|本期目录/Table of Contents|

[1]姚志强.基于深度置信网络的管网泄漏故障诊断方法研究[J].中国安全生产科学技术,2018,14(4):101-106.[doi:10.11731/j.issn.1673-193x.2018.04.016]
 YAO Zhiqiang.Study on fault diagnosis method of pipelines leakage based on deep belief network[J].JOURNAL OF SAFETY SCIENCE AND TECHNOLOGY,2018,14(4):101-106.[doi:10.11731/j.issn.1673-193x.2018.04.016]
点击复制

基于深度置信网络的管网泄漏故障诊断方法研究
分享到:

《中国安全生产科学技术》[ISSN:1673-193X/CN:11-5335/TB]

卷:
14
期数:
2018年4期
页码:
101-106
栏目:
职业安全卫生管理与技术
出版日期:
2018-04-30

文章信息/Info

Title:
Study on fault diagnosis method of pipelines leakage based on deep belief network
文章编号:
1673-193X(2018)-04-0101-06
作者:
姚志强
(中国安全生产科学研究院,北京 100012)
Author(s):
YAO Zhiqiang
(China Academy of Safety Science and Technology, Beijing 100012, China)
关键词:
深度置信网络受限制玻尔兹曼机Softmax分类器对比散度BP
Keywords:
deep belief network restricted Boltzmann machine Softmax classifier contrast divergence BP
分类号:
X937
DOI:
10.11731/j.issn.1673-193x.2018.04.016
文献标志码:
A
摘要:
针对当前管网系统数据量大不利于传统模型方法诊断故障的问题,设计了1种基于深度置信网络的管网故障诊断算法。首先,对管网数据结构以及管网系统运行状态进行分析,选取管网主要数据作为故障诊断网络的输入,确定相应运行状态作为诊断网络输出;其次,设计了基于多个受限制玻尔兹曼机与Softmax分类器级联的深度置信网络,并且利用对比散度算法和BP算法对模型进行预训练与调优,使模型参数达到全局最优;最后,通过实验测试确定所设计的深度置信网络的训练迭代次数与网络层数,使算法诊断准确率达到最优。研究结果表明:提出的基于深度置信网络的管网故障诊断算法对管网故障诊断可以达到良好的诊断结果,泄漏预测准确率在验证集样本上可达96.87%,在管网泄漏检测方面,相较于传统基于模型的方法优势明显。
Abstract:
In view of the large amount of data in the current pipelines system,which is unfavorable to the fault diagnosis of the traditional model method,a pipelines fault diagnosis algorithm based on the deep belief network was designed. Firstly,the data structure of the pipelines network and the operating state of the pipelines system were analyzed,and the main data of the pipelines were selected as the input of the fault diagnosis network,at the same time,the corresponding operating state was determined as the output of the diagnostic network. Secondly,a deep belief network based on the cascade of multiple restricted Boltzmann machines and Softmax classifier was designed,and the model was pre-trained and tuned by using the contrast divergence algorithm and BP algorithm,so as to make the model parameters be the global optimum. Finally,the training iteration times and network layer number of the designed deep belief network were determined by the experimental tests,so as to make the diagnosis accuracy of the algorithm be the optimal. The research results showed that the proposed pipelines fault diagnosis algorithm based on the deep belief network could achieve good diagnosis results for the pipelines fault diagnosis. The accuracy of leakage prediction could reach 96.87% on the verification set sample,and the advantages were obvious compared with the traditional method based on the model in the pipelines leakage detection.

参考文献/References:

[1]张乔. 石油天然气长输管道泄漏检测及定位探讨[J]. 价值工程, 2015(3):75-76. ZHANG Qiao. Discuss on leakage detection and location of long distance pipeline of oil and gas[J]. Value Engineering,2015(3):75-76.
[2]李云鹏,阚哲,王晓蕾,等.基于运动波理论吸油管路故障诊断[J].电子设计工程,2016,24(17):116-118. LI Yunpeng, KAN Zhe, WANG Xiaolei, et al. The fault diagnosis of oil pipeline based on the kinematic wave theory[J]. Electronic Design Engineering,2016, 24(17):116-118.
[3]ZADKARAMI M, SHAHBAZIAN M, SALAHSHOOR K. Pipeline leakage detection and isolation: An integrated approach of statistical and wavelet feature extraction with multi-layer perception neural network (MLPNN)[J]. Journal of Loss Prevention in the Process Industries, 2016, 43:479-487.
[4]ZHAO J, LI D, QI H, et al. The fault diagnosis method of pipeline leakage based on neural network[C]// IEEE .International Conference on Computer, Mechatronics, Control and Electronic Engineering, 2010:322-325.
[5]伦淑娴,张化光,冯健. 自适应模糊神经网络系统在管道泄漏检测中的应用[J]. 石油学报,2004,25(4):101-104. LUN Shuxian,ZHANG Huaguang,FENG Jian.Filering technique based on adaptive fuzzy-neural network applied to leakage detection of pipeline[J]. Acta Petrolei Sinica,2004,25(4):101-104.
[6]ZHANG Q,DONG C L, CUI Y. Dynamic uncertain causality graph for knowledge representation and probabilistic reasoning: statistics base, matrix, and application[J]. IEEE transactions on neural networks and learning systems, 2014, 25(4): 645-663.
[7]王占山,张化光,冯健,等. 长距离流体输送管道泄漏检测与定位技术的现状与展望[J].化工自动化及仪表,2003,30(5):5-10. WANG Zhanshan,ZHANG Huaguang,FENG Jian,et al. Present situation and prospect on leak detection and localization techniques for long-distance fluid transport-pipelines[J].Control and Instruments in Chemical Industry,2003,30(5):5-10.
[8]DONG C T, ZHANG Q, GENG S C. A modeling and probabilistic reasoning method of dynamic uncertain causality graph for industrial fault detection[J]. International Journal of Automation and computing, 2014, 11(3): 288-298.
[9]高强,马艳梅. 深度信念网络(DBN)网络层次数量的研究及应用[J]. 科学技术与工程,2016,16(23):234-238,262. GAO Qiang,MA Yanmei.Research and application of the level of the deep belief network (DBN) [J]. Science Technology and Engineering,2016,16(23):234-238,262.
[10]ZHAO J, WANG Z, ZHAO Y B, et al. Application research of pipeline leakage detection and localization based on information fusion technology[J]. Computer Measurement & Control, 2008, 16(7):923-925.
[11]TAMILSELVAN P, WANG P. Failure diagnosis using deep belief learning based health state classification[J]. Reliability Engineering & System Safety, 2013, 115(7):124-135.
[12]TRAN V T,ALTHOBIANI F,BALL A. An approach to fault detection of reciprocating compressor valves using Teager-Kaiser energy operator and deep belief networks [J]. Expert systems with Applications, 2014,41(9):4113-4122.
[13]WANG F. A feature extraction method for fault classification of rolling bearing based on PCA [J]. Journal of Physics Conference, 2015, 628(1):1-5.
[14]DALTON L, PAGE D L. Grading breast cancer on microarray samples: comparison with Nottingham grade, and use of boosting classification [J]. Histopathology,2012,61(3): 497-508.
[15]WANG X, ZHANG W, QIANG J I. A kernel PCA shape prior and edge based MRF image segmentation[J]. Chinese Journal of Electronics, 2016, 25(5):892-900.

相似文献/References:

备注/Memo

备注/Memo:
中国安全生产科学研究院基本科研业务费专项资金项目(2017JBKY13)
更新日期/Last Update: 2018-05-08