A Temporal Network Based on Characterizing and Extracting Time Series in Copper Smelting for Predicting Matte Grade DOI Creative Commons

Junjia Zhang,

Zhuorui Li, Enzhi Wang

et al.

Sensors, Journal Year: 2024, Volume and Issue: 24(23), P. 7492 - 7492

Published: Nov. 24, 2024

Addressing the issues of low prediction accuracy and poor interpretability in traditional matte grade models, which rely on pre-smelting input assay data for regression, we incorporate process sensors' propose a temporal network based Time to Vector (Time2Vec) convolutional combined with multi-head attention (TCN-TMHA) tackle weak characteristics uncertain periodic information copper smelting process. Firstly, employed maximum coefficient (MIC) criterion select strongly correlated grade. Secondly, used Time2Vec module extract from variables, incorporates time series processing directly into model. Finally, implemented TCN-TMHA specific weighting mechanisms assign weights features prioritize relevant key step features. Experimental results indicate that proposed model yields more accurate predictions content, determination (

Language: Английский

Advancing fractured geothermal system modeling with artificial neural network and bidirectional gated recurrent unit DOI
Yuwei Li, Genbo Peng,

Tong Du

et al.

Applied Energy, Journal Year: 2024, Volume and Issue: 372, P. 123826 - 123826

Published: July 6, 2024

Language: Английский

Citations

25
Review of Machine Learning Methods for Steady State Capacity and Transient Production Forecasting in Oil and Gas Reservoir DOI Creative Commons
Dongyan Fan, S.Y. Lai, Hai Sun

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(4), P. 842 - 842

Published: Feb. 11, 2025

Accurate oil and gas production forecasting is essential for optimizing field development operational efficiency. Steady-state capacity prediction models based on machine learning techniques, such as Linear Regression, Support Vector Machines, Random Forest, Extreme Gradient Boosting, effectively address complex nonlinear relationships through feature selection, hyperparameter tuning, hybrid integration, achieving high accuracy reliability. These maintain relative errors within acceptable limits, offering robust support reservoir management. Recent advancements in spatiotemporal modeling, Physics-Informed Neural Networks (PINNs), agent-based modeling have further enhanced transient forecasting. Spatiotemporal capture temporal dependencies spatial correlations, while PINN integrates physical laws into neural networks, improving interpretability robustness, particularly sparse or noisy data. Agent-based complements these techniques by combining measured data with numerical simulations to deliver real-time, high-precision predictions of dynamics. Despite challenges computational scalability, sensitivity, generalization across diverse reservoirs, future developments, including multi-source lightweight architectures, real-time predictive capabilities, can improve forecasting, addressing the complexities supporting sustainable resource management global energy security.

Language: Английский

Citations

0
EUR Prediction for Shale Gas Wells Based on the ROA-CatBoost-AM Model DOI Creative Commons
Wenping He, Xizhe Li, Yujin Wan

et al.

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(4), P. 2156 - 2156

Published: Feb. 18, 2025

Shale gas is a critical energy resource, and estimating its ultimate recoverable reserves (EUR) key indicator for evaluating the development potential effectiveness of wells. To address challenges in accurately predicting shale EUR, this study analyzed production data from 200 wells CN block. Sixteen factors influencing EUR were considered, geological, engineering, identified using Spearman correlation analysis mutual information methods to exclude highly linearly correlated variables. An attention mechanism was introduced weight input features prior model training, enhancing interpretability feature contributions. The hyperparameters optimized Rabbit Optimization Algorithm (ROA), 10-fold cross-validation employed improve stability reliability evaluation, mitigating overfitting bias. performance four machine learning models compared, optimal selected. results indicated that ROA-CatBoost-AM exhibited superior both fitting accuracy prediction effectiveness. This subsequently applied identifying primary controlling productivity, providing effective guidance practices. dominant forecasts determined by offer valuable references optimizing block strategies.

Language: Английский

Citations

0
Artificial intelligence-driven financial innovation: A robo-advisor system for robust returns across diversified markets DOI
Qing Zhu, Chenyu Han,

Shan Liu

et al.

Expert Systems with Applications, Journal Year: 2025, Volume and Issue: unknown, P. 126881 - 126881

Published: Feb. 1, 2025

Language: Английский

Citations

0
An Intelligent Selection Method of Main Controlling Factors for Tight Gas Reservoirs Productivity Based on Improved Harris Hawk Algorithm DOI
Xiangyu Fan,

Jia Xu,

Chunlan Zhao

et al.

