Pub Date : 2017-05-01DOI: 10.1109/DDCLS.2017.8068171
Xiangfei Wu, Xin Xu, Xiaohui Li, Kai Li, Bohan Jiang
This paper presents a kernel-based extreme learning machine (KELM) modeling method for speed decision making of autonomous land vehicles (ALVs) on rural roads. The model is obtained offline via the KELM algorithm using a small number of typical samples collected by an ALV platform on rural roads from experienced drivers. Compared with other typical machine learning algorithms such as support vector regression and extreme learning machine, the KELM method has the advantages of fast training speed and higher modeling precision. Real-vehicle experiments have been carried out to test the model on an ALV platform on rural roads online. The experimental results demonstrate the effectiveness of the proposed speed decision-making model.
{"title":"A kernel-based extreme learning modeling method for speed decision making of autonomous land vehicles","authors":"Xiangfei Wu, Xin Xu, Xiaohui Li, Kai Li, Bohan Jiang","doi":"10.1109/DDCLS.2017.8068171","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068171","url":null,"abstract":"This paper presents a kernel-based extreme learning machine (KELM) modeling method for speed decision making of autonomous land vehicles (ALVs) on rural roads. The model is obtained offline via the KELM algorithm using a small number of typical samples collected by an ALV platform on rural roads from experienced drivers. Compared with other typical machine learning algorithms such as support vector regression and extreme learning machine, the KELM method has the advantages of fast training speed and higher modeling precision. Real-vehicle experiments have been carried out to test the model on an ALV platform on rural roads online. The experimental results demonstrate the effectiveness of the proposed speed decision-making model.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"26 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":"125984539","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.2017.8068075
Jialu Zhang, Yong Fang, Yuzho Wu
In this paper, we consider iterative learning control(ILC) for discrete-time multi-agent system formation with one-step random time-delay. Random delays during transmission seriously affect the convergence performance of multi-agent formation. Based on one-step random time-delay model, the transition matrix of system is derived, which contains the impact factors of random delays. A learning control scheme is proposed and the convergence of system tracking errors is guaranteed. Simulation results show that the convergence rate is reduced when the probabilities of time-delay are getting higher.
{"title":"An ILC method of formation control for multi-agent system with one-step random time-delay","authors":"Jialu Zhang, Yong Fang, Yuzho Wu","doi":"10.1109/DDCLS.2017.8068075","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068075","url":null,"abstract":"In this paper, we consider iterative learning control(ILC) for discrete-time multi-agent system formation with one-step random time-delay. Random delays during transmission seriously affect the convergence performance of multi-agent formation. Based on one-step random time-delay model, the transition matrix of system is derived, which contains the impact factors of random delays. A learning control scheme is proposed and the convergence of system tracking errors is guaranteed. Simulation results show that the convergence rate is reduced when the probabilities of time-delay are getting higher.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"11 3 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":"134334688","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.2017.8067720
Dazi Li, Wei Wang
Due to the nonaffine nature, tracking control of unknown nonlinear pure feedback system is difficult. Traditional control method based on backstepping has the problem of differential explosion. Reinforcement learning control strategy can avoid this problem. However, the tracking error is relatively large because of lack of system structure information. To overcome this problem, an improved reinforcement learning algorithm with a novel actor network weight correction factor is proposed. This factor can adaptively adjust the weight update rate according to the change of the reference trajectory so that the control policy will be adjusted more timely. Simulation results demonstrate that performance of the controller is improved significantly.
