Pub Date : 2017-05-01DOI: 10.1109/DDCLS.2017.8067718
Ruixuan Wang, Jing Wang, Jinglin Zhou, Haiyan Wu
An improved batch process fault identification approach with kernel exponential discriminant analysis (KEDA) is proposed, in which performance index based on difference degree is given to identify fault classification. This method takes the advantages of both the kernel technology and the exponential discriminant analysis technique. The proposed KEDA method shows powerful ability in dealing with nonlinear, small sample size data and it has a noticeable improvement in classification performance. During the real applications to fault identification, both the normal data model and the fault data model for known faults are established according to the historical data. Then online measurement data is fed into these models to identify the current operation status, i.e., is the system in normal or fault condition, what type of fault occurs, or does new fault appear? Finally, the proposed method is applied to a typical penicillin fermentation process and the simulation results show the effectiveness of the proposed KEDA algorithm and the good performance in fault classification.
{"title":"An improved kernel exponential discriminant analysis for fault identification of batch process","authors":"Ruixuan Wang, Jing Wang, Jinglin Zhou, Haiyan Wu","doi":"10.1109/DDCLS.2017.8067718","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8067718","url":null,"abstract":"An improved batch process fault identification approach with kernel exponential discriminant analysis (KEDA) is proposed, in which performance index based on difference degree is given to identify fault classification. This method takes the advantages of both the kernel technology and the exponential discriminant analysis technique. The proposed KEDA method shows powerful ability in dealing with nonlinear, small sample size data and it has a noticeable improvement in classification performance. During the real applications to fault identification, both the normal data model and the fault data model for known faults are established according to the historical data. Then online measurement data is fed into these models to identify the current operation status, i.e., is the system in normal or fault condition, what type of fault occurs, or does new fault appear? Finally, the proposed method is applied to a typical penicillin fermentation process and the simulation results show the effectiveness of the proposed KEDA algorithm and the good performance in fault classification.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"26 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120915000","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068110
Lingyu Li, Weili Niu
This paper deals with cooperative control problems for multi-agent networks subject to directed dynamic interactions. First, the directed dynamic interactions in these networks are described by directed dynamic graphs which admit adjacency weights of edges in terms of transfer functions. Then a distributed dynamic protocol is proposed based on the nearest neighbor rule, and the Lyapunov-based approach is adopted to perform convergence analysis of networks. Once the adjacency weights are appropriately designed, it is verified that the networks can achieve consensus if and only if the directed dynamic graph contains a spanning tree. Finally, the simulations are given to illustrate the effectiveness of the proposed consensus results for cooperative networks in the presence of directed dynamic interactions.
{"title":"Consensus problems on multi-agent networks with directed dynamic interactions","authors":"Lingyu Li, Weili Niu","doi":"10.1109/DDCLS.2017.8068110","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068110","url":null,"abstract":"This paper deals with cooperative control problems for multi-agent networks subject to directed dynamic interactions. First, the directed dynamic interactions in these networks are described by directed dynamic graphs which admit adjacency weights of edges in terms of transfer functions. Then a distributed dynamic protocol is proposed based on the nearest neighbor rule, and the Lyapunov-based approach is adopted to perform convergence analysis of networks. Once the adjacency weights are appropriately designed, it is verified that the networks can achieve consensus if and only if the directed dynamic graph contains a spanning tree. Finally, the simulations are given to illustrate the effectiveness of the proposed consensus results for cooperative networks in the presence of directed dynamic interactions.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125842389","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068141
Darong Huang, Zhenping Deng, Ling Zhao, Bo Mi
The traffic system is a nonlinear and time-varying complex system with the human body being involved. At the same time, the highly uncertain and nonlinear of short-term traffic flow has been presented due to the complicated factors such as road construction, traffic accident, complex weather and trip distribution. et. So the short-term traffic flow prediction is more difficult than the long-term forecast. First of all, aiming at the uncertain and time-varying characteristics of the short-term traffic flow, the main single-model prediction methods and the combination forecast models are summarized and analyzed. Finally, a combined forecasting method based on Markov chain theory and grey Verhuls model with less data demand and parameter is proposed. The experimental results show that the combined model can obtain a high prediction accuracy.
