Trustworthy and Human Centric neural network approaches for prediction of landfill methane emission and sustainable waste management practices

Waste Management, Journal Year: 2025, Volume and Issue: 195, P. 44 - 54

Published: Jan. 30, 2025

Landfills rank third among the anthropogenic sources of methane gas in atmosphere, hence there is a need for greater emphasis on quantification landfill emission mitigating environmental degradation. However, estimation and prediction challenge as modeling complexity generation involves different chemical, biological physical reactions. Various machine learning techniques lacks providing explainability context addressing uncertainties emission. This work presents novel artificial neural network (ANN) based approach enhancing interpretation prediction. A trustworthy ANN (TANN) using SHapley Additive exPlanations (SHAP) presented this research improving predicted values data seven major producing countries like India, China, Russia, Indonesia, US, EU, Brazil. Further, Human-Centric (HCANN) model two approaches: risks indication physics informed are developed. The HCANN was capable scientific principles well-known LandGEM data. results exhibited close agreement with those produced by LandGEM. Likewise developed factors production rates (MPR), capture system efficiency (GCSE), monitoring reliability (MSR) able to offer intuitive contextual decision understand risk associated unmanaged methane. Proposed TANN approaches valuable tool assessment sustainable waste management practices.

Language: Английский

---

Lignocellulose‐Derived Energy Materials and Chemicals: A Review on Synthesis Pathways and Machine Learning Applications

Small Methods, Journal Year: 2025, Volume and Issue: unknown

Published: April 23, 2025

Abstract Lignocellulose biomass, Earth's most abundant renewable resource, is crucial for sustainable production of high–value chemicals and bioengineered materials, especially energy storage. Efficient pretreatment vital to boost lignocellulose conversion bioenergy biomaterials, cut costs, broaden its energy–sector applications. Machine learning (ML) has become a key tool in this field, optimizing processes, improving decision‐making, driving innovation valorization This review explores main strategies – physical, chemical, physicochemical, biological, integrated methods evaluating their pros cons It also stresses ML's role refining these supported by case studies showing effectiveness. The examines challenges opportunities integrating ML into storage, underlining pretreatment's importance unlocking lignocellulose's full potential. By blending process knowledge with advanced computational techniques, work aims spur progress toward sustainable, circular bioeconomy, particularly storage solutions.

Language: Английский

---

Cropland Suitability Prediction Method Based on Biophysical Variables from Copernicus Data and Machine Learning

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(1), P. 372 - 372

Published: Jan. 2, 2025

The goal of this study was to propose and validate a method for predicting cropland suitability based on biophysical variables machine learning according an FAO land standard using soybean (Glycine max L.) as representative crop, aiming provide alternative geographic information system (GIS)-based multicriteria analysis. peak leaf area index (LAI) the fraction absorbed photosynthetically active radiation (FAPAR) from PROBA-V/Sentinel-3 data were calculated ground-truth agricultural parcels in continental Croatia during 2015–2021. Four regression algorithms, including random forest (RF), support vector (SVM), extreme gradient boosting (XGB), well their combination, evaluated LAI FAPAR entire area, with RF producing highest prediction accuracy R2 range 0.250–0.590. translation K-means classes performed relative-based approach, ranking five resulting relative mean sums values. results proposed approach indicate that it is viable major crops, while minor crops would require higher spatial resolution, such vegetation indices Sentinel-2 imagery.

Language: Английский

---

Industrial Hemp Finola Variety Photosynthetic, Morphometric, Biomechanical, and Yield Responses to K Fertilization Across Different Growth Stages

