Pub Date : 2022-11-25DOI: 10.1109/CAC57257.2022.10055850
Zhixing Zhan, Chen Zhang, Wei Wei, Lin Zeng, S. Xiang
Aiming at the observation of cotton flow conditions in cotton production lines, a cotton image segmentation algorithm with improved DeeplabV3+ network is proposed, which introduces the lightweight network MobileNetV2 as the backbone feature extraction network; replaces the standard convolution in the void space pyramid pooling module with the depth separable convolution to compress the model size, and introduces the channel attention module to capture the image contextual information to effectively improve the segmentation accuracy of the model. The proposed algorithm achieves 96.86% pixel accuracy and 92.14% intersection ratio on the test set, which is 0.70% and 0.22% better than the original version, and the model parameter size is 15.29 MB, which is 92.7% smaller than the previous one, and the prediction time of a single frame is 18.67 ms, which is 65.8% smaller than the previous one. The experimental results show that the algorithm balances the characteristics of accuracy and real-time, and the overall comprehensive performance is optimal.
{"title":"Cotton Image Segmentation Network Based on Improved DeeplabV3+","authors":"Zhixing Zhan, Chen Zhang, Wei Wei, Lin Zeng, S. Xiang","doi":"10.1109/CAC57257.2022.10055850","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055850","url":null,"abstract":"Aiming at the observation of cotton flow conditions in cotton production lines, a cotton image segmentation algorithm with improved DeeplabV3+ network is proposed, which introduces the lightweight network MobileNetV2 as the backbone feature extraction network; replaces the standard convolution in the void space pyramid pooling module with the depth separable convolution to compress the model size, and introduces the channel attention module to capture the image contextual information to effectively improve the segmentation accuracy of the model. The proposed algorithm achieves 96.86% pixel accuracy and 92.14% intersection ratio on the test set, which is 0.70% and 0.22% better than the original version, and the model parameter size is 15.29 MB, which is 92.7% smaller than the previous one, and the prediction time of a single frame is 18.67 ms, which is 65.8% smaller than the previous one. The experimental results show that the algorithm balances the characteristics of accuracy and real-time, and the overall comprehensive performance is optimal.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128206905","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 : 2022-11-25DOI: 10.1109/CAC57257.2022.10055976
Qizhe Xu, Dongxi Lu, Hanyang Zhuang, Chunxiang Wang, Ming Yang
Automated valet parking (AVP) contributes to the reduction in the shortage of the parking resources, which has a significant impact on the efficiency of city transportation. However, in the current situation most AVP systems are implemented through vehicle-based intelligence approaches, leading to not only the problem of high cost in onboard sensor installation but also less global parking efficiency. Therefore, recent research works focusing on infrastructure-based AVP system have become a popular direction. The information of all vehicles is obtained so that the parking of each vehicle can be considered from a global perspective. This paper aims at proposing an intelligent Parking Dispatch System (PDS) to systematically tackle these issues. The PDS is developed to allocate parking space of each vehicle, generate vehicle trajectory to the parking space, compute control commands and also take care of potential collisions and conflicts. The target of this system is to minimize the average waiting time. A simulation platform has been developed to validate the system on the basis of a existent parking lot. The experiment results have shown that the PDS can achieve a high efficiency in coordinating the vehicles in the parking lot.
