基于核典型相关分析的故障检测方法
|
摘要
针对化工过程中存在大量非线性问题,目前存在的主要方法为将核算法与偏最小二乘算法相结合(KPLS),相比于KPLS算法,将核与典型相关分析相结合的方法(KCCA)能够最大化两组变量相关性,以达到更好的检测效果。然而KCCA方法不能准确地将数据空间分解成与关键性能指标(KPI)相关和不相关的部分,从而忽略了剩余空间仍然涉及与KPI相关的一些信息的事实。文中提出了一种新的改进的KCCA (MKCCA)方法,该方法对核矩阵的可计算负载进行奇异值分解(SVD),得到一个投影模型,将核矩阵适当地分解为KPI相关部分和不相关部分,然后设计两个统计量进行故障检测。最后利用田纳西伊士曼(TE)过程验证了所提出方法的有效性和优越性。
Aiming at a large number of nonlinear problems in the chemical process, the main existing method is the combination of the kernel algorithm and the partial least squares(KPLS) algorithm. Compared with KPLS algorithm, the method of combining kernel algorithm and canonical correlation analysis(KCCA) algorithm can maximize the correlation between the two groups of variables to achieve better detection results. However, the current KCCA method cannot accurately decompose the data space into parts that are related and unrelated to the key performance indicator(KPI), thereby it ignores the fact that the remaining space still involves some information related to the KPI. In this paper, an improved KCCA is proposed. The method performs singular value decomposition(SVD) on the calculable loadings of kernel matrix, a projection model is obtained in which the kernel matrix is appropriately decomposed into KPI-related and-unrelated parts, and then two statistics are accordingly designed for fault detection. Finally, the Eastman Eastman(TE) process was used to verify the effectiveness and superiority of the proposed method.
Aiming at a large number of nonlinear problems in the chemical process, the main existing method is the combination of the kernel algorithm and the partial least squares(KPLS) algorithm. Compared with KPLS algorithm, the method of combining kernel algorithm and canonical correlation analysis(KCCA) algorithm can maximize the correlation between the two groups of variables to achieve better detection results. However, the current KCCA method cannot accurately decompose the data space into parts that are related and unrelated to the key performance indicator(KPI), thereby it ignores the fact that the remaining space still involves some information related to the KPI. In this paper, an improved KCCA is proposed. The method performs singular value decomposition(SVD) on the calculable loadings of kernel matrix, a projection model is obtained in which the kernel matrix is appropriately decomposed into KPI-related and-unrelated parts, and then two statistics are accordingly designed for fault detection. Finally, the Eastman Eastman(TE) process was used to verify the effectiveness and superiority of the proposed method.
引文
[1] Juricek B C, Seborg D E, Larimore W E. Fault detection using canonical variate analysis[J]. Industrial&Engineering Chemistry Research, 2004.43(2):458-474.
[2] Li D F. Perspective for smart factory in petrochemical industry[J].Computers&Chemical Engineering. 2016,91:136-148
[3]周东华,李钢,李元,数据驱动的工业过程故障诊断技术:基于主元分析与偏最小二乘的方法[M].科学出版社,2011.Zhou D H, Li G, Li Y. Data Driven Industrial Process Fault Diagnosis Technology——Based on Principal Component Analysis and Partial Least Squares[M]. Beijing:Science Press, 2011.
[4] Lou Z J, Shen D, and Wang Y Q. Preliminary-summation-based principal component analysis for non-Gaussian processes[J].Chemometrics and Intelligent Laboratory Systems, 2015. 146:270-289.
[5] Zhao C H, Wang F L, Lu N Y, et al. Stage-based soft-transition multiple PCA modeling and on-line monitoring strategy for batch processes[J]. Journal of Process Control, 2007. 17(9):728-741.
[6] Fan J C, Qin S J, and Wang Y Q. Online monitoring of nonlinear multivariate industrial processes using filtering KICA-PCA[J].Control Engineering Practice, 2014.22(1):205-216.
[7] Jia Q L, Zhang Y W. Quality-related fault detection approach based on dynamic kernel partial least squares[J]. Chemical Engineering Research&Design, 2016.106:242-252.
[8] Shi H T, Liu J C, Zhang Y, et al. Fault detection method based on relative-transformation partial least squares[J]. Chinese Journal of Scientific Instrument, 2012.33(4):816-822.
[9] Chen Z W, Zhang K, Ding S X, et al. Improved canonical correlation analysis-based fault detection methods for industrial processes[J].Journal of Process Control, 2016.41:26-34.
[10] Chen Z W, Zhang K, Ding S X, et al. Study on small multiplicative fault detection using Canonical Correlation Analysis with the local approach[J]. IFAC-Papers OnLine, 2015,48(21):1414-1419.
[11] Chen Z W. Data-Driven Fault Detection for Industrial Processes[M].Springer Fachmedien Wiesbaden. 2017.
[12] Chen Z W, Ding S X, Zhang K, et al. Canonical correlation analysis-based fault detection methods with application to alumina evaporation process[J]. Control Engineering Practice, 2016. 46:51-58.
[13] Chen Z W,Ding S X,Peng T,et al. Fault Detection for Non-Gaussian Processes Using Generalized Canonical Correlation Analysis and Randomized Algorithms[J]. IEEE Transactions on Industrial Electronics, 2017, PP(99):1-1.
[14]张颖伟,李洪强.基于核偏最小二乘的故障监控与诊断[C]中国控制与决策会议.2009.Zhang Y W, Li H Q. The fault monitoring and diagnosi based on KPLS[C]. Chinese Control and Decision Conference. 2009.
[15] Lai P L, Fyfe C. Kernel and Nonlinear Canonical Correlation Analysis[J]. International Journal of Neural Systems, 2000. 10(5):365-77.
