基于多子空间加权移动窗主成分分析的全厂流程早期故障检测

doi:10.3785/j.issn.1008-973X.2024.10.011

浙江大学学报(工学版)

2024, Vol. 58

Issue (10): 2076-2083 DOI: 10.3785/j.issn.1008-973X.2024.10.011

机械工程、能源工程

基于多子空间加权移动窗主成分分析的全厂流程早期故障检测

宋易盟(

),宋冰*(

),侍洪波,康永波

华东理工大学能源化工过程智能制造教育部重点实验室，上海 200237

Multiple subspace weighted moving window PCA for plant-wide process incipient fault detection

Yimeng SONG(

),Bing SONG*(

),Hongbo SHI,Yongbo KANG

Key Laboratory of Smart Manufacturing in Energy Chemical Process of the Ministry of Education, East China University of Science and Technology, Shanghai 200237, China

全文: PDF(912 KB) HTML

摘要：

早期故障的特征不明显，在全厂流程中比常规故障难检测. 为了提高全厂流程中早期故障的检测率和灵敏度，将检测视角由全局转移至局部，提出基于多子空间加权移动窗主成分分析(PCA)的早期故障检测方法. 使用结合过程知识和数据驱动的双层子空间划分方法，将过程变量划分到不同子空间中. 使用加权的移动窗口增大早期故障的偏移量，将局部离群因子(LOF)算法引入PCA，以便进一步关注数据的局部特征，在每个子空间中建立故障检测模型. 通过贝叶斯推理融合法对各子空间的监测结果进行信息融合，获得分布式监测结果. 通过工业实例验证所提方法的性能. 结果表明，所提方法在全厂流程中有效提升了早期故障检测的准确率和灵敏度.

关键词： 全厂流程; 早期故障检测; 两层子空间划分; 加权移动窗口; 局部离群因子; 贝叶斯推理融合

Abstract:

Incipient faults are difficult to detect in plant-wide processes compared to conventional faults due to the lack of distinctive features. An incipient fault detection method based on multiple subspace weighted moving window principal component analysis (PCA) was proposed by shifting the detection perspective from the global to the local to improve the detection rate and sensitivity of incipient faults in plant-wide processes. Process variables were partitioned into different subspaces using a two-layer subspace partitioning method that combines process knowledge and data-driven approaches. A weighted moving window was used to increase the offset of incipient faults, while a local outlier factor (LOF) algorithm was introduced into PCA to further focus on the local features of the data to model fault detection in each subspace. The monitoring results in each subspace were fused with information by the Bayesian inference fusion method to obtain distributed monitoring results. The proposed method was validated by industrial examples, and the results showed that the method effectively improved the accuracy and detection speed of incipient fault detection in plant-wide processes.

Key words: plant-wide process incipient fault detection two-layer subspace partitioning weighted moving window local outlier factor Bayesian inference fusion

收稿日期: 2024-01-22 出版日期: 2024-09-27

CLC:

TP 277

基金资助: 国家自然科学基金资助项目（62073140, 62073141, 62103149, 62273147）；上海明日之星计划资助项目（21QA1401800）；国家重点研发计划资助项目（2020YFC1522502, 2020YFC1522505）.

通讯作者: 宋冰 E-mail: y30220987@mail.ecust.edu.cn;songbing@ecust.edu.cn

作者简介: 宋易盟（2000—），男，硕士生，从事特征提取、故障检测和故障诊断研究. orcid.org/0009-0007-2543-0348. E-mail：y30220987@mail.ecust.edu.cn

	服务
	把本文推荐给朋友
	加入引用管理器
	E-mail Alert
	作者相关文章
	宋易盟
	宋冰
	侍洪波
	康永波

引用本文:

宋易盟,宋冰,侍洪波,康永波. 基于多子空间加权移动窗主成分分析的全厂流程早期故障检测[J]. 浙江大学学报(工学版), 2024, 58(10): 2076-2083.

Yimeng SONG,Bing SONG,Hongbo SHI,Yongbo KANG. Multiple subspace weighted moving window PCA for plant-wide process incipient fault detection. Journal of ZheJiang University (Engineering Science), 2024, 58(10): 2076-2083.

