Machine learning for the management of biochar yield and properties of biomass sources for sustainable energy

Biofuels Bioproducts and Biorefining, Год журнала: 2024, Номер 18(2), С. 567 - 593

Опубликована: Фев. 5, 2024

Abstract Biochar is emerging as a potential solution for biomass conversion to meet the ever increasing demand sustainable energy. Efficient management systems are needed in order exploit fully of biochar. Modern machine learning (ML) techniques, and particular ensemble approaches explainable AI methods, valuable forecasting properties efficiency biochar properly. Machine‐learning‐based forecasts, optimization, feature selection critical improving techniques. In this research, we explore influences these techniques on accurate yield range sources. We emphasize importance interpretability model, improves human comprehension trust ML predictions. Sensitivity analysis shown be an effective technique finding crucial characteristics that influence synthesis Precision prognostics have far‐reaching ramifications, influencing industries such logistics, technologies, successful use renewable These advances can make substantial contribution greener future encourage development circular biobased economy. This work emphasizes using sophisticated data‐driven methodologies synthesis, usher ecologically friendly energy solutions. breakthroughs hold key more environmentally future.

Язык: Английский

---

Optimizing load-displacement prediction for bored piles with the 3mSOS algorithm and neural networks

Ocean Engineering, Год журнала: 2024, Номер 304, С. 117758 - 117758

Опубликована: Апрель 15, 2024

Язык: Английский

---

Machine learning-based soil–structure interaction analysis of laterally loaded piles through physics-informed neural networks

Acta Geotechnica, Год журнала: 2024, Номер 19(7), С. 4765 - 4790

Опубликована: Янв. 17, 2024

Язык: Английский

---

Trajectory Optimization for Adaptive Deformed Wheels to Overcome Steps Using an Improved Hybrid Genetic Algorithm and an Adaptive Particle Swarm Optimization

Mathematics, Год журнала: 2024, Номер 12(13), С. 2077 - 2077

Опубликована: Июль 2, 2024

Two-wheeled mobile robots with deformed wheels face low stability when climbing steps, and their success rate in overcoming steps is affected by the trajectory. To address these challenges, we propose an improved hybrid genetic adaptive particle swarm optimization (HGAPSO) algorithm to optimize wheels’ trajectory for steps. HGAPSO optimizes maximum minimum values of inertial weight learning factors utilizing region-wide search capabilities algorithm, which substantially improves convergence speed adaptability. Furthermore, analysis motion wheel examination potential interference during operation are used construct a wheel’s center-of-mass route based on fifth-order Bézier curves. Comparative simulation experiments trajectories optimized using different algorithms under same working conditions designed demonstrate efficacy proposed optimizing step. Simulation were conducted deformation various sizes. These then compared unoptimized ones. The results showed that HGAPSO-optimized significantly robot

Язык: Английский

---

A multi-field coupled data-driven surrogate approach for multiphysical damage diagnostic of energy harvesting composite plates

Advances in Engineering Software, Год журнала: 2025, Номер 202, С. 103871 - 103871

Опубликована: Фев. 4, 2025

Язык: Английский

---

Influence of Settlement on Base Resistance of Long Piles in Soft Soil—Field and Machine Learning Assessments

Geotechnics, Год журнала: 2024, Номер 4(2), С. 447 - 469

Опубликована: Май 3, 2024

Understanding the role that settlement can have on base resistance of piles is a crucial matter in design and safety control deep foundations under various buildings infrastructure, especially for long to super-long (60–90 m length) soft soil. This paper presents novel assessment this issue by applying explainable machine learning (ML) techniques robust database (1131 datapoints) fully instrumented pile tests across 37 real-life projects Mekong Delta. The analysis data based conventional methods shows distinct responses rising settlement, as compared short piles. rapidly develop at small threshold (0.015–0.03% pile’s contribute up 50–55% total bearing capacity piles, but it slowly rises over wide range only 20–25% due considerable loss impact depth. Furthermore, leveraging advantages ML methods, results significantly enhance our understanding settlement–base relationship through computations. ML-based prediction method with popular practice codes foundations, further attesting high accuracy reliability newly established model.

Язык: Английский

---

Predicting load–displacement of driven PHC pipe piles using stacking ensemble with Pareto optimization

Engineering Structures, Год журнала: 2024, Номер 316, С. 118574 - 118574

Опубликована: Июль 15, 2024

Язык: Английский

---

Synergistic integration of isogeometric analysis and data-driven modeling for enhanced strip footing design on two-layered clays: Advancing geotechnical engineering practices

Engineering Analysis with Boundary Elements, Год журнала: 2024, Номер 167, С. 105880 - 105880

Опубликована: Июль 24, 2024

Язык: Английский

---

An effective optimum design for passive viscous damping control using FVDs/VWDs in multi-story buildings

Structures, Год журнала: 2024, Номер 67, С. 107004 - 107004

Опубликована: Авг. 7, 2024

Язык: Английский

---

XGBoost-based global sensitivity analysis of ground settlement caused by shield tunneling in dense karst areas

Advanced Engineering Informatics, Год журнала: 2024, Номер 62, С. 102928 - 102928

Опубликована: Окт. 1, 2024

Язык: Английский

---