電力市場の境界条件とカーボンフットプリント要因を考慮した前日電力価格予測手法
A Day-Ahead Electricity Price Forecasting Method Considering the Boundary Conditions of the Electricity Market and the Carbon Footprint Factor (原題)
Jia-Wei Zhang, Han Wang, Ning Zhang, Jie Yan, Chang Ge, Yong-Qian Liu, Jing Zhao
🤖 gxceed AI 要約
日本語
本論文は、電力市場の境界条件と電力炭素フットプリント要因を統合した前日電力価格予測手法を提案する。ピアソン・スピアマン・最大情報係数で特徴量を選別し、CNN-BiLSTM-Multi-Head Attentionモデルで時系列特徴を深掘りする。山西省電力市場の実データで検証し、RMSE6.64%を達成、比較手法より平均5.32%改善した。再エネの市場取引参加に高精度な価格支援を提供する。
English
This paper proposes a day-ahead electricity price forecasting method integrating market boundary conditions and the electricity carbon footprint factor. It screens key features via Pearson, Spearman, and MIC, then applies a CNN-BiLSTM-Multi-Head Attention model for deep time-series mining. Using Shanxi market data, it achieves 6.64% RMSE, a 5.32% average improvement over benchmarks, supporting renewable participation in power markets.
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
炭素フットプリントを価格予測に組み込む視点は、日本での再生可能エネルギー市場統合や非化石価値取引、電力市場改革の議論に示唆を与える。ただし中国市場固有の制度前提が強く、日本の制度への直接適用には注意が必要。
In the global GX context
Integrating carbon footprint into price forecasting aligns with global efforts to link carbon intensity and power market signals, relevant to EU ETS and renewable integration debates. It offers a methodological template for markets where carbon factors increasingly influence electricity prices.
👥 読者別の含意
🔬研究者:炭素要因を組み込んだ電力価格予測の特徴量選択と深層学習アーキテクチャの実証例として参考になる。
🏢実務担当者:電力取引戦略や再エネ調達の価格リスク評価に炭素フットプリント要因を組み込む際の示唆を提供する。
🏛政策担当者:炭素価格シグナルと電力市場設計の連動を検討する際の定量的根拠として参照可能。
📄 Abstract(原文)
Day-ahead electricity price forecasting serves as a crucial foundation for power market entities to devise trading strategies and optimize profits. Addressing the issues of traditional forecasting methods relying heavily on historical data, lacking sufficient characterization of market operating conditions, and neglecting carbon emission factors, this paper proposes a dayahead electricity price forecasting method that integrates market boundary conditions and carbon footprint factors. Firstly, based on Pearson, Spearman, and Maximum Information Coefficient, the correlation between power market boundary conditions and electricity prices is analyzed to screen out key influencing features, while ensuring the key information and reducing the complexity of the model. Besides, the electricity carbon footprint factor is introduced to reflect the impact of energy structure on electricity prices. Then, a CNN-BiLSTM-Multi-Head Attention combined forecasting model is constructed to achieve deep mining of time series features. Taking the actual operating data of Shanxi electricity market as an example for case study, the results show that the proposed method achieves a root mean square error of 6.64% for day-ahead electricity price forecasting, with an average improvement of 5.32% compared to various comparative methods. The research findings can provide high-precision electricity price support for new energy participation in electricity market transactions.
🔗 Provenance — このレコードを発見したソース
- semanticscholar https://doi.org/10.1109/icpsasia70813.2026.11692308first seen 2026-09-26 05:05:49 · last seen 2026-09-29 05:14:27
🔔 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。