{"title":"Parking Dispatch System for Infrastructure-based Automated Valet Parking","authors":"Qizhe Xu, Dongxi Lu, Hanyang Zhuang, Chunxiang Wang, Ming Yang","doi":"10.1109/CAC57257.2022.10055976","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055976","url":null,"abstract":"Automated valet parking (AVP) contributes to the reduction in the shortage of the parking resources, which has a significant impact on the efficiency of city transportation. However, in the current situation most AVP systems are implemented through vehicle-based intelligence approaches, leading to not only the problem of high cost in onboard sensor installation but also less global parking efficiency. Therefore, recent research works focusing on infrastructure-based AVP system have become a popular direction. The information of all vehicles is obtained so that the parking of each vehicle can be considered from a global perspective. This paper aims at proposing an intelligent Parking Dispatch System (PDS) to systematically tackle these issues. The PDS is developed to allocate parking space of each vehicle, generate vehicle trajectory to the parking space, compute control commands and also take care of potential collisions and conflicts. The target of this system is to minimize the average waiting time. A simulation platform has been developed to validate the system on the basis of a existent parking lot. The experiment results have shown that the PDS can achieve a high efficiency in coordinating the vehicles in the parking lot.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128217789","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}
Tactile perception is indispensable for intelligent robots to interact intelligently like humans. Therefore, the effective use of deep learning methods to acquire tactile features has become an important focus of tactile perception research. Satisfactory time-driven characteristics and the ability to process spatiotemporal information efficiently of spiking neural networks are advantageous for event-based data. We apply a temporal spike sequence learning backpropagation method that can handle continuous spikes to improve the spike neural network for tactile object recognition based on event-driven data. We prove the effectiveness of the temporal spike sequence error backpropagation method in practical applications to address the problem of losing temporal information of data using approximate derivatives. In practical application, we have proved the validity of the back propagation method of temporal spike sequence error in solving the problem of losing temporal information of data using approximate derivatives
{"title":"Event-driven Robotic Tactile Data Learning Using Temporal Spike Sequence Backpropagation Method","authors":"Qing Hou, Tingqing Liu, Jing Yang, Xiaoyang Ji, Qinglang Li, Jian Li, Baofan Yin","doi":"10.1109/CAC57257.2022.10055181","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055181","url":null,"abstract":"Tactile perception is indispensable for intelligent robots to interact intelligently like humans. Therefore, the effective use of deep learning methods to acquire tactile features has become an important focus of tactile perception research. Satisfactory time-driven characteristics and the ability to process spatiotemporal information efficiently of spiking neural networks are advantageous for event-based data. We apply a temporal spike sequence learning backpropagation method that can handle continuous spikes to improve the spike neural network for tactile object recognition based on event-driven data. We prove the effectiveness of the temporal spike sequence error backpropagation method in practical applications to address the problem of losing temporal information of data using approximate derivatives. In practical application, we have proved the validity of the back propagation method of temporal spike sequence error in solving the problem of losing temporal information of data using approximate derivatives","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128773715","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 : 2022-11-25DOI: 10.1109/CAC57257.2022.10056022
Zhaoyi Li, Shenghua Dai, Ziyuan Zheng
For the fatigue information conveyed by the upper part posture of the body, 13 feature points of the driver’s upper body in different states in this paper, such as sober fatigue, are collected on the high-speed railway simulator with a monocular camera, and then the sample data are obtained through correlation processing. Each sample in the feature set is continuous data extracted from continuous frames, including angle feature and relative position proportion feature. The training set is used to train the Double-Stream Long Short-Term Memory (LSTM) neural network, and the corresponding and Long Short-Term Memory neural network classifier is obtained. The trained Double-Stream Long Short-Term memory neural network classifier is used to classify the soberness, mild fatigue and severe fatigue of locomotive drivers. The model can achieve a good effect that the average classification accuracy of this model is close to 92.67%, and the F1 score is close to 92.71%, which verify the effectiveness and robustness of the method.
{"title":"Fatigue State Detection of Locomotive Driver Based on Human Posture Features and Double-Stream Long Short-Term Memory Neural Network","authors":"Zhaoyi Li, Shenghua Dai, Ziyuan Zheng","doi":"10.1109/CAC57257.2022.10056022","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10056022","url":null,"abstract":"For the fatigue information conveyed by the upper part posture of the body, 13 feature points of the driver’s upper body in different states in this paper, such as sober fatigue, are collected on the high-speed railway simulator with a monocular camera, and then the sample data are obtained through correlation processing. Each sample in the feature set is continuous data extracted from continuous frames, including angle feature and relative position proportion feature. The training set is used to train the Double-Stream Long Short-Term Memory (LSTM) neural network, and the corresponding and Long Short-Term Memory neural network classifier is obtained. The trained Double-Stream Long Short-Term memory neural network classifier is used to classify the soberness, mild fatigue and severe fatigue of locomotive drivers. The model can achieve a good effect that the average classification accuracy of this model is close to 92.67%, and the F1 score is close to 92.71%, which verify the effectiveness and robustness of the method.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128648166","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 : 2022-11-25DOI: 10.1109/CAC57257.2022.10055403
Xiaoyi Qiao, Jiang Wu
As big data technologies become more prevalent in the energy sector, the importance of data is increasing. Data augmentation techniques can enhance the size and quality of data sets. In the scenario of an integrated energy system, the complex coupling relationship of various forms of energy poses a challenge for load data augmentation, for which a data augmentation method for electricity and thermal coupled load is proposed in this paper. First, an asymmetric Variational Autoencoder (VAE) with KL cost annealing is trained. The encoder part is used as a representation learner to extract the electricity and thermal features, based on which K-means++ is used to cluster the raw data. Then the decoder part generates new samples proportionally according to the clustering results. The experimental results show that the load data generated by this method can retain the overall distribution characteristics and the coupling relationship between electricity and thermal.
