多元过程监控与异常诊断研究
|
摘要
本论文以多元过程为研究对象，以计算机仿真分析为手段，利用联合控制技术、支持向量机与级联相关神经网络等，研究了多元统计过程控制中，多元过程均值向量和协方差矩阵的监控，以及多元过程质量诊断中，发现过程异常，定位过程失控原因的方法和实现技术，以便提高过程产品质量，减少质量缺陷，对提高企业的市场竞争力具有重要的意义。研究将进一步完善补充多元统计过程控制与异常诊断的技术与方法。本论文的主要研究内容包括：
1．多元过程均值向量与协方差矩阵监控。研究能同时监控多元过程均值向量和协方差矩阵的控制图，发现过程异常因素，显得尤为重要。基于此，本文提出了能较大范围有效地监控多元过程的均值向量和协方差矩阵的单个控制图，为企业在质量控制方面提供了较好的应用和参考价值；
2．针对多元过程中特殊的二元过程，本文提出了基于联合T2和VMAX控制图对其监控，研究了该联合控制图的ARL性能，并与联合T2和|S|控制图进行比较，该联合控制图能够有效地监控二元过程的均值向量和协方差矩阵；同时设计了联合单变量控制图，提出了基于非中心卡方分布与联合单变量控制图的二元过程均值向量偏移方向诊断方法，能为企业在线操作人员或质量工程师提供更加详尽的过程变异信息，以便快速的恢复过程的正常状态；
3．多元控制图均值偏移诊断的优化支持向量机方法研究。支持向量机能对小样本很好的学习，但其核函数参数的选择对模型的性能有较大的影响，本文首先利用K折交叉验证方法来选取支持向量机核函数的参数，构建多元控制图均值偏移诊断模型；而后提出应用离子群优化算法优选支持向量机的核函数参数，构建支持向量机模型诊断多元过程异常；最后分析了构建模型的性能，构建的模型能有效地对过程异常进行诊断；
4．级联相关神经网络在制造过程的应用研究。基于反向传播神经网络算法的优缺点，提出基于级联相关神经网络算法的多元过程异常诊断技术，并将该技术应用在多元制造过程中，通过仿真与实证分析表明，该技术能够有效地对过程异常进行诊断，诊断的准确率较高。
This dissertation studies some problems of multivariate statistical process controlusing joint control technology, support vector machine, and cascade correlation neuralnetwork through computer simulation analysis. The problems include monitoringmean vector and covariance matrix of multivariate process, and quality diagnosis ofmultivariate process for finding the process abnormalities, and locating the factor ofthe out-of-control process. This can improve process products quality and reducequality defects. It is meaningful to up-grate market competitiveness of enterprise.Research will further improve and add some technologies and methods of multivariatestatistical process control and diagnosis. The main contents are summarized asfollows:
1. Multivariate process mean vector and covariance matrix monitoring. It is veryimportant to design control charts simultaneously monitoring multivariate processmean vector and covariance matrix, and then identify process assignable factor. Basedon this point, in this dissertation, a single multivariate control chart is presented inorder to effectively monitor mean vector and covariance matrix of multivariateprocess in a big shift interval. This can provide a better application and referencevalue for enterprises with regard to quality control;
2. For a special bivariate process in multivariate processes, this dissertationpresents a joint control chart based on T2and VMAX statistics to monitor the process,studies the ARL performance of the joint control chart, and compare it with the T2and|S|charts. The joint control chart can effectively monitor the mean vector andcovariate matrix; meanwhile a joint single variable control chart is designed. Based onnon-central chi-square distribution of the process and the joint chart, diagnosismethod of mean vector shift direction is proposed. This can provide more details ofprocess variation information for enterprise online operation personnel or qualityengineer, in order to recover the process in-control as soon as possible;
3. Support vector machine optimization method for monitoring mean shift ofmultivariate control chart. Support vector machine can learn from small samples well,but selecting its kernel function parameter has largely effect on model of performance.On the one hand, using K folding cross validation method to select kernel function parameters of support vector machine, it constructs mean shift diagnosis model ofmultivariate control chart; on the other hand, using particle swarm optimizationmethod to select kernel function parameters of support vector machine, mean shiftdiagnosis model of multivariate control chart is proposed. Finally it analyzes theperformance of model. The results show the presented models can effectivelydiagnose out-of-control process;
4. Cascade correlation neural network application in the manufacturing process.Considering advantages and disadvantages of back-propagation neural networkalgorithm, this dissertation proposes multivariate process abnormalities diagnosistechnology based on cascade neural network algorithm, and applies it in multivariatemanufacturing processes. Simulation and empirical analysis suggest that thetechnology can be effective for process abnormalities diagnosis, and diagnosticaccuracy is higher.