{"title":"A short-term traffic flow forecasting method based on Markov chain and grey Verhulst model","authors":"Darong Huang, Zhenping Deng, Ling Zhao, Bo Mi","doi":"10.1109/DDCLS.2017.8068141","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068141","url":null,"abstract":"The traffic system is a nonlinear and time-varying complex system with the human body being involved. At the same time, the highly uncertain and nonlinear of short-term traffic flow has been presented due to the complicated factors such as road construction, traffic accident, complex weather and trip distribution. et. So the short-term traffic flow prediction is more difficult than the long-term forecast. First of all, aiming at the uncertain and time-varying characteristics of the short-term traffic flow, the main single-model prediction methods and the combination forecast models are summarized and analyzed. Finally, a combined forecasting method based on Markov chain theory and grey Verhuls model with less data demand and parameter is proposed. The experimental results show that the combined model can obtain a high prediction accuracy.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"450 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122824027","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068061
Fangming Li, Jian Feng, Senxiang Lu, Jinhai Liu, Yu Yao
Magnetic flux leakage (MFL) inspection is one of the most commonly used nondestructive testing (NDE) technologies. This paper proposes a novel method for classifying the MFL response segments based on convolution neural network (CNN). In order to skip the procedure of saving the normalized MFL segment and save some computing time, a normalization layer is added to the proposed model. Moreover, the rectified linear units (ReLUs) is employed as the activation functions in the convolution layers to allow the proposed model to easily obtain sparse representations. The performance of the proposed model is tested by the real MFL data collected from the experimental pipelines. The results demonstrate that the presented method can achieve a satisfactory accuracy of MFL response segment classification and can be applied to practical application.
{"title":"Convolution neural network for classification of magnetic flux leakage response segments","authors":"Fangming Li, Jian Feng, Senxiang Lu, Jinhai Liu, Yu Yao","doi":"10.1109/DDCLS.2017.8068061","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068061","url":null,"abstract":"Magnetic flux leakage (MFL) inspection is one of the most commonly used nondestructive testing (NDE) technologies. This paper proposes a novel method for classifying the MFL response segments based on convolution neural network (CNN). In order to skip the procedure of saving the normalized MFL segment and save some computing time, a normalization layer is added to the proposed model. Moreover, the rectified linear units (ReLUs) is employed as the activation functions in the convolution layers to allow the proposed model to easily obtain sparse representations. The performance of the proposed model is tested by the real MFL data collected from the experimental pipelines. The results demonstrate that the presented method can achieve a satisfactory accuracy of MFL response segment classification and can be applied to practical application.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"100 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128306313","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068127
B. Chu, D. Owens, C. Freeman, Yanhong Liu
This paper proposes a novel point-to-point iterative learning control (ILC) algorithm for high performance trajectory tracking applications. Based on a successive project formulation of the point-to-point ILC design problem, two point-to-point ILC design algorithms are derived: one algorithm reCovers the norm optimal point to point ILC algorithm with a desirable physical property of converging to the minimum norm (energy) solution, and the other one (interestingly) accelerates convergence speed which could lead to significant reduction in system configuration time/cost. Numerical results are provided to demonstrate the proposed algorithms' effectiveness.
{"title":"Point-to-point ILC with accelerated convergence","authors":"B. Chu, D. Owens, C. Freeman, Yanhong Liu","doi":"10.1109/DDCLS.2017.8068127","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068127","url":null,"abstract":"This paper proposes a novel point-to-point iterative learning control (ILC) algorithm for high performance trajectory tracking applications. Based on a successive project formulation of the point-to-point ILC design problem, two point-to-point ILC design algorithms are derived: one algorithm reCovers the norm optimal point to point ILC algorithm with a desirable physical property of converging to the minimum norm (energy) solution, and the other one (interestingly) accelerates convergence speed which could lead to significant reduction in system configuration time/cost. Numerical results are provided to demonstrate the proposed algorithms' effectiveness.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129279036","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8067717
Minnan Piao, Kai Zhu, Mingwei Sun, Zenghui Wang, Zengqiang Chen
Active Disturbance Rejection Control (ADRC), an innovative control method, has been applied successfully in dealing with internal uncertainties and external disturbances. However, ADRC for time-delay plants is still a challenge due to the restriction on the bandwidth of the extended state observer (ESO). When designing the ESO for time-delay plants, both full-order and reduced-order extended state observers can be utilized and the selection of the proper order and the bandwidth of the ESO is necessary. Therefore we seek to compare the two kinds of observers in terms of the time-delay tolerance of the closed-loop stability and find an appropriate bandwidth to ensure the robustness to the time-delay uncertainty. First, the quantitative bandwidth relationship between the two kinds of observers is established with the equivalent effective bandwidth, and then the time-delay tolerance of the closed-loop stability are calculated for the first- and second-order plants to conduct the comparison, whilst the effects of the bandwidth and other parameters on the time-delay tolerance is analyzed. Furthermore, simulation results for a second-order plant are provided to verify the meaning of this investigation.
