Pub Date : 2022-11-25DOI: 10.1109/CAC57257.2022.10055799
Wanqin Zhang, Degang Xu
LiDAR and stereo cameras are increasingly being used for intelligent perceptual tasks in autonomous vehicles and robotic platforms. However, before the sensors can be used, they usually need to be precisely calibrated both internally and externally considering the calibration affection of the sensor parameters. With the increasing popularity of deep learning (DL), some recent studies have proved the advantages of DL in feature extraction, feature matching and global regression in extrinsic calibration. To improve the accuracy and reduce calibration time, we propose a method for automatic extrinsic calibration of LiDAR and stereo camera based on deep convolutional network. It has the nonlinear mapping ability of neural network to establish the mapping relationship between the target in the LiDAR coordinate system and its image pixel coordinate system. Moreover, the proposed method does not require the resort to any extra calibrator, which reduces some manual steps and compensates some shortcomings of traditional methods. The method can be used for the extrinsic calibration of LiDAR and camera online, which is meaningful for further fusing the sensor data.
{"title":"Extrinsic Calibration of LiDAR-Camera Based on Deep Convolutional Network","authors":"Wanqin Zhang, Degang Xu","doi":"10.1109/CAC57257.2022.10055799","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055799","url":null,"abstract":"LiDAR and stereo cameras are increasingly being used for intelligent perceptual tasks in autonomous vehicles and robotic platforms. However, before the sensors can be used, they usually need to be precisely calibrated both internally and externally considering the calibration affection of the sensor parameters. With the increasing popularity of deep learning (DL), some recent studies have proved the advantages of DL in feature extraction, feature matching and global regression in extrinsic calibration. To improve the accuracy and reduce calibration time, we propose a method for automatic extrinsic calibration of LiDAR and stereo camera based on deep convolutional network. It has the nonlinear mapping ability of neural network to establish the mapping relationship between the target in the LiDAR coordinate system and its image pixel coordinate system. Moreover, the proposed method does not require the resort to any extra calibrator, which reduces some manual steps and compensates some shortcomings of traditional methods. The method can be used for the extrinsic calibration of LiDAR and camera online, which is meaningful for further fusing the sensor data.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"29 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":"115138235","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/CAC57257.2022.10055018
Song Huang, Danhong Zhang, Tuo Zheng, Guangbo Tong, Jianxin Xu, Fangzheng Jia
Due to the current rising in oil prices and energy scarcity, the role of short term load forecasting is critical in basic functioning and scheduling of power systems. The forecasting accuracy of a single model always has its limitations. Therefore, the LSTM-NBEATS model, a combined model combining LSTM and NBEATS by a MAPE (mean absolute percentage error) weighting method is proposed. This model is easy to realize and train, and does not rely on complicated feature engineering. It is applied to hourly load datasets from three European countries, Macedonia (MK), Latvia (LV), and Poland (PL). In this paper, experimental results show that in short term load forecasting the model proposed performs effective.
{"title":"Studies of short-term load forecasting model based on LSTM-NBEATS","authors":"Song Huang, Danhong Zhang, Tuo Zheng, Guangbo Tong, Jianxin Xu, Fangzheng Jia","doi":"10.1109/CAC57257.2022.10055018","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055018","url":null,"abstract":"Due to the current rising in oil prices and energy scarcity, the role of short term load forecasting is critical in basic functioning and scheduling of power systems. The forecasting accuracy of a single model always has its limitations. Therefore, the LSTM-NBEATS model, a combined model combining LSTM and NBEATS by a MAPE (mean absolute percentage error) weighting method is proposed. This model is easy to realize and train, and does not rely on complicated feature engineering. It is applied to hourly load datasets from three European countries, Macedonia (MK), Latvia (LV), and Poland (PL). In this paper, experimental results show that in short term load forecasting the model proposed performs effective.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"42 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":"115378971","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/CAC57257.2022.10055635
Ke Lin, Haiyan Gao, Zhiyong Lin, Rong Fu, Zhichao Chen, Weiqiang Tang
The sliding mode control method with double power exponential reaching law is proposed for the unstable factors such as chattering vibration and overshoot in the sliding mode control of permanent magnet synchronous motor. The method combines the advantages of the traditional reaching law, which enables the motor to approach the sliding mold surface quickly and smoothly from the initial state, and realizes the speed tracking. And the continuous function is chosen to instead of the sign function, which suppresses the chattering phenomenon of the sliding mode control. Comparing the designed new reaching law with the traditional reaching law in simulink, it can be clearly seen that the time from the system movement to the sliding surface is significantly shortened. And at the same time, in the case of uncertain disturbance, the motor can adjust the motion state faster, indicating that the new reaching law has better stability and tracking performance than the traditional reaching law.
