SSM) in P.I Portland cement (PC) to improve their utilization. The pozzolanic activity of ternary composite powder (TCP) was evaluated and their mechanical properties, hydration products, microstructure and environmental impacts were discussed. The results showed that the utilization rate of SS and SSM can be improved and the dependence on cement can be reduced through the synergistic effect between GBFSSS-SSM.

JOURNAL OF ENVIRONMENTAL CHEMICAL ENGINEERING (2023)

Article Green & Sustainable Science & Technology

Influence of waste glass powder as a supplementary cementitious material (SCM) on physical and mechanical properties of cement paste under high temperatures

Xi Jiang et al.

Summary: The effect of using waste glass powder (WGP) as a supplementary material for ordinary Portland cement (OPC) paste specimens was experimentally studied. The addition of WGP improved workability but prolonged the setting time of the OPC paste. It also reduced the degradation of the OPC matrix under high temperatures and mitigated the potential alkaline aggregate reaction (ASR) effect.

JOURNAL OF CLEANER PRODUCTION (2022)

Article Materials Science, Multidisciplinary

A multi-objective optimisation approach for activity excitation of waste glass mortar

Junbo Sun et al.

Summary: Waste glass can be effectively utilized in construction by chemical activation and mechanical grinding. The optimal dosage of chemical activators and the combined activation method were determined. A multi-objective optimization model based on the firefly algorithm and machine learning was used to optimize the strength, alkali-silica reaction expansion, and cost of waste glass mixtures. The Pareto fronts for mortars containing different sizes of waste glass powder were successfully obtained, contributing to the practical application of waste glass in mortar production.

JOURNAL OF MATERIALS RESEARCH AND TECHNOLOGY-JMR&T (2022)

Review Engineering, Environmental

Synthesis and application of geopolymers from municipal waste incineration fly ash (MSWI FA) as raw ingredient-A review

Tian Lan et al.

Summary: MSWI FA as a precursor into geopolymers is significant for environmental protection and resource utilization; the synthesis process parameters have a significant influence on the mechanical properties of products; MSWI FA-based geopolymers have excellent durability and heavy metal solidification ability.

RESOURCES CONSERVATION AND RECYCLING (2022)

Article Chemistry, Physical

Concrete Strength Prediction Using Different Machine Learning Processes: Effect of Slag, Fly Ash and Superplasticizer

Chongchong Qi et al.

Summary: This study utilizes blast furnace slag and fly ash to prepare concrete with a prediction model constructed using random forest and particle swarm optimization methods. The model's performance was evaluated and optimized, and sensitivity analysis revealed the importance of cement in the concrete design. This study fills the research gap in estimating the compressive strength of BFS-FA-superplasticizer concrete and has potential engineering application value.

MATERIALS (2022)

Article Construction & Building Technology

The data-driven research on bond strength between fly ash-based geopolymer concrete and reinforcing bars

Yue Li et al.

Summary: This study predicts and analyzes the bond strength between fly ash-based concrete and rebars using machine learning methods, demonstrating that machine learning models have better prediction performance. Slag content, compressive strength, reactivity modulus, and cover thickness to rebar diameter ratio are identified as important factors influencing bond strength.

CONSTRUCTION AND BUILDING MATERIALS (2022)

Article Green & Sustainable Science & Technology

Effect of incorporating municipal solid waste incinerated bottom ash in alkali-activated fly ash concrete subjected to accelerated CO2 curing

David Suescum-Morales et al.

Summary: The effect of incorporating bottom ash from municipal solid waste incineration (MIBA) as partial replacement for fly ash (FA) on alkali-activated mortars was studied. The use of MIBA resulted in a decrease of calcium silicate hydrate gel (C-S-H) and sodium aluminosilicate hydrate (N-A-S-H), affecting the macrostructural behavior of the mortars. The accelerated CO2 curing stage improved the macrostructural behavior of the mortars.

JOURNAL OF CLEANER PRODUCTION (2022)

Article Chemistry, Physical

Recycling of Cement Industry Waste for Alkali-Activated Materials Production

Madina Salamanova et al.

Summary: This paper presents a comprehensive study on the structure formation of alkali-activated materials based on cement production waste. The research results provide an alternative to expensive and energy-intensive Portland cement, offering eco-friendly and durable concrete and reinforced concrete composites.

MATERIALS (2022)

Article Construction & Building Technology

Effect of granulated phosphorus slag on physical, mechanical and microstructural characteristics of Class F fly ash based geopolymer

Yanhai Wang et al.

Summary: The addition of granulated phosphorus slag (GPS) in Class F fly ash (FFA) based geopolymer can reduce setting time, increase compressive strength, and create a denser and more homogeneous microstructure. Utilizing scanning electron microscopy and other tools, it was observed that GPS generated calcium silicate hydrate (C-S-H) gel, coexisting with sodium aluminate silicate hydrate (N-A-S-H) gel, resulting in improved mechanical properties of the derived geopolymer.

CONSTRUCTION AND BUILDING MATERIALS (2021)

Article Chemistry, Multidisciplinary

Machine Learning Approaches for Prediction of the Compressive Strength of Alkali Activated Termite Mound Soil

Assia Aboubakar Mahamat et al.

Summary: The unconventional usage of earth-based materials in construction makes it challenging to accurately estimate their properties, with traditional materials procedures falling short in accuracy. To predict the compressive strength of these materials, support vector machines, artificial neural networks, and linear regression models were used, with support vector machines outperforming the other methods in terms of R-2 score and root mean square error.

APPLIED SCIENCES-BASEL (2021)