1．多元过程均值向量与协方差矩阵监控。研究能同时监控多元过程均值向量和协方差矩阵的控制图，发现过程异常因素，显得尤为重要。基于此，本文提出了能较大范围有效地监控多元过程的均值向量和协方差矩阵的单个控制图，为企业在质量控制方面提供了较好的应用和参考价值；
2．针对多元过程中特殊的二元过程，本文提出了基于联合T2和VMAX控制图对其监控，研究了该联合控制图的ARL性能，并与联合T2和|S|控制图进行比较，该联合控制图能够有效地监控二元过程的均值向量和协方差矩阵；同时设计了联合单变量控制图，提出了基于非中心卡方分布与联合单变量控制图的二元过程均值向量偏移方向诊断方法，能为企业在线操作人员或质量工程师提供更加详尽的过程变异信息，以便快速的恢复过程的正常状态；
3．多元控制图均值偏移诊断的优化支持向量机方法研究。支持向量机能对小样本很好的学习，但其核函数参数的选择对模型的性能有较大的影响，本文首先利用K折交叉验证方法来选取支持向量机核函数的参数，构建多元控制图均值偏移诊断模型；而后提出应用离子群优化算法优选支持向量机的核函数参数，构建支持向量机模型诊断多元过程异常；最后分析了构建模型的性能，构建的模型能有效地对过程异常进行诊断；
4．级联相关神经网络在制造过程的应用研究。基于反向传播神经网络算法的优缺点，提出基于级联相关神经网络算法的多元过程异常诊断技术，并将该技术应用在多元制造过程中，通过仿真与实证分析表明，该技术能够有效地对过程异常进行诊断，诊断的准确率较高。
This dissertation studies some problems of multivariate statistical process controlusing joint control technology, support vector machine, and cascade correlation neuralnetwork through computer simulation analysis. The problems include monitoringmean vector and covariance matrix of multivariate process, and quality diagnosis ofmultivariate process for finding the process abnormalities, and locating the factor ofthe out-of-control process. This can improve process products quality and reducequality defects. It is meaningful to up-grate market competitiveness of enterprise.Research will further improve and add some technologies and methods of multivariatestatistical process control and diagnosis. The main contents are summarized asfollows:
1. Multivariate process mean vector and covariance matrix monitoring. It is veryimportant to design control charts simultaneously monitoring multivariate processmean vector and covariance matrix, and then identify process assignable factor. Basedon this point, in this dissertation, a single multivariate control chart is presented inorder to effectively monitor mean vector and covariance matrix of multivariateprocess in a big shift interval. This can provide a better application and referencevalue for enterprises with regard to quality control;
2. For a special bivariate process in multivariate processes, this dissertationpresents a joint control chart based on T2and VMAX statistics to monitor the process,studies the ARL performance of the joint control chart, and compare it with the T2and|S|charts. The joint control chart can effectively monitor the mean vector andcovariate matrix; meanwhile a joint single variable control chart is designed. Based onnon-central chi-square distribution of the process and the joint chart, diagnosismethod of mean vector shift direction is proposed. This can provide more details ofprocess variation information for enterprise online operation personnel or qualityengineer, in order to recover the process in-control as soon as possible;
3. Support vector machine optimization method for monitoring mean shift ofmultivariate control chart. Support vector machine can learn from small samples well,but selecting its kernel function parameter has largely effect on model of performance.On the one hand, using K folding cross validation method to select kernel function parameters of support vector machine, it constructs mean shift diagnosis model ofmultivariate control chart; on the other hand, using particle swarm optimizationmethod to select kernel function parameters of support vector machine, mean shiftdiagnosis model of multivariate control chart is proposed. Finally it analyzes theperformance of model. The results show the presented models can effectivelydiagnose out-of-control process;
4. Cascade correlation neural network application in the manufacturing process.Considering advantages and disadvantages of back-propagation neural networkalgorithm, this dissertation proposes multivariate process abnormalities diagnosistechnology based on cascade neural network algorithm, and applies it in multivariatemanufacturing processes. Simulation and empirical analysis suggest that thetechnology can be effective for process abnormalities diagnosis, and diagnosticaccuracy is higher.
引文
[1]Alt F B, Multivariate Quality Control, In: S. Kotz, N.L. Johnson, EditorsEncyclopedia of Statistical Sciences:6th New York, N.Y.: John Wiley&sons,1985,110-122.
[2]Spiring F A, Cheng S W, An Alternate Variables Control Chart: The Univariateand Multivariate Case, Stat Sinica,1998,8:273-287.
[3]Khoo M B C, A New Bivariate Control Chart to Monitor the MultivariateProcess Mean and Variance Simultaneously, Quality Engineering,2005,17(1):109-118.
[4]Chen G M, Cheng S W, Xie H S, A New Multivariate Control Chart forMonitoring Both Location and Dispersion, Commun Stat-Simul C,2005,34(1):203-217.
[5]Cheng S W, Mao H, A Multivariate Semi-Circle Control Chart for VariablesData, Quality Technology&Quantitive Management,2008,5(4):331-338.
[6]Bersimis S, Psarakis S, Panaretos J, Multivariate Statistical Process ControlCharts: An Overview, Qual Reliab Eng Int,2007,23(5):517-543.
[7]Hotelling H, Multivariate Quality Control-Illustrated by the Air Testing ofSample Bombsights, Techniques of statistical analysis,Eisenhart C,Hastay MW,Wallis WA(eds.).McGraw-Hill:New York,1947:111-184.
[8]Woodall W H, Ncube M M, Multivariate CUSUM Quality-Control Procedures,Technometrics,1985:285-292.
[9]Crosier R B, Multivariate Generalizations of Cumulative Sum Quality-ControlSchemes, Technometrics,1988:291-303.
[10]C. A. Lowry W H W, A Multivariate Exponentially Weighted Moving AverageControl Chart, Technometrics,1992,34(1):772-785.
[11]Sullivan J H, Woodall W H, A Comparison of Multivariate Control Charts forIndividual Observations, J Qual Technol,1996,28(4):398-408.
[12]Vargas N J A, Robust Estimation in Multivariate Control Charts for IndividualObservations, J Qual Technol,2003,35(4):367-376.
[13]Jensen W A, Birch J B, Woodall W H, High Breakdown Estimation Methods forPhase I Multivariate Control Charts, Qual Reliab Eng Int,2007,23(5):615-629.
[14]Variyath A M, Steiner S H, A Multivariate Robust Control Chart for IndividualObservations, J Qual Technol,2009,41(3):259-271.
[15]Rocke D M, Robust Control Charts, Technometrics,1989:173-184.
[16]Dou Y, Sa P, One-Sided Control Charts for the Mean of Positively SkewedDistributions, Total Quality Management,2002,13(7):1021-1033.
[17]Niang N, Multidimensional Methods for Statistical Process Control: SomeContributions of Robust Statistics, Multivariate Total Quality Control,2002:137-163.
[18]Rousseeuw P J, Van Zomeren B C, Unmasking Multivariate Outliers andLeverage Points, J Am Stat Assoc,1990:633-639.
[19]Rousseeuw P J, Van Driessen K, A Fast Algorithm for the Minimum CovarianceDeterminant Estimator, Technometrics,1999:212-223.
[20]Alfaro J L, Ortega J F, A Robust Alternative to Hotelling's T^2Control ChartUsing Trimmed Estimators, Qual Reliab Eng Int,2008,24(5):601-611.
[21]Djauhari M A, Mashuri M, Herwindiati D E, Multivariate Process VariabilityMonitoring, Commun Stat-Theor M,2008,37(11):1742-1754.
[22]Lowry C A, Montgomery D C, A Review of Multivariate Control Charts, IieTrans,1995,27(6):800-810.
[23]Alt F B, Smith N D, Multivariate Process Control, Handbook of Statistics,1988,7:333-351.
[24]Aparisi F, Jabaioyes J, Carrion A, Statistical Properties of the Lsi MultivariateControl Chart, Commun Stat-Theor M,1999,28(11):2671-2686.
