Reinforcement learning, as an effective framework for solving continuous decision tasks in machine learning, has been widely used in manipulator decision control. However, for manipulator grasping tasks in complex environments, it is difficult for intelligence to improve performance by exploring to obtain high-quality interaction samples. In addition, the training models of reinforcement learning usually lack task generalization and need to be relearned to adapt to task changes. To address these issues, researchers have proposed transfer learning that uses external prior knowledge to help the target task to improve the reinforcement learning process. In this paper, the transfer of the manipulator grasping source task to the grasping target task based on the deep Q-network algorithm is achieved by constructing lateral connections between fully convolutional neural networks using Densenet. Experimental results in the CoppeliaSim simulation environment show that the methods successfully achieve inter-task transfer by constructing lateral connections between fully convolutional neural networks. The validated transfer reinforcement learning approach improves the effectiveness of task training while reducing the complexity of the network due to lateral connections.
{"title":"Transfer Reinforcement Learning of Robotic Grasping Training using Neural Networks with Lateral Connections","authors":"Wenxiao Wang, Xiaojuan Wang, Renqiang Li, Haosheng Jiang, Ding Liu, X. Ping","doi":"10.1109/DDCLS58216.2023.10166333","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166333","url":null,"abstract":"Reinforcement learning, as an effective framework for solving continuous decision tasks in machine learning, has been widely used in manipulator decision control. However, for manipulator grasping tasks in complex environments, it is difficult for intelligence to improve performance by exploring to obtain high-quality interaction samples. In addition, the training models of reinforcement learning usually lack task generalization and need to be relearned to adapt to task changes. To address these issues, researchers have proposed transfer learning that uses external prior knowledge to help the target task to improve the reinforcement learning process. In this paper, the transfer of the manipulator grasping source task to the grasping target task based on the deep Q-network algorithm is achieved by constructing lateral connections between fully convolutional neural networks using Densenet. Experimental results in the CoppeliaSim simulation environment show that the methods successfully achieve inter-task transfer by constructing lateral connections between fully convolutional neural networks. The validated transfer reinforcement learning approach improves the effectiveness of task training while reducing the complexity of the network due to lateral connections.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115335318","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166034
Yimin Li, Yanfang Chen, Tianru Li, Jingtao Lao, Xuefang Li
The present work develops a DDPG-based path planning algorithm that integrates the artificial potential field method into reinforcement learning to learn and generate an obstacle-free path quickly and autonomously. The vehicle kinematic model is adopted to describe the motion of autonomous vehicles, and the potential field function of obstacles, road boundaries as well as reference waypoints are considered to construct rewards of reinforcement learning, which enables the vehicle to realize the tradeoff between avoiding obstacles, preventing driving off the road and following the reference route. In contrast to the existent path planning algorithms, the proposed approach is able to learn autonomously in different driving environments, which is more suitable to autonomous vehicles. Moreover, simulations are provided to further demonstrate the effectiveness and adaptability of the proposed algorithm.
{"title":"DDPG-Based Path Planning Approach for Autonomous Driving","authors":"Yimin Li, Yanfang Chen, Tianru Li, Jingtao Lao, Xuefang Li","doi":"10.1109/DDCLS58216.2023.10166034","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166034","url":null,"abstract":"The present work develops a DDPG-based path planning algorithm that integrates the artificial potential field method into reinforcement learning to learn and generate an obstacle-free path quickly and autonomously. The vehicle kinematic model is adopted to describe the motion of autonomous vehicles, and the potential field function of obstacles, road boundaries as well as reference waypoints are considered to construct rewards of reinforcement learning, which enables the vehicle to realize the tradeoff between avoiding obstacles, preventing driving off the road and following the reference route. In contrast to the existent path planning algorithms, the proposed approach is able to learn autonomously in different driving environments, which is more suitable to autonomous vehicles. Moreover, simulations are provided to further demonstrate the effectiveness and adaptability of the proposed algorithm.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"376 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123088548","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166983
Debiao Chang, Rongmin Cao, Zhongsheng Hou, Jihui Jia, Yifan Li
The position tracking accuracy of a two-dimensional linear motor is the most important accuracy index in the servo motion process of a two-dimensional linear motor, and it is of great significance to the servo motion process of two-dimensional linear motor modeling and control. Aiming at the problem that the complex dynamic characteristics of the two-dimensional linear motor are difficult to carry out conventional mechanism modeling and other disturbances such as friction impedance during its movement a compensation scheme founded on the combination of tight format dynamic linearization model-free adaptive control and active disturbance rejection control technology is proposed, according to the data-driven control idea. The scheme provides an idea for solving the problem of friction disturbance of two-dimensional linear motors. After establishing the mathematical model of a two-dimensional linear motor, the scheme uses Matlab to simulate the algorithm. Then, owing to the influence of many adjustable parameters on the performance of the controller, and the problems of time-consuming and unsatisfactory optimization of many parameters, the controller parameters are optimized based on a genetic algorithm to improve the efficiency of parameter tuning.
{"title":"Model-free Active Disturbance Rejection Control of Two-dimensional Linear Motor Based on Multi-parameter Genetic Optimization","authors":"Debiao Chang, Rongmin Cao, Zhongsheng Hou, Jihui Jia, Yifan Li","doi":"10.1109/DDCLS58216.2023.10166983","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166983","url":null,"abstract":"The position tracking accuracy of a two-dimensional linear motor is the most important accuracy index in the servo motion process of a two-dimensional linear motor, and it is of great significance to the servo motion process of two-dimensional linear motor modeling and control. