Computational Optimization of Ceramic Waste-Based Concrete Mixtures: A Comprehensive Analysis of Machine Learning Techniques

Archives of Computational Methods in Engineering, Год журнала: 2025, Номер unknown

Опубликована: Фев. 12, 2025

Язык: Английский

---

Ultrasonic detection and deep learning for high-precision concrete strength prediction

Journal of Building Engineering, Год журнала: 2025, Номер unknown, С. 112372 - 112372

Опубликована: Март 1, 2025

Язык: Английский

---

Usage of machine learning methods for forecasting the strength of environmentally friendly geopolymer concrete

Journal of Ambient Intelligence and Humanized Computing, Год журнала: 2025, Номер unknown

Опубликована: Март 22, 2025

Язык: Английский

---

Optimization of Material Composition for Improving Mechanical Properties of Fly Ash-Slag-Based Geopolymers: A Deep Learning Approach

Langmuir, Год журнала: 2025, Номер unknown

Опубликована: Апрель 9, 2025

Geopolymer is regarded as a novel type of eco-friendly material that may replace cement. To improve the prediction accuracy mechanical properties fly ash-slag-based geopolymer (FASGG), well optimize composition and mix design, this study utilizes seven key parameters variables, compressive flexural strengths were outputs. Deep learning techniques applied to train predict 600 sets experimental data, developing predictive model MK-CNN-GRU, which integrated Maximal Information Coefficient-K-median algorithm, Convolutional Neural Network, Gated Recurrent Unit algorithms. Results indicated ranking input related with strength was curing age, Ca/Si ratio, ash-to-slag Si/Al water-to-binder alkali activator modulus, equivalent. Three classical models selected benchmarks for predicting at different ages. The MK-CNN-GRU could fully exploit internal features data learn its variation patterns, resulting in more stable performance. An ablation submodels confirms considers temporal dependencies, long- short-term features, local dependencies hierarchical feature representations within data. Experimental suggested an exponential relationship between FASGG. predictions effectively captured variations, demonstrating good generalization ability applicability. This enhances estimation regarding behavior FASGG, offering theoretical framework refining design.

Язык: Английский

---

Optimizing compressive strength prediction in eco-friendly recycled concrete via artificial intelligence models

Multiscale and Multidisciplinary Modeling Experiments and Design, Год журнала: 2024, Номер 8(1)

Опубликована: Ноя. 7, 2024

Язык: Английский

---