[25]Mason R L, Chou Y M, Sullivan J Het al., Systematic Patterns in T^2Charts, JQual Technol,2003,35(1):47-58.
[26]Ryan T P, Statistical Methods for Quality Improvement: Wiley,2011,.
[27]Surtihadi J, Raghavachari M, Runger G, Multivariate Control Charts for ProcessDispersion, Int J Prod Res,2004,42(15):2993-3009.
[28]Healy J D, A Note On Multivariate Cusum Procedures, Technometrics,1987:409-412.
[29]Pignatiello J J, Runger G C, Comparisons of Multivariate CUSUM Charts, JQual Technol,1990,22(3):173-186.
[30]Woodall W H, Ncube M M, Multivariate CUSUM Quality-Control Procedures,Technometrics,1985:285-292.
[31]Hawkins D M, Multivariate Quality Control Based On Regression-AdjustedVariables, Technometrics,1991:61-75.
[32]Hauck D J, Runger G C, Montgomery D C, Multivariate Statistical ProcessMonitoring and Diagnosis with Grouped Regression‐Adjusted Variables,Commun Stat-Simul C,1999,28(2):309-328.
[33]Kramer H G, Schmid L V, EWMA Charts for Multivariate Time Series,Sequential analysis,1997,16(2):131-154.
[34]Hawkins D M, Choi S, Lee S, A General Multivariate Exponentially WeightedMoving-Average Control Chart, J Qual Technol,2007,39(2):118-125.
[35]Zou C, Tsung F, Directional Mewma Schemes for Multistage ProcessMonitoring and Diagnosis, J Qual Technol,2008,40(4):407-427.
[36]Qiu P, Hawkins D, A Rank-Based Multivariate CUSUM Procedure,Technometrics,2001,43(2):120-132.
[37]Qiu P, Hawkins D, A Nonparametric Multivariate Cumulative Sum Procedurefor Detecting Shifts in All Directions, Journal of the Royal Statistical Society:Series D (The Statistician),2003,52(2):151-164.
[38]Chan L K, Zhang J, Cumulative Sum Control Charts for the Covariance Matrix,Stat Sinica,2001,11(3):767-790.
[39]Yeh A B, Lin D, Zhou H Het al., A Multivariate Exponentially WeightedMoving Average Control Chart for Monitoring Process Variability, J Appl Stat,2003,30(5):507-536.
[40]Yeh A, Huwang L, Wu Y F, A Likelihood-Ratio-Based EWMA Control Chartfor Monitoring Variability of Multivariate Normal Processes, Iie Trans,2004,36(9):865-879.
[41]Yeh A B, Huwang L, Wu C, A Multivariate EWMA Control Chart forMonitoring Process Variability with Individual Observations, IIE Trans,2005,37(11):1023-1035.
[42]Hawkins D M, Maboudou-Tchao E M, Multivariate Exponentially WeightedMoving Covariance Matrix, Technometrics,2008,50(2):155-166.
[43]Yeh A B, Lin D K J, A New Variables Control Chart for SimultaneouslyMonitoring Multivariate Process Mean and Variability, International Journal ofReliability Quality and Safety Engineering,2002,9:41-60.
[44]Reynolds M R, Cho G Y, Multivariate Control Charts for Monitoring the MeanVector and Covariance Matrix, J Qual Technol,2006,38(3):230-253.
[45]Zamba K D, Hawkins D M, A Multivariate Change-Point Model for Change inMean Vector and/Or Covariance Structure, J Qual Technol,2009,41(3):285-303.
[46]Reynolds M R, Stoumbos Z G, Combinations of Multivariate Shewhart andMewma Control Charts for Monitoring the Mean Vector and Covariance Matrix,J Qual Technol,2008,40(4):381-393.
[47]Yeh A B, Lin D K J, Mcgrath R N, Multivariate Control Charts for MonitoringCovariance Matrix: A Review, Quality Technology and QuantitativeManagement,2006,3(4):415-436.
[48]Cheng S W, Thaga K, Single Variables Control Charts: An Overview, QualReliab Eng Int,2006,22(7):811-820.
[49]张公绪，质量控制与诊断七十年，质量与可靠性，1998，4：23-27.
[50]卓德保，徐济超，质量诊断技术及其应用综述，系统工程学报，2008，23(3)：338-346.
[51]Fuchs C, Benjamini Y, Multivariate Profile Charts for Statistical Process Control,Technometrics,1994:182-195.
[52]Sparks R, Adolphson A, Phatak A, Multivariate Process Monitoring Using theDynamic Biplot, Int Stat Rev,1997,65(3):325-349.
[53]Jackson J E, Wiley J, A User's Guide to Principal Components: Wiley OnlineLibrary,1991,.
[54]Kourti T, Macgregor J F, Multivariate SPC Methods for Process and ProductMonitoring, J Qual Technol,1996,28(4):409-428.
[55]Maravelakis P E, Bersimis S, Panaretos J et al., Identifying the Out of ControlVariable in a Multivariate Control Chart, Commun Stat-Theor M,2002,31(12):2391-2408.
[56]Choi S W, Martin E B, Morris A J et al., Fault Detection Based On aMaximum-Likelihood Principal Component Analysis (PCA) Mixture, Ind EngChem Res,2005,44(7):2316-2327.
[57]Mason R L, Tracy N D, Young J C, Decomposition of T^2for MultivariateControl Chart Interpretation, J Qual Technol,1995,27(2):99-1108.
[58]Mason R L, Tracy N D, Young J C, A Practical Approach for InterpretingMultivariate T^2Control Chart Signals, J Qual Technol,1997,29(4):396-406.
[59]Mason R L, Young J C, Improving the Sensitivity of the T^2Statistic inMultivariate Process Control, J Qual Technol,1999,31(2):155-165.
[60]Murphy B J, Selecting Out of Control Variables with the T^2MultivariateQuality Control Procedure, The Statistician,1987:571-581.
[61]Sullivan J H, Stoumbos Z G, Mason R Let al., Step-Down Analysis for Changesin the Covariance Matrix and Other Parameters, J Qual Technol,2007,39(1):66-84.
[62]Wang K, Jiang W, High-Dimensional Process Monitoring and Fault IsolationVia Variable Selection, J Qual Technol,2009,41(3):247-258.
