GX & Decarbonization Research

Wang-Chen Yang, H. Arian, Cui-Cui Luo

This paper proposes a constrained residual learning framework for portfolio optimization in China and Europe's carbon markets. It applies machine learning to capture carbon price dynamics and optimize risk-return trade-offs, offering improv…

Samrana Afzal, Rimsha Kanwal, Habibullah +1

This study introduces 'algorithmic power' to assess AI's implications for Pakistan's climate diplomacy. Proposing a framework of data, analytical, and diplomatic capacities, it uses composite indices (ARI, DLR, AIS) to estimate Pakistan's l…

Hannes Cordes

This systematic review maps NLP tasks used to analyze sustainability reports, categorizing 160 studies hierarchically. It finds keyword search, content analysis, and topic modeling are most common, and highlights limitations such as linguis…

Rongyu Ji

This qualitative case study of Microsoft examines how ESG information disclosure affects capital market performance in AI enterprises. It finds that completeness, openness, and AI-specific governance of ESG disclosure are high, with respons…

Dara Ghahremani

Replication materials for a doctoral study on how institutional, state, corporate, and foreign ownership affect ESG performance and annual-report ESG disclosure among European listed banks. Includes Python/Stata code for panel analysis, tex…

Léo Staccioli, José Gallego, Milena Nasner +1

This study assesses carbon footprint of defect repair in wind turbine blade manufacturing across three configurations: baseline, optimized infusion, and sensor+ML. Optimized infusion cuts repair CO2 by 30.4%, ML by 19.5%, less than its 32.1…

Atul Khanna, Abhishek Shukla

As AI workloads grow, data center electricity demand and carbon emissions rise sharply. This paper argues from an operations perspective—not hardware design—that workload placement, energy- and carbon-aware scheduling, right-sizing, inciden…

Krishna Rao, Andrew Dumit, Shaena Ulissi +7

PCFBench is the first benchmark that decomposes product carbon footprint (PCF) estimation into six independently evaluable tasks, with 614 expert-labeled items probing decomposition, retrieval, ontology matching, and numerical extraction. T…

Junjie Cai, Xijian Li, Shoukun Chen

This study proposes a method combining COMSOL multi-physics simulation and machine learning to accurately predict coal rock gas extraction volume, targeting the Qinglong Coal Mine in Guizhou. The XGBoost-LSTM hybrid model achieved the best …

Oyetola Ogunkunle, Emmanuel Uche, Kinsgley I. Okere +1

This paper systematically evaluates AI contributions to low-carbon energy production using bibliometric analysis and SWOT appraisal. It identifies research trends, key themes, and strengths, weaknesses, opportunities, and threats, offering …

Bing Zou, Xiang Yang, Yuhan Jin +1

This study proposes a multicollinearity-reduced conditional variational autoencoder (MCLRCVAE) to generate joint wind-solar scenarios for low-carbon dispatch of industrial integrated energy systems. Representative scenarios are selected via…

Bilal M, Latif F, Hasnain M +3

This paper systematically reviews the role of machine learning (ML) in designing Metal-Organic Frameworks (MOFs) for CO2 capture. It evaluates the latest models across four key areas: process-level application, mechanistic interpretable mod…

Islam MS, Sarkar J

Eco-ITAD is an on-premise system that diagnoses hardware health using S.M.A.R.T. telemetry and computes avoided Scope 3 emissions using DEFRA factors. It generates compliance certificates in volatile memory, eliminating persistent writes. E…

Karli Eka Setiawan, Marvel Martawidjaja, Hayyun Lisdiana

This study proposes a CNN-based image classification approach to automatically detect faults in solar panels. Using a public Kaggle dataset with six classes, Inception-V3 achieved the highest accuracy of 91% and best F1-scores in four categ…

Sahrin, Alfin, Utami, Erna, Shoffiana, Nur +1

This study develops LSTM-based models to predict power output of polycrystalline solar panels using environmental data. Comparing pure LSTM, CNN-LSTM, LSTM-AE, and GWO-LSTM, the GWO-LSTM achieves the highest accuracy (R²=0.98, MAPE=4.3%), d…

Surathunmanun, Surasak, Ongsakul, Weerakorn, Singh, Jai Govind +1

This paper proposes a novel framework (CTT–ViT–Transformer) integrating satellite cloud top temperature imagery with Vision Transformer and Transformer models for short-term solar PV forecasting in sensor-constrained microgrids. Evaluated o…

Ghasemzadeh, Nima, Javaherian, Amirreza, Yari, Mortaza +3

This study proposes a medium-scale biogas-fueled gas turbine poly-generation system supplying electricity, heating, cooling, and water. Multi-objective optimization using machine learning and Grey Wolf algorithms improves efficiency, cost, …

Meng X.

This paper proposes a methodology using Fuzzy AHP and Fuzzy WASPAS to evaluate ESG criteria and formulate green finance investment strategies. It provides a quantitative framework for investors to incorporate ESG factors into investment dec…

Nguyen Q.

This paper analyzes how data quality of Scope 3 emissions affects the prediction accuracy of machine learning models. It evaluates the impact of data variability on model performance, highlighting the potential and limitations of ML for cor…

R.Kaladevi, V.Umarani, V. N +1

This study builds an ML and explainable AI system to assess green loan eligibility. Using synthetic data, multiple models were evaluated, with Gradient Boosting achieving high accuracy. SHAP analysis identified credit score and energy effic…