A Proposal of On-Line Detection of New Faults and Automatic Learning in Fault Diagnosis

In this paper a new approach of automatic learning for a fault diagnosis system using fuzzy clustering techniques is presented. The proposal presents an off-line learning stage, for training the classifier to diagnose the initial known faults and the normal operation state. In this stage, the data are firstly pre-processed to eliminate outliers and reducing the confusion in the classification process by using the Density Objective Fuzzy C-Means (DOFCM) algorithm. Later on, the Kernel Fuzzy C-Means (KFCM) algorithm is used to achieve greater separability among the classes, and reducing the classification errors. Finally, a step is developed to optimize the two parameters used in the algorithms in the training stage using the Differential Evolution algorithm. After the training, the classifier is used on-line (recognition stage) in order to process every new sample taken from the process. In this stage, a novel fault detection algorithm is applied. The algorithm analyzes the observations which are not classified in the known classes and belonging to a window of time to determine if they constitute a new class, probably representative of a new fault or if they are noise. If a new class is identified, a procedure is developed to incorporate it to the known classes by the classifier. The approach proposed was validated using an illustrative example. The results obtained indicate the feasibility of the proposal.

[1]  Miin-Shen Yang,et al.  A cluster validity index for fuzzy clustering , 2005, Pattern Recognit. Lett..

[2]  James C. Bezdek,et al.  Pattern Recognition with Fuzzy Objective Function Algorithms , 1981, Advanced Applications in Pattern Recognition.

[3]  Hans-Peter Kriegel,et al.  A Density-Based Algorithm for Discovering Clusters in Large Spatial Databases with Noise , 1996, KDD.

[4]  Jian Xiao,et al.  A novel chaotic particle swarm optimization based fuzzy clustering algorithm , 2012, Neurocomputing.

[5]  Yan Yan Pang,et al.  Fault Diagnosis Method Based on KPCA and Selective Neural Network Ensemble , 2014 .

[6]  Steven X. Ding,et al.  Model-based Fault Diagnosis Techniques: Design Schemes, Algorithms, and Tools , 2008 .

[7]  Jicong Fan,et al.  Fault detection and diagnosis of non-linear non-Gaussian dynamic processes using kernel dynamic independent component analysis , 2014, Inf. Sci..

[8]  Le Hoang Son,et al.  Picture fuzzy clustering for complex data , 2016, Eng. Appl. Artif. Intell..

[9]  Ujjwal Maulik,et al.  Validity index for crisp and fuzzy clusters , 2004, Pattern Recognit..

[10]  Krzysztof Patan,et al.  Artificial Neural Networks for the Modelling and Fault Diagnosis of Technical Processes , 2008 .

[11]  Orhan Kesemen,et al.  Fuzzy c-means clustering algorithm for directional data (FCM4DD) , 2016, Expert Syst. Appl..

[12]  V. Steffen,et al.  Solution of inverse radiative transfer problems in two-layer participating media with differential evolution , 2010 .

[13]  Inseok Hwang,et al.  A Survey of Fault Detection, Isolation, and Reconfiguration Methods , 2010, IEEE Transactions on Control Systems Technology.

[14]  Adrián Rodríguez Ramos,et al.  An approach to multiple fault diagnosis using fuzzy logic , 2019, J. Intell. Manuf..

[15]  Witold Pedrycz,et al.  Fuzzy C-Means clustering of incomplete data based on probabilistic information granules of missing values , 2016, Knowl. Based Syst..

[16]  R. Weber,et al.  A Rough-Fuzzy approach for Support Vector Clustering , 2016, Inf. Sci..

[17]  Raghunathan Rengaswamy,et al.  A review of process fault detection and diagnosis: Part II: Qualitative models and search strategies , 2003, Comput. Chem. Eng..

[18]  Moshe Kam,et al.  A noise-resistant fuzzy c means algorithm for clustering , 1998, 1998 IEEE International Conference on Fuzzy Systems Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98CH36228).

[19]  Chee Peng Lim,et al.  A modified fuzzy min-max neural network for data clustering and its application to power quality monitoring , 2015, Appl. Soft Comput..

[20]  Dao-Qiang Zhang,et al.  A novel kernelized fuzzy C-means algorithm with application in medical image segmentation , 2004, Artif. Intell. Medicine.

