Gasification process prediction using a novel and reliable metaheuristic algorithm coupled with the K-nearest neighbors DOI
Yuan-Fang Li, Ming Cao

Chemical Product and Process Modeling, Journal Year: 2025, Volume and Issue: unknown

Published: March 5, 2025

Abstract The present work introduces a new method for forecasting the formation of CH 4 and C 2 H n gases in gasification biomass. K-nearest neighbors (KNN) algorithm is utilized as base model, while two innovative optimization techniques, Artificial Rabbits Optimization (ARO) Smell Agent (SAO), are employed to enhance overall performance achieve optimal results. goal this investigation create prediction model that can reliably accurately anticipate amount produced during By combining strengths KNN with capabilities ARO SAO, proposed approach aims overcome existing limitations process predictions. experimental results demonstrate effectiveness combined predicting estimating quantities produced. integration SAO enables better leading improved accuracy reliability outcomes. Additionally, suggested models was thoroughly evaluated assessed utilizing evaluators. Remarkably, KNSA (combination SAO) achieved highest R values 0.994 0.995 , correspondingly, which demonstrates methods. conclusion study contributes field biomass gasification, it methodology used further improving its through implementation techniques. Further optimizations may now be opened, set insights derived from research curiosity-driven scholars practitioners renewable energy production.

Language: Английский

Hydrogen Enhancement in Syngas Through Biomass Steam Gasification: Assessment with Machine Learning Models DOI Creative Commons

Yunye Shi,

Diego Mauricio Yepes Maya, Electo Eduardo Silva Lora

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(5), P. 1200 - 1200

Published: Feb. 28, 2025

Artificial intelligence (AI), particularly supervised machine learning, has revolutionized the biofuel industry by enhancing feedstock selection, predicting fluid compositions, optimizing operations, and streamlining decision-making. These algorithms outperform traditional models accurately handling complex, high-dimensional data more efficiently cost-effectively. This study assesses effectiveness of various learning in engineering, focusing on a comparative analysis artificial neural networks (ANNs), support vector machines (SVMs), tree-based models, regularized regression models. The results show that random forest (RF) excel syngas composition its lower heating value (LHV), achieving high precision with training testing RMSE values below 0.2 R-squared close to 1. A detailed SHAP identified steam-to-biomass ratio (SBR) as most critical factor these predictions while also noting significant impact temperature conditions. underscores importance thermal parameters gasification supports systematic integration AI production enhance predictive accuracy.

Language: Английский

Citations

0
Gasification process prediction using a novel and reliable metaheuristic algorithm coupled with the K-nearest neighbors DOI
Yuan-Fang Li, Ming Cao

Chemical Product and Process Modeling, Journal Year: 2025, Volume and Issue: unknown

Published: March 5, 2025

Abstract The present work introduces a new method for forecasting the formation of CH 4 and C 2 H n gases in gasification biomass. K-nearest neighbors (KNN) algorithm is utilized as base model, while two innovative optimization techniques, Artificial Rabbits Optimization (ARO) Smell Agent (SAO), are employed to enhance overall performance achieve optimal results. goal this investigation create prediction model that can reliably accurately anticipate amount produced during By combining strengths KNN with capabilities ARO SAO, proposed approach aims overcome existing limitations process predictions. experimental results demonstrate effectiveness combined predicting estimating quantities produced. integration SAO enables better leading improved accuracy reliability outcomes. Additionally, suggested models was thoroughly evaluated assessed utilizing evaluators. Remarkably, KNSA (combination SAO) achieved highest R values 0.994 0.995 , correspondingly, which demonstrates methods. conclusion study contributes field biomass gasification, it methodology used further improving its through implementation techniques. Further optimizations may now be opened, set insights derived from research curiosity-driven scholars practitioners renewable energy production.

Language: Английский

Citations

0