链接本文:

https://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2024.10.011 或 https://www.zjujournals.com/eng/CN/Y2024/V58/I10/2076

图 1 基于多子空间加权移动窗主成分分析模型示意图

表 1 田纳西-伊斯曼过程的双层子空间划分结果

图 2 故障4基于不同算法的仿真结果

图 3 故障13基于不同算法的仿真结果

图 4 故障8基于不同算法的仿真结果

图 5 故障18基于不同算法的仿真结果

表 2 故障工况的误报率和故障检测率

1	SILVA A F, VERCRUYSSE J, VERVAET C, et al In-depth evaluation of data collected during a continuous pharmaceutical manufacturing process: a multivariate statistical process monitoring approach[J]. Journal of Pharmaceutical Sciences, 2019, 108 (1): 439- 450 doi: 10.1016/j.xphs.2018.07.033
2	TAO Y, SHI H, SONG B, et al A distributed adaptive monitoring method for performance indicator in large-scale dynamic process[J]. IEEE Transactions on Industrial Informatics, 2023, 19 (10): 10425- 10433 doi: 10.1109/TII.2023.3240732
3	LIU D, SHANG J, CHEN M Principal component analysis-based ensemble detector for incipient faults in dynamic processes[J]. IEEE Transactions on Industrial Informatics, 2021, 17 (8): 5391- 5401 doi: 10.1109/TII.2020.3031496
4	贾瑶, 李帅, 柴天佑金矿生产全流程控制系统设计与实现[J]. 控制工程, 2022, 29 (5): 873- 887 JIA Yao, LI Shuai, CHAI Tianyou Design and implementation of whole process control system for gold ore production[J]. Control Engineering of China, 2022, 29 (5): 873- 887
5	ZHOU J, ZHANG S, WANG J A dual robustness projection to latent structure method and its application[J]. IEEE Transactions on Industrial Electronics, 2021, 68 (2): 1604- 1614 doi: 10.1109/TIE.2020.2970664
6	MA J, ZHANG J Progress of process monitoring for the multi-mode process: a review[J]. Applied Sciences, 2022, 12 (14): 7207 doi: 10.3390/app12147207
7	WANG B, YAN X, JIANG Q, et al Generalized dice’s coefficient-based multi-block principal component analysis with bayesian inference for plant-wide process monitoring[J]. Journal of Chemometrics, 2015, 29 (3): 165- 178 doi: 10.1002/cem.2687
8	CHEN X, WANG J, DING S X Complex system monitoring based on distributed least squares method[J]. IEEE Transactions on Automation Science and Engineering, 2021, 18 (14): 1892- 1900
9	JIANG Q, YAN X, HUANG B Performance-driven distributed PCA process monitoring based on fault-relevant variable selection and bayesian inference[J]. IEEE Transactions on Industrial Electronics, 2015, 63 (1): 377- 386
10	KONG X, YANG Z, LUO J, et al Extraction of reduced fault subspace based on KDICA and its application in fault diagnosis[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 3505212
11	王光, 单发顺, 钱禹丞, 等基于集成学习传递熵的化工过程微小故障检测方法[J]. 化工学报, 2023, 74 (7): 2967- 2978 WANG Guang, SHAN Fashun, QIAN Yucheng, et al Incipient fault detection method for chemical process based on ensemble learning transfer entropy[J]. CIESC Journal, 2023, 74 (7): 2967- 2978
12	CHEN Z, YANG C, PENG T, et al A cumulative canonical correlation analysis-based sensor precision degradation detection method[J]. IEEE Transactions on Industrial Electronics, 2019, 8 (66): 6321- 6330
13	WONG P K, YANG Z, VONG C M, et al Real-time fault diagnosis for gas turbine generator systems using extreme learning machine[J]. Neurocomputing, 2014, 128: 249- 257
14	DENG X, CAI P, CAO Y, et al Two-step localized kernel principal component analysis based incipient fault diagnosis for nonlinear[J]. Industrial and Engineering Chemistry Research, 2020, 59 (13): 5956- 5968 doi: 10.1021/acs.iecr.9b06826
15	PILARIO K E S, CAO Y Canonical variate dissimilarity analysis for process incipient fault detection[J]. IEEE Transactions on Industrial Informatics, 2018, 14 (12): 5308- 5315 doi: 10.1109/TII.2018.2810822
16	QIN Y, YAN Y, JI H, et al Recursive correlative statistical analysis method with sliding windows for incipient fault detection[J]. IEEE Transactions on Industrial Electronics, 2022, 69 (4): 4185- 4194 doi: 10.1109/TIE.2021.3070521
17	FENG Z, LI Y, XIAO B, et al Process monitoring of abnormal working conditions in the zinc roasting process with an ALD-based LOF-PCA method[J]. Process Safety and Environmental Protection, 2022, 161: 640- 650 doi: 10.1016/j.psep.2022.03.064
18	ZENG J, HUANG W, WANG Z, et al Mutual information-based sparse multiblock dissimilarity method for incipient fault detection and diagnosis in plant-wide process[J]. Journal of Process Control, 2019, 83: 63- 76 doi: 10.1016/j.jprocont.2019.09.004
19	CAO Y, YUAN X, WANG Y, et al Hierarchical hybrid distributed PCA for plant-wide monitoring of chemical processes[J]. Control Engineering Practice, 2021, 111: 104784 doi: 10.1016/j.conengprac.2021.104784
20	BROWN G, POCOCK A, ZHAO M J, et al Conditional likelihood maximisation: a unifying framework for information theoretic feature selection[J]. The Journal Of Machine Learning Research, 2012, 13 (1): 27- 66
21	AMIN M T, KHAN F, AHMED S, et al A data-driven Bayesian network learning method for process fault diagnosis[J]. Process Safety and Environmental Protection, 2021, 150: 110- 122 doi: 10.1016/j.psep.2021.04.004
22	CHEN H, JIANG B, ZHANG T, et al Data-driven and deep learning-based detection and diagnosis of incipient faults with application to electrical traction systems[J]. Neurocomputing, 2020, 396: 429- 437 doi: 10.1016/j.neucom.2018.07.103
23	KONG X, BI Y, GLASS D Detecting anomalies in sequential data augmented with new features[J]. Artificial Intelligence Review, 2020, 53 (1): 625- 652 doi: 10.1007/s10462-018-9671-x
24	JIANG Q, YAN X Nonlinear plant-wide process monitoring using MI-spectral clustering and Bayesian inference-based multiblock KPCA[J]. Journal of Process Control, 2015, 32: 38- 50 doi: 10.1016/j.jprocont.2015.04.014
25	DOWNS J J, VOGEL E F A plant-wide industrial process control problem[J]. Computers and Chemical Engineering, 1993, 17 (3): 245- 255 doi: 10.1016/0098-1354(93)80018-I