{"title":"A Clustering-Generative Model Based Method for Load Data Augmentation","authors":"Xiaoyi Qiao, Jiang Wu","doi":"10.1109/CAC57257.2022.10055403","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055403","url":null,"abstract":"As big data technologies become more prevalent in the energy sector, the importance of data is increasing. Data augmentation techniques can enhance the size and quality of data sets. In the scenario of an integrated energy system, the complex coupling relationship of various forms of energy poses a challenge for load data augmentation, for which a data augmentation method for electricity and thermal coupled load is proposed in this paper. First, an asymmetric Variational Autoencoder (VAE) with KL cost annealing is trained. The encoder part is used as a representation learner to extract the electricity and thermal features, based on which K-means++ is used to cluster the raw data. Then the decoder part generates new samples proportionally according to the clustering results. The experimental results show that the load data generated by this method can retain the overall distribution characteristics and the coupling relationship between electricity and thermal.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129010078","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}
Structural monitoring systems are increasingly used in bridge engineering because real environmental and structural response data can be obtained directly. In order to accurately assess bridge conditions and provide basic data for new bridge design, it is important to ensure the quality of the data, and when data are missing, various methods are needed to reconstruct the missing data. In this paper, we propose an EMD-BiLSTM model to reconstruct the missing deflection data by predicting the original data and the decomposed subsequences. The core of this method is to make the data subsequence more correlated by using EMD decomposition, and to obtain the before-and-after correlation of the data subsequence by BiLSTM. The EMD-BiLSTM model can effectively reconstruct the missing bridge deflection data with a root-mean-square error of 0.07759. The subseries of the original data decomposed by EMD improves the prediction accuracy of the BiLSTM model, and the BiLSTM also outperforms other machine learning algorithms to obtain more features of the data.
{"title":"Research and application of EMD-BiLSTM in bridge data reconstruction","authors":"Mingzhi Xue, Funian Li, Xingsheng Yu, Junfeng Yan, Zhidan Chen","doi":"10.1109/CAC57257.2022.10055127","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055127","url":null,"abstract":"Structural monitoring systems are increasingly used in bridge engineering because real environmental and structural response data can be obtained directly. In order to accurately assess bridge conditions and provide basic data for new bridge design, it is important to ensure the quality of the data, and when data are missing, various methods are needed to reconstruct the missing data. In this paper, we propose an EMD-BiLSTM model to reconstruct the missing deflection data by predicting the original data and the decomposed subsequences. The core of this method is to make the data subsequence more correlated by using EMD decomposition, and to obtain the before-and-after correlation of the data subsequence by BiLSTM. The EMD-BiLSTM model can effectively reconstruct the missing bridge deflection data with a root-mean-square error of 0.07759. The subseries of the original data decomposed by EMD improves the prediction accuracy of the BiLSTM model, and the BiLSTM also outperforms other machine learning algorithms to obtain more features of the data.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129025444","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 : 2022-11-25DOI: 10.1109/CAC57257.2022.10055730
Fengwei Yang, Kexin Zhang
A generalized predictive control technique for hybrid magnetic bearings is put forth in this study. Firstly, a nonlinear model is established for the radial freedom of hybrid magnetic bearing, and is then linearized through small perturbation method at the equilibrium point, it is used to develop a generalized predictive controller. Then, the technique’s results proposed on the radial freedom floating control is analyzed together with the interference suppression ability, meanwhile the effect of main control parameters and the performance of the system are analyzed and compared, which contrast the conventional PID control approach. The simulation’s findings indicate that the generalized predictive control system presented has a better dynamic performance and robustness.