[16] Shardt Y A W, Hao H, Ding S X. A New Soft-Sensor-Based Process Monitoring Scheme Incorporating Infrequent KPI Measurements[J].IEEE Transactions on Industrial Electronics, 2015,62(6):3843-3851.
[17] Zhou D H, Li G, Qin S J. Total projection to latent structures for process monitoring[J]. Aiche Journal, 2010.56(1):168-178.
[18] Jiao J F, Zhao N, Wang G, et al. A nonlinear quality-related fault detection approach based on modified kernel partial least squares[J].ISA Trans, 2016. 66:275-283.
[19] Zhu Q Q, Liu Q, Qin S J. Quality-relevant fault detection of nonlinear processes based on kernel concurrent canonical correlation analysis[C]. 2017 American Control Conference(ACC)IEEE, 2017.
[20] Yin S, Zhu X P, Kaynak O. Improved PLS Focused on Key-Performance-Indicator-Related Fault Diagnosis[J]. IEEE Transactions on Industrial Electronics, 2015,62(3):1651-1658.
[21]陈志文,彭涛,阳春华,等.基于改进的典型相关分析的故障检测方法[J].山东大学学报(工学版),2017,(05):49-55.Chen Z Y, Peng T, Yang C H, et al. A fault detection method based on modified canonical correlation analysis[J]. Journal of Shandong University(Engineering Science), 2017. 62(3):1651-1658
[2] Li D F. Perspective for smart factory in petrochemical industry[J].Computers&Chemical Engineering. 2016,91:136-148
[3]周东华,李钢,李元,数据驱动的工业过程故障诊断技术:基于主元分析与偏最小二乘的方法[M].科学出版社,2011.Zhou D H, Li G, Li Y. Data Driven Industrial Process Fault Diagnosis Technology——Based on Principal Component Analysis and Partial Least Squares[M]. Beijing:Science Press, 2011.
[4] Lou Z J, Shen D, and Wang Y Q. Preliminary-summation-based principal component analysis for non-Gaussian processes[J].Chemometrics and Intelligent Laboratory Systems, 2015. 146:270-289.
[5] Zhao C H, Wang F L, Lu N Y, et al. Stage-based soft-transition multiple PCA modeling and on-line monitoring strategy for batch processes[J]. Journal of Process Control, 2007. 17(9):728-741.
[6] Fan J C, Qin S J, and Wang Y Q. Online monitoring of nonlinear multivariate industrial processes using filtering KICA-PCA[J].Control Engineering Practice, 2014.22(1):205-216.
[7] Jia Q L, Zhang Y W. Quality-related fault detection approach based on dynamic kernel partial least squares[J]. Chemical Engineering Research&Design, 2016.106:242-252.
[8] Shi H T, Liu J C, Zhang Y, et al. Fault detection method based on relative-transformation partial least squares[J]. Chinese Journal of Scientific Instrument, 2012.33(4):816-822.
[9] Chen Z W, Zhang K, Ding S X, et al. Improved canonical correlation analysis-based fault detection methods for industrial processes[J].Journal of Process Control, 2016.41:26-34.
[10] Chen Z W, Zhang K, Ding S X, et al. Study on small multiplicative fault detection using Canonical Correlation Analysis with the local approach[J]. IFAC-Papers OnLine, 2015,48(21):1414-1419.
[11] Chen Z W. Data-Driven Fault Detection for Industrial Processes[M].Springer Fachmedien Wiesbaden. 2017.
[12] Chen Z W, Ding S X, Zhang K, et al. Canonical correlation analysis-based fault detection methods with application to alumina evaporation process[J]. Control Engineering Practice, 2016. 46:51-58.
[13] Chen Z W,Ding S X,Peng T,et al. Fault Detection for Non-Gaussian Processes Using Generalized Canonical Correlation Analysis and Randomized Algorithms[J]. IEEE Transactions on Industrial Electronics, 2017, PP(99):1-1.
[14]张颖伟,李洪强.基于核偏最小二乘的故障监控与诊断[C]中国控制与决策会议.2009.Zhang Y W, Li H Q. The fault monitoring and diagnosi based on KPLS[C]. Chinese Control and Decision Conference. 2009.
[15] Lai P L, Fyfe C. Kernel and Nonlinear Canonical Correlation Analysis[J]. International Journal of Neural Systems, 2000. 10(5):365-77.
[16] Shardt Y A W, Hao H, Ding S X. A New Soft-Sensor-Based Process Monitoring Scheme Incorporating Infrequent KPI Measurements[J].IEEE Transactions on Industrial Electronics, 2015,62(6):3843-3851.
[17] Zhou D H, Li G, Qin S J. Total projection to latent structures for process monitoring[J]. Aiche Journal, 2010.56(1):168-178.
[18] Jiao J F, Zhao N, Wang G, et al. A nonlinear quality-related fault detection approach based on modified kernel partial least squares[J].ISA Trans, 2016. 66:275-283.
[19] Zhu Q Q, Liu Q, Qin S J. Quality-relevant fault detection of nonlinear processes based on kernel concurrent canonical correlation analysis[C]. 2017 American Control Conference(ACC)IEEE, 2017.
[20] Yin S, Zhu X P, Kaynak O. Improved PLS Focused on Key-Performance-Indicator-Related Fault Diagnosis[J]. IEEE Transactions on Industrial Electronics, 2015,62(3):1651-1658.
[21]陈志文,彭涛,阳春华,等.基于改进的典型相关分析的故障检测方法[J].山东大学学报(工学版),2017,(05):49-55.Chen Z Y, Peng T, Yang C H, et al. A fault detection method based on modified canonical correlation analysis[J]. Journal of Shandong University(Engineering Science), 2017. 62(3):1651-1658