摘要：This paper proposes a method for optimizing the supply resilience of natural gas pipeline networks, driven by demand-side dynamics and data science. The method is divided into two main components: user demand characteristic modeling and system supply resilience optimization modeling. In the user demand characteristic modeling phase, preprocessed user demand data is used, combining the Tabular Variational Autoencoder (TVAE) with probability density distribution curve fitting to provide an in-depth characterization of user demand patterns. For the system supply resilience optimization modeling, constraints are established based on the functional characteristics of the system＇s components, and specific objective functions are designed for different operational scenarios. Additionally, the Latin Hypercube Sampling (LHS) method is employed to capture fluctuations in user demand. Finally, this paper introduces a set of evaluation indicators for gas supply resilience and validates the proposed methodology through five scenario-based case studies. The results confirm the effectiveness and feasibility of this approach in improving the resilience of natural gas pipeline systems. ? 2025 Elsevier Ltd

ISSN号：0951-8320

卷、期、页：卷261

发表日期：2025-09-01

期刊分区（SCI为中科院分区）：一区

收录情况：SCI（科学引文索引印刷版）,EI（工程索引）,SCIE（科学引文索引网络版）

发表期刊名称：Reliability Engineering and System Safety

参与作者：麦克

通讯作者：赵志伟

第一作者：杨兆铭,苏怀,张劲军

论文类型：期刊论文

论文概要：赵志伟,杨兆铭,苏怀,麦克,张劲军，A methodology of natural gas pipeline network system supply resilience optimization: Based on demand-side and data science-driven approach，Reliability Engineering and System Safety，2025，卷261

论文题目：A methodology of natural gas pipeline network system supply resilience optimization: Based on demand-side and data science-driven approach