BCLH2Pro: A novel computational tools approach for hydrogen production prediction via machine learning in biomass chemical looping processes

Energy and AI, Год журнала: 2024, Номер 18, С. 100414 - 100414

Опубликована: Авг. 13, 2024

This study optimizes biomass chemical looping processes (BCLpro), a technique for converting to energy, through machine learning (ML) sustainable energy production. The proposes an integrated Fe2O3-based ฺBCLpro combining steam gasification H2 Aspen Plus is used as the primary tool generate extensive datasets covering 24 types with 18 feature inputs in supervised model. A methodology involving K-Nearest Neighbors (KNN), Extreme Gradient Boosting (XGB), Light Machine (LGBM), Support Vector (SVM), Random Forest (RF), and CatBoost (CB) algorithms was employed predict yields BCLpro, utilizing 10-fold cross-validation robust model evaluation. Findings highlight CB algorithm's superior performance, achieving up 98% predictive accuracy, carbon content, reducer temperature, Fe2O3/Al2O3 mass ratio identified crucial features. algorithm has been developed into user-friendly tool, BCLH2Pro, accessible via web server. designed assist reducing costs, optimizing selection, planning operational conditions maximize yield BCLpro systems. Access can be obtained following link: http://bclh2pro.pythonanywhere.com/.

Язык: Английский

---

The prediction of biodiesel production yield from transesterification of vegetable oils with Machine learning

Results in Engineering, Год журнала: 2024, Номер unknown, С. 103236 - 103236

Опубликована: Окт. 1, 2024

Язык: Английский

---

Enhancing waste cooking oil biodiesel yield and characteristics through machine learning, response surface methodology, and genetic algorithms for optimal utilization in CI engines

International Journal of Green Energy, Год журнала: 2023, Номер 21(6), С. 1345 - 1365

Опубликована: Сен. 3, 2023

ABSTRACTThe current study seeks to predict and optimize the biodiesel production yield physicochemical properties of waste cooking oil. Using Box-Behnken design (BBD), L46 orthogonal arrays were developed for experimentation at five factors three levels. Furthermore, different boosting algorithms (AdaBoost, ExtraTrees, Gradient-Boosting regression) used develop an ML-based prognostic model using experimental data. Based on coefficient determination (R2) (0.997 yield, 0.999 kinematic viscosity, 0.996 calorific value, flash point), has most accurate predictions, followed by ExtraTrees AdaBoost. Through implementation GA, optimal conditions obtained, including a molar ratio 7.24:1, catalyst concentrations 1.49 wt.%, reaction temperatures 65°C, times 59.95 minutes, stirring speeds 733.32 rpm. Experimental validation these optimized is closely aligned with predicted values. evaluates engine performance emission parameters assess impact biodiesel/diesel blends (B10, B20, B30) compared pure diesel. The utilization demonstrates satisfactory performance, effectively reducing carbon monoxide (CO) hydrocarbon (HC) emissions. However, nitrogen oxide (NOx) emissions increase diesel.KEYWORDS: Biodiesel synthesisoptimal controlmachine learninggenetic algorithmphysiochemical propertiesengine Disclosure statementThe authors declare that they have no known competing financial interests or personal relationships could appeared influence work reported in this paper.Consent publicationYesAdditional informationFundingThe author(s) there funding associated featured article.

Язык: Английский

---

Artificial intelligence and machine learning models application in biodiesel optimization process and fuel properties prediction

Sustainable Energy Technologies and Assessments, Год журнала: 2024, Номер 73, С. 104097 - 104097

Опубликована: Ноя. 29, 2024

Язык: Английский

---

Prediction of Oil–Water Two-Phase Flow Patterns Based on Bayesian Optimisation of the XGBoost Algorithm

Processes, Год журнала: 2024, Номер 12(8), С. 1660 - 1660

Опубликована: Авг. 7, 2024

With the continuous advancement of petroleum extraction technologies, importance horizontal and inclined wells in reservoir exploitation has been increasing. However, accurately predicting oil–water two-phase flow regimes is challenging due to complexity subsurface fluid patterns. This paper introduces a novel approach address this challenge by employing extreme gradient boosting (XGBoost, version 2.1.0) optimised through Bayesian techniques (using Bayesian-optimization library, 1.4.3) predict regimes. The integration optimisation aims enhance efficiency parameter tuning precision predictive models. methodology commenced with experimental studies utilising multiphase simulation apparatus gather data across spectrum water cut rate, well inclination angles, rates. Flow patterns were meticulously recorded via direct visual inspection, these empirical datasets subsequently used train validate both conventional XGBoost model its Bayesian-optimised counterpart. A total 64 collected, 48 sets for training 16 testing, divided 3:1 ratio. findings highlight marked improvement accuracy model, achieving testing 93.8%, compared 75% traditional model. Precision, recall, F1-score metrics also showed significant improvements: increased from 0.806 0.938, recall 0.875 0.873 0.938. further supported results, (BO-XGBoost) an 0.948 Comparative analyses demonstrate that enhanced capabilities algorithm. Shapley additive explanations (SHAP) analysis revealed rates, daily rates most features contributing predictions. study confirms efficacy superiority algorithm regimes, offering robust effective investigating complex dynamics. research outcomes are crucial improving predictions introducing innovative technical approaches within domain engineering. work lays foundational stone application studies.

Язык: Английский

---

Breaking Barriers in Biodiesel: From Feedstock Challenges to Technological Advancements

Chemistry Africa, Год журнала: 2025, Номер unknown

Опубликована: Май 3, 2025

Язык: Английский

---

Application of Artificial Intelligence and Image Processing for the Cultivation of Chlorella sp. Using Tubular Photobioreactors

ACS Omega, Год журнала: 2024, Номер 9(46), С. 46017 - 46029

Опубликована: Ноя. 7, 2024

By integrating innovative technologies to enhance the efficiency and sustainability of production, this study specifies establishment a cutting-edge growing system for

Язык: Английский

---

Pilot scale production of biodiesel from Madhuca indica and comparative techno-economic analysis

Environmental Science and Pollution Research, Год журнала: 2025, Номер unknown

Опубликована: Фев. 7, 2025

Язык: Английский

---

Machine learning-driven predictive frameworks for optimizing chemical strategies in Microcystis aeruginosa mitigation

Journal of Water Process Engineering, Год журнала: 2025, Номер 71, С. 107235 - 107235

Опубликована: Фев. 12, 2025

Язык: Английский

---

Analytical optimization of biodiesel synthesis from seasonal mixed seed oils using Bara Gokhru nano-biocatalyst

Bioresource Technology Reports, Год журнала: 2025, Номер unknown, С. 102066 - 102066

Опубликована: Фев. 1, 2025

Язык: Английский

---