{"title":"Improving reinforcement learning output feedback control for unknown nonlinear pure feedback system","authors":"Dazi Li, Wei Wang","doi":"10.1109/DDCLS.2017.8067720","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8067720","url":null,"abstract":"Due to the nonaffine nature, tracking control of unknown nonlinear pure feedback system is difficult. Traditional control method based on backstepping has the problem of differential explosion. Reinforcement learning control strategy can avoid this problem. However, the tracking error is relatively large because of lack of system structure information. To overcome this problem, an improved reinforcement learning algorithm with a novel actor network weight correction factor is proposed. This factor can adaptively adjust the weight update rate according to the change of the reference trajectory so that the control policy will be adjusted more timely. Simulation results demonstrate that performance of the controller is improved significantly.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"25 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":"128850325","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.2017.8068102
Lin Zhao, Jinpeng Yu
This paper studies the adaptive bipartite consensus tracking problems for second-order coopetition multi-agent systems with input saturation. A fuzzy-based command filtered backstepping scheme is developed, which can guarantee the bipartite position tracking errors converging to the desired neighborhood and all the closed-loop signals are bounded although the nonlinear dynamics are unknown and the input saturation exists. An example is included to verify the proposed method.
{"title":"Adaptive bipartite consensus tracking control for coopetition multi-agent systems with input saturation","authors":"Lin Zhao, Jinpeng Yu","doi":"10.1109/DDCLS.2017.8068102","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068102","url":null,"abstract":"This paper studies the adaptive bipartite consensus tracking problems for second-order coopetition multi-agent systems with input saturation. A fuzzy-based command filtered backstepping scheme is developed, which can guarantee the bipartite position tracking errors converging to the desired neighborhood and all the closed-loop signals are bounded although the nonlinear dynamics are unknown and the input saturation exists. An example is included to verify the proposed method.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"3 6","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2017-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131436240","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.2017.8068069
Yu Pang, L. Jia, Zhan Liu, Q. Gao
Many sets of wind turbines of the wind farm in Shan Xi province run above the rated wind speed, especially in the condition of wind speed 17m/s or above, wind turbine nacelle occurs vibration in the vertical direction of transmission chain which is characterized emergency, intermittent, accidental, and distinctive. Moreover, vibration cycle is not obvious and vibration strength is large. Severe vibration does harm to wind turbine that then will be able to lead wind turbine halt. According to this phenomenon, a method of emergency fault diagnosis for wind turbine nacelle based on empirical mode decomposition (EMD) is presented in this paper to discriminate a variety of factors carefully that have led to excessive vibration. In particular, the results are shown in this paper that strong tower shadow effect may cause excessive vibration of wind turbine nacelle, and then gives rise to shut down. In the meantime, curve theory analysis of the blade's aerodynamic characteristics is deduced in this paper. It demonstrates that the proposed method EMD works well in the face of fault diagnosis for wind turbine nacelle with a better overall performance.
{"title":"Emergency fault diagnosis for wind turbine nacelle","authors":"Yu Pang, L. Jia, Zhan Liu, Q. Gao","doi":"10.1109/DDCLS.2017.8068069","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068069","url":null,"abstract":"Many sets of wind turbines of the wind farm in Shan Xi province run above the rated wind speed, especially in the condition of wind speed 17m/s or above, wind turbine nacelle occurs vibration in the vertical direction of transmission chain which is characterized emergency, intermittent, accidental, and distinctive. Moreover, vibration cycle is not obvious and vibration strength is large. Severe vibration does harm to wind turbine that then will be able to lead wind turbine halt. According to this phenomenon, a method of emergency fault diagnosis for wind turbine nacelle based on empirical mode decomposition (EMD) is presented in this paper to discriminate a variety of factors carefully that have led to excessive vibration. In particular, the results are shown in this paper that strong tower shadow effect may cause excessive vibration of wind turbine nacelle, and then gives rise to shut down. In the meantime, curve theory analysis of the blade's aerodynamic characteristics is deduced in this paper. It demonstrates that the proposed method EMD works well in the face of fault diagnosis for wind turbine nacelle with a better overall performance.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"37 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":"130279978","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 adaptive elastic net has been widely studied in the microarray classification due to the elegant performances in gene selection. However, the classification accuracy will be affected if the noise is included. As such, this paper proposes a weighted adaptive elastic net for the binary microarray classification with noise by using the distances from the sample points to both class centers. Furthermore, the performance of adaptive gene selection is proved and the solution path algorithm is developed. Finally, the results on two cancer data added 4 additional samples illustrate that the weighted adaptive elastic net can achieve considerable classification accuracy and select the genes related with diseases.
