Pub Date : 2019-07-01DOI: 10.1109/SAFEPROCESS45799.2019.9213447
Xiuhua Tai, Tianxu Guo, Maoyin Chen, Junfeng Zhang, Donghua Zhou
The electric multiple unit (EMU) has become one of the most important components in strategic transportation in China. The braking system should receive special attention from both academic and practical perspectives. Understanding the fault of EMU braking system is one of the key points before doing research on fault detection. In this paper, we give a very brief introduction of the electro-pneumatic brake control structure and the network topology of EMU fault detection and diagnosis(FDD) module. Finally, a PCA based fault detection algorithm is proposed and the efficiency of the algorithm is verified through the experiment operated on a certain model of EMU braking system.
{"title":"Understanding the Fault in EMU Braking System","authors":"Xiuhua Tai, Tianxu Guo, Maoyin Chen, Junfeng Zhang, Donghua Zhou","doi":"10.1109/SAFEPROCESS45799.2019.9213447","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213447","url":null,"abstract":"The electric multiple unit (EMU) has become one of the most important components in strategic transportation in China. The braking system should receive special attention from both academic and practical perspectives. Understanding the fault of EMU braking system is one of the key points before doing research on fault detection. In this paper, we give a very brief introduction of the electro-pneumatic brake control structure and the network topology of EMU fault detection and diagnosis(FDD) module. Finally, a PCA based fault detection algorithm is proposed and the efficiency of the algorithm is verified through the experiment operated on a certain model of EMU braking system.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"1 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":"129641766","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/SAFEPROCESS45799.2019.9213431
Fengtian Kuang, Darong Huang
Aiming at the shortcomings of low prediction accuracy due to the randomness and complexity of power load data, this paper bring up a power load prediction method on the strength of VMD and dynamic adjustment BP. Firstly, for the redundant information and trend components contained in the original data of the power load, the VMD decomposed component reconstruction is used to remove the trend component and the redundant information. Secondly, after the VMD detrended, there is a disadvantage that the fixed points in traditional BP neural network prediction may cause low accuracy, the dynamic adjustment of nodes is designed to achieve the optimal prediction. Finally, based on the electric load data provided by Chongqing Tongnan Electric Power Co., Ltd., the prediction model put forward in this paper is used to estimate the electric load. The comparison of the example simulation results shows that the predicted values of the VMD and the dynamically adjusted BP cooperative electric load forecasting method are closer to the real one. The load value and the prediction error are lower, which is a better short-term power load forecasting method.
{"title":"Power Load Prediction Method Based on VMD and Dynamic Adjustment BP","authors":"Fengtian Kuang, Darong Huang","doi":"10.1109/SAFEPROCESS45799.2019.9213431","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213431","url":null,"abstract":"Aiming at the shortcomings of low prediction accuracy due to the randomness and complexity of power load data, this paper bring up a power load prediction method on the strength of VMD and dynamic adjustment BP. Firstly, for the redundant information and trend components contained in the original data of the power load, the VMD decomposed component reconstruction is used to remove the trend component and the redundant information. Secondly, after the VMD detrended, there is a disadvantage that the fixed points in traditional BP neural network prediction may cause low accuracy, the dynamic adjustment of nodes is designed to achieve the optimal prediction. Finally, based on the electric load data provided by Chongqing Tongnan Electric Power Co., Ltd., the prediction model put forward in this paper is used to estimate the electric load. The comparison of the example simulation results shows that the predicted values of the VMD and the dynamically adjusted BP cooperative electric load forecasting method are closer to the real one. The load value and the prediction error are lower, which is a better short-term power load forecasting method.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"109 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":"133226286","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/SAFEPROCESS45799.2019.9213307
Xu Chen, Xiao He
With the increasing number of deep-sea manned submersibles being on service, fault diagnosis for their control systems has become an indispensable task. Current hardware solutions of fault diagnosis are usually designed for a specific category of devices, which would lead to huge manual and economic costs when applied to deep-sea submersibles composed of subsystems varying in interface. In order to avoid the difficulties, a wireless-based fault diagnosis hardware solution is proposed which is applicable for systems with different electrical features. Moreover, it provides several functions including fault diagnosis, simulation of fault injection and direct control of actuators, and it not only possesses scalability for further analysis of target system but also expandability for applications besides deep-sea submersibles.
