Pub Date : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455626
Youdao Ma, Wenhan Zhang, Xinyang Liu, Zhenhua Wang, Yi Shen
This paper studies the data-driven fault symptoms generation and augmentation for satellite attitude control system via an approximate model technique and a generative adversarial network. An approximate model is determined to fit the input and output data of satellite attitude control system. Based on the designed model, a small number of addictive fault symptoms and multiplicative fault symptoms are generated. To obtain abundant symptom data, the generative adversarial network is introduced to augment the fault symptoms. Finally, numerical simulation results are presented to demonstrate the effectiveness of the proposed method.
{"title":"Data-Driven Fault Symptoms Generation and Augmentation for Satellite Attitude Control System","authors":"Youdao Ma, Wenhan Zhang, Xinyang Liu, Zhenhua Wang, Yi Shen","doi":"10.1109/DDCLS52934.2021.9455626","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455626","url":null,"abstract":"This paper studies the data-driven fault symptoms generation and augmentation for satellite attitude control system via an approximate model technique and a generative adversarial network. An approximate model is determined to fit the input and output data of satellite attitude control system. Based on the designed model, a small number of addictive fault symptoms and multiplicative fault symptoms are generated. To obtain abundant symptom data, the generative adversarial network is introduced to augment the fault symptoms. Finally, numerical simulation results are presented to demonstrate the effectiveness of the proposed method.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"76 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133875714","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 order to attenuate the influence of the uncertainties of high altitude parafoil and environment on trajectory tracking control, active disturbance rejection control (ADRC) is used to regulate the trajectory of the high-altitude wind power parafoil. Linear extended state observer (LESO) is designed to estimate and compensate for nonlinear disturbances of the system. The simulation results show that this method has good control precision and fast-tracking velocity.
{"title":"Trajectory Tracking Control of High-Altitude Wind Power Parafoil","authors":"Xinyu Long, Mingwei Sun, Minnan Piao, Shengfei Liu, Zengqiang Chen","doi":"10.1109/DDCLS52934.2021.9455649","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455649","url":null,"abstract":"In order to attenuate the influence of the uncertainties of high altitude parafoil and environment on trajectory tracking control, active disturbance rejection control (ADRC) is used to regulate the trajectory of the high-altitude wind power parafoil. Linear extended state observer (LESO) is designed to estimate and compensate for nonlinear disturbances of the system. The simulation results show that this method has good control precision and fast-tracking velocity.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"172 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123010773","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 : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455479
Jin Yang, Qishui Zhong, Kaibo Shi, S. Zhong, Shengzhi Han
In this paper, the sampled-data consensus problem of nonlinear multiagent systems (MASs) with time-varying delays is investigated. Compared with the widely used sampled-data controller, a proportional integral type (PI-type) protocol utilizing the information of neighbors considering the effects of memory delay is adopted. Then, by adequately considering characteristic about the time-varying delays, an improved time-varying quadratic type of Lyapunov-Krasovskii functional (LKF) is developed. Besides, augmented state vectors and two-sided looped-functional approach are adopting to constructed the LKF, some relaxed matrices in the LKF are not necessarily positive definite. Furthermore, some sufficient criteria are derived to ensure the consistency of the MASs. By solving a series of linear matrix inequalities, the desired memory PI-type sampled-data control gain matrices are obtained. Finally, the numerical examples are presented to illustrate the theoretical results.
