Integrating Teletraffic Theory with Neural Networks for Quality-of-Service Evaluation in Mobile Networks DOI
Yin-Chi Chan, Jingjin Wu, Eric W. M. Wong

et al.

Published: Jan. 1, 2023

In mobile cellular design, one important quality-of-service metric is the blocking probability. Using computer simulation for studying probability quite time-consuming. Furthermore, existing teletraffic models such as Information Exchange Surrogate Approximation (IESA) only give a rough estimate of Another common approach, direct evaluation using neural networks (NN), performs poorly when extrapolating to network conditions outside training set. This paper addresses shortcomings and NN-based approaches by combining both approaches, creating what we call IESA-NN. IESA-NN, an NN used tuning parameter, which in turn via modified IESA approach. other words, approach still forms core with techniques improve accuracy parameter. Simulation results show that IESA-NN better than previous based on or theory alone. particular, even cannot produce good value example not experienced set, final generally accurate due bounds set underlying theory.

Language: Английский

Prediction of oil and gas pipeline failures through machine learning approaches: A systematic review DOI Creative Commons
Abdulnaser M. Al-Sabaeei, Hitham Alhussian, Said Jadid Abdulkadir

et al.

Energy Reports, Journal Year: 2023, Volume and Issue: 10, P. 1313 - 1338

Published: Aug. 16, 2023

Pipelines are vital for transporting oil and gas, but leaks can have serious consequences such as fires, injuries, pollution, property damage. Therefore, preserving pipeline integrity is crucial a safe sustainable energy supply. The rapid progress of machine learning (ML) technologies provides an advantageous opportunity to develop predictive models that effectively tackle these challenges. This review article mainly focuses on the novelty using deep techniques, specifically artificial neural networks (ANNs), support vector machines (SVMs) hybrid (HML) algorithms, predicting different failures in gas industry. In contrast existing noncomprehensive reviews defects, this explicitly addresses application ML parameters, data reliability purpose. surveys research specific area, offering coherent discussion identifying motivations challenges associated with types defects pipelines. also includes bibliometric analysis literature, highlighting common investigated failures, experimental tests. It in-depth details, summarized tables, failure types, commonly used resources, critical discussions. Based comprehensive aforementioned, it was found approaches, ANNs SVMs, accurately predict compared conventional methods. However, highly recommended combine multiple algorithms enhance accuracy prediction time further. Comparing based field, experimental, simulation various establish reliable cost-effective monitoring systems entire network. systematic expected aid understanding gaps provide options other researchers interested failures.

Language: Английский

Citations

33
Prediction of external corrosion rate for buried oil and gas pipelines: a novel deep learning with DNN and attention mechanism method DOI

Yu Guang,

Wenhe Wang, Hongwei Song

et al.

International Journal of Pressure Vessels and Piping, Journal Year: 2024, Volume and Issue: 209, P. 105218 - 105218

Published: May 24, 2024

Language: Английский

Citations

14
Residual strength prediction of corroded pipelines based on physics-informed machine learning and domain generalization DOI Creative Commons
Tingting Wu, Xingyuan Miao,

Fulin Song

et al.

npj Materials Degradation, Journal Year: 2025, Volume and Issue: 9(1)

Published: Feb. 14, 2025

Language: Английский

Citations

1
Energy pipeline degradation condition assessment using predictive analytics – challenges, issues, and future directions DOI Creative Commons
Muhammad Nihal Hussain, Tieling Zhang, Richard Dwight

et al.

Journal of Pipeline Science and Engineering, Journal Year: 2024, Volume and Issue: 4(3), P. 100178 - 100178

Published: Feb. 13, 2024

It is of paramount importance to ensure the safe operation energy pipelines for pipeline owners and operators. Therefore, effective condition assessment imperative. For this purpose, there are a great number models developed using various techniques. How select modeling approach associated techniques get out most effectiveness model under with limited monitoring data experience remains big concern This paper provides comprehensive review approaches degradation assessment. The primary motivation behind pivotal role in integrity management proliferation techniques, including statistical modeling, stochastic processes, machine learning, deep used assessing degradation. work aims identify assess challenges gaps inherent utilization these approaches. By systematically analyzing current state research practice, not only highlights strengths limitations but also offers insights into future opportunities enhancing practice field management. Our analysis valuable researchers, practitioners, policymakers domain facilitates better understanding complexities intricacies assessment, ultimately contributing development more robust strategies safeguarding pipelines.

