基于因果关系的故障传播路径辨识方法研究

doi:10.12305/j.issn.1001-506X.2023.12.40

摘要/Abstract

摘要：

针对故障传播路径辨识问题, 提出了一种基于因果关系的故障传播路径辨识方法, 从因果关系的角度揭示了故障发生及传播的内涵。利用系统中故障发生的因果性, 确定故障发生时受影响的变量, 构建故障相关变量集合; 通过因果关系指示指标确定故障相关变量中各个变量的因果性, 构建因果矩阵; 提出保可达性的赋权有向图最小生成树算法, 根据因果矩阵对相关变量之间的因果性进行图示化表达, 确定故障相关变量之间的传播影响过程, 实现故障传播路径的辨识。所提方法在双带通滤波器电路上进行了实验验证, 实验结果表明了所提方法能够正确筛选故障相关变量集合, 分析变量之间的因果关系, 辨识出故障传播路径, 同时所提方法在时间成本上相较于常用的传递熵方法具有一定的优势。

关键词: 因果关系, 结构因果模型, 故障传播路径, 有向图最小生成树

Abstract:

Aiming at the fault propagation path identification problem, a causality-based fault propagation path identification method is proposed, which reveals the occurrence and propagation connotation of faults from the perspective of causality. The causality of fault occurrence in system is used to determine the variables affected by fault occurrence and to construct the set of failure-related variables. The causality of each variable is determined by the causality relationship indicator index, and the causality matrix is constructed. A weighted directed graph minimum spanning tree algorithm with reachability is proposed, and the causality among related variables is graphically expressed according to causality matrix, and the propagation influence process between fault related variables is determined to realize the identification of fault propagation path. The proposed method is verified by experiments on the circuit of double bandpass filter. The experimental results show that the proposed method can correctly screen the set of fault related variables, analyze the causal relationship between variables, and identify the fault propagation path. Meanwhile, the proposed method has certain advantages over the commonly used transfer entropy method in terms of the time cost.

Key words: causality, structural causal model, fault propagation path, minimum spanning tree of directed graph

中图分类号:

TP306+.3

吕佳朋, 史贤俊, 秦亮, 赵超轮. 基于因果关系的故障传播路径辨识方法研究[J]. 系统工程与电子技术, 2023, 45(12): 4090-4100.

Jiapeng LYU, Xianjun SHI, Liang QIN, Chaolun ZHAO. Research on fault propagation path identification method based on causality[J]. Systems Engineering and Electronics, 2023, 45(12): 4090-4100.