[63]Yu J B, Xi L F, A Neural Network Ensemble-Based Model for On-LineMonitoring and Diagnosis of Out-of-Control Signals in MultivariateManufacturing Processes, Expert Syst Appl,2009,36(1):909-921.
[64]Niaki S, Abbasi B, Fault Diagnosis in Multivariate Control Charts UsingArtificial Neural Networks, Qual Reliab Eng Int,2005,21(8):825-840.
[65]Guh R S, On-Line Identification and Quantification of Mean Shifts in BivariateProcesses Using a Neural Network-Based Approach, Qual Reliab Eng Int,2007,23(3):367-385.
[66]Yu J B, Xi L F, Zhou X J, Intelligent Monitoring and Diagnosis ofManufacturing Processes Using an Integrated Approach of KBANN and GA,Comput Ind,2008,59(5):489-501.
[67]Hou T H, Liu W L, Lin L, Intelligent Remote Monitoring and Diagnosis ofManufacturing Processes Using an Integrated Approach of Neural Networks andRough Sets, J Intell Manuf,2003,14(2):239-253.
[68]Karim M, Halgamuge S, Smith A et al., Manufacturing Yield Improvement byClustering: Springer,2006,526-234.
[69]Wang C H, Recognition of Semiconductor Defect Patterns Using SpatialFiltering and Spectral Clustering, Expert Syst Appl,2008,34(3):1914-1923.
[70]Guh R S, Shiue Y R, An Effective Application of Decision Tree Learning forOn-Line Detection of Mean Shifts in Multivariate Control Charts, Comput IndEng,2008,55(2):475-493.
[71]Wang C H, Separation of Composite Defect Patterns On Wafer Bin Map UsingSupport Vector Clustering, Expert Syst Appl,2009,36(2):2554-2561.
[72]Zhang F, Luk T, A Data Mining Algorithm for Monitoring PCB AssemblyQuality, Electronics Packaging Manufacturing, IEEE Transactions on,2007,30(4):299-305.
[73]Wu R C, Chen R S, Chian S S, Design of a Product Quality Control SystemBased On the Use of Data Mining Techniques, IIE Trans,2006,38(1):39-51.
[74]Lau H C W, Ho G T S, Chu K Fet al., Development of an Intelligent QualityManagement System Using Fuzzy Association Rules, Expert Syst Appl,2009,36(2):1801-1815.
[75]董华，杨世元，吴德会，基于模糊支持向量机的小批量生产质量智能预测方法，系统工程理论与实践，2007，27(3)：98-104.
[76]杨世元，吴德会，苏海涛，基于PCA和SVM的控制图失控模式智能识别方法，系统仿真学报，2006，18(5)：1314-1318.
[77]邓晓刚，田学民，基于核规范变量分析的非线性故障诊断方法，控制与决策，2006，21(10)：1109-1113.
[78]Sun R, Tsung F, Qu L, Evolving Kernel Principal Component Analysis for FaultDiagnosis, Comput Ind Eng,2007,53(2):361-371.
[79]Ho G, Lau H, Lee Cet al., An Intelligent Production Workflow Mining Systemfor Continual Quality Enhancement, The International Journal of AdvancedManufacturing Technology,2006,28(7):792-809.
[80]Rokach L, Maimon O, Data Mining for Improving the Quality of Manufacturing:A Feature Set Decomposition Approach, J Intell Manuf,2006,17(3):285-299.
[81]Su C T, Wong J T, Designing MIMO Controller by Neuro-Traveling ParticleSwarm Optimizer Approach, Expert Syst Appl,2007,32(3):848-855.
[82]Chong I G, Albin S L, Jun C H, A Data Mining Approach to ProcessOptimization without an Explicit Quality Function, Iie Trans,2007,39(8):795-804.
[83]Dengiz O, Smith A E, Nettleship I, Two-Stage Data Mining for FlawIdentification in Ceramics Manufacture, Int J Prod Res,2006,44(14):2839-2851.
[84]Kwon Y, Ertekin Y M, Feature-Based Rule Induction in Machining OperationUsing Rough Set Theory for Quality Assurance, Robot Cim-Int Manuf,2005,21(6):559-567.
[85]杜福洲，唐晓青，基于PCA的多元质量控制与诊断方法研究，制造业自动化，2006，28(8)：10-13.
[86]吴以凡，艾丽君，欧阳树生等，面向钢铁生产过程质量控制的动态数据挖掘方法，冶金自动化，2006，30(4)：6-10.
[87]杜世昌，奚立峰，潘尔顺，面向质量控制的制造系统可诊断性分析，机械工程学报，2006，42(5)：149-154.
[88]Tracy N D, Young J C, Mason R L, Multivariate Control Charts for IndividualObservations, J Qual Technol,1992,24(2):88-95.
[89]Hawkins D M, The Detection of Errors in Multivariate Data Using PrincipalComponents, J Am Stat Assoc,1974:340-344.
[90]Montgomery D C, Introduction to Statistical Quality Control: Wiley-India,2007,.
[91]Jackson J E, Principal Components and Factor Analysis: Part I-PrincipalComponents, J Qual Technol,1980,12(4):201-213.
[92]Jackson J E, Multivariate Quality Control, Commun Stat-Theor M,1985,14(11):2657-2688.
[93]Niaki S, Nezhad M, Decision-Making in Detecting and Diagnosing Faults ofMultivariate Statistical Quality Control Systems, Int J Adv Manuf Tech,2009,42(7-8):713-724.
[94]Seber G A F, Multivariate Observations: Wiley Online Library,1984,.
[95]Alt F B, Smith N D, Multivariate Process Control,1988,333-351.
[96]Hecht-Nielsen R, Theory of the Backpropagation Neural Network: IEEE,1989,593-605.
[97]Aoki T, Ceravolo R, De Stefano Aet al., Seismic Vulnerability Assessment ofChemical Plants through Probabilistic Neural Networks, Reliab Eng Syst Safe,2002,77(3):263-268.
[98]Vapnik V N, The Nature of Statistical Learning Theory: Springer-Verlag NewYork Inc,2000,.
[99]Chen G M, Cheng S W, Xie H S, A New Multivariate Control Chart forMonitoring Both Location and Dispersion, Commun Stat-Simul C,2005,34(1):203-217.