Agronomy, Journal Year: 2025, Volume and Issue: 15(2), P. 496 - 496

Published: Feb. 19, 2025

The growing interest in Cannabis sativa as a highly used crop is present worldwide. There are limited data about the effect of potassium (K) fertilizer on industrial hemp yield for dual purposes (seed and stem production). current study aimed to investigate influence adding two different K fertilizers, KCl K2SO4, at growth stages (flowering ripening) productivity chlorophyll fluorescence (ChlF) sativa, variety Finola. Before sowing, treatments were applied: K1—100 kg ha−1 (60% K) K2—100 K2SO4 (52% K, S 17%). OJIP (O stands “origin” (minimal fluorescence), P “peak” (maximum J I inflection points between O levels) recorded ChlF transients individual parameters during vegetation. At harvest, morphology (plant height, weight diameter, seed yield), tensile strength, modulus elasticity determined. results show sensitivity minimal (F0) maximal (Fm), electron transport from QA intersystem acceptors (ET0/(TR0 − ET0)), flux until PSI (RE0/RC) fertilization. that described (ET0/RC, ψE0, φE0), performance index absorption basis (PIABS, TR0/DI0, φP0), dissipation (DI0/RC), photosystem (φR0 δR0/(1 δR0)) had reaction only stage, indicating change their activity aging plants. average height was 67.5 cm, diameter 0.41 cm. sources did not significantly nor dry (on 12.2 t ha−1) 1.85 ha−1). strength stems highest with (53.32 MPa) lowest (49.25 MPa). stiffness by 5 GPa all treatments. In general, photosynthetic this varied more than formulations. Moreover, based study, it can be recommended use both dual-purpose production since no significant found morphometric biomechanical well agronomic parameters.

Language: Английский

---

Comparative Evaluation of Ensemble Machine Learning Models for Methane Production from Anaerobic Digestion

Fermentation, Journal Year: 2025, Volume and Issue: 11(3), P. 130 - 130

Published: March 7, 2025

This study provides a comparative evaluation of several ensemble model constructions for the prediction specific methane yield (SMY) from anaerobic digestion. From authors’ knowledge based on existing research, present their accuracy and utilization in digestion modeling relative to individual machine learning methods is incomplete. Three input datasets compiled samples using agricultural forestry lignocellulosic residues previous studies were used this study. A total six five evaluated per dataset, whose was assessed robust 10-fold cross-validation 100 repetitions. Ensemble models outperformed one out three terms accuracy. They also produced notably lower coefficients variation root-mean-square error (RMSE) than most accurate (0.031 0.393 dataset A, 0.026 0.272 B, 0.021 0.217 AB), being much less prone randomness training test data split. The optimal generally benefited higher number included, as well diversity principles. Since reporting final fitting single split-sample approach highly randomness, adoption multiple repetitions proposed standard future studies.

Language: Английский

---

Review of energy self-circulation systems integrating biogas utilization with Powerfuels production in global livestock industry

Bioresource Technology, Journal Year: 2024, Volume and Issue: 408, P. 131193 - 131193

Published: July 31, 2024

Language: Английский

---

Anaerobic digestion of lignocellulosic biomass: Process intensification and artificial intelligence

Renewable and Sustainable Energy Reviews, Journal Year: 2024, Volume and Issue: 210, P. 115264 - 115264

Published: Dec. 24, 2024

Language: Английский

---

A Comprehensive Evaluation of Machine Learning Algorithms for Digital Soil Organic Carbon Mapping on a National Scale

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(21), P. 9990 - 9990

Published: Nov. 1, 2024

The aim of this study was to narrow the research gap ambiguity in which machine learning algorithms should be selected for evaluation digital soil organic carbon (SOC) mapping. This performed by providing a comprehensive assessment prediction accuracy 15 frequently used SOC mapping based on studies indexed Web Science Core Collection (WoSCC), basis algorithm selection future studies. Two areas, including mainland France and Czech Republic, were 2514 400 samples from LUCAS 2018 dataset. Random Forest first ranked (mainland) then Republic regarding accuracy; coefficients determination 0.411 0.249, respectively, accordance with its dominant appearance previous WoSCC. Additionally, K-Nearest Neighbors Gradient Boosting Machine regression indicated, relative their frequency WoSCC, that they are underrated more considered Future consider areas not strictly related human-made administrative borders, as well interpretable ensemble approaches.

Language: Английский

---

Exploring interactive effects of environmental and microbial factors on food waste anaerobic digestion performance: Interpretable machine learning models

Bioresource Technology, Journal Year: 2024, Volume and Issue: 416, P. 131762 - 131762

Published: Nov. 7, 2024

Language: Английский

---