{"title":"Quantitative relationship in terms of time-delay tolerance of two kinds of extended state observers","authors":"Minnan Piao, Kai Zhu, Mingwei Sun, Zenghui Wang, Zengqiang Chen","doi":"10.1109/DDCLS.2017.8067717","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8067717","url":null,"abstract":"Active Disturbance Rejection Control (ADRC), an innovative control method, has been applied successfully in dealing with internal uncertainties and external disturbances. However, ADRC for time-delay plants is still a challenge due to the restriction on the bandwidth of the extended state observer (ESO). When designing the ESO for time-delay plants, both full-order and reduced-order extended state observers can be utilized and the selection of the proper order and the bandwidth of the ESO is necessary. Therefore we seek to compare the two kinds of observers in terms of the time-delay tolerance of the closed-loop stability and find an appropriate bandwidth to ensure the robustness to the time-delay uncertainty. First, the quantitative bandwidth relationship between the two kinds of observers is established with the equivalent effective bandwidth, and then the time-delay tolerance of the closed-loop stability are calculated for the first- and second-order plants to conduct the comparison, whilst the effects of the bandwidth and other parameters on the time-delay tolerance is analyzed. Furthermore, simulation results for a second-order plant are provided to verify the meaning of this investigation.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129144940","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068084
Xiaoxia Han, Lei Ma, Kun Feng
The distributed Coverage control of networked heterogeneous mobile robots is proposed. The heterogeneous robots are driven to their centroid of multiplicatively-weighted Voronoi regions under the action of the proposed control law, the target position is modified in real-time based on the density distribution of the environment. Two types driving mechanisms are considered in this paper, different control methods are used to drive different robots to their target positions. Simulative and experimental results are given to verify the effectiveness of the improved Coverage algorithm.
{"title":"Distributed coverage control of networked heterogeneous robots","authors":"Xiaoxia Han, Lei Ma, Kun Feng","doi":"10.1109/DDCLS.2017.8068084","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068084","url":null,"abstract":"The distributed Coverage control of networked heterogeneous mobile robots is proposed. The heterogeneous robots are driven to their centroid of multiplicatively-weighted Voronoi regions under the action of the proposed control law, the target position is modified in real-time based on the density distribution of the environment. Two types driving mechanisms are considered in this paper, different control methods are used to drive different robots to their target positions. Simulative and experimental results are given to verify the effectiveness of the improved Coverage algorithm.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"20 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129203105","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068099
Yu-ping Qin, Fengfeng Qin, Q. Leng, Aihua Zhang
Aim to multiclass text categorization problem, a classification algorithm based on multiconlitron and 1-a-r method is presented. 1-a-r method is used to convert a multiclass categorization problem to several binary problems. Multiconlitron is constructed for each binary problem in input space. For the text to be classified, its class is decided by multiconlitrons. The classification experiments are made on the Reuters 21578. Experimental results indicate that the proposed algorithm has better classification performance compare with 1-a-r SVMs.