[100]Aparisi F, Hotelling's T^2Control Chart with Adaptive Sample Sizes, Int J ProdRes,1996,34(10):2853-2862.
[101]Costa A, Machado M, A New Chart Based On Sample Variances for Monitoringthe Covariance Matrix of Multivariate Processes, The International Journal ofAdvanced Manufacturing Technology,2009,41(7):770-779.
[102]Machado M, Costa A, The Double Sampling and the EWMA Charts Based Onthe Sample Variances, Int J Prod Econ,2008,114(1):134-148.
[103]Machado M, Costa A, Rahim M A, The Synthetic Control Chart Based On TwoSample Variances for Monitoring the Covariance Matrix, Qual Reliab Eng Int,2009,25(5):595-606.
[104]Zhang G, Chang S, Multivariate EWMA Control Charts Using IndividualObservations for Process Mean and Variance Monitoring and Diagnosis, Int JProd Res,2008,46(24):6855-6881.
[105]Kourti T, Macgregor J F, Multivariate SPC Methods for Process and ProductMonitoring, J Qual Technol,1996,28(4):409-428.
[106]Mason R L, Champ C W, Tracy N Det al., Assessment of Multivariate ProcessControl Techniques, J Qual Technol,1997,29(2):140-143.
[107]Fuchs C, Benjamini Y, Multivariate Profile Charts for Statistical Process Control,Technometrics,1994:182-195.
[108]Machado M, Costa A, Monitoring the Mean Vector and the Covariance Matrixof Bivariate Process, Brazilian Journal of Operations&Production Management,2008,5(1):47-62.
[109]Cheng C S, Cheng H P, Identifying the Source of Variance Shifts in theMultivariate Process Using Neural Networks and Support Vector Machines,Expert Syst Appl,2008,35(1):198-206.
[110]Shao Y, Lu C J, Wang Y C, Determination of the Fault Quality Variables of aMultivariate Process Using Independent Component Analysis and Support VectorMachine, Advances in Data Mining. Applications and Theoretical Aspects,2010:338-349.
[111]Venkat Venkatasubramanian R R S N, A Review of Process Fault Detection andDiagnosis Part III: Process History Based Methods, Computers and ChemicalEngineering,2003,(27):327-346.
[112]N. V V，统计学习理论的本质，北京:清华大学出版社，2000.
[113]邓乃扬，田英杰，数据挖掘中的新方法：支持向量机，北京:科学出版社，2004.
[114]Burges C J C, A Tutorial On Support Vector Machines for Pattern Recognition,Data Min Knowl Disc,1998,(2):121-167.
[115]Hsu C W L C, A Comparison of Methods for Multiclass Support VectorMachines, Ieee T Neural Networ,2002,(13):415-425.
[116]Eberhart Rc K J, A New Optimizer Using Particle Swarm Theory.: In:Proc.ISMMH S.,1995,39-43.
[117]V Ranaee A E R G, Application of the PSOSVM Model for Recognition ofControl Chart Patterns, Isa T,2010,(49):577-586.
[118]Wang C H, Kuo W, Qi H, An Integrated Approach for Process MonitoringUsing Wavelet Analysis and Competitive Neural Network, Int J Prod Res,2007,45(1):227-244.
[119]Ribeiro B, Support Vector Machines for Quality Monitoring in a PlasticInjection Molding Process, Systems, Man, and Cybernetics, Part C: Applicationsand Reviews, IEEE Transactions on,2005,35(3):401-410.
[120]Guh R S, A Hybrid Learning-Based Model for On-Line Detection and Analysisof Control Chart Patterns, Comput Ind Eng,2005,49(1):35-62.
[121]Rumelhart D E, Hintont G E, Williams R J, Learning Representations byBack-Propagating Errors, Nature,1986,323(6088):533-536.
[122]Fahlman S E, The Cascade-Correlation Learning Architecture: DTIC Document,1990,.
[123]Chandra B, Varghese P P, Applications of Cascade Correlation Neural Networksfor Cipher System Identification,2007,311-314.
[124]Fausett L V, Fundamentals of Neural Networks: Prentice-Hall Englewood Cliffs,NJ,1994,.
[125]Haykin S, Neural Networks: A Comprehensive Foundation,1999:Prentice-Hall,.
[126]Dowla F U, Rogers L L, Solving Problems in Environmental Engineering andGeosciences with Artificial Neural Networks: The MIT Press,1995,.
[127]Fahlman S E, The Cascade-Correlation Learning Architecture: DTIC Document,1990,.
[128]Rumelhart D E, Learning Internal Representations by Error Propagation: DTICDocument,1985,.
[2]Spiring F A, Cheng S W, An Alternate Variables Control Chart: The Univariateand Multivariate Case, Stat Sinica,1998,8:273-287.
[3]Khoo M B C, A New Bivariate Control Chart to Monitor the MultivariateProcess Mean and Variance Simultaneously, Quality Engineering,2005,17(1):109-118.
[4]Chen G M, Cheng S W, Xie H S, A New Multivariate Control Chart forMonitoring Both Location and Dispersion, Commun Stat-Simul C,2005,34(1):203-217.
[5]Cheng S W, Mao H, A Multivariate Semi-Circle Control Chart for VariablesData, Quality Technology&Quantitive Management,2008,5(4):331-338.
[6]Bersimis S, Psarakis S, Panaretos J, Multivariate Statistical Process ControlCharts: An Overview, Qual Reliab Eng Int,2007,23(5):517-543.
[7]Hotelling H, Multivariate Quality Control-Illustrated by the Air Testing ofSample Bombsights, Techniques of statistical analysis,Eisenhart C,Hastay MW,Wallis WA(eds.).McGraw-Hill:New York,1947:111-184.
[8]Woodall W H, Ncube M M, Multivariate CUSUM Quality-Control Procedures,Technometrics,1985:285-292.
[9]Crosier R B, Multivariate Generalizations of Cumulative Sum Quality-ControlSchemes, Technometrics,1988:291-303.
[10]C. A. Lowry W H W, A Multivariate Exponentially Weighted Moving AverageControl Chart, Technometrics,1992,34(1):772-785.
[11]Sullivan J H, Woodall W H, A Comparison of Multivariate Control Charts forIndividual Observations, J Qual Technol,1996,28(4):398-408.
[12]Vargas N J A, Robust Estimation in Multivariate Control Charts for IndividualObservations, J Qual Technol,2003,35(4):367-376.
