Fault detection method based on minimum sufficient statistics pattern analysis

doi:10.11949/j.issn.0438-1157.20171054

Abstract

Abstract:

Recently, the statistic pattern analysis (SPA) has been used with widespread applications in the field of fault detection. Its essence is to use data statistics matrix for process monitoring instead of the original data matrix. However, SPA lacks reasonable method in choosing the statistics variables, also complex nonlinear interactions exist among these statistics variables. As a result, fault detection cannot be processed by using ordinary principal component analysis (PCA) algorithm. In order to solve these problems, a new minimum sufficient statistics pattern analysis (MSSPA) fault detection method is proposed. This method first eliminates the correlations among variables by performing an orthogonal transformation of the raw data matrix, and then estimates the probability density function of the single variables or joint probability density function of multiple variables, so as to acquire the minimum sufficient statistic of original data, and construct the statistic matrix with it. The introduction of minimum sufficient statistics is also beneficial to handle the problem of non-Gaussian distribution of the raw data. Finally, the feasibility and validity of this method for fault detection are verified by testing on the Tennessee Eastmann (TE) process.

Key words: principal component analysis, algorithm, process systems, statistics pattern analysis, minimum sufficient statistics, fault detection

摘要：

统计量模式分析（SPA）最近在故障检测领域取得了广泛应用，其实质是用数据的统计量矩阵来代替原始数据矩阵进行故障检测，然而其统计量的选取存在盲目性且各统计量之间存在复杂的非线性关联关系，难以满足后续应用主成分分析（PCA）完成故障检测所需的基本条件。为了解决这个问题，提出了基于最小充分统计量模式分析的故障检测方法（MSSPA）。该方法首先将原始数据矩阵进行正交变换以消除变量之间的关联性，然后估计出每个变量的概率密度函数或者多个变量的联合概率密度函数，进而求出原始数据的最小充分统计量，并用最小充分统计量来构造统计量矩阵。最小充分统计量的引入还能够有效应对数据的非高斯分布问题。最后，通过在TE过程上的仿真测试验证了该方法用于故障检测的可行性和有效性。

关键词: 主元分析, 算法, 过程系统, 统计量模式分析, 最小充分统计量, 故障检测

CLC Number:

TP277

SUN Shuanzhu, DONG Shun, JIANG Yefeng, ZHOU Ting, LI Yiguo. Fault detection method based on minimum sufficient statistics pattern analysis[J]. CIESC Journal, 2018, 69(3): 1228-1237.

孙栓柱, 董顺, 江叶峰, 周挺, 李益国. 基于最小充分统计量模式分析的故障检测方法[J]. 化工学报, 2018, 69(3): 1228-1237.

