Pub Date : 2019-07-01DOI: 10.1109/SAFEPROCESS45799.2019.9213380
Li Feng, Meng Deng, Shuiqing Xu, Ke Zhang
In this study, a sensor fault estimation framework is proposed for linear repetitive system. Firstly, the problem of sensor fault estimation is converted to state estimation via state redefinition. Then, state estimation is realized by conventional state observer. The uniformly convergence of error extended system is guaranteed by asymptotic stability. Afterwards, iterative learning law is presented for fault estimation. And the optimal function is designed for the iterative convergence. Finally, Linear matrix inequalities (LMIs) is utilized to obtain the specific feasible solution, thus to improve the performance of proposed method. Further, a numerical example is provided to demonstrate the effectiveness of the developed method.
{"title":"Sensor Fault Estimation via Iterative Learning Scheme for Linear Repetitive System","authors":"Li Feng, Meng Deng, Shuiqing Xu, Ke Zhang","doi":"10.1109/SAFEPROCESS45799.2019.9213380","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213380","url":null,"abstract":"In this study, a sensor fault estimation framework is proposed for linear repetitive system. Firstly, the problem of sensor fault estimation is converted to state estimation via state redefinition. Then, state estimation is realized by conventional state observer. The uniformly convergence of error extended system is guaranteed by asymptotic stability. Afterwards, iterative learning law is presented for fault estimation. And the optimal function is designed for the iterative convergence. Finally, Linear matrix inequalities (LMIs) is utilized to obtain the specific feasible solution, thus to improve the performance of proposed method. Further, a numerical example is provided to demonstrate the effectiveness of the developed method.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121303306","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/SAFEPROCESS45799.2019.9213433
Chao Jia, Hanwen Zhang
It is important to predict the remaining useful life (RUL) for evaluating the performance of industrial equipment. Many simple and complex methods have been proposed to predict RUL based on stochastic processes. However, these methods have different prediction accuracies. The uncertainty associated with using one of these methods instead of another is called statistical model uncertainty. Therefore, some problems naturally arise: How can we reduce the uncertainty among different methods? Is it possible to obtain a more exact prediction of RUL, compared with the individual method? In this study, we apply a Bayesian model aggregation (BMA) approach to solve these problems. For a Wiener degradation process with unknown parameters, assume that there are $P$ types of methods to predict RUL, for example, maximum likelihood estimation (MLE), stochastic Newton algorithm (SNA), and Kalman filter (KF)- based methods. Then, there are 2P- 1 distinct combinations of these $P$ types of methods, each with a corresponding statistical model and an estimated parameter vector. BMA can statistically combine these estimated parameter vectors through a weighted average, and thus, the probability density function (PDF) of RUL can be obtained. BMA can be successfully applied to realistic bearing data, and simulation results show that BMA achieves higher prediction accuracy than an individual method.