{"title":"Memory-Based PI-Type Sampled-Data Consensus Control for Nonlinear Multiagent Systems with Time-Varying Delays","authors":"Jin Yang, Qishui Zhong, Kaibo Shi, S. Zhong, Shengzhi Han","doi":"10.1109/DDCLS52934.2021.9455479","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455479","url":null,"abstract":"In this paper, the sampled-data consensus problem of nonlinear multiagent systems (MASs) with time-varying delays is investigated. Compared with the widely used sampled-data controller, a proportional integral type (PI-type) protocol utilizing the information of neighbors considering the effects of memory delay is adopted. Then, by adequately considering characteristic about the time-varying delays, an improved time-varying quadratic type of Lyapunov-Krasovskii functional (LKF) is developed. Besides, augmented state vectors and two-sided looped-functional approach are adopting to constructed the LKF, some relaxed matrices in the LKF are not necessarily positive definite. Furthermore, some sufficient criteria are derived to ensure the consistency of the MASs. By solving a series of linear matrix inequalities, the desired memory PI-type sampled-data control gain matrices are obtained. Finally, the numerical examples are presented to illustrate the theoretical results.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124674380","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}
Wireless capsule endoscopy (WCE) has been widely used in the detection of digestive tract diseases because of its painlessness and convenience. Accurate classification of WCE abnormal images is very crucial for the diagnosis and treatment of early gastrointestinal tumors, while it remains challenging due to the ambiguous boundary between lesions and normal tissues. In order to overcome the above limitations, a three-branch effectively fused attention guided convolutional neural network (EFAG-CNN) which imitates the practical diagnosis process is proposed. Specifically, global features and local images with suppressed background noise are generated by branch1 and local features are extracted by branch2 based on the local images. What's more, an effective attention feature fusion (EAFF) module is devised and inserted into branch3 to make the final prediction, which helps adaptively capture more discriminative features for classification. EAFF can integrate the representative features from branch1 and branch2 better than other methods. Furthermore, we propose a joint loss function to enhance the classification performance of branch2. Extensive experimental results demonstrate that the overall classification accuracy of the proposed method on the public Kvasir dataset reaches 96.50%, which is superior to the state-of-the-art deep learning methods.
{"title":"EFAG-CNN: Effectively fused attention guided convolutional neural network for WCE image classification","authors":"Jing Cao, Jiafeng Yao, Zhibo Zhang, Shan Cheng, Sheng Li, Jinhui Zhu, Xiongxiong He, Qianru Jiang","doi":"10.1109/DDCLS52934.2021.9455575","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455575","url":null,"abstract":"Wireless capsule endoscopy (WCE) has been widely used in the detection of digestive tract diseases because of its painlessness and convenience. Accurate classification of WCE abnormal images is very crucial for the diagnosis and treatment of early gastrointestinal tumors, while it remains challenging due to the ambiguous boundary between lesions and normal tissues. In order to overcome the above limitations, a three-branch effectively fused attention guided convolutional neural network (EFAG-CNN) which imitates the practical diagnosis process is proposed. Specifically, global features and local images with suppressed background noise are generated by branch1 and local features are extracted by branch2 based on the local images. What's more, an effective attention feature fusion (EAFF) module is devised and inserted into branch3 to make the final prediction, which helps adaptively capture more discriminative features for classification. EAFF can integrate the representative features from branch1 and branch2 better than other methods. Furthermore, we propose a joint loss function to enhance the classification performance of branch2. Extensive experimental results demonstrate that the overall classification accuracy of the proposed method on the public Kvasir dataset reaches 96.50%, which is superior to the state-of-the-art deep learning methods.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132408526","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 : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455537
D. Shen, Jiaxi Qian
With the rapid development of communication technology, network control is widely used. In the process of wireless transmission, a signal may be affected by the attenuation channel. In this paper, we review the recent advances in learning control with fading channels. We first study the case that the fading channel statistics are known, then we turn to the unknown case. We also make some comparisons among these results to illustrate the newly developed techniques. This review paper may assist the readers in understanding the progress of the researches on the design of fading channel algorithms as well as the related issues in multiplicative randomness.
{"title":"Recent Advances in Iterative Learning Control with Fading Channel","authors":"D. Shen, Jiaxi Qian","doi":"10.1109/DDCLS52934.2021.9455537","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455537","url":null,"abstract":"With the rapid development of communication technology, network control is widely used. In the process of wireless transmission, a signal may be affected by the attenuation channel. In this paper, we review the recent advances in learning control with fading channels. We first study the case that the fading channel statistics are known, then we turn to the unknown case. We also make some comparisons among these results to illustrate the newly developed techniques. This review paper may assist the readers in understanding the progress of the researches on the design of fading channel algorithms as well as the related issues in multiplicative randomness.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134473804","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 : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455685
Shaoxue Jing
The ARMAX model is widely used in industrial modeling. However, the traditional stochastic information gradient algorithm for ARMAX identification needs less computation, but its convergence speed is too slow. To accelerate the algorithm, we propose a two-step algorithm based on a gradient acceleration strategy. The first step is to replace the error scalar with the error vector, and the second step is to introduce a momentum related to the gradient. The simulation results show that the proposed algorithm can obtain more accurate estimation and the convergence speed is greatly improved.
