基于 Newton-GNN 的天然气管道系统 稳态仿真方法.
Saved in:
| Title: | 基于 Newton-GNN 的天然气管道系统 稳态仿真方法. |
|---|---|
| Alternate Title: | A steady-state simulation method for natural gas pipeline systems based on Newton-GNN. |
| Authors: | 杨毅1 yangyi@pipechina.com.cn, 刁洪涛1, 陈洁1, 侯本权1, 贾文龙2, 林友志2, 李长俊2 |
| Source: | Chemical Engineering of Oil & Gas / Shi You Yu Tian Ran Qi Hua Gong. Jun2026, Vol. 55 Issue 3, p164-174. 11p. |
| Subject Terms: | *Graph neural networks, *Newton-Raphson method, *Generalization, *Natural gas pipelines, *Pipeline maintenance & repair, *Computer simulation |
| Abstract (English): | Objective Simulation constitutes a core technology for intelligent pipeline network dispatching. Data-driven approaches leverage large-scale datasets and pre-trained models to significantly reduce real-time simulation latency. Method This study introduces a steady-state simulation framework based on graph neural networks (GNN). First, a graph representation of the pipeline system was constructed using cross-sections and adjacency matrices. Subsequently, inspired by the Newton-Raphson method widely employed in mechanistic simulations, the original GNN was reconstructed by designing a linear incremental update function, leading to the development of the Newton-GNN model. Finally, pipeline system simulation was achieved by mimicking the inverse Jacobian matrix and linear iteration steps through learned approximations. Result Validation demonstrates that: First, across 6 000 diverse cases, the proposed model enhances accuracy by 72.3% compared to the baseline GNN, with a relative deviation as low as 0.06%. Second, in generalization tests, the model adapts effectively to various scenarios with parameter variations. Third, in two case studies, including the XYX and FR-SH sections, the proposed model yields relative deviations of 2.17% and 1.70% for mainline pressure predictions, respectively. Conclusion The Newton-GNN model exhibits high accuracy and strong generalization ability, making it promising for large-scale natural gas pipeline network simulations. [ABSTRACT FROM AUTHOR] |
| Abstract (Chinese): | 目的 仿真是管网智能调度的核心技术, 数据驱动方法利用海量数据和前置训练, 可以显著缩短实时仿真时间。 方法 提出基于图神经网络 (graph neural network, GNN) 的稳态仿真方法。首先, 利用管道截面和邻接矩阵构建图结构; 其次, 受机理仿真的牛顿迭代法启发, 重构原始 GNN, 设计了线性增量更新函数, 提出了 Newton-GNN 模型; 最后, 通过模型学习 并比拟雅可比矩阵求逆与线性迭代过程, 实现管道系统仿真。结果 ①在 6000 组案例中, 模型较原始 GNN 精度提升了 72.3% (相对偏差 0.06% );②泛化性分析中, 模型可以适应多种参数变化; ③在 XYX 和 FR-SH 段两个实例验证中, 主干线压 力相对偏差为 2.17% 和 1.70%。结论 该模型具有较高精度和优异的泛化性, 可应用于大规模天然气管网系统仿真。 [ABSTRACT FROM AUTHOR] |
| Database: | Energy & Power Source |
| FullText | Links: – Type: pdflink Text: Availability: 0 |
|---|---|
| Header | DbId: enr DbLabel: Energy & Power Source An: 195131745 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: 基于 Newton-GNN 的天然气管道系统 稳态仿真方法. – Name: TitleAlt Label: Alternate Title Group: TiAlt Data: A steady-state simulation method for natural gas pipeline systems based on Newton-GNN. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22杨毅%22">杨毅</searchLink><relatesTo>1</relatesTo><i> yangyi@pipechina.com.cn</i><br /><searchLink fieldCode="AR" term="%22刁洪涛%22">刁洪涛</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22陈洁%22">陈洁</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22侯本权%22">侯本权</searchLink><relatesTo>1</relatesTo><br /><searchLink fieldCode="AR" term="%22贾文龙%22">贾文龙</searchLink><relatesTo>2</relatesTo><br /><searchLink fieldCode="AR" term="%22林友志%22">林友志</searchLink><relatesTo>2</relatesTo><br /><searchLink fieldCode="AR" term="%22李长俊%22">李长俊</searchLink><relatesTo>2</relatesTo> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22Chemical+Engineering+of+Oil+%26+Gas+%2F+Shi+You+Yu+Tian+Ran+Qi+Hua+Gong%22">Chemical Engineering of Oil & Gas / Shi You Yu Tian Ran Qi Hua Gong</searchLink>. Jun2026, Vol. 55 Issue 3, p164-174. 