Study on molten salt on torrefaction and subsequent pyrolysis of elm branches

Industrial Crops and Products, Journal Year: 2024, Volume and Issue: 222, P. 119672 - 119672

Published: Sept. 17, 2024

Language: Английский

---

Algae’s potential as a bio-mass source for bio-fuel production: MLR vs. ANN models analyses

Fuel, Journal Year: 2025, Volume and Issue: 395, P. 134853 - 134853

Published: March 28, 2025

Language: Английский

---

Advancing Energy Recovery: Evaluating Torrefaction Temperature Effects on Food Waste Properties from Fruit and Vegetable Processing

Processes, Journal Year: 2025, Volume and Issue: 13(1), P. 208 - 208

Published: Jan. 13, 2025

Most organic waste from food production is still not used for energy production. From the perspective of production, one option to valorise properties waste. The fruit juice industry growing rapidly and generates large amounts One main wastes in processing peach pits apple peels. aim this study was analyse influence torrefaction temperature on waste, namely peels, pea shells, order improve their value determine potential further use valorisation as a renewable source. different temperatures heating (HHV), mass yield (MY) (EY) better understand behavior thermal individual selected samples. process carried out at 250 °C, 350 °C 450 °C. obtained biomass compared with dried biomass. For HHV after (28 kJ/kg), MY decreased by (66–34%), while EY fell (97–83%). Peach pits, despite higher (18 achieved low (38–89%) (59–99%), which reduces efficiency biochar Pea peels had (82–97%) lower (11 but high ash content limits wider use. results confirm that, increasing temperature, all biomasses decrease, consequence degradation hemicellulose cellulose loss volatile compounds. In most cases, improved resistance moisture adsorption, related that causes structural changes. showed hydrophobic Temperature seen have great impact efficiency. Apple generally highest yield.

Language: Английский

---

Integrating Advanced Machine Learning Models for Accurate Prediction of Porosity and Permeability in Fractured and Vuggy Carbonate Reservoirs: Insights from the Tarim Basin, Northwestern, China

SPE Journal, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 27

Published: April 1, 2025

Summary Accurate prediction of porosity and permeability in fractured vuggy carbonate reservoirs is crucial for optimizing hydrocarbon recovery but remains challenging due to their extreme heterogeneity anisotropy. Traditional methods often struggle capture the complex geological variability, leading suboptimal reservoir characterization. To address this, we propose a novel hybrid machine learning (ML) framework that integrates particle swarm optimization (PSO), mixed-effects random forest (MERF), ensemble models, such as light gradient boosting (LightGBM), (XGBoost), (RF). These models were trained validated using leave-one well-out cross-validation (LOO-CV) train-test split method, leveraging geophysical well-log data from Tarim Basin’s reservoirs. Among three PSO-MERF-LightGBM outperformed others, achieving an R² 0.9752 root mean square error (RMSE) 0.0606 R2 0.9983 RMSE 0.00473 during testing. Moreover, model demonstrates exceptional computational efficiency, completing processing just 11 seconds 9 seconds, respectively. This marks significant reduction computation time compared with other making it highly efficient alternative. results confirm its superior ability nonlinear relationships spatial variability. The study how advanced ML techniques can enhance characterization, improving decision-making subsurface resource management. Future research should extend this settings validate broader applicability.

Language: Английский

---

Modeling of Global and Individual Kinetic Parameters in Wheat Straw Torrefaction: Particle Swarm Optimization and Its Impact on Elemental Composition Prediction

Algorithms, Journal Year: 2025, Volume and Issue: 18(5), P. 283 - 283

Published: May 13, 2025

With the growing demand for sustainable energy solutions, biomass torrefaction has emerged as a crucial technology converting agricultural waste into high-value biofuels. This work develops dual kinetic modeling using global and individual parameters combined particle swarm optimization (PSO) to predict densification based on elemental composition (CHNO) high heating values (HHVs). The are calculated from experiments conducted at 250 °C, 275 300 obtained by adjusting experimental points each temperature. A two-step model was used optimized achieve exceptional adjustment accuracy (98.073–99.999%). were carried out in an inert atmosphere of nitrogen with rate 20 °C/min 100 min residence time. results demonstrate trade-off: while provide superior (an average fit 99.516%) predicting degradation weight loss, offer better predictions composition, errors 2.129% (carbon), 1.038% (hydrogen), 9.540% (nitrogen), 3.997% (oxygen). Furthermore, it been found that determining temperature higher than maximum peak observed derivative thermogravimetric (DTG) curve (275 °C), is possible behavior process within 250–325 °C range R-squared value corresponding error lower 3%. approach significantly reduces number required twelve only four relying single isothermal condition parameter estimation.