{"title":"Sliding Mode Control Strategy For PMSM Based on Improved Reaching Law","authors":"Ke Lin, Haiyan Gao, Zhiyong Lin, Rong Fu, Zhichao Chen, Weiqiang Tang","doi":"10.1109/CAC57257.2022.10055635","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055635","url":null,"abstract":"The sliding mode control method with double power exponential reaching law is proposed for the unstable factors such as chattering vibration and overshoot in the sliding mode control of permanent magnet synchronous motor. The method combines the advantages of the traditional reaching law, which enables the motor to approach the sliding mold surface quickly and smoothly from the initial state, and realizes the speed tracking. And the continuous function is chosen to instead of the sign function, which suppresses the chattering phenomenon of the sliding mode control. Comparing the designed new reaching law with the traditional reaching law in simulink, it can be clearly seen that the time from the system movement to the sliding surface is significantly shortened. And at the same time, in the case of uncertain disturbance, the motor can adjust the motion state faster, indicating that the new reaching law has better stability and tracking performance than the traditional reaching law.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"209 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":"115464198","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/cac57257.2022.10055815
Jiaying Duan, H. Shen, T. Shi, Chao Li, Guoyuan Yang, Yongmei Chen
With the development of display technology, screenless display technology represented by augmented reality (AR) technology, virtual reality (VR) technology, intelligent projection technology and (pseudo) holographic technology, in the construction of smart stations, can be used in passenger identity verification, equipment inspection, staff training, onsite navigation and passenger service. This paper first classifies and summarizes the technical principles of several screenless display technologies, and elaborates on specific technologies in detail. Subsequently, we have combed and summarized the application of several screenless display technologies in railway intelligent passenger stations in detail. Based on the actual demand of railway passenger stations, according to the current situation and limitations of the use of screen display technology in railway station equipment operation and maintenance and passenger services, we can try to study and apply new screenless display technologies, such as AR, VR and intelligent projection to improve the intelligence level of railway stations.
{"title":"Application research and prospect of screenless display technologies in railway intelligent station","authors":"Jiaying Duan, H. Shen, T. Shi, Chao Li, Guoyuan Yang, Yongmei Chen","doi":"10.1109/cac57257.2022.10055815","DOIUrl":"https://doi.org/10.1109/cac57257.2022.10055815","url":null,"abstract":"With the development of display technology, screenless display technology represented by augmented reality (AR) technology, virtual reality (VR) technology, intelligent projection technology and (pseudo) holographic technology, in the construction of smart stations, can be used in passenger identity verification, equipment inspection, staff training, onsite navigation and passenger service. This paper first classifies and summarizes the technical principles of several screenless display technologies, and elaborates on specific technologies in detail. Subsequently, we have combed and summarized the application of several screenless display technologies in railway intelligent passenger stations in detail. Based on the actual demand of railway passenger stations, according to the current situation and limitations of the use of screen display technology in railway station equipment operation and maintenance and passenger services, we can try to study and apply new screenless display technologies, such as AR, VR and intelligent projection to improve the intelligence level of railway stations.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"42 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":"124415297","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/CAC57257.2022.10055246
Yang Xu, Xingyu Yang, Jingtao Zhang, Jianjie Yang, Jie Tang
The development of Bluetooth technology and mobile communication technology makes it possible for smartphones to replace traditional car keys. In order to achieve high-precision positioning and accurate identification of smartphones in the vehicle environment, this paper adopts a location fingerprint method based on Bluetooth Low Energy (BLE) technology. In the stage of offline fingerprint database construction, a fusion filtering algorithm based on moving average and Kalman filtering is used to obtain more stable data. In the phase of location and identification, a binary K-means clustering based on the similarity of position coordinates combined with weighted K-nearest neighbor (WKNN) algorithm is proposed to locate and identify smartphones. The experimental results show that the proposed algorithm improves the recognition accuracy of smartphones by 8.8% compared with the K-nearest neighbor (KNN) algorithm, and reduces the positioning error from 1.56m to 0.41m.