{"title":"Research on Generalized Predictive Control of Hybrid Magnetic Bearing System","authors":"Fengwei Yang, Kexin Zhang","doi":"10.1109/CAC57257.2022.10055730","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055730","url":null,"abstract":"A generalized predictive control technique for hybrid magnetic bearings is put forth in this study. Firstly, a nonlinear model is established for the radial freedom of hybrid magnetic bearing, and is then linearized through small perturbation method at the equilibrium point, it is used to develop a generalized predictive controller. Then, the technique’s results proposed on the radial freedom floating control is analyzed together with the interference suppression ability, meanwhile the effect of main control parameters and the performance of the system are analyzed and compared, which contrast the conventional PID control approach. The simulation’s findings indicate that the generalized predictive control system presented has a better dynamic performance and robustness.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"167 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129277798","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 : 2022-11-25DOI: 10.1109/CAC57257.2022.10054862
Hongbin Pan, Xinghui Xiao, Ziwei Huang, Siqi Peng
To address the issue that the traditional Quadric Error Metrics (QEM) simplification algorithm cannot effectively maintain the two crucial visual features of model details and edges, this paper improved the algorithm and proposed a simplification algorithm based on the information constraint of model local area features. The algorithm considered the changes in the average area of the neighborhood grid, the bending degree of the region, and the quality factor of the grid before and after grid simplification, and the amount of information from these changes is combined with the quadratic error measure to form a composite simplification error value. A simpler detection scheme is also given based on the characteristics of the model boundaries and sharp feature areas. The detection results are used as one of the conditions for simplification to avoid oversimplification of the model detail feature areas and protection of the model edges. The experimental findings demonstrate that, compared to the QEM simplification algorithm, this algorithm successfully suppressed the rise in simplification error while retaining model detail characteristics, improving the quality of the simplified model mesh.
{"title":"Improved QEM simplification algorithm based on local area feature information constraint","authors":"Hongbin Pan, Xinghui Xiao, Ziwei Huang, Siqi Peng","doi":"10.1109/CAC57257.2022.10054862","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10054862","url":null,"abstract":"To address the issue that the traditional Quadric Error Metrics (QEM) simplification algorithm cannot effectively maintain the two crucial visual features of model details and edges, this paper improved the algorithm and proposed a simplification algorithm based on the information constraint of model local area features. The algorithm considered the changes in the average area of the neighborhood grid, the bending degree of the region, and the quality factor of the grid before and after grid simplification, and the amount of information from these changes is combined with the quadratic error measure to form a composite simplification error value. A simpler detection scheme is also given based on the characteristics of the model boundaries and sharp feature areas. The detection results are used as one of the conditions for simplification to avoid oversimplification of the model detail feature areas and protection of the model edges. The experimental findings demonstrate that, compared to the QEM simplification algorithm, this algorithm successfully suppressed the rise in simplification error while retaining model detail characteristics, improving the quality of the simplified model mesh.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124715122","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 : 2022-11-25DOI: 10.1109/CAC57257.2022.10055163
Linguo Chai, Xiangyang Liu, W. Shangguan, Xu Li, B. Cai, Yue Cao
In order to meet the data requirements of the virtual simulation test of autonomous driving, we use the camera’s single-sample video data, interpolate frames to generate multiple camera simulation data. And then realize the simulation of extended front-end perception function at the data level. This paper proposes a video sampling frame simulation reconstruction mechanism based on sampling vehicle pose information in real scenes. Calculate the target position of the simulated vehicle according to the simulation requirements, and establish a simulation node on the sampling path. Taking the position difference between the simulated node and the real node as the offset, the DAIN algorithm is used to insert the image data into the target node. It is possible to realize simulation data generation of autonomous driving camera with variable vehicle speed/sampling frequency. Combined with the camera’s internal and external parameters, the coordinate system transformation of the annotation results is carried out to realize the inheritance of the annotation results of the simulation data. This paper combines the nuScenes open source database to test the authenticity of the synchronous simulation data results. The results show that the mean SSIM of the synchronous simulation image data and the real data is 0.71684, indicating that the simulation data has high authenticity. Based on yolov4, the perceptual function of the simulated data is verified. The average value of the SSIM of the simulated image data and the real data recognition frame is 0.984504795, and the perceptual recognition results are similar. The synchronous mapping result of the marked 3D-BOX box to the simulation data space is correct. Camera simulation data can well meet the needs of autonomous driving development and testing in perception and recognition. It can provide data support for autonomous driving simulation test.