{"title":"Microarray classification with noise via weighted adaptive elastic net","authors":"Juntao Li, Jingxuan Wang, Yuhan Zheng, Huimin Xiao","doi":"10.1109/DDCLS.2017.8068109","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068109","url":null,"abstract":"The adaptive elastic net has been widely studied in the microarray classification due to the elegant performances in gene selection. However, the classification accuracy will be affected if the noise is included. As such, this paper proposes a weighted adaptive elastic net for the binary microarray classification with noise by using the distances from the sample points to both class centers. Furthermore, the performance of adaptive gene selection is proved and the solution path algorithm is developed. Finally, the results on two cancer data added 4 additional samples illustrate that the weighted adaptive elastic net can achieve considerable classification accuracy and select the genes related with diseases.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"36 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":"133943097","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.2017.8068114
Yuan Gao, H. Hu, L. Yu, H. Yuan, X. Dai
Considering the time-varying scaling function matrix and system disturbances, a new sliding mode control strategy is proposed to realize modified function projective synchronization (MFPS) of two different fractional-order hyperchaotic systems, meanwhile improve the control robustness of synchronization system. From the MFPS error equations, combining a proper fractional-order exponential reaching raw, an active controller for MFPS is derived out via sliding mode control technology. By mean of the stability theorem, the asymptotic stability of synchronization error system is proved. Simulation results of the MFPS between fractional-order hyperchaoticLorenz system and Chen system demonstrate the validity of the presented method.
{"title":"Modified function projective synchronization of fractional-order hyperchaotic systems based on active sliding mode control","authors":"Yuan Gao, H. Hu, L. Yu, H. Yuan, X. Dai","doi":"10.1109/DDCLS.2017.8068114","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068114","url":null,"abstract":"Considering the time-varying scaling function matrix and system disturbances, a new sliding mode control strategy is proposed to realize modified function projective synchronization (MFPS) of two different fractional-order hyperchaotic systems, meanwhile improve the control robustness of synchronization system. From the MFPS error equations, combining a proper fractional-order exponential reaching raw, an active controller for MFPS is derived out via sliding mode control technology. By mean of the stability theorem, the asymptotic stability of synchronization error system is proved. Simulation results of the MFPS between fractional-order hyperchaoticLorenz system and Chen system demonstrate the validity of the presented method.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"159 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":"132842028","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.2017.8068117
Hongqi Zhang, Linqing Wang, Jun Zhao, Wei Wang
Blast furnace gas (BFG) system of steel enterprise generally accompanies with multi-dimension and nonlinear features. It's a hard assignment for energy scheduling operators to make real-time scheduling decision when monitoring such system. In this study, a novel dimensionality reduction method named Space Direction Neighborhood Preserving Embedding (SDNPE) is proposed for the BFG system monitoring and scheduling units determination. To maintain the system dynamic characteristic in the low dimension space, such method constructs a neighborhood graph that searches for nearest neighbors with respect to both the neighbors in spatial scales and fluctuation tendency of the gas flow data. Then, for the BFG system monitoring and scheduling units determination, Hotelling's T2 chart and score chart are constructed upon the SDNPE model. Experiments with real-time data of an iron enterprise in China demonstrated the effectiveness of the proposed method.