References 30

[1]	GE Z, SONG Z, GAO F. Review of recent research on data-based process monitoring[J]. Industrial & Engineering Chemistry Research, 2013, 52(10):3543-3562.
[2]	RUSSELL E L, CHIANG L H, BRAATZ R D. Fault detection in industrial processes using canonical variate analysis and dynamic principal component analysis[J]. Chemometrics & Intelligent Laboratory Systems, 2000, 51(1):81-93.
[3]	SANG W C, LEE C, LEE J M, et al. Fault detection and identification of nonlinear processes based on kernel PCA[J]. Chemometrics & Intelligent Laboratory Systems, 2005, 75(1):55-67.
[4]	LI W, YUE H H, VALLE-CERVANTES S, et al. Recursive PCA for adaptive process monitoring[J]. Journal of Process Control, 2000, 10(5):471-486.
[5]	GE Z, ZHANG M, SONG Z. Nonlinear process monitoring based on linear subspace and Bayesian inference[J]. Journal of Process Control, 2010, 20(5):676-688.
[6]	HE Q P, WANG J. Statistics pattern analysis:a new process monitoring framework and its application to semiconductor batch processes[J]. AIChE Journal, 2015, 57 (1):107-121.
[7]	WANG J, HE Q P. Multivariate statistical process monitoring based on statistics pattern analysis[J]. Industrial & Engineering Chemistry Research, 2010, 49(17):7858-7869.
[8]	常鹏, 王普, 高学金,等. 基于统计量模式分析的MKPLS间歇过程监控与质量预报[J]. 仪器仪表学报, 2014, 35(6):1409-1416. CHANG P, WANG P, GAO X J, et al. Batch process monitoring and quality prediction based on statistics pattern analysis and MKPLS[J]. Chinese Journal of Scientific Instrument, 2014, 35(6):1409-1416.
[9]	常鹏, 王普, 高学金. 基于统计量模式分析的T-KPLS间歇过程故障监控[J]. 化工学报, 2015,66(1):265-271. CHANG P, WANG P, GAO X J. Fault monitoring batch process based on statistics pattern analysis of T-KPLS[J]. CIESC Journal, 2015, 66(1):265-271.
[10]	张汉元, 田学民, 邓晓刚. 基于SPA相似系数的故障识别方法[J]. 化工学报, 2013, 64(12):4503-4508. ZHANG H Y, TIAN X M, DENG X G. Fault identification method based on SPA similarity factor[J]. CIESC Journal, 2013, 64(12):4503-4508.
[11]	SHEN Y, DING S X, HAGHANI A, et al. A comparison study of basic data-driven fault diagnosis and process monitoring methods on the benchmark Tennessee Eastman process[J]. Journal of Process Control, 2012, 22(9):1567-1581.
[12]	DOWNS J J, VOGEL E F. A plant-wide industrial process control problem[J]. Computers & Chemical Engineering, 1993, 17(3):245-255.
[13]	RAO C R. Sufficient statistics and minimum variance estimates[J]. Mathematical Proceedings of the Cambridge Philosophical Society, 1949, 45(2):213-218.
[14]	COVER T M, THOMAS J A. Elements of Information Theory[M]. 2nd ed. Hoboken:John Wiley & Sons, 2012:221
[15]	OLIVE D J. Statistical Theory and Inference[M]. New York:Springer International Publishing, 2014:101-122
[16]	AYECH N, CHAKOUR C, HARKAT M F. New adaptive moving window PCA for process monitoring[J]. IFAC Proceedings Volumes, 2012, 45(20):606-611.
[17]	SHANG J, CHEN M, JI H, et al. Recursive transformed component statistical analysis for incipient fault detection[J]. Automatica, 2017, 80:313-327.
[18]	HAIMI H, MULAS M, CORONA F, et al. Adaptive data-derived anomaly detection in the activated sludge process of a large-scale wastewater treatment plant[J]. Engineering Applications of Artificial Intelligence, 2016, 52:65-80.
[19]	FISHER R A. Statistical methods for research workers[M]//Breakthroughs in Statistics. New York:Springer, 1992:66-70.
[20]	ŠKACH J, PUN?OCHÁ? I. Active fault detection:a comparison of probabilistic methods[C]//Journal of Physics:Conference Series. Bristol:IOP Publishing, 2015, 1742-6596.
[21]	SEVERSON K, CHAIWATANODOM P, BRAATZ R D. Perspectives on process monitoring of industrial systems[J]. Annual Reviews in Control, 2016, 42:190-200.
[22]	GE Z, SONG Z. Multivariate Statistical Process Control:Process Monitoring Methods and Applications[M]. New York:Springer Science & Business Media, 2012:18-20
[23]	CHIANG L H, RUSSELL E L, BRAATZ R D. Fault Detection and Diagnosis in Industrial Systems[M]. London:Springer London, 2001:103-109
[24]	LIU Y, ZHANG G, XU B. Compressive sparse principal component analysis for process supervisory monitoring and fault detection[J]. Journal of Process Control, 2017, 50:1-10.
[25]	ZHANG K, SHARDT Y A W, CHEN Z, et al. Using the expected detection delay to assess the performance of different multivariate statistical process monitoring methods for multiplicative and drift faults[J]. ISA Transactions, 2017, 67:56-66.
[26]	BAKDI A, KOUADRI A. A new adaptive PCA based thresholding scheme for fault detection in complex systems[J]. Chemometrics & Intelligent Laboratory Systems, 2017, 162:83-93.
[27]	RUSSELL E L, CHIANG L H, BRAATZ R D. Data-driven Methods for Fault Detection and Diagnosis in Chemical Processes[M]. New York:Springer Science & Business Media, 2012:84-90
[28]	薄翠梅, 韩晓春, 易辉,等. 基于聚类选择k近邻的LLE算法及故障检测[J]. 化工学报, 2016, 67(3):925-930. BO C M, HAN X C, YI H, et al. Neighborhood selection of LLE based on cluster for fault detection[J]. CIESC Journal, 2016, 67(3):925-930.
[29]	ZHANG Y. Fault detection and diagnosis of nonlinear processes using improved kernel independent component analysis (KICA) and support vector machine (SVM)[J]. Industrial & Engineering Chemistry Research, 2008, 47(18):6961-6971.
[30]	BERNAL-DE-LÁZARO J M, LLANES-SANTIAGO O, PRIETO-MORENO A, et al. Enhanced dynamic approach to improve the detection of small-magnitude faults[J]. Chemical Engineering Science, 2016, 146:166-179