Deterministic Drivers of City-Scale Green Hydrogen Production: A Physics-Guided Data-Driven Study
都市スケールのグリーン水素生産の決定要因：物理誘導データ駆動研究 (AI 翻訳)
J. Tarra, S. Gowthami, R. Swetha, Attada Jyothsna, Kanaka Raju Kalla, Ankamma Rao Jonnalagadda, G. Ramarao, Trinadha Burle, Budi Srinivasa Rao
🤖 gxceed AI 要約
日本語
本研究は2,535地点のデータを用いて、都市レベルでのグリーン水素生産量の主要決定要因を解析。再生可能エネルギー補完性指数（RCI）とエネルギー-水素変換効率指数（EHCI）という2つの物理誘導指標を導入し、線形回帰モデルが高い精度（R²=0.9989）で予測可能であることを示した。SHAP分析によりEHCIが最も重要な予測因子であり、設備容量拡大よりも効率改善が効果的なレバーであることを明らかにした。
English
This study analyzes the dominant drivers of city-level green hydrogen production using a dataset of 2,535 locations. It introduces two physics-guided indices: Renewable Complementarity Index (RCI) and Energy-to-Hydrogen Conversion Index (EHCI). A linear model achieves near-perfect accuracy (R²=0.9989), and SHAP analysis identifies EHCI as the most influential predictor, highlighting efficiency improvements over capacity expansion.
Unofficial AI-generated summary based on the public title and abstract. Not an official translation.
📝 gxceed 編集解説 — Why this matters
日本のGX文脈において
日本は水素基本戦略に基づきグリーン水素の導入を推進している。本研究成果は、都市計画や水素インフラ整備において、再生可能エネルギーと水素変換効率の相互作用を考慮した効率的な水素生産拠点の設計に活用できる。
In the global GX context
Globally, green hydrogen is critical for decarbonization. This study provides a physics-guided, interpretable modeling framework that can be applied to urban hydrogen planning worldwide, emphasizing the role of system efficiency over mere capacity expansion.
👥 読者別の含意
🔬研究者:Demonstrates the value of physics-guided feature engineering for interpretable ML in energy systems.
🏢実務担当者:Provides actionable insights on prioritizing efficiency improvements in hydrogen plant design.
🏛政策担当者:Supports policies that incentivize efficiency over capacity expansion in hydrogen production.
📄 Abstract（原文）
Green hydrogen production is a key pathway for decarbonizing energy systems, yet its city-scale output depends on interactions among renewable resources, conversion efficiency, and auxiliary energy demands. This study analyzes a validated dataset of 2,535 locations to identify the dominant drivers of city-level green hydrogen production using locally available variables. Exploratory analysis shows that hydrogen output does not scale linearly with aggregate renewable power, highlighting the role of system-level efficiency and resource interactions. Two physics-guided indices are introduced: a Renewable Complementarity Index (RCI) to quantify solar-wind synergy and an Energy-to-Hydrogen Conversion Index (EHCI) to represent effective conversion efficiency accounting for desalination demand. Linear Regression and Random Forest models are developed for prediction. The linear model achieves near-perfect accuracy (R2=0.9989) and outperforms the ensemble model across error metrics, indicating dominance of deterministic, efficiency-driven relationships. SHAP analysis identifies EHCI as the most influential predictor, with raw renewable resource variables contributing secondarily. These results demonstrate that physics-guided feature engineering enables accurate, interpretable, and parsimonious modeling of urban green hydrogen production and highlights efficiency improvements as a more effective lever than capacity expansion.
🔗 Provenance — このレコードを発見したソース
- semanticscholar https://doi.org/10.1109/iciscn67954.2026.11566285first seen 2026-06-26 05:42:03
🔔 こうした論文の新着を逃したくない方は キーワードアラート に登録（無料・3キーワードまで）。
gxceed は公開メタデータに基づく研究支援データセットです。要約・翻訳・解説は AI 支援で生成されています。 最終的な解釈・検証は利用者が原典資料に基づいて行うことを前提とします。