Language: Английский

---

Adsorption Capacity Prediction and Optimization of Electrospun Nanofiber Membranes for Estrogenic Hormone Removal Using Machine Learning Algorithms

Polymers for Advanced Technologies, Journal Year: 2024, Volume and Issue: 35(11)

Published: Nov. 1, 2024

ABSTRACT This study focuses on developing four machine learning (ML) models (Gaussian process regression (GPR), support vector (SVM), decision tree (DT), and ensemble (ELT)) optimized hyperparameters tuned via genetic algorithm (GA) particle swarm optimization (PSO) to analyze predict the adsorption capacity of estrogenic hormones. These hormones are a serious cause fish femininity various forms cancer in humans. Their electrospun nanofibers offers sustainable relatively environmentally friendly solution compared nanoparticle adsorbents, which require secondary treatment. The intricate task is find relationship between input parameters obtain optimum conditions, requires an efficient ML model. GPR integrated GA hybrid model performed most accurate precise results with R 2 = 0.999 RMSE 2.4052e −06 , followed by ELT (0.9976 4.3458e −17 ), DT (0.9586 2.4673e −16 SVM (0.7110 0.0639). 2D 3D partial dependence plots showed temperature, dosage, initial concentration, contact time, pH as vital parameters. Additionally, Shapley's analysis further revealed time dosage sensitive Finally, user‐friendly graphical user interface (GUI) was developed predictor utilizing (GPR‐GA), were experimentally validated maximum error < 3.3% for all tests. Thus, GUI can legitimately work any desired material given conditions efficiently monitor removal concentration simultaneously at wastewater treatment plants.

Language: Английский

---

Carbon Capture and Storage Optimization with Machine Learning using an ANN model

E3S Web of Conferences, Journal Year: 2024, Volume and Issue: 588, P. 01003 - 01003

Published: Jan. 1, 2024

The purpose of this study is to evaluate the accuracy predictions regarding work capacity CO 2 and selectivity MOF, using machine learning methodologies in relation /N . A dataset was used that includes numerous characteristics MOFs for development a neural network model. factors determined operational included pore size, surface area, chemical composition, among others. model demonstrated its by evaluating ; mean absolute errors were 25 0.8 mmol/g, respectively. correlation Analysis showed fairly negative (-0.014) between makeup very positive ( 0.029) area amount size. Thus, gas absorbability not top-dependent exclusively; size material contribute as well. More research should be carried out capability on predicting nature different Flow Object Models (MOFs) with an aim increasing efficiency, precision dependability models.

Language: Английский

---

Comparative Synthesis of Copper Nanoparticles Using Various Reduction Methods: Size Control, Stability, and Environmental Considerations

E3S Web of Conferences, Journal Year: 2024, Volume and Issue: 588, P. 02002 - 02002

Published: Jan. 1, 2024

The present work investigates three strategies for the production of copper nanoparticles (CuNPs): sodium borohydride reduction, ascorbic acid and reduction without reducing agent. Analyzed were size distribution, stability, ecological sustainability potential produced nanoparticles. method yielded most uniform diminutive nanoparticles, with an average diameter 8 ± 2 nm. This characteristic made it optimal selection applications necessitating meticulous control dimensions, such as in fields electronics catalysis. Although resulted formation considerably bigger measuring 15 5 nm, provided a much more environmentally friendly manufacturing approach that was well-suited biological applications. experiments showed stabilizers might be advantageous lowering ions, technique agent biggest least consistent 25 results indicate modulating incurs both advantages disadvantages. Among options considered, offers although is friendly. For purpose enhancing particle stability improving nanoparticle production, future study should investigate agents optimize reaction parameters.

Language: Английский

---

RESEARCH ON THE CONTROL SYSTEM OF MOBILE STRAW COMPACTION MOLDING MACHINE BASED ON PSO-ELM-GPC MODEL

INMATEH Agricultural Engineering, Journal Year: 2024, Volume and Issue: unknown, P. 652 - 661

Published: Dec. 21, 2024

To address the issue of mutual influence and coupling between main shaft speed feeding amount mobile straw compaction molding machine, which is beneficial for intelligent operation molding, this paper designs a PSO-ELM-GPC control model. This model integrates Particle Swarm Optimization (PSO) algorithm, Extreme Learning Machine (ELM), Generalized Predictive Control (GPC). It uses ELM optimized by PSO to predict output amount, adjusts input GPC controller based on deviation weight adjustment unit. Field simulation experiments show that maximum dynamic 1.72%, from target value 1.52%. The 1.22%, 1.42%. designed in can promptly correct uncertainties caused disturbances.

Language: Английский

---