[13]Jensen W A, Birch J B, Woodall W H, High Breakdown Estimation Methods forPhase I Multivariate Control Charts, Qual Reliab Eng Int,2007,23(5):615-629.
[14]Variyath A M, Steiner S H, A Multivariate Robust Control Chart for IndividualObservations, J Qual Technol,2009,41(3):259-271.
[15]Rocke D M, Robust Control Charts, Technometrics,1989:173-184.
[16]Dou Y, Sa P, One-Sided Control Charts for the Mean of Positively SkewedDistributions, Total Quality Management,2002,13(7):1021-1033.
[17]Niang N, Multidimensional Methods for Statistical Process Control: SomeContributions of Robust Statistics, Multivariate Total Quality Control,2002:137-163.
[18]Rousseeuw P J, Van Zomeren B C, Unmasking Multivariate Outliers andLeverage Points, J Am Stat Assoc,1990:633-639.
[19]Rousseeuw P J, Van Driessen K, A Fast Algorithm for the Minimum CovarianceDeterminant Estimator, Technometrics,1999:212-223.
[20]Alfaro J L, Ortega J F, A Robust Alternative to Hotelling's T^2Control ChartUsing Trimmed Estimators, Qual Reliab Eng Int,2008,24(5):601-611.
[21]Djauhari M A, Mashuri M, Herwindiati D E, Multivariate Process VariabilityMonitoring, Commun Stat-Theor M,2008,37(11):1742-1754.
[22]Lowry C A, Montgomery D C, A Review of Multivariate Control Charts, IieTrans,1995,27(6):800-810.
[23]Alt F B, Smith N D, Multivariate Process Control, Handbook of Statistics,1988,7:333-351.
[24]Aparisi F, Jabaioyes J, Carrion A, Statistical Properties of the Lsi MultivariateControl Chart, Commun Stat-Theor M,1999,28(11):2671-2686.
[25]Mason R L, Chou Y M, Sullivan J Het al., Systematic Patterns in T^2Charts, JQual Technol,2003,35(1):47-58.
[26]Ryan T P, Statistical Methods for Quality Improvement: Wiley,2011,.
[27]Surtihadi J, Raghavachari M, Runger G, Multivariate Control Charts for ProcessDispersion, Int J Prod Res,2004,42(15):2993-3009.
[28]Healy J D, A Note On Multivariate Cusum Procedures, Technometrics,1987:409-412.
[29]Pignatiello J J, Runger G C, Comparisons of Multivariate CUSUM Charts, JQual Technol,1990,22(3):173-186.
[30]Woodall W H, Ncube M M, Multivariate CUSUM Quality-Control Procedures,Technometrics,1985:285-292.
[31]Hawkins D M, Multivariate Quality Control Based On Regression-AdjustedVariables, Technometrics,1991:61-75.
[32]Hauck D J, Runger G C, Montgomery D C, Multivariate Statistical ProcessMonitoring and Diagnosis with Grouped Regression‐Adjusted Variables,Commun Stat-Simul C,1999,28(2):309-328.
[33]Kramer H G, Schmid L V, EWMA Charts for Multivariate Time Series,Sequential analysis,1997,16(2):131-154.
[34]Hawkins D M, Choi S, Lee S, A General Multivariate Exponentially WeightedMoving-Average Control Chart, J Qual Technol,2007,39(2):118-125.
[35]Zou C, Tsung F, Directional Mewma Schemes for Multistage ProcessMonitoring and Diagnosis, J Qual Technol,2008,40(4):407-427.
[36]Qiu P, Hawkins D, A Rank-Based Multivariate CUSUM Procedure,Technometrics,2001,43(2):120-132.
[37]Qiu P, Hawkins D, A Nonparametric Multivariate Cumulative Sum Procedurefor Detecting Shifts in All Directions, Journal of the Royal Statistical Society:Series D (The Statistician),2003,52(2):151-164.
[38]Chan L K, Zhang J, Cumulative Sum Control Charts for the Covariance Matrix,Stat Sinica,2001,11(3):767-790.
[39]Yeh A B, Lin D, Zhou H Het al., A Multivariate Exponentially WeightedMoving Average Control Chart for Monitoring Process Variability, J Appl Stat,2003,30(5):507-536.
[40]Yeh A, Huwang L, Wu Y F, A Likelihood-Ratio-Based EWMA Control Chartfor Monitoring Variability of Multivariate Normal Processes, Iie Trans,2004,36(9):865-879.
[41]Yeh A B, Huwang L, Wu C, A Multivariate EWMA Control Chart forMonitoring Process Variability with Individual Observations, IIE Trans,2005,37(11):1023-1035.
[42]Hawkins D M, Maboudou-Tchao E M, Multivariate Exponentially WeightedMoving Covariance Matrix, Technometrics,2008,50(2):155-166.
[43]Yeh A B, Lin D K J, A New Variables Control Chart for SimultaneouslyMonitoring Multivariate Process Mean and Variability, International Journal ofReliability Quality and Safety Engineering,2002,9:41-60.
[44]Reynolds M R, Cho G Y, Multivariate Control Charts for Monitoring the MeanVector and Covariance Matrix, J Qual Technol,2006,38(3):230-253.
[45]Zamba K D, Hawkins D M, A Multivariate Change-Point Model for Change inMean Vector and/Or Covariance Structure, J Qual Technol,2009,41(3):285-303.
[46]Reynolds M R, Stoumbos Z G, Combinations of Multivariate Shewhart andMewma Control Charts for Monitoring the Mean Vector and Covariance Matrix,J Qual Technol,2008,40(4):381-393.
[47]Yeh A B, Lin D K J, Mcgrath R N, Multivariate Control Charts for MonitoringCovariance Matrix: A Review, Quality Technology and QuantitativeManagement,2006,3(4):415-436.
[48]Cheng S W, Thaga K, Single Variables Control Charts: An Overview, QualReliab Eng Int,2006,22(7):811-820.
[49]张公绪，质量控制与诊断七十年，质量与可靠性，1998，4：23-27.
[50]卓德保，徐济超，质量诊断技术及其应用综述，系统工程学报，2008，23(3)：338-346.
[51]Fuchs C, Benjamini Y, Multivariate Profile Charts for Statistical Process Control,Technometrics,1994:182-195.