{"title":"RUL Prediction: Reducing Statistical Model Uncertainty Via Bayesian Model Aggregation","authors":"Chao Jia, Hanwen Zhang","doi":"10.1109/SAFEPROCESS45799.2019.9213433","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213433","url":null,"abstract":"It is important to predict the remaining useful life (RUL) for evaluating the performance of industrial equipment. Many simple and complex methods have been proposed to predict RUL based on stochastic processes. However, these methods have different prediction accuracies. The uncertainty associated with using one of these methods instead of another is called statistical model uncertainty. Therefore, some problems naturally arise: How can we reduce the uncertainty among different methods? Is it possible to obtain a more exact prediction of RUL, compared with the individual method? In this study, we apply a Bayesian model aggregation (BMA) approach to solve these problems. For a Wiener degradation process with unknown parameters, assume that there are $P$ types of methods to predict RUL, for example, maximum likelihood estimation (MLE), stochastic Newton algorithm (SNA), and Kalman filter (KF)- based methods. Then, there are 2P- 1 distinct combinations of these $P$ types of methods, each with a corresponding statistical model and an estimated parameter vector. BMA can statistically combine these estimated parameter vectors through a weighted average, and thus, the probability density function (PDF) of RUL can be obtained. BMA can be successfully applied to realistic bearing data, and simulation results show that BMA achieves higher prediction accuracy than an individual method.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126344761","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/SAFEPROCESS45799.2019.9213378
Ping Ma, Hongli Zhang, Cong Wang
In recent years, intelligent fault diagnosis models based on machine learning used for intelligent condition monitoring and diagnosis have achieved considerable success. However, in the current research, the diagnosis process is based on an assumption that the same feature distribution exists between training data and testing data. Regrettably, in real application, training data and testing data are often from diverse domains, the difference in feature distributions is often prevalent; in this case, the traditional diagnostic models lack adaptability. To address this issue, this work proposed a diagnosis framework based on domain adaptation. This framework is inspired by the domain adaptation ability of transfer learning, in that the model trained by the labeled data in source domain can be transferred to diagnose a new but similar target data. The domain adaptation algorithm transfer component analysis (TCA) and its improved algorithm- improved transfer component analysis (ITCA) are embedded into this framework, respectively, to verify its applicability. An experiment was conducted on the datasets of bearing to demonstrate the applicability and practicability of the proposed transfer framework. The results show that the proposed method presents high accuracy in the transfer task of bearing fault diagnosis under different conditions across diverse experimental positions and fault types.
{"title":"Improved Transfer Component Analysis and It Application for Bearing Fault Diagnosis Across Diverse Domains","authors":"Ping Ma, Hongli Zhang, Cong Wang","doi":"10.1109/SAFEPROCESS45799.2019.9213378","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213378","url":null,"abstract":"In recent years, intelligent fault diagnosis models based on machine learning used for intelligent condition monitoring and diagnosis have achieved considerable success. However, in the current research, the diagnosis process is based on an assumption that the same feature distribution exists between training data and testing data. Regrettably, in real application, training data and testing data are often from diverse domains, the difference in feature distributions is often prevalent; in this case, the traditional diagnostic models lack adaptability. To address this issue, this work proposed a diagnosis framework based on domain adaptation. This framework is inspired by the domain adaptation ability of transfer learning, in that the model trained by the labeled data in source domain can be transferred to diagnose a new but similar target data. The domain adaptation algorithm transfer component analysis (TCA) and its improved algorithm- improved transfer component analysis (ITCA) are embedded into this framework, respectively, to verify its applicability. An experiment was conducted on the datasets of bearing to demonstrate the applicability and practicability of the proposed transfer framework. The results show that the proposed method presents high accuracy in the transfer task of bearing fault diagnosis under different conditions across diverse experimental positions and fault types.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126444825","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/SAFEPROCESS45799.2019.9213336
M. Ma, Wei Dong, Xinya Sun, Xingquan Ji
The catenary of the high-speed rail power supply system is greatly affected by the weather during operation. Once it breaks down, there will be serious consequences. Besides, the mechanism of failure risk of catenary is complex so that it's difficult to analyze. Aiming at such characteristics, this paper proposes a dynamic flashover risk probability calculation method combining characteristic quantity based on Bayesian network. In this paper, the flashover risk propagation chain of the catenary in the humid and polluted environment is established and the probability mathematical model of the risk propagation process is given. In addition, the mechanism of risk propagation is used to establish the functional relation between the monitored characteristic quantity and the risk probability. Then the functional relation is used as the dynamic condition probability of Bayesian network to calculate the dynamic probability of the whole risk. The consequences of rail station passenger congestion caused by catenary flashover in bad weather are analyzed and the severity of consequence is determined to assess the dynamic risk level.