11p. – Name: Subject Label: Subject Terms Group: Su Data: *<searchLink fieldCode="DE" term="%22Graph+neural+networks%22">Graph neural networks</searchLink><br />*<searchLink fieldCode="DE" term="%22Newton-Raphson+method%22">Newton-Raphson method</searchLink><br />*<searchLink fieldCode="DE" term="%22Generalization%22">Generalization</searchLink><br />*<searchLink fieldCode="DE" term="%22Natural+gas+pipelines%22">Natural gas pipelines</searchLink><br />*<searchLink fieldCode="DE" term="%22Pipeline+maintenance+%26+repair%22">Pipeline maintenance & repair</searchLink><br />*<searchLink fieldCode="DE" term="%22Computer+simulation%22">Computer simulation</searchLink> – Name: Abstract Label: Abstract (English) Group: Ab Data: Objective Simulation constitutes a core technology for intelligent pipeline network dispatching. Data-driven approaches leverage large-scale datasets and pre-trained models to significantly reduce real-time simulation latency. Method This study introduces a steady-state simulation framework based on graph neural networks (GNN). First, a graph representation of the pipeline system was constructed using cross-sections and adjacency matrices. Subsequently, inspired by the Newton-Raphson method widely employed in mechanistic simulations, the original GNN was reconstructed by designing a linear incremental update function, leading to the development of the Newton-GNN model. Finally, pipeline system simulation was achieved by mimicking the inverse Jacobian matrix and linear iteration steps through learned approximations. Result Validation demonstrates that: First, across 6 000 diverse cases, the proposed model enhances accuracy by 72.3% compared to the baseline GNN, with a relative deviation as low as 0.06%. Second, in generalization tests, the model adapts effectively to various scenarios with parameter variations. Third, in two case studies, including the XYX and FR-SH sections, the proposed model yields relative deviations of 2.17% and 1.70% for mainline pressure predictions, respectively. Conclusion The Newton-GNN model exhibits high accuracy and strong generalization ability, making it promising for large-scale natural gas pipeline network simulations. [ABSTRACT FROM AUTHOR] – Name: Abstract Label: Abstract (Chinese) Group: Ab Data: 目的 仿真是管网智能调度的核心技术, 数据驱动方法利用海量数据和前置训练, 可以显著缩短实时仿真时间。 方法 提出基于图神经网络 (graph neural network, GNN) 的稳态仿真方法。首先, 利用管道截面和邻接矩阵构建图结构; 其次, 受机理仿真的牛顿迭代法启发, 重构原始 GNN, 设计了线性增量更新函数, 提出了 Newton-GNN 模型; 最后, 通过模型学习 并比拟雅可比矩阵求逆与线性迭代过程, 实现管道系统仿真。结果 ①在 6000 组案例中, 模型较原始 GNN 精度提升了 72.3% (相对偏差 0.06% );②泛化性分析中, 模型可以适应多种参数变化; ③在 XYX 和 FR-SH 段两个实例验证中, 主干线压 力相对偏差为 2.17% 和 1.70%。结论 该模型具有较高精度和优异的泛化性, 可应用于大规模天然气管网系统仿真。 [ABSTRACT FROM AUTHOR] |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=enr&AN=195131745 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.3969/j.issn.1007-3426.2026.03.018 Languages: – Code: chi Text: Chinese PhysicalDescription: Pagination: PageCount: 11 StartPage: 164 Subjects: – SubjectFull: Graph neural networks Type: general – SubjectFull: Newton-Raphson method Type: general – SubjectFull: Generalization Type: general – SubjectFull: Natural gas pipelines Type: general – SubjectFull: Pipeline maintenance & repair Type: general – SubjectFull: Computer simulation Type: general Titles: – TitleFull: 基于 Newton-GNN 的天然气管道系统 稳态仿真方法. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: 杨毅 – PersonEntity: Name: NameFull: 刁洪涛 – PersonEntity: Name: NameFull: 陈洁 – PersonEntity: Name: NameFull: 侯本权 – PersonEntity: Name: NameFull: 贾文龙 – PersonEntity: Name: NameFull: 林友志 – PersonEntity: Name: NameFull: 李长俊 IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 06 Text: Jun2026 Type: published Y: 2026 Identifiers: – Type: issn-print Value: 10073426 Numbering: – Type: volume Value: 55 – Type: issue Value: 3 Titles: – TitleFull: Chemical Engineering of Oil & Gas / Shi You Yu Tian Ran Qi Hua Gong Type: main |
| ResultId | 1 |