{"title":"Identification of an ARMAX model based on a momentum-accelerated multi-error stochastic information gradient algorithm","authors":"Shaoxue Jing","doi":"10.1109/DDCLS52934.2021.9455685","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455685","url":null,"abstract":"The ARMAX model is widely used in industrial modeling. However, the traditional stochastic information gradient algorithm for ARMAX identification needs less computation, but its convergence speed is too slow. To accelerate the algorithm, we propose a two-step algorithm based on a gradient acceleration strategy. The first step is to replace the error scalar with the error vector, and the second step is to introduce a momentum related to the gradient. The simulation results show that the proposed algorithm can obtain more accurate estimation and the convergence speed is greatly improved.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"37 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134461137","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}
Bearing Remaining Useful Life (RUL) prediction has important meaning in the mechanical maintenance. However, the existing RUL algorithms cannot achieve stable prediction. Therefore, an improved bearing health monitoring algorithm based on Bidirectional Long Short-Term Memory (BiLSTM) integrating Conditional Random Field (BiLSTM-CRF) is proposed. The empirical mode decomposition (EMD) algorithm is used to decompose the bearing diagnostic signal into several intrinsic mode function (IMF) components. Moreover, the effective IMF component is selected to reconstruct the signal by combining the crosscorrelation coefficient and kurtosis criterion. Through the reconstructed signal extracting the time-frequency features into a feature vector, the feature data with lower dimension can be got. Then, the feature with lower dimension as inputs and RUL status as the output are used to train the BiLSTM-CRF model, which can achieve more accurate predictions. Finally, the XJTU-SY bearing data is used to verify the effectiveness of the proposed algorithm. Experiments show that this proposed method can get the best performance comparing with the convolutional neural networks and the Long Short-Term Memory.
{"title":"Bearing Health Monitoring Based on the Improved BiISTM-CRF","authors":"Zhiqiang Geng, Xin Zhang, Yongming Han, Chengmei Zhang, Kai Chen, Feng Xie","doi":"10.1109/ddcls52934.2021.9455471","DOIUrl":"https://doi.org/10.1109/ddcls52934.2021.9455471","url":null,"abstract":"Bearing Remaining Useful Life (RUL) prediction has important meaning in the mechanical maintenance. However, the existing RUL algorithms cannot achieve stable prediction. Therefore, an improved bearing health monitoring algorithm based on Bidirectional Long Short-Term Memory (BiLSTM) integrating Conditional Random Field (BiLSTM-CRF) is proposed. The empirical mode decomposition (EMD) algorithm is used to decompose the bearing diagnostic signal into several intrinsic mode function (IMF) components. Moreover, the effective IMF component is selected to reconstruct the signal by combining the crosscorrelation coefficient and kurtosis criterion. Through the reconstructed signal extracting the time-frequency features into a feature vector, the feature data with lower dimension can be got. Then, the feature with lower dimension as inputs and RUL status as the output are used to train the BiLSTM-CRF model, which can achieve more accurate predictions. Finally, the XJTU-SY bearing data is used to verify the effectiveness of the proposed algorithm. Experiments show that this proposed method can get the best performance comparing with the convolutional neural networks and the Long Short-Term Memory.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131663741","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}
Aiming at the difficulty in identifying small fault of gear, a gear diagnosis method was proposed based on integrated empirical mode decomposition (EEMD), cloud model, support vector machine, and particle swarm optimization (PSO-SVM). Firstly, the vibration signal was decomposed into several IMF components by EEMD, and the backward cloud generator calculation was performed on the IMF components to obtain the digital characteristics of the cloud model. Then, the digital features obtained and the frequency domain features and time-domain features obtained after linear reconstruction were constructed as feature vectors, which were dimensionalized by principal component analysis. Finally, the features after dimensionality reduction are input into PSO-SVM for classification training and testing. The results show that this method can effectively complete gear fault diagnosis and has a higher recognition rate.
