Hybrid multiscale forecasting of SRU sulfur gas concentrations using VMD CEEMDAN and optimized PatchTST
VMD、CEEMDAN、最適化PatchTSTを用いたSRU硫黄ガス濃度のハイブリッドマルチスケール予測 (AI 翻訳)
Wenzhe Sun, Longhao Li, Binglin Lu, Lijun Jiang, Jie Zhang
🤖 gxceed AI 要約
日本語
本研究は、硫黄回収装置（SRU）におけるH2SおよびSO2濃度の高精度予測のため、VMD、CEEMDAN、PatchTST-PSA、PIMOを組み合わせたマルチスケールフレームワークを提案。イタリアの精油所データで検証し、6つのベースラインモデルより優れた性能を示した。
English
This study proposes a hybrid multiscale framework combining VMD, CEEMDAN, an enhanced PatchTST with ProbSparse attention, and projection iterative optimization for forecasting H2S and SO2 concentrations in SRUs. Experiments on Italian refinery data show improved RMSE, MAE, MAPE, and R2 over six baselines, demonstrating robustness for low-carbon industrial operations.
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
日本の製油所においてもSRUの低炭素運転は重要だが、本論文は工学的な予測手法に焦点を当てており、GX政策や開示枠組みとの直接的な接点は少ない。
In the global GX context
This paper addresses process optimization for low-carbon sulfur recovery, but it is a technical engineering study with limited direct relevance to global GX disclosure or climate finance.
👥 読者別の含意
🔬研究者:A novel hybrid model combining VMD, CEEMDAN, and PatchTST for time-series forecasting in industrial processes.
🏢実務担当者:Can be applied to optimize SRU operations for energy efficiency and emissions reduction.
📄 Abstract（原文）
Summary The sulfur recovery process in SRUs is highly nonlinear and non-stationary, making accurate forecasting of H2S and SO2 concentrations challenging yet crucial for efficient, low-carbon operation. Many existing models fail to handle multi-scale fluctuations, high-frequency noise, and complex variable couplings, limiting their accuracy. This study presents a multi-scale framework combining variational mode decomposition (VMD), complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), an enhanced patch time-series transformer with ProbSparse attention (PatchTST-PSA), and projection iterative modeling optimization (PIMO). VMD decomposes the concentration series into intrinsic mode functions, and CEEMDAN suppresses noise while preserving dynamics. PatchTST-PSA captures nonlinear variable interactions, while PIMO optimizes hyperparameters. Experiments on SRU data from an Italian refinery demonstrate that the framework provides improved results in RMSE, MAE, MAPE, and R2 compared to six baseline models, highlighting its robustness and industrial relevance.
🔗 Provenance — このレコードを発見したソース
- semanticscholar https://doi.org/10.1016/j.isci.2026.114986first seen 2026-05-06 00:13:42
🔔 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。