{"title":"A Dynamic Risk Analysis Method for High-speed Railway Catenary Based on Bayesian Network","authors":"M. Ma, Wei Dong, Xinya Sun, Xingquan Ji","doi":"10.1109/SAFEPROCESS45799.2019.9213336","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213336","url":null,"abstract":"The catenary of the high-speed rail power supply system is greatly affected by the weather during operation. Once it breaks down, there will be serious consequences. Besides, the mechanism of failure risk of catenary is complex so that it's difficult to analyze. Aiming at such characteristics, this paper proposes a dynamic flashover risk probability calculation method combining characteristic quantity based on Bayesian network. In this paper, the flashover risk propagation chain of the catenary in the humid and polluted environment is established and the probability mathematical model of the risk propagation process is given. In addition, the mechanism of risk propagation is used to establish the functional relation between the monitored characteristic quantity and the risk probability. Then the functional relation is used as the dynamic condition probability of Bayesian network to calculate the dynamic probability of the whole risk. The consequences of rail station passenger congestion caused by catenary flashover in bad weather are analyzed and the severity of consequence is determined to assess the dynamic risk level.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"130 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131937354","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
To guarantee the normal workflow and accurate brightness adjustment, it is important to predict fault of brightness sensor in rail vehicle compartment LED lighting system. In this paper, a BRB (belief rule base) based fault prediction model is proposed to accurate brightness adjustment and reliability based on the analysis of the failure mechanism of the brightness sensor in the rail vehicle compartment LED lighting system. The fault prediction model based on BRB can make full use of the system's expert prior knowledge, which can fuse the system feature quantity to achieve accurate fault prediction of the brightness sensor. In this process, the parameters of the model are updated by iterative estimation algorithm to compensate for the inaccuracy of expert knowledge. Finally, in order to verify the validity and accuracy of the proposed model, a case is studied by using the proposed prediction model for brightness sensor module in the rail vehicle compartment LED lighting system, which shows that the method can accurately predict the faults with qualitative knowledge and quantitative information.
{"title":"Fault Prediction of Brightness Sensor based on BRB in Rail Vehicle Compartment LED Lighting System","authors":"Xiaojing Yin, Guangxu Shi, Bangcheng Zhang, Shiyuan Lv, Yubo Shao","doi":"10.1109/SAFEPROCESS45799.2019.9213347","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213347","url":null,"abstract":"To guarantee the normal workflow and accurate brightness adjustment, it is important to predict fault of brightness sensor in rail vehicle compartment LED lighting system. In this paper, a BRB (belief rule base) based fault prediction model is proposed to accurate brightness adjustment and reliability based on the analysis of the failure mechanism of the brightness sensor in the rail vehicle compartment LED lighting system. The fault prediction model based on BRB can make full use of the system's expert prior knowledge, which can fuse the system feature quantity to achieve accurate fault prediction of the brightness sensor. In this process, the parameters of the model are updated by iterative estimation algorithm to compensate for the inaccuracy of expert knowledge. Finally, in order to verify the validity and accuracy of the proposed model, a case is studied by using the proposed prediction model for brightness sensor module in the rail vehicle compartment LED lighting system, which shows that the method can accurately predict the faults with qualitative knowledge and quantitative information.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"42 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134321077","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/SAFEPROCESS45799.2019.9213340
Jiaqi Mao, Fuyang Chen, B. Jiang, Li Wang
Taking the traction motor of CRH2 high-speed train as the research object, this paper proposes a composite fault diagnosis method based on park vector module for the composite fault of rotor broken bar and air gap eccentricity. Firstly, the current noise is reduced with the improved empirical mode decomposition method; and the three phase stator current is converted to park vector using the extension park vector method, to effectively avoid the case in which the composite fault features are submerged by the fundamental frequency characteristics; Secondly, the park vector module of stator current is transformed by fast Fourier transform, and compound fault features are extracted in frequency domain. Finally, the fault feature is put into the decision tree classifier to estimate the fault degree. The data of CRH2 semi-physical simulation platform are used to verify the validity of this method.