{"title":"On multiclass text classification algorithm based on 1-a-r and multiconlitron","authors":"Yu-ping Qin, Fengfeng Qin, Q. Leng, Aihua Zhang","doi":"10.1109/DDCLS.2017.8068099","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068099","url":null,"abstract":"Aim to multiclass text categorization problem, a classification algorithm based on multiconlitron and 1-a-r method is presented. 1-a-r method is used to convert a multiclass categorization problem to several binary problems. Multiconlitron is constructed for each binary problem in input space. For the text to be classified, its class is decided by multiconlitrons. The classification experiments are made on the Reuters 21578. Experimental results indicate that the proposed algorithm has better classification performance compare with 1-a-r SVMs.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116516946","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}
The system health management technology uses observation data, system models, and relative intelligent algorithms to monitor system anomaly, evaluate system degradation, predict residual life, and further determine the corresponding maintenance and operation strategy. Recently, the in-orbit management work of spacecraft in China focuses on fault detection and diagnosis, whereas the spacecraft's health management technology stays in the framework study status. For the spacecraft's health monitoring, the theory and algorithm are important to support practical engineering. Therefore, this paper introduces the profust-reliability-based health monitoring method into the in-orbit management work of spacecraft. This method analyzes telemetry parameters of spacecraft, and calculates the profust reliability of telemetry parameters, component, and system, respectively. Then, the health level of the spacecraft is classified according to the calculated profust reliability value. This method provides a new thought to the in-orbit management work of spacecraft. The simulation result shows that this method can effectively evaluate the health status of spacecraft system.
{"title":"A profust-reliability-based health monitoring technique of spacecraft","authors":"Bo-neng Tan, Hai-Yan Yu, Jinlun Zhou, Danni Nian, Zhiyao Zhao","doi":"10.1109/DDCLS.2017.8068115","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068115","url":null,"abstract":"The system health management technology uses observation data, system models, and relative intelligent algorithms to monitor system anomaly, evaluate system degradation, predict residual life, and further determine the corresponding maintenance and operation strategy. Recently, the in-orbit management work of spacecraft in China focuses on fault detection and diagnosis, whereas the spacecraft's health management technology stays in the framework study status. For the spacecraft's health monitoring, the theory and algorithm are important to support practical engineering. Therefore, this paper introduces the profust-reliability-based health monitoring method into the in-orbit management work of spacecraft. This method analyzes telemetry parameters of spacecraft, and calculates the profust reliability of telemetry parameters, component, and system, respectively. Then, the health level of the spacecraft is classified according to the calculated profust reliability value. This method provides a new thought to the in-orbit management work of spacecraft. The simulation result shows that this method can effectively evaluate the health status of spacecraft system.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"106 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124086539","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 : 2017-05-01DOI: 10.1109/DDCLS.2017.8068149
Zhe Dong, Wenjuan Liu, Yueheng Li, Jie Han, Mengjiao Chen
Ozone is considered as one of the strongest oxidizing agent, yet it leaves no residues that are harmful to global environment. In this paper, the close loop control of ozone generator has been studied. The main concern of this issue is to achieve desired ozone concentration. Due to the ozone generation process is a complex nonlinear multivariable system, which is difficult to model and regulate, thus a date-driven neuro-control method is adopted to construct the dynamics of the system, and the adaptive dynamic programming algorithm(ADP) is used for controller design and optimization. According to the hardware-in-loop simulation, the ozone generation process can be effectively approximated by the neuro-network model, and the concentration and flow rate of ozone can be tracked by the ADP controller.
{"title":"Data-driven neuro-optimal tracking control of ozone generation process based on adaptive dynamic programming","authors":"Zhe Dong, Wenjuan Liu, Yueheng Li, Jie Han, Mengjiao Chen","doi":"10.1109/DDCLS.2017.8068149","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068149","url":null,"abstract":"Ozone is considered as one of the strongest oxidizing agent, yet it leaves no residues that are harmful to global environment. In this paper, the close loop control of ozone generator has been studied. The main concern of this issue is to achieve desired ozone concentration. Due to the ozone generation process is a complex nonlinear multivariable system, which is difficult to model and regulate, thus a date-driven neuro-control method is adopted to construct the dynamics of the system, and the adaptive dynamic programming algorithm(ADP) is used for controller design and optimization. According to the hardware-in-loop simulation, the ozone generation process can be effectively approximated by the neuro-network model, and the concentration and flow rate of ozone can be tracked by the ADP controller.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115806402","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：