Environmental Risk Mitigation via Deep Learning Modeling of Compressive Strength in Green Concrete Incorporating Incinerator Ash DOI Creative Commons

Amin Amraee,

Seyed Azim Hosseini, Farshid Farokhizadeh

и другие.

Buildings, Год журнала: 2025, Номер 15(7), С. 1103 - 1103

Опубликована: Март 28, 2025

Green concrete uses incinerator ash or lightweight as a substitute for cement. It retains the properties of conventional concrete. Initial laboratory tests have determined optimum mix design, weight variation, and compressive strength. Defined an environmentally friendly material, green reduces pollution improves environmental conditions during production. This study incorporates ash, toxic byproduct waste disposal, into production through phased numerical approach. A database deep learning modeling was created using Convolutional Neural Networks (CNNs) Multi-Verse Optimizer (MVO) algorithm. After evaluating efficiency structure model MATLAB coding, focus shifted to analyzing sensitivity input parameters on output parameter training, evaluation. The initial results indicate significant effect strength In addition, show that regression coefficient (R) 90% reflects accuracy current design. error index, which is also reported, shows applied method achieves optimal performance, with average 0.14. analysis introduced among five parameters, cement (W) has greatest influence strength, indicated by statistical group distances from baseline, percentage values, values.

Язык: Английский

The Role of Lightweight AI Models in Supporting a Sustainable Transition to Renewable Energy: A Systematic Review DOI Creative Commons
Tymoteusz Miller, Irmina Durlik, Ewelina Kostecka

и другие.

Energies, Год журнала: 2025, Номер 18(5), С. 1192 - 1192

Опубликована: Фев. 28, 2025

The transition from fossil fuels to renewable energy (RE) sources is an essential step in mitigating climate change and ensuring environmental sustainability. However, large-scale deployment of renewables accompanied by new challenges, including the growing demand for rare-earth elements, need recycling end-of-life equipment, rising footprint digital tools—particularly artificial intelligence (AI) models. This systematic review, following Preferred Reporting Items Systematic Reviews Meta-Analyses (PRISMA) guidelines, explores how lightweight, distilled AI models can alleviate computational burdens while supporting critical applications systems. We examined empirical conceptual studies published between 2010 2024 that address energy, circular economy paradigm, model distillation low-energy techniques. Our findings indicate adopting significantly reduce consumption data processing, enhance grid optimization, support sustainable resource management across lifecycle infrastructures. review concludes highlighting opportunities challenges policymakers, researchers, industry stakeholders aiming integrate principles into RE strategies, emphasizing urgent collaborative solutions incentivized policies encourage low-footprint innovation.

Язык: Английский

Процитировано

1
Environmental Risk Mitigation via Deep Learning Modeling of Compressive Strength in Green Concrete Incorporating Incinerator Ash DOI Creative Commons

Amin Amraee,

Seyed Azim Hosseini, Farshid Farokhizadeh

и другие.

Buildings, Год журнала: 2025, Номер 15(7), С. 1103 - 1103

Опубликована: Март 28, 2025

Green concrete uses incinerator ash or lightweight as a substitute for cement. It retains the properties of conventional concrete. Initial laboratory tests have determined optimum mix design, weight variation, and compressive strength. Defined an environmentally friendly material, green reduces pollution improves environmental conditions during production. This study incorporates ash, toxic byproduct waste disposal, into production through phased numerical approach. A database deep learning modeling was created using Convolutional Neural Networks (CNNs) Multi-Verse Optimizer (MVO) algorithm. After evaluating efficiency structure model MATLAB coding, focus shifted to analyzing sensitivity input parameters on output parameter training, evaluation. The initial results indicate significant effect strength In addition, show that regression coefficient (R) 90% reflects accuracy current design. error index, which is also reported, shows applied method achieves optimal performance, with average 0.14. analysis introduced among five parameters, cement (W) has greatest influence strength, indicated by statistical group distances from baseline, percentage values, values.

Язык: Английский

Процитировано

0