Fault detection and diagnosis for chillers using MSPCA-KECA

doi:10.11949/j.issn.0438-1157.20161239

Abstract

Abstract:

There are differences among different levels of the same type of the fault, which may cause misdiagnose. A fault diagnosis strategy based on multi-scale principal component analysis and kernel entropy component analysis (MSPCA-KECA) is proposed. Taking the features extracted by MSPCA as the input of KECA classifier can be used for fault online detection as well as automatic identification. MSPCA combines wavelet multi-scale analysis with principal component analysis to select the scales which contain fault-related information, and then use PCA to extract the fault-related features, extracting the similarity among different levels of the same type of fault and the difference among different faults, which can improve the ability of fault diagnosis. The combination of KECA and Cauchy-Schwarz (CS) statistics extract and express the angular structure of different kinds of faults, which is good for fault classification. The control limit here is achieved by support vector data description (SVDD) for the unacquainted distribution of the statistics. Through the simulation of ASHRAR 1043-RP chiller data, the feasibility and effectiveness of the MSPCA-KECA method are verified.

Key words: fault diagnosis, MSPCA, KECA, chillers

摘要：

针对冷水机组同类型不同等级故障的变量变化存在差异会造成误诊断的问题，提出一种基于多尺度主元分析-核熵成分分析（MSPCA-KECA）的故障诊断策略。MSPCA提取故障特征，其输出作为KECA分类器的输入，实现故障的实时监测与自动诊断。首先，改进的MSPCA算法通过将小波多尺度分析与主元分析相结合，筛选故障信息可能存在的尺度直接重构并采用PCA提取故障特征，获取不同类型故障之间差异的同时也保留了同类型但不同等级故障之间的相似性，提高故障诊断的可靠性。之后建立KECA非线性分类器并引入一种新的监测统计量——散度测度统计量，使降维后不同特征信息之间呈现显著的角度差异，易于分类。最后，采用支持向量数据描述（SVDD）算法确定新统计量的控制限，以克服无法获知统计量分布的问题。通过对冷水机组数据的仿真研究，验证了MSPCA-KECA方法的可行性及有效性。

关键词: 故障诊断, 多尺度主元分析, 核熵成分分析, 冷水机组

CLC Number:

TP277

QI Yongsheng, ZHANG Haili, WANG Lin, GAO Xuejin, LU Chenxi. Fault detection and diagnosis for chillers using MSPCA-KECA[J]. CIESC Journal, 2017, 68(4): 1499-1508.

齐咏生, 张海利, 王林, 高学金, 陆晨曦. 基于MSPCA-KECA的冷水机组故障监测及诊断[J]. 化工学报, 2017, 68(4): 1499-1508.

