On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

169,051 papers · 148 categories

Trend · papers per month

count % of corpus

11 results for “methane”

MethaneMapper: Spectral Absorption aware Hyperspectral Transformer for Methane Detectioneess.IV

MethaneMapper detects methane emissions with high accuracy and reduced model size.

problem Challenges in detecting and quantifying methane emissions from AVIRIS-NG data.

method Spectral absorption wavelength aware transformer network, introducing two novel modules.

result Achieves 0.63 mAP in detection and reduces model size by 5x.

methane detection spectral transformer AVIRIS-NG climate change model size reduction

2023-04-05abs ↗pdf ↗

Dispersion based Recurrent Neural Network Model for Methane Monitoring in Albertan Tailings Pondsstat.AP

Model predicts methane emissions from oil sands tailing ponds, suggesting significant environmental impact.

problem Estimating methane emissions from inactive oil sands tailing ponds.

method Physics constrained machine learning model using real-time weather data and laboratory experiments.

result Active oil sands tailing ponds emit between 950 to 1500 tonnes of methane per year, equivalent to 6000 gasoline vehicles.

methane emissions oil sands tailing ponds machine learning greenhouse gases climate impact

2024-11-11abs ↗pdf ↗

Heat Transfer Prediction for Methane in Regenerative Cooling Channels with Neural Networkscs.LG

ANN predicts methane heat transfer in rocket engines efficiently.

problem Accurate heat transfer prediction for supercritical methane in rocket engine cooling channels.

method Artificial neural networks trained on CFD simulation data.

result ANN model predicts maximum wall temperature with convincing precision.

neural networks heat transfer methane rocket engine regenerative cooling

2019-07-24abs ↗pdf ↗

Toward Machine Learned Highly Reduce Kinetic Models For Methane/Air Combustionphysics.chem-ph

Compact models for methane/air combustion reduce complexity without sacrificing accuracy.

problem Creating accurate, computationally efficient models for methane combustion.

method Data-oriented three-step methodology: 1) Remove non-essential species, 2) Numerically optimize to key species profiles, 3) Machine learning to refine parameters.

result Produced 19 and 15 species compact models that outperform current state-of-the-art models in accuracy and range of conditions.

methane combustion kinetic models machine learning reduced models turbine combustion

2021-03-15abs ↗pdf ↗

Low-Dimensional High-Fidelity Kinetic Models for NOX Formation by a Compute Intensification Methodphysics.chem-ph

Compact models for NOX formation during methane combustion are created using a new algorithm.

problem Creating accurate models for NOX formation during complex combustion processes.

method Adapted Machine Learning Optimization of Chemical Kinetics (MLOCK) algorithm with Latin Square method for virtual reaction network generation.

result Compact models with high fidelity (>75%) in reproducing industry-defined performance targets are generated.

NOX formation kinetic models machine learning methane combustion compact models

2022-02-21abs ↗pdf ↗

Targeted Source Detection for Environmental Dataeess.SP

Paper proposes a technique to detect and predict sources of contaminants in complex systems.

problem Difficulty in identifying sources of contaminants in coupled natural and human systems.

method Developed a technique for simultaneous source detection and prediction.

result Outperforms other approaches in detecting potential groundwater contamination.

contaminants source detection water pollution shale gas development groundwater contamination

2019-08-29abs ↗pdf ↗

Learning Radiative Transfer Models for Climate Change Applications in Imaging Spectroscopycs.LG

Efficient neural network speeds up radiative transfer model processing for climate change analysis.

problem Computational inefficiency in processing spectroscopy data for climate change analysis.

method Developed an algorithm using neural networks to emulate radiative transfer models (RTMs), significantly reducing processing time.

result Multifold speedup in processing time for radiative transfer models, enabling their use with high volume imaging spectrometers.

climate change radiative transfer models neural networks spectroscopy imaging

2019-06-08abs ↗pdf ↗

Deep Learning Surrogates for Real-Time Gas Emission Inversioncs.LG

Deep learning speeds up real-time emission monitoring.

problem Real-time greenhouse gas emission monitoring under transient conditions.

method Bayesian inference with deep learning surrogate of CFD outputs.

result Near-real-time predictions with orders-of-magnitude faster runtimes.

real-time monitoring greenhouse gases deep learning Bayesian inference spatio-temporal inversion

2025-06-17abs ↗pdf ↗

Interpretable Data-driven Methods for Subgrid-scale Closure in LES for Transcritical LOX/GCH4 Combustionphysics.flu-dyn

Study assesses data-driven and physics-based SGS models for transcritical combustion.

problem Challenges in simulating high-pressure combustion systems due to complex fluid behaviors.

method Comparison of physics-based and random forest machine learning models in turbulent transcritical non-premixed flames.

result Random forest models can effectively model subgrid stresses, providing insight into their formation.

transcritical combustion LES subgrid-scale modeling random forest machine learning

2021-03-11abs ↗pdf ↗

An Ensemble-Based Deep Framework for Estimating Thermo-Chemical State Variables from Flamelet Generated Manifoldscs.LG

Improved deep learning framework for estimating combustion variables.

problem Accurately estimating thermo-chemical state variables in turbulent combustion.

method Introducing deep ensembles to approximate posterior distribution of quantities of interest, using Flamelets or Points strategies.

result ChemTab Deep Ensembles provide more accurate representation of source energy and key species source terms.

turbulent combustion deep learning ensemble methods flamelet generated manifolds uncertainty quantification

2022-11-25abs ↗pdf ↗

Toward Development of Machine Learned Techniques for Production of Compact Kinetic Modelsphysics.chem-ph

Automated method creates compact chemical models from detailed ones, reducing complexity and improving accuracy.

problem Creating accurate low-dimensional chemical kinetic models from detailed ones is time-consuming and requires expert knowledge.

method Machine Learned Optimisation of Chemical Kinetics (MLOCK) algorithm systematically perturbs sub-models to find optimal compact models.

result Compact models (15 species) retain ~87% fidelity to detailed models, outperforming previous methods.

chemical kinetics kinetic models machine learning automated optimization combustion

2022-02-16abs ↗pdf ↗