{"title":"Design of a Fault Diagnosis System for the “JiaoLong” Deep-sea Manned Vehicle","authors":"Xu Chen, Xiao He","doi":"10.1109/SAFEPROCESS45799.2019.9213307","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213307","url":null,"abstract":"With the increasing number of deep-sea manned submersibles being on service, fault diagnosis for their control systems has become an indispensable task. Current hardware solutions of fault diagnosis are usually designed for a specific category of devices, which would lead to huge manual and economic costs when applied to deep-sea submersibles composed of subsystems varying in interface. In order to avoid the difficulties, a wireless-based fault diagnosis hardware solution is proposed which is applicable for systems with different electrical features. Moreover, it provides several functions including fault diagnosis, simulation of fault injection and direct control of actuators, and it not only possesses scalability for further analysis of target system but also expandability for applications besides deep-sea submersibles.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"80 1 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":"116351986","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/SAFEPROCESS45799.2019.9213245
Ming Huang, Daqi Zhu, Zhenzhong Chu
In this paper, a thruster fault tolerant control combines with trajectory tracking control method is applied for 4500-m Human Occupied Vehicle. First, the tracking control method and thruster configuration of a human occupied vehicle with 4500m operation depth is simply introduced. Then control allocation problem of underwater vehicle is described, thruster forces reconstructed during control allocation. Finally, introduce a hybrid fault tolerant control method, this hybrid method is designed based on weighted pseudo-inverse matrixes and quantum particle swarm optimization (QPSO), compared with the classical weighted pseudo-inverse fault tolerant control, and simulations results illustrate the performance of the thruster fault tolerant control strategy.
{"title":"Thruster Fault Tolerant Control Scheme for 4500-m Human Occupied Vehicle","authors":"Ming Huang, Daqi Zhu, Zhenzhong Chu","doi":"10.1109/SAFEPROCESS45799.2019.9213245","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213245","url":null,"abstract":"In this paper, a thruster fault tolerant control combines with trajectory tracking control method is applied for 4500-m Human Occupied Vehicle. First, the tracking control method and thruster configuration of a human occupied vehicle with 4500m operation depth is simply introduced. Then control allocation problem of underwater vehicle is described, thruster forces reconstructed during control allocation. Finally, introduce a hybrid fault tolerant control method, this hybrid method is designed based on weighted pseudo-inverse matrixes and quantum particle swarm optimization (QPSO), compared with the classical weighted pseudo-inverse fault tolerant control, and simulations results illustrate the performance of the thruster fault tolerant control strategy.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"11 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":"128253154","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/safeprocess45799.2019.9213424
Guangwu Chen, Yijian Yu, Dongfeng Xing, Juhau Yang
With the rapid development of Chinese railways, railway station signal control system has developed rapidly with the help of the fourth generation of all-electronic interlocking system. According to the control circuit and switching state in switch module of electronic interlocking system and monitor switching current, analysis the monitoring machine of turnout active current, the characteristic input value of turnout is extracted and turnout fault model is established. Firstly, data training and test is classified by BP neural network, then strong classifier is constructed by optimized Adaboost, the matching classification between turnout characteristic quantity and turnout fault type is carried out. After simulation, when BP neural network algorithm is used alone, the fault diagnosis rate is 90.2%, while the strong classification effect of BP_Adaboost algorithm can improve accuracy of turnout fault diagnosis by 95.8%, and the accuracy of latter is 5% higher than that of the former. The method validity is verified, which provides important research significance for turnout fault diagnosis of all-electronic interlocking system.
{"title":"Turnout Fault Diagnosis Algorithms of Full-Electronic Interlocking System Based on BP_Adaboost","authors":"Guangwu Chen, Yijian Yu, Dongfeng Xing, Juhau Yang","doi":"10.1109/safeprocess45799.2019.9213424","DOIUrl":"https://doi.org/10.1109/safeprocess45799.2019.9213424","url":null,"abstract":"With the rapid development of Chinese railways, railway station signal control system has developed rapidly with the help of the fourth generation of all-electronic interlocking system. According to the control circuit and switching state in switch module of electronic interlocking system and monitor switching current, analysis the monitoring machine of turnout active current, the characteristic input value of turnout is extracted and turnout fault model is established. Firstly, data training and test is classified by BP neural network, then strong classifier is constructed by optimized Adaboost, the matching classification between turnout characteristic quantity and turnout fault type is carried out. After simulation, when BP neural network algorithm is used alone, the fault diagnosis rate is 90.2%, while the strong classification effect of BP_Adaboost algorithm can improve accuracy of turnout fault diagnosis by 95.8%, and the accuracy of latter is 5% higher than that of the former. The method validity is verified, which provides important research significance for turnout fault diagnosis of all-electronic interlocking system.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"92 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":"128640427","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/SAFEPROCESS45799.2019.9213440
Wei Wu, Yunfeng Kang, L. Yao
In this paper, a learning observer (LO) based manipulators sensor fault diagnosis (FD) scheme is proposed. The dynamic model of the manipulator is taken as the research object and the effects of the disturbance is considered. When the fault occurs in the sensor, a learning observer is designed to obtain the fault information. Correspondingly the stability analysis of the observation error system is carried out using Lyapunov stability theorem. Then, a sliding mode fault tolerant controller is designed to make the manipulator can track the desired trajectory. Finally, a simulation example is given to prove the effectiveness of the algorithm.