Energy & Fuels, Journal Year: 2025, Volume and Issue: unknown

Published: March 21, 2025

Language: Английский

Citations

0
A Hybrid Tabular-Spatial-Temporal Model with 3D Geomodel for Production Prediction in Shale Gas Formations DOI
Muming Wang, Hai Wang, Gang Hui

et al.

SPE Journal, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 13

Published: March 1, 2025

Summary The evolution of shale gas production has reshaped North America’s energy profile. Using the vast amounts data generated from and operations, machine learning offers significant advantages in forecasting performance optimization. In this study, we propose a pioneering hybrid model that integrates tabular, spatial, temporal modalities to enhance unconventional reservoirs. Despite traditional methods, such as artificial neural networks (ANN) extreme gradient boosting (XGBoost), which rely solely on tabular for training prediction, novel 3D parameterization method. This approach tokenizes formation property distribution into three-axis tensors, enabling more comprehensive representation spatial data. For established 3D-convolutional network (3D-CNN) with an attention mechanism module process created modality, used long short-term memory (LSTM) accept dynamic input predict monthly simultaneously. Data total 677 wells Duvernay Formation were collected, preprocessed, fed according based their modality. results show combining three achieved impressive level accuracy, coefficient determination (R2) 0.8771, surpassing (0.7841) tabular-spatial (0.8230) modality models. addition, applied global optimization further by optimizing architecture each hyperparameters, 1.88% improvement empirical design. These advancements set new benchmark predictive modeling reservoirs, highlighting importance using different improving forecast prediction.

Language: Английский

Citations

0
Forecasting day-ahead electric power prices with functional data analysis DOI Creative Commons
Faheem Jan, Hasnain Iftikhar, Muhammad Junaid Tahir

et al.

Frontiers in Energy Research, Journal Year: 2025, Volume and Issue: 13

Published: March 28, 2025

Day-ahead electricity prices in today’s competitive electric power markets have complex features such as high frequency, volatility, non-linearity, non-stationarity, mean reversion, multiple periodicities, and calendar effects. These complicated make price forecasting difficult. To address this, this research examines the application of functional data analysis to day-ahead prices. Compared classical time series approaches, is more appealing since it anticipates daily profile, allowing for short-term projections. This technique uses a autoregressive ( F AR) with exogenous predictors id="m2">X ) model predict next-day In addition, standard time-series models, including (AR) id="m4">

Language: Английский

Citations

0
Prediction of Waste Sludge Production in Municipal Wastewater Treatment Plants by Deep-Learning Algorithms with Antioverfitting Strategies DOI
Juanjuan Chen, Weixiang Chao, Yixuan Wang

et al.

ACS ES&T Engineering, Journal Year: 2025, Volume and Issue: unknown

Published: April 3, 2025

Language: Английский

Citations

0
An Innovative Method for Evaluating the Fracturability of Continental Shale Oil Reservoirs Based on Machine Learning and Logging Data DOI

Yu Mei,

Junbin Chen, Yin Qi

et al.

SPE Journal, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 23

Published: April 1, 2025

Summary Fracturability evaluation is an important task before hydraulic fracturing, and machine learning (ML) methods have been applied to petroleum-related studies but not fracturability evaluation. In this study, we present a novel workflow based on ML that focuses the Chang 7 continental shale oil reservoirs in Ordos Basin, aiming generate comprehensive index (KC) guide selection of sections clusters fracturing. Owing lithological differences between reservoirs, brittleness-based are poorly adapted research area. Hence, integrate geological sweet spots, brittleness, difficulty forming complex fracture network reservoir. The typical factors influencing include saturation (SOG), porosity (POR), permeability (PERM), Young’s modulus (YM), Poisson’s ratio (PR), Mode-I toughness (KIC), Mode-II (KIIC), horizontal stress difference coefficient (HSDC). Furthermore, powerful nonlinear dimension reduction capability kernel principal component analysis (KPCA) used main characteristics each effect. To verify adaptability KPCA-based method, KC compared with logging interpretations microseismic events. Considering substantial spatial correlation data, hybrid neural [convolutional (CNN)-multihead attention (MHA)-bidirectional long short-term memory (BiLSTM)] presented simplify intermediate computation procedure directly use data predict KC. CNN excels at extracting local features, MHA enables model focus more task-relevant BiLSTM captures bidirectional dependencies data. experimental results show CNN-MHA-BiLSTM outperforms other networks testing set can better handle hidden patterns.

Language: Английский

Citations

0
Spatio-temporal prediction of total energy consumption in multiple regions using explainable deep neural network DOI
Shiliang Peng, Lin Fan, Zhang Li

et al.

Energy, Journal Year: 2024, Volume and Issue: 301, P. 131526 - 131526

Published: May 3, 2024

Language: Английский

Citations

3