Language: Английский

Citations

6
Corroded submarine pipeline degradation prediction based on theory-guided IMOSOA-EL model DOI
Xingyuan Miao, Hong Zhao

Reliability Engineering & System Safety, Journal Year: 2023, Volume and Issue: 243, P. 109902 - 109902

Published: Dec. 21, 2023

Language: Английский

Citations

15
Maximum pitting corrosion depth prediction of buried pipeline based on theory-guided machine learning DOI
Xingyuan Miao, Hong Zhao

International Journal of Pressure Vessels and Piping, Journal Year: 2024, Volume and Issue: 210, P. 105259 - 105259

Published: July 11, 2024

Language: Английский

Citations

4
Corrosion Area Detection and Depth Prediction using Machine Learning DOI Creative Commons

Eun-Young Son,

Dayeon Jeong,

Minjae Oh

et al.

International Journal of Naval Architecture and Ocean Engineering, Journal Year: 2024, Volume and Issue: 16, P. 100617 - 100617

Published: Jan. 1, 2024

Language: Английский

Citations

4
A mechanistic model for flow accelerated corrosion prediction of a 90° carbon steel elbow in CO2 environments DOI Creative Commons

Dongrun Li,

Fang Zheng,

Tai Ma

et al.

Journal of Pipeline Science and Engineering, Journal Year: 2025, Volume and Issue: unknown, P. 100258 - 100258

Published: Jan. 1, 2025

Language: Английский

Citations

0
A novel multimodal deep learning framework for predicting residual strength of corroded rectangular hollow-section columns DOI
Yujia Zhang, Yu Zhou, Yu Zhou

et al.

Engineering Applications of Artificial Intelligence, Journal Year: 2025, Volume and Issue: 149, P. 110554 - 110554

Published: March 22, 2025

Language: Английский

Citations

0
Explainable Ensemble Learning Model for Residual Strength Forecasting of Defective Pipelines DOI Creative Commons

Hongbo Liu,

Xiangzhao Meng

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(7), P. 4031 - 4031

Published: April 6, 2025

The accurate prediction of the residual strength defective pipelines is a critical prerequisite for ensuring safe operation oil and gas pipelines, it holds significant implications pipeline’s remaining service life preventive maintenance. Traditional machine learning algorithms often fail to comprehensively account correlative factors influencing exhibit limited capability in extracting nonlinear features from data, suffer insufficient predictive accuracy. Furthermore, models typically lack interpretability. To address these issues, this study proposes hybrid model based on Bayesian optimization (BO) eXtreme Gradient Boosting (XGBoost). This approach resolves issues excessive iterations high computational costs associated with conventional hyperparameter methods, significantly enhancing model’s performance. performance evaluated using mainstream metrics such as Mean Absolute Percentage Error (MAPE), Coefficient Determination (R2), Root Square (RMSE), robustness analysis, overfitting grey relational analysis. enhance interpretability predictions, reveal significance features, confirm prior domain knowledge, Shapley additive explanations (SHAP) are employed conduct relevant research. results indicate that, compared Random Forest, LightGBM, Support Vector Machine, gradient boosting regression tree, Multi-Layer Perceptron, BO-XGBoost exhibits best performance, MAPE, R2, RMSE values 5.5%, 0.971, 1.263, respectively. Meanwhile, proposed demonstrates highest robustness, least tendency overfitting, most relation degree value. SHAP analysis reveals that ranked descending order importance, defect depth (d), wall thickness (t), yield (σy), external diameter (D), length (L), tensile (σu), width (w). development contributes improving integrity management provides decision support intelligent fields.

Language: Английский

Citations

0