[52]Sparks R, Adolphson A, Phatak A, Multivariate Process Monitoring Using theDynamic Biplot, Int Stat Rev,1997,65(3):325-349.
[53]Jackson J E, Wiley J, A User's Guide to Principal Components: Wiley OnlineLibrary,1991,.
[54]Kourti T, Macgregor J F, Multivariate SPC Methods for Process and ProductMonitoring, J Qual Technol,1996,28(4):409-428.
[55]Maravelakis P E, Bersimis S, Panaretos J et al., Identifying the Out of ControlVariable in a Multivariate Control Chart, Commun Stat-Theor M,2002,31(12):2391-2408.
[56]Choi S W, Martin E B, Morris A J et al., Fault Detection Based On aMaximum-Likelihood Principal Component Analysis (PCA) Mixture, Ind EngChem Res,2005,44(7):2316-2327.
[57]Mason R L, Tracy N D, Young J C, Decomposition of T^2for MultivariateControl Chart Interpretation, J Qual Technol,1995,27(2):99-1108.
[58]Mason R L, Tracy N D, Young J C, A Practical Approach for InterpretingMultivariate T^2Control Chart Signals, J Qual Technol,1997,29(4):396-406.
[59]Mason R L, Young J C, Improving the Sensitivity of the T^2Statistic inMultivariate Process Control, J Qual Technol,1999,31(2):155-165.
[60]Murphy B J, Selecting Out of Control Variables with the T^2MultivariateQuality Control Procedure, The Statistician,1987:571-581.
[61]Sullivan J H, Stoumbos Z G, Mason R Let al., Step-Down Analysis for Changesin the Covariance Matrix and Other Parameters, J Qual Technol,2007,39(1):66-84.
[62]Wang K, Jiang W, High-Dimensional Process Monitoring and Fault IsolationVia Variable Selection, J Qual Technol,2009,41(3):247-258.
[63]Yu J B, Xi L F, A Neural Network Ensemble-Based Model for On-LineMonitoring and Diagnosis of Out-of-Control Signals in MultivariateManufacturing Processes, Expert Syst Appl,2009,36(1):909-921.
[64]Niaki S, Abbasi B, Fault Diagnosis in Multivariate Control Charts UsingArtificial Neural Networks, Qual Reliab Eng Int,2005,21(8):825-840.
[65]Guh R S, On-Line Identification and Quantification of Mean Shifts in BivariateProcesses Using a Neural Network-Based Approach, Qual Reliab Eng Int,2007,23(3):367-385.
[66]Yu J B, Xi L F, Zhou X J, Intelligent Monitoring and Diagnosis ofManufacturing Processes Using an Integrated Approach of KBANN and GA,Comput Ind,2008,59(5):489-501.
[67]Hou T H, Liu W L, Lin L, Intelligent Remote Monitoring and Diagnosis ofManufacturing Processes Using an Integrated Approach of Neural Networks andRough Sets, J Intell Manuf,2003,14(2):239-253.
[68]Karim M, Halgamuge S, Smith A et al., Manufacturing Yield Improvement byClustering: Springer,2006,526-234.
[69]Wang C H, Recognition of Semiconductor Defect Patterns Using SpatialFiltering and Spectral Clustering, Expert Syst Appl,2008,34(3):1914-1923.
[70]Guh R S, Shiue Y R, An Effective Application of Decision Tree Learning forOn-Line Detection of Mean Shifts in Multivariate Control Charts, Comput IndEng,2008,55(2):475-493.
[71]Wang C H, Separation of Composite Defect Patterns On Wafer Bin Map UsingSupport Vector Clustering, Expert Syst Appl,2009,36(2):2554-2561.
[72]Zhang F, Luk T, A Data Mining Algorithm for Monitoring PCB AssemblyQuality, Electronics Packaging Manufacturing, IEEE Transactions on,2007,30(4):299-305.
[73]Wu R C, Chen R S, Chian S S, Design of a Product Quality Control SystemBased On the Use of Data Mining Techniques, IIE Trans,2006,38(1):39-51.
[74]Lau H C W, Ho G T S, Chu K Fet al., Development of an Intelligent QualityManagement System Using Fuzzy Association Rules, Expert Syst Appl,2009,36(2):1801-1815.
[75]董华，杨世元，吴德会，基于模糊支持向量机的小批量生产质量智能预测方法，系统工程理论与实践，2007，27(3)：98-104.
[76]杨世元，吴德会，苏海涛，基于PCA和SVM的控制图失控模式智能识别方法，系统仿真学报，2006，18(5)：1314-1318.
[77]邓晓刚，田学民，基于核规范变量分析的非线性故障诊断方法，控制与决策，2006，21(10)：1109-1113.
[78]Sun R, Tsung F, Qu L, Evolving Kernel Principal Component Analysis for FaultDiagnosis, Comput Ind Eng,2007,53(2):361-371.
[79]Ho G, Lau H, Lee Cet al., An Intelligent Production Workflow Mining Systemfor Continual Quality Enhancement, The International Journal of AdvancedManufacturing Technology,2006,28(7):792-809.
[80]Rokach L, Maimon O, Data Mining for Improving the Quality of Manufacturing:A Feature Set Decomposition Approach, J Intell Manuf,2006,17(3):285-299.
[81]Su C T, Wong J T, Designing MIMO Controller by Neuro-Traveling ParticleSwarm Optimizer Approach, Expert Syst Appl,2007,32(3):848-855.
[82]Chong I G, Albin S L, Jun C H, A Data Mining Approach to ProcessOptimization without an Explicit Quality Function, Iie Trans,2007,39(8):795-804.
[83]Dengiz O, Smith A E, Nettleship I, Two-Stage Data Mining for FlawIdentification in Ceramics Manufacture, Int J Prod Res,2006,44(14):2839-2851.
[84]Kwon Y, Ertekin Y M, Feature-Based Rule Induction in Machining OperationUsing Rough Set Theory for Quality Assurance, Robot Cim-Int Manuf,2005,21(6):559-567.
[85]杜福洲，唐晓青，基于PCA的多元质量控制与诊断方法研究，制造业自动化，2006，28(8)：10-13.
[86]吴以凡，艾丽君，欧阳树生等，面向钢铁生产过程质量控制的动态数据挖掘方法，冶金自动化，2006，30(4)：6-10.