{"title":"Learning Observer Based Fault Diagnosis and Fault Tolerant Control for Manipulators with Sensor Fault","authors":"Wei Wu, Yunfeng Kang, L. Yao","doi":"10.1109/SAFEPROCESS45799.2019.9213440","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213440","url":null,"abstract":"In this paper, a learning observer (LO) based manipulators sensor fault diagnosis (FD) scheme is proposed. The dynamic model of the manipulator is taken as the research object and the effects of the disturbance is considered. When the fault occurs in the sensor, a learning observer is designed to obtain the fault information. Correspondingly the stability analysis of the observation error system is carried out using Lyapunov stability theorem. Then, a sliding mode fault tolerant controller is designed to make the manipulator can track the desired trajectory. Finally, a simulation example is given to prove the effectiveness of the algorithm.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"1 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":"124217413","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/SAFEPROCESS45799.2019.9213370
Changli Liu, Yi Zhang, Xiao He
Safety assessment is of great importance to the deep-sea manned submersible, but little literature has been reported on this topic. The goal of this paper is to work out an effective tool for the safety assessment of the deep-sea manned submersible according to the study of JiaoLong, which is the first manned submersible that can dive more than 7,000 meters in China. In this paper, a relatively new subsystem division of the manned submersible is introduced firstly. Furthermore, a BN-based safety assessment method is proposed which combines the Bayesian Network (BN) and data-driven fault detection algorithms. Based on the BN, qualitative and quantitative analysis can both be implemented. Moreover, real-time safety assessment can be realized by combining data-driven fault detection algorithms. The proposed method is verified on the JiaoLong manned submersible by constructing and analyzing the BN. Also, an example of the propeller fault detection using kernel principal component analysis (KPCA) is displayed to illustrate how to employ the proposed method in real-time.
{"title":"Safety Assessment of the JiaoLong Deep-sea Manned Submersible based on Bayesian Network","authors":"Changli Liu, Yi Zhang, Xiao He","doi":"10.1109/SAFEPROCESS45799.2019.9213370","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213370","url":null,"abstract":"Safety assessment is of great importance to the deep-sea manned submersible, but little literature has been reported on this topic. The goal of this paper is to work out an effective tool for the safety assessment of the deep-sea manned submersible according to the study of JiaoLong, which is the first manned submersible that can dive more than 7,000 meters in China. In this paper, a relatively new subsystem division of the manned submersible is introduced firstly. Furthermore, a BN-based safety assessment method is proposed which combines the Bayesian Network (BN) and data-driven fault detection algorithms. Based on the BN, qualitative and quantitative analysis can both be implemented. Moreover, real-time safety assessment can be realized by combining data-driven fault detection algorithms. The proposed method is verified on the JiaoLong manned submersible by constructing and analyzing the BN. Also, an example of the propeller fault detection using kernel principal component analysis (KPCA) is displayed to illustrate how to employ the proposed method in real-time.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"1 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":"130325199","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}
In this paper, a new control allocation method which is based on the improved grey wolf optimizer (IGWO) algorithm is proposed for redundant control of aircraft with multiple actuators. Firstly, we introduce the ADMIRE model which is an aircraft with multiple actuators. Then, the controller based on the linear quadratic regulator (LQR) theory is designed, and the control allocation method based on IGWO algorithm is introduced. Finally, to prove the method is effectiveness, the actuator without failure and the actuator with loss of effectiveness failure are both considered in the simulation. The results show that the attitude angle control of aircraft with multiple actuators can be realized by this method.
{"title":"A New Control Allocation Method Based on the Improved Grey Wolf Optimizer Algorithm for Aircraft with Multiple Actuators","authors":"Wendong Gai, Chengxian Sun, Yecheng Zhou, Jing Zhang","doi":"10.1109/SAFEPROCESS45799.2019.9213444","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213444","url":null,"abstract":"In this paper, a new control allocation method which is based on the improved grey wolf optimizer (IGWO) algorithm is proposed for redundant control of aircraft with multiple actuators. Firstly, we introduce the ADMIRE model which is an aircraft with multiple actuators. Then, the controller based on the linear quadratic regulator (LQR) theory is designed, and the control allocation method based on IGWO algorithm is introduced. Finally, to prove the method is effectiveness, the actuator without failure and the actuator with loss of effectiveness failure are both considered in the simulation. The results show that the attitude angle control of aircraft with multiple actuators can be realized by this method.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"36 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":"127164937","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/SAFEPROCESS45799.2019.9213442
Yang Wang, D. Ling, Weidong Yang, Bo Tao, Ying Zheng
In order for the fault detection of processes with noise and nonlinearity, a method based on Ensemble Empirical Mode Decomposition (EEMD) and Support Vector Data Description (SVDD) is proposed. In this work, EEMD-based denoising method is utilized to remove the noise from the original dataset. The SVDD model is then developed to handle the nonlinear data for fault detection. The proposed method contains three steps. Firstly, the original dataset is decomposed into a series of Intrinsic Mode Functions (IMFs) by the EEMD method. Each IMF characterizes the corresponding scale information of the data. Secondly, the original data is reconstructed using the partial reconstruction denoising method. Only the relevant IMFs which mostly contain useful information are retained, and the IMFs that primarily carry noise are discarded. The optimal number of relevant IMFs is selected based on the Signal-to-Noise Ratio (SNR). Finally, the SVDD model is constructed on the reconstructed data to detect faults. The effectiveness of the proposed method is demonstrated by a numerical example. The results show the proposed method performs better compared with other existing methods.