References 30

[1]	赵春晖, 王福利, 姚远. 基于时段的间歇过程统计建模在线监测及质量预报[J]. 自动化学报, 2010, 36(3):366-374.ZHAO C H, WANG F L, YAO Y. Phase-based statistical modeling, online monitoring and quality prediction for batch processes[J]. Acta Automatica Sinica, 2010, 36(3):366-374.
[2]	ALDRICH C, AURET L. Unsupervised Process Monitoring and Fault Diagnosis with Machine Learning Methods[M]. London:Springer London, 2013:371.
[3]	蒋丽英, 王树青. 基于MPCA-MDPLS的间歇过程的故障诊断[J]. 化工学报, 2005, 56(3):482-486.JIANG L Y, WANG S Q. Fault diagnosis for batch processes based on MPCA-MDPLS[J]. Journal of Chemical Industry and Engineering (China), 2005, 56(3):482-486.
[4]	张成, 李元. 基于统计模量分析间歇过程故障监测方法研究[J]. 仪器仪表学报, 2013, 23(9):2103-2110.ZHANG C, LI Y. Study on the fault-detection method in batch process based on statistical pattern analysis[J]. Chinese Journal of Scientific Instrument, 2013, 23(9):2103-2110.
[5]	GALICIA H J, HE Q P, WANG J. Statistics pattern analysis-fault detection and diagnosis[C]//NICK S. Proceedings of the ISA Automation Week. AL, USA:FOCAPO 2012, 2012:325-330.
[6]	WANG J, HE Q P. Multivariate statistical process monitoring based on statistics pattern analysis[J]. Industrial & Engineering Chemistry Research, 2010, 49(17):7858-7869.
[7]	GAO X, HOU J. An improved SVM integrated GS-PCA fault diagnosis approach of Tennessee Eastman process[J]. Neurocomputing, 2016, 174:906-911.
[8]	梁晴晴, 韩华, 崔晓钰. 基于主元分析-概率神经网络的制冷系统故障诊断[J]. 化工学报, 2016, 67(3):1023-1031.LIANG Q Q, HAN H, CUI X Y. Fault diagnosis for refrigeration system based on PCA-PNN[J]. CIESC Journal, 2016, 67(3):1023-1031.
[9]	DING S X. Data-driven design of monitoring and diagnosis systems for dynamic processes:a review of subspace technique based schemes and some recent results[J]. Journal of Process Control, 2014, 24:431-449.
[10]	YAN K, SHEN W, MULUMBA T, et al. ARX model based fault detection and diagnosis for chillers using support vector machines[J]. Energy and Buildings, 2014, (81):287-295.
[11]	KARIMI I, SALAHSHOOR K. A new fault detection and diagnosis approach for a distillation column system based on a combined PCA and ANFIS scheme[J]. IEEE Conference on Cybernetics & Intelligent Systems, 2012, 34(1):13-21.
[12]	HONG J J, ZHANG J. Progressive PCA modeling for enhanced fault diagnosis in a batch process[C]//IEEE. International Conference on Control, Automation and Systems 2010. United States:IEEE Computer Society, 2010:713-718.
[13]	XIE S K, LAWNIZAK A T, LIO' P, et al. Feature extraction by multi-scale principal component analysis and classification in spectral domain[J]. Engineering, 2013, 5:268-271.
[14]	WANG D, ROMAGNOLI J A. Robust multi-scale principal components analysis with applications to process monitoring[J]. Journal of Process Control, 2005, 15:869-882.
[15]	JENSSEN R. Kernel entropy component analysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(5):847-860.
[16]	JIANG Q C, YAN X F, LÜ Z M, et al. Fault detection in nonlinear chemical process based on kernel entropy component analysis and angular structure[J]. Korean Journal of Chemical Engineering, 2013, 30(6):1181-1186.
[17]	SHI J, JIANG Q K, ZHANG Q, et al. Sparse kernel entropy component analysis for dimensionality reduction of biomedical data[J]. Neurocomputing, 2015, 168:930-940.
[18]	GÓMEZ-CHOVA L, JENSSEN R, CAMPS-VALLS G. Kernel entropy component analysis for remote sensing image clustering[J]. IEEE Geoscience and Remote Sensing Letters, 2012, 9(2):312-316.
[19]	LIU X, WU X J. ECA and 2DECA:Entropy contribution based methods for face recognition inspired by KECA[C]//IEEE. 2011 International Conference of Soft Computing and Pattern Recognition. United States:IEEE Computer Society, 2011:544-549.
[20]	YANG Y H, LI X L, LIU X Z, et al. Wavelet kernel entropy component analysis with application to industrial process monitoring[J]. Neurocomputing, 2015, 147:395-402.