[87]杜世昌，奚立峰，潘尔顺，面向质量控制的制造系统可诊断性分析，机械工程学报，2006，42(5)：149-154.
[88]Tracy N D, Young J C, Mason R L, Multivariate Control Charts for IndividualObservations, J Qual Technol,1992,24(2):88-95.
[89]Hawkins D M, The Detection of Errors in Multivariate Data Using PrincipalComponents, J Am Stat Assoc,1974:340-344.
[90]Montgomery D C, Introduction to Statistical Quality Control: Wiley-India,2007,.
[91]Jackson J E, Principal Components and Factor Analysis: Part I-PrincipalComponents, J Qual Technol,1980,12(4):201-213.
[92]Jackson J E, Multivariate Quality Control, Commun Stat-Theor M,1985,14(11):2657-2688.
[93]Niaki S, Nezhad M, Decision-Making in Detecting and Diagnosing Faults ofMultivariate Statistical Quality Control Systems, Int J Adv Manuf Tech,2009,42(7-8):713-724.
[94]Seber G A F, Multivariate Observations: Wiley Online Library,1984,.
[95]Alt F B, Smith N D, Multivariate Process Control,1988,333-351.
[96]Hecht-Nielsen R, Theory of the Backpropagation Neural Network: IEEE,1989,593-605.
[97]Aoki T, Ceravolo R, De Stefano Aet al., Seismic Vulnerability Assessment ofChemical Plants through Probabilistic Neural Networks, Reliab Eng Syst Safe,2002,77(3):263-268.
[98]Vapnik V N, The Nature of Statistical Learning Theory: Springer-Verlag NewYork Inc,2000,.
[99]Chen G M, Cheng S W, Xie H S, A New Multivariate Control Chart forMonitoring Both Location and Dispersion, Commun Stat-Simul C,2005,34(1):203-217.
[100]Aparisi F, Hotelling's T^2Control Chart with Adaptive Sample Sizes, Int J ProdRes,1996,34(10):2853-2862.
[101]Costa A, Machado M, A New Chart Based On Sample Variances for Monitoringthe Covariance Matrix of Multivariate Processes, The International Journal ofAdvanced Manufacturing Technology,2009,41(7):770-779.
[102]Machado M, Costa A, The Double Sampling and the EWMA Charts Based Onthe Sample Variances, Int J Prod Econ,2008,114(1):134-148.
[103]Machado M, Costa A, Rahim M A, The Synthetic Control Chart Based On TwoSample Variances for Monitoring the Covariance Matrix, Qual Reliab Eng Int,2009,25(5):595-606.
[104]Zhang G, Chang S, Multivariate EWMA Control Charts Using IndividualObservations for Process Mean and Variance Monitoring and Diagnosis, Int JProd Res,2008,46(24):6855-6881.
[105]Kourti T, Macgregor J F, Multivariate SPC Methods for Process and ProductMonitoring, J Qual Technol,1996,28(4):409-428.
[106]Mason R L, Champ C W, Tracy N Det al., Assessment of Multivariate ProcessControl Techniques, J Qual Technol,1997,29(2):140-143.
[107]Fuchs C, Benjamini Y, Multivariate Profile Charts for Statistical Process Control,Technometrics,1994:182-195.
[108]Machado M, Costa A, Monitoring the Mean Vector and the Covariance Matrixof Bivariate Process, Brazilian Journal of Operations&Production Management,2008,5(1):47-62.
[109]Cheng C S, Cheng H P, Identifying the Source of Variance Shifts in theMultivariate Process Using Neural Networks and Support Vector Machines,Expert Syst Appl,2008,35(1):198-206.
[110]Shao Y, Lu C J, Wang Y C, Determination of the Fault Quality Variables of aMultivariate Process Using Independent Component Analysis and Support VectorMachine, Advances in Data Mining. Applications and Theoretical Aspects,2010:338-349.
[111]Venkat Venkatasubramanian R R S N, A Review of Process Fault Detection andDiagnosis Part III: Process History Based Methods, Computers and ChemicalEngineering,2003,(27):327-346.
[112]N. V V，统计学习理论的本质，北京:清华大学出版社，2000.
[113]邓乃扬，田英杰，数据挖掘中的新方法：支持向量机，北京:科学出版社，2004.
[114]Burges C J C, A Tutorial On Support Vector Machines for Pattern Recognition,Data Min Knowl Disc,1998,(2):121-167.
[115]Hsu C W L C, A Comparison of Methods for Multiclass Support VectorMachines, Ieee T Neural Networ,2002,(13):415-425.
[116]Eberhart Rc K J, A New Optimizer Using Particle Swarm Theory.: In:Proc.ISMMH S.,1995,39-43.
[117]V Ranaee A E R G, Application of the PSOSVM Model for Recognition ofControl Chart Patterns, Isa T,2010,(49):577-586.
[118]Wang C H, Kuo W, Qi H, An Integrated Approach for Process MonitoringUsing Wavelet Analysis and Competitive Neural Network, Int J Prod Res,2007,45(1):227-244.
[119]Ribeiro B, Support Vector Machines for Quality Monitoring in a PlasticInjection Molding Process, Systems, Man, and Cybernetics, Part C: Applicationsand Reviews, IEEE Transactions on,2005,35(3):401-410.
[120]Guh R S, A Hybrid Learning-Based Model for On-Line Detection and Analysisof Control Chart Patterns, Comput Ind Eng,2005,49(1):35-62.
[121]Rumelhart D E, Hintont G E, Williams R J, Learning Representations byBack-Propagating Errors, Nature,1986,323(6088):533-536.
[122]Fahlman S E, The Cascade-Correlation Learning Architecture: DTIC Document,1990,.
[123]Chandra B, Varghese P P, Applications of Cascade Correlation Neural Networksfor Cipher System Identification,2007,311-314.
[124]Fausett L V, Fundamentals of Neural Networks: Prentice-Hall Englewood Cliffs,NJ,1994,.
[125]Haykin S, Neural Networks: A Comprehensive Foundation,1999:Prentice-Hall,.
[126]Dowla F U, Rogers L L, Solving Problems in Environmental Engineering andGeosciences with Artificial Neural Networks: The MIT Press,1995,.
[127]Fahlman S E, The Cascade-Correlation Learning Architecture: DTIC Document,1990,.
[128]Rumelhart D E, Learning Internal Representations by Error Propagation: DTICDocument,1985,.