{"title":"Space direction neighborhood preserving embedding-based monitoring and scheduling guidance for blast furnace gas system","authors":"Hongqi Zhang, Linqing Wang, Jun Zhao, Wei Wang","doi":"10.1109/DDCLS.2017.8068117","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068117","url":null,"abstract":"Blast furnace gas (BFG) system of steel enterprise generally accompanies with multi-dimension and nonlinear features. It's a hard assignment for energy scheduling operators to make real-time scheduling decision when monitoring such system. In this study, a novel dimensionality reduction method named Space Direction Neighborhood Preserving Embedding (SDNPE) is proposed for the BFG system monitoring and scheduling units determination. To maintain the system dynamic characteristic in the low dimension space, such method constructs a neighborhood graph that searches for nearest neighbors with respect to both the neighbors in spatial scales and fluctuation tendency of the gas flow data. Then, for the BFG system monitoring and scheduling units determination, Hotelling's T2 chart and score chart are constructed upon the SDNPE model. Experiments with real-time data of an iron enterprise in China demonstrated the effectiveness of the proposed method.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"40 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":"129440087","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.2017.8068088
Lei Liu, Zejin Feng, Cunwu Han
A class of linear singularly perturbed system and the optimal problem of the upper bound of the perturbed parameter based on the genetic algorithm are considered. Firstly, the problem of the asymptotically stability is studied in the term of Lyapunov stability theory based on the Linear Matrix Inequality (LMI). Then, the standard problem of the upper perturbed parameter to be optimized is presented. Thirdly, the optimization algorithm for the upper bound of the perturbed parameter in the linear singularly perturbed system is given based on the genetic algorithm. Lastly, two numerical examples are provided to demonstrate the effectiveness and feasibility of the proposed methods.
{"title":"Optmization for the upper bound of the perturbed parameter in singularly perturbed system based on genetic algorithm","authors":"Lei Liu, Zejin Feng, Cunwu Han","doi":"10.1109/DDCLS.2017.8068088","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068088","url":null,"abstract":"A class of linear singularly perturbed system and the optimal problem of the upper bound of the perturbed parameter based on the genetic algorithm are considered. Firstly, the problem of the asymptotically stability is studied in the term of Lyapunov stability theory based on the Linear Matrix Inequality (LMI). Then, the standard problem of the upper perturbed parameter to be optimized is presented. Thirdly, the optimization algorithm for the upper bound of the perturbed parameter in the linear singularly perturbed system is given based on the genetic algorithm. Lastly, two numerical examples are provided to demonstrate the effectiveness and feasibility of the proposed methods.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"220 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":"122643158","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.2017.8068082
Yinlei Wen, Huaguang Zhang, Jinhai Liu, Fangming Li
In the practice applications of defect detecting, large amounts of data need to be analyzed. In this paper, a new analysis method is developed based on adaboost algorithm. By using neural networks with a fixed structure, a series of models are built which may be not accurate. Error rates of the models are computed to gain and adjust the weights of every model. A higher accurate model is built by the models and weights. Compared with traditional neural network method, this adaboost based method does not need to adjust the node numbers of neural networks. In addition, it remains accuracy and reduces complexity. Finally, an example is given to demonstrate the effectiveness and advantages of the methods.
{"title":"A novel adaboost based algorithm for processing defect big data","authors":"Yinlei Wen, Huaguang Zhang, Jinhai Liu, Fangming Li","doi":"10.1109/DDCLS.2017.8068082","DOIUrl":"https://doi.org/10.1109/DDCLS.2017.8068082","url":null,"abstract":"In the practice applications of defect detecting, large amounts of data need to be analyzed. In this paper, a new analysis method is developed based on adaboost algorithm. By using neural networks with a fixed structure, a series of models are built which may be not accurate. Error rates of the models are computed to gain and adjust the weights of every model. A higher accurate model is built by the models and weights. Compared with traditional neural network method, this adaboost based method does not need to adjust the node numbers of neural networks. In addition, it remains accuracy and reduces complexity. Finally, an example is given to demonstrate the effectiveness and advantages of the methods.","PeriodicalId":419114,"journal":{"name":"2017 6th Data Driven Control and Learning Systems (DDCLS)","volume":"2006 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":"128200992","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：