{"title":"Interpolation and simulation of autonomous driving camera data for vehicle position synchronization","authors":"Linguo Chai, Xiangyang Liu, W. Shangguan, Xu Li, B. Cai, Yue Cao","doi":"10.1109/CAC57257.2022.10055163","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055163","url":null,"abstract":"In order to meet the data requirements of the virtual simulation test of autonomous driving, we use the camera’s single-sample video data, interpolate frames to generate multiple camera simulation data. And then realize the simulation of extended front-end perception function at the data level. This paper proposes a video sampling frame simulation reconstruction mechanism based on sampling vehicle pose information in real scenes. Calculate the target position of the simulated vehicle according to the simulation requirements, and establish a simulation node on the sampling path. Taking the position difference between the simulated node and the real node as the offset, the DAIN algorithm is used to insert the image data into the target node. It is possible to realize simulation data generation of autonomous driving camera with variable vehicle speed/sampling frequency. Combined with the camera’s internal and external parameters, the coordinate system transformation of the annotation results is carried out to realize the inheritance of the annotation results of the simulation data. This paper combines the nuScenes open source database to test the authenticity of the synchronous simulation data results. The results show that the mean SSIM of the synchronous simulation image data and the real data is 0.71684, indicating that the simulation data has high authenticity. Based on yolov4, the perceptual function of the simulated data is verified. The average value of the SSIM of the simulated image data and the real data recognition frame is 0.984504795, and the perceptual recognition results are similar. The synchronous mapping result of the marked 3D-BOX box to the simulation data space is correct. Camera simulation data can well meet the needs of autonomous driving development and testing in perception and recognition. It can provide data support for autonomous driving simulation test.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124964759","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 : 2022-11-25DOI: 10.1109/CAC57257.2022.10055909
Zhichao Chen, Haiyan Gao, Ke Lin, Rong Fu, Zhiyong Lin, Weiqiang Tang
In order to strengthen the tracking performance and robustness of permanent magnet synchronous motor (PMSM) system, a deadbeat predictive current control (DPCC) based on extreme learning machine (ELM) is come up. Since PMSM is susceptible to uncertainties such as external disturbances and parameter changes, the uncertainty factors are introduced in the mathematical model. The uncertainty of the system is approximated by the ELM, then the speed tracking of the permanent magnet synchronous motor is realized, and the stability is certificated by establishing the Lyapunov function. In addition, DPCC method of PMSM is proposed, which is equivalent to high-gain proportional control and improves the performance of the PMSM. Finally, the simulation experiments are carried out in nominal case and parameter mismatch case respectively, through the comparative study of system simulation, the results indicate that contrast with the traditional control method, The ELM-DPCC proposed in this paper has better speed tracking performance and robustness.
{"title":"PMSM Deadbeat Predictive Current Control Based on Extreme Learning Machine","authors":"Zhichao Chen, Haiyan Gao, Ke Lin, Rong Fu, Zhiyong Lin, Weiqiang Tang","doi":"10.1109/CAC57257.2022.10055909","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055909","url":null,"abstract":"In order to strengthen the tracking performance and robustness of permanent magnet synchronous motor (PMSM) system, a deadbeat predictive current control (DPCC) based on extreme learning machine (ELM) is come up. Since PMSM is susceptible to uncertainties such as external disturbances and parameter changes, the uncertainty factors are introduced in the mathematical model. The uncertainty of the system is approximated by the ELM, then the speed tracking of the permanent magnet synchronous motor is realized, and the stability is certificated by establishing the Lyapunov function. In addition, DPCC method of PMSM is proposed, which is equivalent to high-gain proportional control and improves the performance of the PMSM. Finally, the simulation experiments are carried out in nominal case and parameter mismatch case respectively, through the comparative study of system simulation, the results indicate that contrast with the traditional control method, The ELM-DPCC proposed in this paper has better speed tracking performance and robustness.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129457560","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：