Pressure and Temperature Prediction of Oil Pipeline Networks Based on a Mechanism-Data Hybrid Driven Method DOI Creative Commons
Faming Gong, Xingfang Zhao, Chengze Du

et al.

Information, Journal Year: 2024, Volume and Issue: 15(11), P. 709 - 709

Published: Nov. 5, 2024

To ensure the operational safety of oil transportation stations, it is crucial to predict impact pressure and temperature before crude enters pipeline network. Accurate predictions enable assessment pipeline’s load-bearing capacity prevention potential incidents. Most existing studies primarily focus on describing modeling mechanisms flow process. However, monitoring data can be skewed by factors such as instrument aging friction, leading inaccurate when relying solely mechanistic or data-driven approaches. address these limitations, this paper proposes a Temporal-Spatial Three-stream Temporal Convolutional Network (TS-TTCN) model that integrates knowledge with methods. Building upon Networks (TCN), TS-TTCN synthesizes insights into transport process establish hybrid driving mechanism. In temporal dimension, incorporates real-time operating parameters applies convolution techniques capture time-series characteristics spatial constructs directed topological map based network’s node structure characterize features. Data analysis experimental results show (TTCN) model, which uses Tanh activation function, achieves an error rate below 5%. By analyzing validating from Dongying station, proposed proves more stable, reliable, accurate under varying conditions.

Language: Английский

Power prediction considering NWP wind speed error tolerability: A strategy to improve the accuracy of short-term wind power prediction under wind speed offset scenarios DOI
Mao Yang, Yunfeng Guo, Tao Huang

et al.

Applied Energy, Journal Year: 2024, Volume and Issue: 377, P. 124720 - 124720

Published: Oct. 19, 2024

Language: Английский

Citations

8
Forecasting China's agricultural carbon emissions: A comparative study based on deep learning models DOI Creative Commons
Tiantian Xie, Zetao Huang, Tao Tan

et al.

Ecological Informatics, Journal Year: 2024, Volume and Issue: 82, P. 102661 - 102661

Published: June 3, 2024

Given the critical urgency to combat escalating climate crisis and continuous rise in agricultural carbon emissions (ACE) China, accurately forecasting their future trends is crucial. This research employs emission factor method assess ACE throughout mainland China from 1993 2021. To refine our approach, both statistical neural network methodologies were utilized pinpoint key factors influencing ACE. We crafted models incorporating deep learning techniques traditional methods. Notably, Tree-structured Parzen Estimator Bayesian Optimization (TPEBO) algorithm was applied optimize Long Short-Term Memory (LSTM) networks, culminating creation of a superior integrated TPEBO-LSTM model that demonstrated strong performance across various datasets. The outcomes suggest 24 provinces are expected reach zenith before 2030, primarily driven by farm operations, as well livestock poultry manure management. result provides significant tool for assessing different regions, offering insights crucial targeted mitigation strategies.

Language: Английский

Citations

7
A Novel Transformer-CNN Approach for Predicting Soil Properties from LUCAS Vis-NIR Spectral Data DOI Creative Commons
Liying Cao, Miao Sun, Zhicheng Yang

et al.

Agronomy, Journal Year: 2024, Volume and Issue: 14(9), P. 1998 - 1998

Published: Sept. 2, 2024

Soil, a non-renewable resource, requires continuous monitoring to prevent degradation and support sustainable agriculture. Visible-near-infrared (Vis-NIR) spectroscopy is rapid cost-effective method for predicting soil properties. While traditional machine learning methods are commonly used modeling Vis-NIR spectral data, large datasets may benefit more from advanced deep techniques. In this study, based on the library LUCAS, we aimed enhance regression model performance in property estimation by combining Transformer convolutional neural network (CNN) techniques predict 11 properties (clay, silt, pH CaCl2, H2O, CEC, OC, CaCO3, N, P, K). The Transformer-CNN accurately predicted most properties, outperforming other (partial least squares (PLSR), random forest (RFR), vector (SVR), Long Short-Term Memory (LSTM), ResNet18) with 10–24 percentage point improvement coefficient of determination (R2). excelled N (R2 = 0.94–0.96, RPD > 3) performed well clay, sand, K 0.77–0.85, 2 < 3). This study demonstrates potential enhancing prediction, although future work should aim optimize computational efficiency explore wider range applications ensure its utility different agricultural settings.

Language: Английский

Citations

6
Efficiency assessment and scenario simulation of the water-energy-food system in the Yellow river basin, China DOI
Chenjun Zhang, Xiangyang Zhao, Changfeng Shi

et al.