{"title":"Composite Fault Diagnosis of Rotor Broken Bar and Air Gap Eccentricity Based on Park Vector Module and Decision Tree Algorithm","authors":"Jiaqi Mao, Fuyang Chen, B. Jiang, Li Wang","doi":"10.1109/SAFEPROCESS45799.2019.9213340","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213340","url":null,"abstract":"Taking the traction motor of CRH2 high-speed train as the research object, this paper proposes a composite fault diagnosis method based on park vector module for the composite fault of rotor broken bar and air gap eccentricity. Firstly, the current noise is reduced with the improved empirical mode decomposition method; and the three phase stator current is converted to park vector using the extension park vector method, to effectively avoid the case in which the composite fault features are submerged by the fundamental frequency characteristics; Secondly, the park vector module of stator current is transformed by fast Fourier transform, and compound fault features are extracted in frequency domain. Finally, the fault feature is put into the decision tree classifier to estimate the fault degree. The data of CRH2 semi-physical simulation platform are used to verify the validity of this method.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"62 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131648061","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/SAFEPROCESS45799.2019.9213411
Yi Zhang, Zhongjun Ding, Changli Liu, Haibin Qi, Qingxin Zhao, Jie Huang, Xiao He
Fault monitoring for manned deep-sea submersibles has great significance for the safety of the pilots and equipment of submersibles. Based on an analysis of the existing fault monitoring methods and fault content of the manned deep-sea submersible JIAOLONG, faults actually occurred in JIAOLONG in the recent years are investigated in detail. Safety-oriented fault categorization for manned deep-sea submersible JIAOLONG is proposed. Possible research directions of fault monitoring techniques of manned deep-sea submersible are discussed.
{"title":"Safety-Oriented Fault Monitoring for Manned Deep-Sea Submersibles","authors":"Yi Zhang, Zhongjun Ding, Changli Liu, Haibin Qi, Qingxin Zhao, Jie Huang, Xiao He","doi":"10.1109/SAFEPROCESS45799.2019.9213411","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213411","url":null,"abstract":"Fault monitoring for manned deep-sea submersibles has great significance for the safety of the pilots and equipment of submersibles. Based on an analysis of the existing fault monitoring methods and fault content of the manned deep-sea submersible JIAOLONG, faults actually occurred in JIAOLONG in the recent years are investigated in detail. Safety-oriented fault categorization for manned deep-sea submersible JIAOLONG is proposed. Possible research directions of fault monitoring techniques of manned deep-sea submersible are discussed.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133376483","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/SAFEPROCESS45799.2019.9213259
Yuan Wang, Zhanshan Wang
This paper explores the intermediate observer-based fault estimation problem for nonlinear system with actuator faults, sensor faults and input disturbances. First, for sake of handling sensor faults conveniently, the system is transformed into augmented form. Second, the intermediate observer is utilized to simultaneously estimate the states, faults and input disturbances, which overcomes the constraint of observer matching condition. The estimation of input disturbances is introduced to raise the accuracy of fault estimation. Finally, by means of Lyapunov stability theory, the estimation errors are proved to be uniformly ultimately bounded. Simulations are given to validate the effectiveness and advantages of the developed approach.