{"title":"Research on Digital Key Positioning Method Based on Bluetooth Low Energy","authors":"Yang Xu, Xingyu Yang, Jingtao Zhang, Jianjie Yang, Jie Tang","doi":"10.1109/CAC57257.2022.10055246","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055246","url":null,"abstract":"The development of Bluetooth technology and mobile communication technology makes it possible for smartphones to replace traditional car keys. In order to achieve high-precision positioning and accurate identification of smartphones in the vehicle environment, this paper adopts a location fingerprint method based on Bluetooth Low Energy (BLE) technology. In the stage of offline fingerprint database construction, a fusion filtering algorithm based on moving average and Kalman filtering is used to obtain more stable data. In the phase of location and identification, a binary K-means clustering based on the similarity of position coordinates combined with weighted K-nearest neighbor (WKNN) algorithm is proposed to locate and identify smartphones. The experimental results show that the proposed algorithm improves the recognition accuracy of smartphones by 8.8% compared with the K-nearest neighbor (KNN) algorithm, and reduces the positioning error from 1.56m to 0.41m.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"11 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":"123082525","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}
This paper discusses an adaptive neural tracking control of permanent magnet synchronous motors subject to input saturation and output constraint. The difficulty is to consider output constraints and input saturation. Firstly, the adaptive dynamic surface control design process is systematized by embedding many existing tools into the classical backstepping framework. Then, a nonlinear transformation function is proposed to transform the output constrained system into an unconstrained system. Furthermore, using radial basis function neural networks to Process the unknown terms, the Gaussian error function is utilized to describe the continuously differentiable asymmetric saturation nonlinearity. It turns out that all signals in the proposed scheme are bounded. The simulation results are provided to further show the feasibility of the proposed method.
{"title":"Neural Adaptive Dynamic Surface Control of PMSMs with Input Saturation and output constraint","authors":"Dongchao Lv, Shaobo Li, T. Zhang, Fengbin Wu, Menghan Li, Chao Zheng","doi":"10.1109/CAC57257.2022.10055960","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055960","url":null,"abstract":"This paper discusses an adaptive neural tracking control of permanent magnet synchronous motors subject to input saturation and output constraint. The difficulty is to consider output constraints and input saturation. Firstly, the adaptive dynamic surface control design process is systematized by embedding many existing tools into the classical backstepping framework. Then, a nonlinear transformation function is proposed to transform the output constrained system into an unconstrained system. Furthermore, using radial basis function neural networks to Process the unknown terms, the Gaussian error function is utilized to describe the continuously differentiable asymmetric saturation nonlinearity. It turns out that all signals in the proposed scheme are bounded. The simulation results are provided to further show the feasibility of the proposed method.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"41 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":"116867059","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/CAC57257.2022.10055045
Sicheng Shao, Zheng Lu, Bin Zhang, Xuetao Zhang
As Transformer is more widely used in the domain of Computer Vision (CV), modern techniques for Change Detection (CD) have also begun to use Transformer structures, including Bitemporal Image Transformer (BIT). Although BIT shows excellent performance due to its efficient context modeling ability, the simple backbone network and the Cross-Entropy (CE) loss it uses still have room for improvement. In this paper, we propose a Feature Pyramid Network of Change Detection (FPN-CD) and a Change Detection focal (CDF) loss to address the shortcomings of the BIT method. Meanwhile, the outcomes of ablation experiments performed on two CD datasets attest to the method's efficacy.
{"title":"Feature Enhancement and Reweighting for Transformer-Based Change Detection","authors":"Sicheng Shao, Zheng Lu, Bin Zhang, Xuetao Zhang","doi":"10.1109/CAC57257.2022.10055045","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055045","url":null,"abstract":"As Transformer is more widely used in the domain of Computer Vision (CV), modern techniques for Change Detection (CD) have also begun to use Transformer structures, including Bitemporal Image Transformer (BIT). Although BIT shows excellent performance due to its efficient context modeling ability, the simple backbone network and the Cross-Entropy (CE) loss it uses still have room for improvement. In this paper, we propose a Feature Pyramid Network of Change Detection (FPN-CD) and a Change Detection focal (CDF) loss to address the shortcomings of the BIT method. Meanwhile, the outcomes of ablation experiments performed on two CD datasets attest to the method's efficacy.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"15 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":"117215853","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/CAC57257.2022.10054770
Min Zhang, Yungang Liu, Yuan Wang, Fengzhong Li, Lin Chen
With the increasing complexity of UAV missions, path planning, as a key issue, is receiving more and more attention. Currently, most of the literatures related to this problem are concerned about off-line path planning. However, the dynamic and complex environment with uncertainty makes it more challenging for UAVs to complete their missions autonomously, safely and quickly, which calls for path planning in real time. In the paper, several representative algorithms for UAV real-time path planning are reviewed, from perspective on path searching and trajectory optimization. Therein, Artificial Potential Field (APF) method, Markov Decision Process (MDP) based method and Artificial Neural Network (ANN) algorithm are set forth, while their performance, fusion and improvement are analyzed. Finally, we propose a series of challenging real-time path planning problems for future research.