{"title":"A Fault Detection Method with Ensemble Empirical Mode Decomposition and Support Vector Data Description","authors":"Yang Wang, D. Ling, Weidong Yang, Bo Tao, Ying Zheng","doi":"10.1109/SAFEPROCESS45799.2019.9213442","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213442","url":null,"abstract":"In order for the fault detection of processes with noise and nonlinearity, a method based on Ensemble Empirical Mode Decomposition (EEMD) and Support Vector Data Description (SVDD) is proposed. In this work, EEMD-based denoising method is utilized to remove the noise from the original dataset. The SVDD model is then developed to handle the nonlinear data for fault detection. The proposed method contains three steps. Firstly, the original dataset is decomposed into a series of Intrinsic Mode Functions (IMFs) by the EEMD method. Each IMF characterizes the corresponding scale information of the data. Secondly, the original data is reconstructed using the partial reconstruction denoising method. Only the relevant IMFs which mostly contain useful information are retained, and the IMFs that primarily carry noise are discarded. The optimal number of relevant IMFs is selected based on the Signal-to-Noise Ratio (SNR). Finally, the SVDD model is constructed on the reconstructed data to detect faults. The effectiveness of the proposed method is demonstrated by a numerical example. The results show the proposed method performs better compared with other existing methods.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"1 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":"129098022","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/SAFEPROCESS45799.2019.9213324
Guo-Zhu Wang, Zhi-Yong Du, Yong-Tao Hu, Yuan Li
In modern chemical processes, varieties of fault detection and diagnosis methods have been used for ensuring process safety and product quality widely. As an important branch, fault detection and diagnosis methods based on data-driven are effective in large-scale chemical processes. However, they do not often show superior performance owing to the self-limitations and the characteristics of process data, such as nonlinearity, non-Gaussian, and multi-operating mode. To cope with these issues, k-NN (k-Nearest Neighbor) fault detection method and its extension have been developed in recent years. Nevertheless, these methods are used for fault detection mainly, few papers can be found about fault diagnosis. In this paper, a novel abnormal variables identification method is proposed, this method uses k-NN distance contribution analysis theory to evaluate which variables are most likely to be abnormal, meanwhile, the feasibility of this method is verified by contribution decomposition theory. The proposed search strategy can guarantee that all abnormal variables are found in each sample. The reliability and validity of the proposed method are verified by a numerical example and the Continuous Stirred Tank Reactor system.
{"title":"Fault Diagnosis of Chemical Processes Based on k-NN Distance Contribution Analysis Method","authors":"Guo-Zhu Wang, Zhi-Yong Du, Yong-Tao Hu, Yuan Li","doi":"10.1109/SAFEPROCESS45799.2019.9213324","DOIUrl":"https://doi.org/10.1109/SAFEPROCESS45799.2019.9213324","url":null,"abstract":"In modern chemical processes, varieties of fault detection and diagnosis methods have been used for ensuring process safety and product quality widely. As an important branch, fault detection and diagnosis methods based on data-driven are effective in large-scale chemical processes. However, they do not often show superior performance owing to the self-limitations and the characteristics of process data, such as nonlinearity, non-Gaussian, and multi-operating mode. To cope with these issues, k-NN (k-Nearest Neighbor) fault detection method and its extension have been developed in recent years. Nevertheless, these methods are used for fault detection mainly, few papers can be found about fault diagnosis. In this paper, a novel abnormal variables identification method is proposed, this method uses k-NN distance contribution analysis theory to evaluate which variables are most likely to be abnormal, meanwhile, the feasibility of this method is verified by contribution decomposition theory. The proposed search strategy can guarantee that all abnormal variables are found in each sample. The reliability and validity of the proposed method are verified by a numerical example and the Continuous Stirred Tank Reactor system.","PeriodicalId":353946,"journal":{"name":"2019 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes (SAFEPROCESS)","volume":"55 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":"126441040","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：