图/表 17

图1

图2

图3

图4

图5

表1

表2

表3

图6

表4

表5

表6

表7

图7

图8

图9

图10

参考文献 32

1	LIUH,PID C,QIUS Y,et al.Data-driven identification model for associated fault propagation path[J].Measurement,2022,188,110628. doi: 10.1016/j.measurement.2021.110628
2	AMINM T,KHANF,IMTIAZS.Fault detection and pathway analysis using a dynamic bayesian network[J].Chemical Engineering Science,2019,195(23):777-790.
3	OHLEFH,BINROTHW,HABOUSHR.Statistical model for a failure mode and effects analysis and its application to computer fault-tracing[J].IEEE Trans.on Reliability,1978,27(1):16-22.
4	HUANGX L,GAOJ M,JIANGH Q,et al.Fault root cause tracing of complicated equipment based on fault graph[J].Proceedings of Institution of Mechanical Engineers Part E: Journal of Process Mechanical Engineering,2012,227(1):17-32.
5	YUJ,RASHIDM M.A novel dynamic Bayesian network-based networked process monitoring approach for fault detection, propagation identification, and root cause diagnosis[J].AIChE Journal,2013,59(7):2348-2365. doi: 10.1002/aic.14013
6	DUANS Y,ZHAOC H,WUM.Multiscale partial symbolic transfer entropy for time-delay root cause diagnosis in nonstationary industrial processes[J].IEEE Trans.on Industrial Electronics,2022,70(2):2015-2025.
7	OLIVEIRAE E,MIGUÉISV L,BORGESJ L.Automatic root cause analysis in manufacturing: an overview & conceptualization[J].Journal of Intelligent Manufacturing,2023,34(5):2061-2078. doi: 10.1007/s10845-022-01914-3
8	马亮,彭开香,董洁.工业过程故障根源诊断与传播路径识别技术综述[J].自动化学报,2022,48(7):1650-1663. doi: 10.16383/j.aas.c200257
	MAL,PENGK X,DONGJ.Review of root cause diagnosis and propagation path identification techniques for faults in industrial processes[J].Acta Automatica Sinica,2022,48(7):1650-1663. doi: 10.16383/j.aas.c200257
9	ALIH,MAULUDA S,ZABIRIH,et al.Multiscale principal component analysis-signed directed graph based process monitoring and fault diagnosis[J].ACS Omega,2022,7(11):9496-9512. doi: 10.1021/acsomega.1c06839
10	唐鹏,彭开香,董洁.一种新颖的深度因果图建模及其故障诊断方法[J].自动化学报,2022,48(6):1616-1624. doi: 10.16383/j.aas.c200996
	TANGP,PENGK X,DONGJ.A novel method for deep casuality graph modeling and fault diagnosis[J].Acta Automatica Sinica,2022,48(6):1616-1624. doi: 10.16383/j.aas.c200996
11	MAL,DONGJ,PENGK X.A novel key performance indicator oriented hierarchical monitoring and propagation path identification framework for complex industrial processes[J].ISA Transactions,2020,96,1-13. doi: 10.1016/j.isatra.2019.06.004
12	LANDMANR,KORTELAJ,SUNQ,et al.Fault propagation analysis of oscillations in control loops using data-driven causality and plant connectivity[J].Computers & Chemical Engineering,2014,71(4):446-456.
13	SHOJAIEA,FOXE B.Granger causality: a review and recent advances[J].Annual Review of Statistics and its Application,2022,9,289-319. doi: 10.1146/annurev-statistics-040120-010930
14	BAUERM,THORNHILLN F,MEABURNA.Specifying the directionality of fault propagation paths using transfer entropy[J].IFAC Proceedings Volumes,2004,37(9):203-208. doi: 10.1016/S1474-6670(17)31812-8
15	DUANP,YANGF,SHAHS L,et al.Transfer zero-entropy and its application for capturing cause and effect relationship between variables[J].IEEE Trans.on Control Systems Technology,2015,23(3):855-867. doi: 10.1109/TCST.2014.2345095
16	何飞,杜学飞,王朝俊.基于对比格兰杰因果关系的热轧带钢头部拉窄根因诊断[J].中国机械工程,2020,31(19):2340-2346. doi: 10.3969/j.issn.1004-132X.2020.19.010
	HEF,DUX F,WANGC J.Root cause diagnosis of head narrowing in hot rolled strip based on comparative granger causality[J].China Mechanical Egnineering,2020,31(19):2340-2346. doi: 10.3969/j.issn.1004-132X.2020.19.010
17	SUJ J,WANGD Z,ZHANGY N,et al.Capturing causality for fault diagnosis based on multi-valued alarm series using transfer entropy[J].Entropy,2019,19(12):663.
18	DUANP,CHENT W,SHAHS L.Direct causality detection via the transfer entropy approach[J].IEEE Trans.on Control Systems Technology,2013,21(6):2052-2066.
19	BARNETTL,BARRETTA B,SETHA K.Granger causality and transfer entropy are equivalent for Gaussian variables[J].Physical Review Letters,2009,103(23):238701.
20	LANDMANR,JÄMSÄ-JOUNELAS L.Fault propagation analysis by implementing nearest neighbors method using process connectivity[J].IEEE Trans.on Control Systems Technology,2019,27(5):2058-2067.
21	SMIRNOVD A.Spurious causalities with transfer entropy[J].Physical Review E,2013,87(4):042912.
22	LEEH,KIMC,LIMS,et al.Data-driven fault diagnosis for chemical processes using transfer entropy and graphical lasso[J].Computers & Chemical Engineering,2020,142(2):107064.
23	PEARLJ.Causality: model, reasoning, and inference[M].Cambridge, UK:Cambridge University Press,2009.
24	PEARLJ,MACKENZIED.The book of why: the new science of cause and effect[M].New York:Basic Books,2018.
25	KUANGK,LIL,GENGZ,et al.Causal inference[J].Engineering,2020,6(3):253-263.
26	LIUY Y,ZHIX C,SHENGL,et al.A survey on causal inference[J].ACM Trans.on Knowledge Discovery from Data,2021,15(5):74.
27	SPIRTESP,ZHANGK.Causal discovery and inference: concepts and recent methodological advances[J].Applied Informatics,2016,3(1):3.
28	GUOR C,CHENGL,LIJ D,et al.A survey of learning causality with data: Problems and methods[J].ACM Computer Surveys,2020,53(4):75.
29	DANIUSIS P, JANZING D, MOOIJ J, et al. Inferring deterministic causal relations[EB/OL]. [2022-12-05]. https://arxiv.org/abs/1203.3475.
30	穆钢,陈奇,刘洪波,等.揭示电力系统运行数据中因果关系的逆信息熵因果推理方法[J].中国电机工程学报,2022,42(15):5406-5416.
	MUG,CHENQ,LIUH B,et al.A reciprocal information entropy causal inference method for exploring the cause-effect relationship in power system operation data[J].Proceedings of the CSEE,2022,42(15):5406-5416.
31	孙凌宇,冷明,谭云兰,等.赋权有向图的最小生成树算法[J].计算机工程,2010,36(2):61-63, 66.
	SUNL Y,LENGM,TANY L,et al.Minimum spanning tree algotirhm of weighted directed graph[J].Computer Engineering,2010,36(2):61-63, 66.
32	贾邵华. 基于冷启动SVDD技术的模拟电路模块级故障诊断方法研究[D]. 哈尔滨: 哈尔滨工业大学, 2020.
	JIA S H. Research on fault diagnosis of analog circuit modules based on cold start SVDD theropy[D]. Harbin: Harbin Institute of Technology, 2020.

故障编号	元件	故障值
f₀	正常	-
f₁	C1	10 nf
f₂	R8	50 kΩ
f₃	R18	断路
f₄	R18	20 kΩ

测试点	故障编号
测试点	f₁	f₂	f₃	f₄
1	0.223 5	0.001 7	0.033 5	0.039 2
2	0.269 5	0.326 3	0.049 8	0.007 3
3	0.009 2	0.017 6	0.008 6	0.005 5
4	0.151 1	0.144 8	0.002 3	0.013 4
5	0.181 4	0.257 2	0.107 7	0.472 1

故障编号	故障相关变量辨识结果
故障编号	本文方法	基于贡献图方法
f₁	测试点1、测试点2、测试点4、测试点5	测试点1、测试点2、测试点4、测试点5
f₂	测试点2、测试点4、测试点5	-
f₃	测试点5	-
f₄	测试点5	测试点5

测试点	测试点
测试点	1	2	4	5
1	-	-0.007 5	-0.007 1	-0.014 4
2	0.007 5	-	-0.014 6	-0.021 9
4	0.007 1	0.014 6	-	-0.007 3
5	0.014 4	0.021 9	0.007 3	-

测试点	测试点
测试点	2	4	5
2	-	-0.366 8	-0.148 9
4	0.366 8	-	-0.217 8
5	0.148 9	0.217 8	-