{"title":"Real-time Path Planning Algorithms for Autonomous UAV","authors":"Min Zhang, Yungang Liu, Yuan Wang, Fengzhong Li, Lin Chen","doi":"10.1109/CAC57257.2022.10054770","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10054770","url":null,"abstract":"With the increasing complexity of UAV missions, path planning, as a key issue, is receiving more and more attention. Currently, most of the literatures related to this problem are concerned about off-line path planning. However, the dynamic and complex environment with uncertainty makes it more challenging for UAVs to complete their missions autonomously, safely and quickly, which calls for path planning in real time. In the paper, several representative algorithms for UAV real-time path planning are reviewed, from perspective on path searching and trajectory optimization. Therein, Artificial Potential Field (APF) method, Markov Decision Process (MDP) based method and Artificial Neural Network (ANN) algorithm are set forth, while their performance, fusion and improvement are analyzed. Finally, we propose a series of challenging real-time path planning problems for future research.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"801 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"120935565","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/CAC57257.2022.10055314
Min Li, Ke Xu, Yu Li, Dong Li
Continuous event monitoring for large-scale wireless sensor networks (WSNs) is one of the important applications. In these applications, clusters should be efficiently and dynamically organized and of appropriate size. This paper proposes a cooperative transmission scheme with dynamic clustering based on splitting and reorganization for monitoring applications. When no event occurs, a static network model with a hierarchical clustering tree for large-scale WSNs is proposed. Once the event was triggered, a dynamic clustering scheme based on splitting and reorganization according to the characteristics of event diffusion is proposed. Moreover, selection cooperation is introduced into the data transmission of cluster head(CH) and a cooperative transmission protocol by utilizing a feedback mechanism is designed to enhance the transmission reliability of the event area. Simulation results show that the proposed scheme can effectively improve the transmission reliability for wireless sensor networks driven by diffuse events.
{"title":"A Cooperative Transmission Scheme with Dynamic Clustering Based on Splitting and Reorganization for Event-Driven WSNs","authors":"Min Li, Ke Xu, Yu Li, Dong Li","doi":"10.1109/CAC57257.2022.10055314","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10055314","url":null,"abstract":"Continuous event monitoring for large-scale wireless sensor networks (WSNs) is one of the important applications. In these applications, clusters should be efficiently and dynamically organized and of appropriate size. This paper proposes a cooperative transmission scheme with dynamic clustering based on splitting and reorganization for monitoring applications. When no event occurs, a static network model with a hierarchical clustering tree for large-scale WSNs is proposed. Once the event was triggered, a dynamic clustering scheme based on splitting and reorganization according to the characteristics of event diffusion is proposed. Moreover, selection cooperation is introduced into the data transmission of cluster head(CH) and a cooperative transmission protocol by utilizing a feedback mechanism is designed to enhance the transmission reliability of the event area. Simulation results show that the proposed scheme can effectively improve the transmission reliability for wireless sensor networks driven by diffuse events.","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":"121034415","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 effectiveness of bridge condition assessment relies on the reliability of monitoring data, which can have multiple types of anomaly patterns in complex environments, and the accuracy and timeliness of traditional pattern recognition anomaly detection schemes do not meet the needs of practice. To address the multiple types of anomalies in bridge monitoring data, nine features that fit the bridge data situation are constructed using high-frequency acceleration data, and anomaly detection is performed using a pattern recognition LSTM neural network that is sensitive to time series, and run in real time by Flux in the Ganjiang Special Bridge monitoring system. The experimental results show that this scheme achieves a high level of accuracy for each category of anomaly detection, with a 4.53% improvement in accuracy compared to the PRNN neural network scheme. The overall detection time of real-time samples in practice is about 1.10s, and the overall anomaly detection accuracy reaches 99.65%, which meets the need for timeliness and accuracy of bridge anomaly detection system in practice.
{"title":"Research and application of pattern recognition LSTM based bridge data anomaly detection","authors":"Zheng Gao, Funian Li, Xingsheng Yu, Junfeng Yan, Zhidan Chen","doi":"10.1109/CAC57257.2022.10054767","DOIUrl":"https://doi.org/10.1109/CAC57257.2022.10054767","url":null,"abstract":"The effectiveness of bridge condition assessment relies on the reliability of monitoring data, which can have multiple types of anomaly patterns in complex environments, and the accuracy and timeliness of traditional pattern recognition anomaly detection schemes do not meet the needs of practice. To address the multiple types of anomalies in bridge monitoring data, nine features that fit the bridge data situation are constructed using high-frequency acceleration data, and anomaly detection is performed using a pattern recognition LSTM neural network that is sensitive to time series, and run in real time by Flux in the Ganjiang Special Bridge monitoring system. The experimental results show that this scheme achieves a high level of accuracy for each category of anomaly detection, with a 4.53% improvement in accuracy compared to the PRNN neural network scheme. The overall detection time of real-time samples in practice is about 1.10s, and the overall anomaly detection accuracy reaches 99.65%, which meets the need for timeliness and accuracy of bridge anomaly detection system in practice.","PeriodicalId":287137,"journal":{"name":"2022 China Automation Congress (CAC)","volume":"17 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":"121340781","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：