{"title":"A Gear Fault Diagnosis Method Based on EEMD Cloud Model and PSO_SVM","authors":"Yunhui Ou, Darong Huang, Chengchong Hu, Haiyang Hao, J. Gong, Ling Zhao","doi":"10.1109/DDCLS52934.2021.9455486","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455486","url":null,"abstract":"Aiming at the difficulty in identifying small fault of gear, a gear diagnosis method was proposed based on integrated empirical mode decomposition (EEMD), cloud model, support vector machine, and particle swarm optimization (PSO-SVM). Firstly, the vibration signal was decomposed into several IMF components by EEMD, and the backward cloud generator calculation was performed on the IMF components to obtain the digital characteristics of the cloud model. Then, the digital features obtained and the frequency domain features and time-domain features obtained after linear reconstruction were constructed as feature vectors, which were dimensionalized by principal component analysis. Finally, the features after dimensionality reduction are input into PSO-SVM for classification training and testing. The results show that this method can effectively complete gear fault diagnosis and has a higher recognition rate.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122212072","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 : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455688
Jia Wang, Ying Yang
This paper studies the stability performance recovery for linear systems with input and output constraints. In particular, the model predictive controller is formulated based on the nominal model to cope with constraints. The multiplicative fault-induced performance degradation is detected by the stability margin. For the purpose to recover the stability performance, the model of the faulty plant is identified with the aid of the process data, then, the model predictive controller is reconfigured based on the identified model.
{"title":"Model Predictive Control-based Stability Performance Recovery","authors":"Jia Wang, Ying Yang","doi":"10.1109/DDCLS52934.2021.9455688","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455688","url":null,"abstract":"This paper studies the stability performance recovery for linear systems with input and output constraints. In particular, the model predictive controller is formulated based on the nominal model to cope with constraints. The multiplicative fault-induced performance degradation is detected by the stability margin. For the purpose to recover the stability performance, the model of the faulty plant is identified with the aid of the process data, then, the model predictive controller is reconfigured based on the identified model.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123803329","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 : 2021-05-14DOI: 10.1109/DDCLS52934.2021.9455469
Baiyan Liu, Dan Su, Mei Liu, Yang Shi, Shuai Li
It is necessary to make physical constraints on the joints for the redundant robot motion control in order to avoid damage. In this paper, a discrete-time neural network model with minimum kinetic energy as the performance index is proposed, which has predominant convergence performance. Then, a solution in robot motion control is studied and further transformed into a dynamic quadratic programming (QP) with equality and inequality constraints. In addition, for solving the formulated QP problem, a continuous-time neural network model is designed by introducing the Lagrange multiplier method, and a discrete-time neural network model is obtained by the Euler forward difference formula. Moreover, the simulations on robot motion control are carried out, and the simulative results further substantiate the superiority, thus extending a solution for motion control of redundant robots with double-bound constraints.
{"title":"MKE Scheme for the Control of Dynamic Constrained Redundant Robots Based on Discrete-time Neural Network","authors":"Baiyan Liu, Dan Su, Mei Liu, Yang Shi, Shuai Li","doi":"10.1109/DDCLS52934.2021.9455469","DOIUrl":"https://doi.org/10.1109/DDCLS52934.2021.9455469","url":null,"abstract":"It is necessary to make physical constraints on the joints for the redundant robot motion control in order to avoid damage. In this paper, a discrete-time neural network model with minimum kinetic energy as the performance index is proposed, which has predominant convergence performance. Then, a solution in robot motion control is studied and further transformed into a dynamic quadratic programming (QP) with equality and inequality constraints. In addition, for solving the formulated QP problem, a continuous-time neural network model is designed by introducing the Lagrange multiplier method, and a discrete-time neural network model is obtained by the Euler forward difference formula. Moreover, the simulations on robot motion control are carried out, and the simulative results further substantiate the superiority, thus extending a solution for motion control of redundant robots with double-bound constraints.","PeriodicalId":325897,"journal":{"name":"2021 IEEE 10th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"217 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-05-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124277142","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：