[21]	JIA X M, MENG Q H, JING Y Q, et al. A new method combining KECA-LDA with ELM for classification of Chinese liquors using electronic nose[J]. IEEE Sensors Journal, 2016, 16(22):8010-8017.
[22]	SHI J, JIANG Q, MAO R, et al. FR-KECA:fuzzy robust kernel entropy component analysis[J]. Neurocomputing, 2015, 149:1415-1423.
[23]	YANG Y H, LI X L, LIU X Z, et al. Wavelet kernel entropy component analysis with application to industrial process monitoring[J]. Neurocomputing, 2015, 147(5):395-402.
[24]	JIANG Q C, YAN X F, LÜ Z M, et al. Fault detection in nonlinear chemical processes based on kernel entropy component analysis and angular structure[J]. Korean Journal of Chemical Engineering, 2013, 30(6):1181-1186.
[25]	YANG Y H, LI H Q, LI C L, et al. Kernel entropy component analysis based process monitoring method with process subsystem division[C]//IEEE. 201527th Chinese Control and Decision Conference. USA:IEEE, 2015:2730-2734.
[26]	JENSSEN R, ELTOFT T. A new information theoretic analysis of sum-of-squared-error kernel clustering[J]. Neurocomputing, 2008, 72(1/2/3):23-32.
[27]	JENSSEN R, PRINCIPE J, ERDOGMUSD, et al. The Cauchy-Schwarz divergence and Parzen windowing:connections to graph theory and Mercer kernel[J]. Journal of the Franklin Institute, 2006, 343(6):614-629.
[28]	李冠男, 胡云鹏, 陈焕新, 等. 基于SVDD的冷水机组传感器故障检测及效率分析[J]. 化工学报, 2015, 66(5):1815-1820.LI G N, HU Y P, CHEN H X, et al. SVDD-based chiller sensor fault detection method and its detection efficiency[J]. CIESC Journal, 2015, 66(5):1815-1820.
[29]	谢磊, 刘雪芹, 张建明, 等. 基于NGPP-SVDD的非高斯过程监控及其应用研究[J]. 自动化学报, 2009, 35(1):107-112.XIE L, LIU X Q, ZHANG J M, et al. Non-Gaussian process monitoring based on NGPP-SVDD[J]. Acta Auomatica Sinica, 2009, 35(1):107-112.
[30]	COMSTOCK M C, AUN J E. Development and validation of a mechanistic, dynamic model for a vapor compression centrifugal liquid chiller[D]. USA:ASHRAE, 2002.ting, 2015, 147:395-402.
[21]	JIA X M,MENG Q H,JING Y Q,et al. A New Method Combining KECA-LDA With ELM for Classification of Chinese Liquors Using Electronic Nose[J].IEEE Sensors Journal,2016,16(22):8010-8017.
[22]	SHI J, JIANG Q, MAO R, et al. FR-KECA:Fuzzy robust kernel entropy component analysis[J].Neurocomputing,2015,149:1415-1423.
[23]	YANG Y H, LI X L, LIU X Z,et al. Wavelet kernel entropy component analysis with application to industrial process monitoring[J]. Neurocomputing, 2015, 147(5):395-402.
[24]	JIANG Q C,YAN X F, LV Z M,et al. Fault detection in nonlinear chemical processes based on kernel entropy component analysis and ngular structure[J]. Korean Journal of Chemical Engineering, 2013, 30(6):1181-1186.
[25]	YANG Y H,LI H Q,LI C L,et al. Kernel Entropy Component Analysis Based Process Monitoring Method with Process Subsystem Division[C]//IEEE.201527th Chinese Control and Decision Conference.USA:IEEE,2015:2730-2734.
[26]	JENSSEN R, ELTOFT T. A New Information Theoretic Analysis of Sum-of-Squared-Error Kernel Clustering[J]. Neurocomputing, 2008, 72(1/2/3):23-32.
[27]	JENSSEN R,PRINCIPE J,ERDOGMUSD, et al. The Cauchy-Schwarz Divergence and Parzen Windowing:Connections to Graph Theory and Mercer Kernel[J]. Journal of the Franklin Institute, 2006, 343(6):614-629.
[28]	李冠男,胡云鹏,陈焕新等. 基于SVDD的冷水机组传感器故障检测及效率分析[J].化工学报, 2015,66(5):1815-1820. LI G N, HU Y P,CHEN H X, et al.SVDD-based chiller sensor fault detection method and its detection efficiency[J].CIESC Journal, 2015,66(5):1815-1820.
[29]	谢磊,刘雪芹,张建明等. 基于NGPP-SVDD的非高斯过程监控及其应用研究[J].自动化学报, 2009,35(1):107-112. XIE L,LIU X Q,ZHANG J M, et al. Non-Gaussian process monitoring based on NGPP-SVDD[J]. ACTA AUOMATICA SINICA, 2009, 35(1):107-112.
[30]	COMSTOCK M C, BRAUN J E. Development And Validation Of A Mechanistic, Dynamic Model For A Vapor Compression Centrifugal Liquid Chiller[D].USA:ASHRAE,2002.