{"title":"Intermediate Observer-based Fault Estimation for Nonlinear System with Input Disturbances","authors":"Yuan Wang, Zhanshan Wang","doi":"10.1109/SAFEPROCESS45799.2019.9213259","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213259","url":null,"abstract":"This paper explores the intermediate observer-based fault estimation problem for nonlinear system with actuator faults, sensor faults and input disturbances. First, for sake of handling sensor faults conveniently, the system is transformed into augmented form. Second, the intermediate observer is utilized to simultaneously estimate the states, faults and input disturbances, which overcomes the constraint of observer matching condition. The estimation of input disturbances is introduced to raise the accuracy of fault estimation. Finally, by means of Lyapunov stability theory, the estimation errors are proved to be uniformly ultimately bounded. Simulations are given to validate the effectiveness and advantages of the developed approach.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133425367","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/SAFEPROCESS45799.2019.9213393
Y. Liu, Xiuqing Wang, X. Ren, Feng Lyu
Deep convolution neural networks (DCNNs) are popular deep neural networks and are widely used in object recognition, handwriting recognition, image processing, and so on. In this paper, manipulator fault classifier based on DCNNs is proposed, and the sensor data from force and torque sensors are preprocessed and reconstructed into a new form that is suitable for the input of DCNNs. The experimental results show that the designed classifier can effectively distinguish time-series sensor data from the manipulator's normal state and various fault states. The proposed method aids in measurement, allowing the manipulator to recover from the fault state to normal working state, and is useful for enhancing the executive capability of manipulators.
{"title":"Deep Convolution Neural Networks for the Classification of Robot Execution Failures","authors":"Y. Liu, Xiuqing Wang, X. Ren, Feng Lyu","doi":"10.1109/SAFEPROCESS45799.2019.9213393","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213393","url":null,"abstract":"Deep convolution neural networks (DCNNs) are popular deep neural networks and are widely used in object recognition, handwriting recognition, image processing, and so on. In this paper, manipulator fault classifier based on DCNNs is proposed, and the sensor data from force and torque sensors are preprocessed and reconstructed into a new form that is suitable for the input of DCNNs. The experimental results show that the designed classifier can effectively distinguish time-series sensor data from the manipulator's normal state and various fault states. The proposed method aids in measurement, allowing the manipulator to recover from the fault state to normal working state, and is useful for enhancing the executive capability of manipulators.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132456108","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2019-07-01DOI: 10.1109/SAFEPROCESS45799.2019.9213244
Guobo Liao, Han Zhou, Yanxia Li, H. Yin, Y. Chai
Fault detection and identification could minimize unexpected degradation of system and further avoid dangerous situation. Due to the rapid development of sensor technology as well as the Internet, exponential data could be collected, resulting in that data-driven based fault diagnosis method receives increasing attention. However, most works often learned low dimensional representations so that they couldn't preserve the real local geometric structure of original data. This might degrade fault diagnosis capabilities. In this paper, a novel semi-supervised constraints propagation based approach for fault diagnosis was proposed. The key point was to spread the linking information of supervised data to its neighbors via constraints propagation. Accordingly, the propagated similarity matrix could correctly reflect the structure of the samples. Further, with the aid of propagated matrix, sample indexes were learned via singular value decomposition and support vector machine were utilized to identify the type of faults. The effectiveness of the proposed methods was demonstrated through the experimental results, compared with other popular fault diagnosis methods.
{"title":"A Semi-supervised Constraints Propagation Based Method for Fault Diagnosis","authors":"Guobo Liao, Han Zhou, Yanxia Li, H. Yin, Y. Chai","doi":"10.1109/SAFEPROCESS45799.2019.9213244","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213244","url":null,"abstract":"Fault detection and identification could minimize unexpected degradation of system and further avoid dangerous situation. Due to the rapid development of sensor technology as well as the Internet, exponential data could be collected, resulting in that data-driven based fault diagnosis method receives increasing attention. However, most works often learned low dimensional representations so that they couldn't preserve the real local geometric structure of original data. This might degrade fault diagnosis capabilities. In this paper, a novel semi-supervised constraints propagation based approach for fault diagnosis was proposed. The key point was to spread the linking information of supervised data to its neighbors via constraints propagation. Accordingly, the propagated similarity matrix could correctly reflect the structure of the samples. Further, with the aid of propagated matrix, sample indexes were learned via singular value decomposition and support vector machine were utilized to identify the type of faults. The effectiveness of the proposed methods was demonstrated through the experimental results, compared with other popular fault diagnosis methods.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"337 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133118542","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：