Energy, Journal Year: 2024, Volume and Issue: 305, P. 132279 - 132279

Published: July 2, 2024

Language: Английский

Citations

5
Wind power prediction through acoustic data-driven online modeling and active wake control DOI

Bingchuan Sun,

Mingxu Su, Jié He

et al.

Energy Conversion and Management, Journal Year: 2024, Volume and Issue: 319, P. 118920 - 118920

Published: Aug. 17, 2024

Language: Английский

Citations

5
BiLSTM-InceptionV3-Transformer-fully-connected model for short-term wind power forecasting DOI
Linfei Yin,

Yujie Sun

Energy Conversion and Management, Journal Year: 2024, Volume and Issue: 321, P. 119094 - 119094

Published: Sept. 25, 2024

Language: Английский

Citations

5
Two-stage correction prediction of wind power based on numerical weather prediction wind speed superposition correction and improved clustering DOI Creative Commons
Mao Yang, Yunfeng Guo, Fulin Fan

et al.

Energy, Journal Year: 2024, Volume and Issue: 302, P. 131797 - 131797

Published: May 27, 2024

Language: Английский

Citations

4
A novel dynamic nonlinear non-Gaussian approach for fault detection and diagnosis DOI
Yihan Ma, Fei Ye, Dazi Li

et al.

Neurocomputing, Journal Year: 2025, Volume and Issue: unknown, P. 129403 - 129403

Published: Jan. 1, 2025

Language: Английский

Citations

0
Composite denoising-based LSTM prediction method of supercapacitor performance degradation law and remaining useful life DOI

Yanming Zhao,

Jinhao Wu,

Yongbo Zhu

et al.

Circuit World, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 14, 2025

Purpose This paper aims to reduce the impact of noise on prediction accuracy remaining useful life (RUL) for supercapacitor. First, Savitzky–Golay (SG) smoothing filter method (Savitzky and Golay, 1964) is used eliminate local small fluctuation high-frequency noises that are generated by capacity drop rebound during charging discharging process Then, variational mode decomposition (VMD) large caused internal temperature change supercapacitor chemical reaction Its parameters optimized using marine predators algorithm (MPA), sequence after denoising reconstructed. Finally, long short term memory neural networks (LSTM) predict performance degradation law (PDL) reconstructed sequence, then comparative analysis conducted with other methods, which results show this improves effectively, provides theoretical support timely accurately understanding PDL RUL backup power supply. Design/methodology/approach SG VMD MPA, LSTM accurate Findings These factors will bring different types service supply, such as regeneration, differences rate, supercapacitor, external electromagnetic interference. Therefore, proposes an supercapacitor’s based composite denoising, divided into three stages: smoothing, reduction prediction. noises, MPA-VMD nonlinear nonstationary noises. reconstructed, methods carried out. The SG-VMD-LSTM has higher accuracy, can improve safety reliability wind turbine operation under severe conditions. Originality/value

Language: Английский

Citations

0
Enhanced Electricity Forecasting for Smart Buildings Using a TCN‐Bi‐LSTM Deep Learning Model DOI Open Access
Sandeep Kumar Gautam,

V.K. Shrivastava,

Sandeep S. Udmale

et al.

Expert Systems, Journal Year: 2025, Volume and Issue: 42(3)

Published: Jan. 30, 2025

ABSTRACT Integration of sensor technology and advanced software empowers consumers to manage energy usage proactively. This proactive approach yields positive impacts at both micro macro levels, benefiting individuals contributing broader environmental conservation efforts. By leveraging predictive models, can make informed decisions that serve their interests promote a greener more sustainable future for all. Thus, consumption (EC) prediction is crucial effective resource management. In this study, we propose an innovative deep‐learning predict EC, focusing specifically on smart buildings. Our model utilises hybrid deep learning architecture effectively capture low high information patterns present in multivariate time series data various sensors deployed buildings numerous influencing factors. To address the nonlinear dynamic nature data, our combines neural network (DNN) with sequential (DLS). Specifically, temporal convolutional networks (TCN) within DNN family are employed extract trends from while DLS model, which consists Bi‐directional Long Short‐term Memory Networks (Bi‐LSTM), learn these effectively. Consequently, framework leverages related EC shared feature representation. validate approach, extensively evaluate using dataset office building Berkeley, California. Experimental results demonstrate achieves satisfactory accuracy prediction. For 7‐h horizon TS R 2 0.97 realised proposed model. confirmed by 1.65% improvement transiting univariate supports multiple modalities.

Language: Английский

Citations

0