[21]  Orestes Llanes-Santiago,et al.  A variant of the particle swarm optimization for the improvement of fault diagnosis in industrial systems via faults estimation , 2014, Eng. Appl. Artif. Intell..

[22]  Orestes Llanes-Santiago,et al.  Optimizing kernel methods to reduce dimensionality in fault diagnosis of industrial systems , 2015, Comput. Ind. Eng..

[23]  Qiang Wang,et al.  Robust level set image segmentation algorithm using local correntropy-based fuzzy c-means clustering with spatial constraints , 2016, Neurocomputing.

[24]  Anjana Gosain,et al.  A density oriented fuzzy C-means clustering algorithm for recognising original cluster shapes from noisy data , 2011 .

[25]  Orestes Llanes-Santiago,et al.  Enhanced dynamic approach to improve the detection of small-magnitude faults , 2016 .

[26]  Cesar A. Uribe,et al.  Unsupervised Feature Selection Based on Fuzzy Clustering for Fault Detection of the Tennessee Eastman Process , 2012, IBERAMIA.

[27]  Ali Sarosh,et al.  A novel KFCM based fault diagnosis method for unknown faults in satellite reaction wheels. , 2012, ISA transactions.

[28]  Raghunathan Rengaswamy,et al.  A review of process fault detection and diagnosis: Part I: Quantitative model-based methods , 2003, Comput. Chem. Eng..

[29]  Anjana Gosain,et al.  Performance Analysis of Various Fuzzy Clustering Algorithms: A Review , 2016 .

[30]  James M. Keller,et al.  A possibilistic fuzzy c-means clustering algorithm , 2005, IEEE Transactions on Fuzzy Systems.

[31]  Rajesh N. Davé,et al.  Robust clustering methods: a unified view , 1997, IEEE Trans. Fuzzy Syst..

[32]  Khashayar Khorasani,et al.  Dynamic neural network-based fault diagnosis of gas turbine engines , 2014, Neurocomputing.

[33]  Jiangping Wang,et al.  Vibration-based fault diagnosis of pump using fuzzy technique , 2006 .

[34]  Dao-Qiang Zhang,et al.  Clustering Incomplete Data Using Kernel-Based Fuzzy C-means Algorithm , 2003, Neural Processing Letters.

[35]  Rajesh N. Davé,et al.  Characterization and detection of noise in clustering , 1991, Pattern Recognit. Lett..

[36]  Joseph Aguilar-Martin,et al.  Automaton based on fuzzy clustering methods for monitoring industrial processes , 2013, Eng. Appl. Artif. Intell..

[37]  Jacek M. Leski,et al.  Fuzzy c-ordered-means clustering , 2016, Fuzzy Sets Syst..

[38]  James M. Keller,et al.  A possibilistic approach to clustering , 1993, IEEE Trans. Fuzzy Syst..

[39]  Chi-Man Vong,et al.  Simultaneous-fault detection based on qualitative symptom descriptions for automotive engine diagnosis , 2014, Appl. Soft Comput..

[40]  Qin Ming Liu,et al.  The Study of Fault Diagnosis Based on Particle Swarm Optimization Algorithm , 2009, Comput. Inf. Sci..

[41]  Anjana Gosain,et al.  Robust kernelized approach to clustering by incorporating new distance measure , 2013, Eng. Appl. Artif. Intell..

[42]  Meng Joo Er,et al.  Sequential fuzzy clustering based dynamic fuzzy neural network for fault diagnosis and prognosis , 2016, Neurocomputing.

[43]  Si-Zhao Joe Qin,et al.  Survey on data-driven industrial process monitoring and diagnosis , 2012, Annu. Rev. Control..

[44]  Sachin C. Patwardhan,et al.  A possibilistic clustering approach to novel fault detection and isolation , 2006 .

[45]  Orestes Llanes-Santiago,et al.  Fault Diagnosis in Industrial Systems Using Bioinspired Cooperative Strategies , 2010, NICSO.

[46]  Orestes Llanes-Santiago,et al.  The fault diagnosis inverse problem with Ant Colony Optimization and Ant Colony Optimization with dispersion , 2014, Appl. Math. Comput..