[1]	孔祥玉,陈雅琳,罗家宇,杨治艳. 基于动态内并行潜结构投影的故障检测方法[J]. 浙江大学学报(工学版), 2023, 57(7): 1297-1306.
[2]	孔祥玉,王晓兵,李红增,罗家宇. 面向工业平稳/非平稳复杂系统的在线故障监测技术[J]. 浙江大学学报(工学版), 2021, 55(10): 1856-1866.
[3]	刘兴,余建波. 注意力卷积GRU自编码器及其在工业过程监控的应用[J]. 浙江大学学报(工学版), 2021, 55(9): 1643-1651.
[4]	李炜,张建军. 攻击与故障共存的ICPS综合安全控制方法[J]. 浙江大学学报(工学版), 2021, 55(6): 1185-1198.
[5]	谭晓栋,罗建禄,李庆,邱静. 机械系统的故障演化测试性建模及预计[J]. 浙江大学学报(工学版), 2016, 50(3): 442-448.
[6]	陈特欢, 徐巍华, 许超, 谢磊. 基于PSO的管道泄漏模型反问题求解及敏感性分析[J]. 浙江大学学报(工学版), 2014, 48(10): 1850-1855.
[7]	李德骏, 杨竣程, 林冬冬, 杨灿军, 金波, 陈鹰. 单片机与CPLD技术的海底接驳盒电能监控系统[J]. J4, 2012, 46(8): 1369-1374.
[8]	于保华, 杨世锡,周晓峰. 基于故障-测点互信息的传感器多目标优化配置[J]. J4, 2012, 46(1): 156-162.
[9]	段斌, 梁军, 费正顺, 杨敏, 胡斌. 基于GA-ANN的非线性半参数建模方法[J]. J4, 2011, 45(6): 977-983.
[10]	徐贵斌, 周东华. 基于在线学习神经网络的状态依赖型故障预测[J]. J4, 2010, 44(7): 1251-1254.
[11]	杜文莉, 王坤, 钱锋. 基于特征空间降维的溶剂脱水分离过程监控[J]. J4, 2010, 44(7): 1255-1259.
[12]	曹玉苹, 田学民. 基于信息散度的过程故障检测与诊断[J]. J4, 2010, 44(7): 1315-1320.
[13]	林勇, 周晓军, 杨先勇, 等. 基于双谱识别和人工免疫网络的智能故障检测[J]. J4, 2009, 43(10): 1777-1782.

Viewed

Full text

Abstract

Cited

Shared

Discussed