自动化学报2017，Vol.43Issue(8)：1370-1382,13.DOI:10.16383/j.aas.2017.c160435
基于大维数据驱动的油气管网泄漏监控模糊决策方法
A Large Dimensional Data-driven Fuzzy Detection Method for Oil-gas Pipeline Network Leakage
摘要
Abstract
It is difficult to build an exact math model for the oil-gas pipeline network because there are multiple state variables and complex processing technology. A leakage detection method using fuzzy detection based on large dimensional data for oil pipeline network is proposed, which can monitor oil pipeline network operational status in real time. Firstly, from the view-point of cyber-physical system (CPS), existing oil pipeline network data containing network topology, physical data like valve opening, and operating data like pressure and flow are used to build a data-driven model without dimensionality reduction. Then, as for fuzzy decision conditions, the eigenvalue spectrum distribution and ring law based on spectral analysis of large dimensional random matrices are used to judge the network operating status. When the topological structure of oil pipeline network is dynamically altered, the fuzzy method can solve the problem of high rate false alarm. Finally, the effectiveness of the proposed method is demonstrated through simulation and case study.关键词
大维数据驱动/信息物理系统/谱分析/模糊决策/油气管网Key words
Large dimensional data-driven/cyber-physical system (CPS)/spectral analysis/fuzzy detection/oil-gas pipeline network引用本文复制引用
马大中,胡旭光,孙秋野..基于大维数据驱动的油气管网泄漏监控模糊决策方法[J].自动化学报,2017,43(8):1370-1382,13.基金项目
国家自然科学基金重大项目(61627809), 国家自然科学基金重点项目(61433004), 国家自然科学基金(61473069, 61573094), 中央高校基本科研业务费专项基金(N160404005) 资助Supported by the Major Program of National Natural Science Foundation of China (61627809), the Key Program of National Natural Science Foundation of China (61433004), National Natural Science Foundation of China (61473069, 61573094) and Fundamental Research Funds for the Central Universities (N160404005) （61627809）
