International Journal of Hydrogen Energy, Год журнала: 2024, Номер unknown
Опубликована: Ноя. 1, 2024
Язык: Английский
International Journal of Hydrogen Energy, Год журнала: 2024, Номер unknown
Опубликована: Ноя. 1, 2024
Язык: Английский
Digital Chemical Engineering, Год журнала: 2024, Номер unknown, С. 100209 - 100209
Опубликована: Дек. 1, 2024
Язык: Английский
Процитировано
3Digital Chemical Engineering, Год журнала: 2025, Номер unknown, С. 100234 - 100234
Опубликована: Март 1, 2025
Язык: Английский
Процитировано
0Journal of Chemical Technology & Biotechnology, Год журнала: 2025, Номер unknown
Опубликована: Апрель 23, 2025
Abstract BACKGROUND Dual‐reflux pressure swing adsorption (DR‐PSA) is an advanced variant of conventional PSA systems, capable simultaneously producing two high‐purity products with high recovery. However, inherent trade‐offs among purity, productivity, and cycle work necessitate multi‐objective optimization (MOO) to balance performance. This study investigates the MOO DR‐PSA process for biogas upgrading using artificial neural networks (ANNs) as surrogate models. Single‐hidden‐layer ANNs were developed various structures training algorithms integrated into non‐dominated sorting genetic algorithm‐II solve problems. RESULTS Different performance metrics require distinct ANN architectures algorithms, indicating no universal model selection. The Adam‐ selu‐ 446 CH 4 purity in light product (LP) relu ‐398 work, both on a large dataset, exhibited accuracy, 0.9975 0.9959, respectively. Meanwhile, elu ‐212 bed small achieved 0.9988. Pareto front maximizing LP productivity revealed sharp knee region, significant trade‐off. Under constrained fronts emerged at different levels (>91% 91.5%), fewer feasible solutions higher purity. Ultimately, optimized conditions resulted enhancement by least 17%, reduction much 34%, 30%, compared base case. CONCLUSION ANN‐assisted optimizes performance, offering insights development, mapping, identification operational regions. © 2025 Society Chemical Industry (SCI).
Язык: Английский
Процитировано
0Energy, Год журнала: 2025, Номер unknown, С. 136512 - 136512
Опубликована: Май 1, 2025
Язык: Английский
Процитировано
0Energies, Год журнала: 2024, Номер 17(19), С. 4804 - 4804
Опубликована: Сен. 25, 2024
Coke oven gas (COG), as a by-product of the coking process and mixture with high hydrogen content, is an important potential component sustainable economy industry. Ongoing studies analyses are looking at many opportunities for utilization coke gas, including production hydrogen, methanol or other chemicals. However, it not to forget that all processes this may pose hazard humans environment. This due physicochemical properties COG content flammable gases such methane carbon monoxide in its composition. Potential hazardous events also related toxic substances composition gas. The publication focuses on occurrence fire explosion result uncontrolled release purified from installation. zones associated these phenomena presented concerning different levels concentration influence selected factors range zones. Zones human deaths reached maximum about 130 m site failure, depending composition, level damage parameters did occur. zone injury 12 m.
Язык: Английский
Процитировано
1International Journal of Hydrogen Energy, Год журнала: 2024, Номер 91, С. 718 - 727
Опубликована: Окт. 17, 2024
Язык: Английский
Процитировано
1International Journal of Hydrogen Energy, Год журнала: 2024, Номер unknown
Опубликована: Ноя. 1, 2024
Язык: Английский
Процитировано
0