Adaptive fault diagnosis of HVCBs based on P-SVDD and P-KFCM

The proposed adaptive fault diagnosis model could contribute to resolving the lack of new knowledge acquisition and real-time model classification updating ability in current studies.The design of P-SVDD could provide a new solution for SVDD parameters optimization, which can guarantee the detection accuracy of SVDD efficiently.The design of P-KFCM could improve the clustering accuracy by avoiding being trapped in the local optimum and overcome the shortcoming of poor performance in the imbalanced classification.The design of the joint work between P-KFCM and its cluster validity could provide a facility to acquire a new fault category in the adaptive fault diagnosis model for updating its model classification. High voltage circuit breakers (HVCBs) are among most important pieces of equipment in the power system, and thus its fault diagnosis is quite essential for efficient operation. Most related research can only diagnose the known fault. However, the diagnosis may fail to work when an unknown fault occurs. Aimed to address the lack of new knowledge acquisition and real-time model classification updating, a novel method based on particle swarm optimization-support vector domain description (P-SVDD) and particle swarm optimization-kernel-based fuzzy c-means (P-KFCM) is proposed for adaptive fault diagnosis of HVCBs in this paper. In the proposed method, P-SVDD can detect the unknown fault sample by particle swarm optimization (PSO) parameter optimization while P-KFCM is used in known sample category recognition and its modified partition coefficient (MPC) cluster validity is used in unknown fault category search. The proposed method's operation process is introduced in detail, and the principles underlying the adaptive fault diagnosis model are discussed as well. In engineering application of an online monitoring system with fault diagnosis, the simulation results based on the measured HVCB closing coil current show that the proposed adaptive model can acquire new knowledge and update model classification in real-time with a higher diagnostic accuracy, compared with the existing algorithms.

[1]  Robert P. W. Duin,et al.  Support vector domain description , 1999, Pattern Recognit. Lett..

[2]  Rajesh N. Davé,et al.  Validating fuzzy partitions obtained through c-shells clustering , 1996, Pattern Recognit. Lett..

[3]  Zhongyi Hu,et al.  A PSO and pattern search based memetic algorithm for SVMs parameters optimization , 2013, Neurocomputing.

[4]  Wang Li-xin.  On-line Self-learning Fault Diagnosis for Circuit Breakers Based on Artificial Immune Network , 2009 .

[5]  Hongbo Xu,et al.  An intelligent fault identification method of rolling bearings based on LSSVM optimized by improved PSO , 2013 .

[6]  Guo Xue.  Mechanical Fault Diagnosis of High Voltage Circuit Breakers Based on Opening/Closing Coil Current Parameters , 2013 .

[7]  Robert Sabourin,et al.  Leave-One-Out-Training and Leave-One-Out-Testing Hidden Markov Models for a Handwritten Numeral Recognizer: The Implications of a Single Classifier and Multiple Classifications , 2009, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  Saeid Homayouni,et al.  An Improved FCM Algorithm Based on the SVDD for Unsupervised Hyperspectral Data Classification , 2013, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[9]  Hui Luo,et al.  A SVDD approach of fuzzy classification for analog circuit fault diagnosis with FWT as preprocessor , 2011, Expert Syst. Appl..

[10]  Haibo Chen,et al.  Reliability estimation of high voltage SF6 circuit breakers by statistical analysis on the basis of the field data , 2013 .

[11]  Changshui Zhang,et al.  Solving one-class problem with outlier examples by SVM , 2015, Neurocomputing.

[12]  Yi-Hung Liu,et al.  Fast Support Vector Data Descriptions for Novelty Detection , 2010, IEEE Transactions on Neural Networks.

[13]  Wu Yang,et al.  Mechanical condition recognition of medium-voltage vacuum circuit breaker based on mechanism dynamic features simulation and ANN , 2005 .

[14]  Bo Wang,et al.  The machine abnormal degree detection method based on SVDD and negative selection mechanism , 2013 .

[15]  Xiaowei Yang,et al.  A Kernel Fuzzy c-Means Clustering-Based Fuzzy Support Vector Machine Algorithm for Classification Problems With Outliers or Noises , 2011, IEEE Transactions on Fuzzy Systems.

[16]  Miin-Shen Yang,et al.  Alternative c-means clustering algorithms , 2002, Pattern Recognit..

[17]  Chengliang Liu,et al.  Fault Diagnosis of Plunger Pump in Truck Crane Based on Relevance Vector Machine with Particle Swarm Optimization Algorithm , 2013 .

[18]  Anton Janssen,et al.  International Surveys on Circuit-Breaker Reliability Data for Substation and System Studies , 2014, IEEE Transactions on Power Delivery.

[19]  Matti Lehtonen,et al.  Circuit-Breaker Automated Failure Tracking Based on Coil Current Signature , 2014, IEEE Transactions on Power Delivery.

[20]  Fan Yang,et al.  Support vector machine with genetic algorithm for machinery fault diagnosis of high voltage circuit breaker , 2011 .

[21]  Lijun Yang,et al.  A multiclass SVM-based classifier for transformer fault diagnosis using a particle swarm optimizer with time-varying acceleration coefficients , 2013 .

[22]  S. X. Yang,et al.  An Adaptive Approach Based on KPCA and SVM for Real-Time Fault Diagnosis of HVCBs , 2011, IEEE Transactions on Power Delivery.

[23]  Feng Zhao,et al.  Fuzzy clustering algorithms with self-tuning non-local spatial information for image segmentation , 2013, Neurocomputing.

[24]  Changyin Sun,et al.  A weighted LS-SVM approach for the identification of a class of nonlinear inverse systems , 2009, Science in China Series F: Information Sciences.

[25]  Ruijin Liao,et al.  An Integrated Decision-Making Model for Condition Assessment of Power Transformers Using Fuzzy Approach and Evidential Reasoning , 2011, IEEE Transactions on Power Delivery.

[26]  Xinghuo Yu,et al.  Internal model control based on a novel least square support vector machines for MIMO nonlinear discrete systems , 2010, Neural Computing and Applications.

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

[28]  Xiaohui Li,et al.  RMP neural network based dissolved gas analyzer for fault diagnostic of oil-filled electrical equipment , 2011, IEEE Transactions on Dielectrics and Electrical Insulation.

[29]  M. Lehtonen,et al.  Priority Assessment of Online Monitoring Investment for Power System Circuit Breakers—Part I: Qualitative-Quantitative Approach , 2013, IEEE Transactions on Power Delivery.

[30]  C. K. Babulal,et al.  Maximum loadability limit of power system using hybrid differential evolution with particle swarm optimization , 2012 .

[31]  Mei Fe.  Application of Particle Swarm Fused KFCM and Classification Model of SVM for Fault Diagnosis of Circuit Breaker , 2013 .

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

[33]  Ming Cheng,et al.  Fault diagnosis of wind turbine based on multisensors information fusion technology , 2014 .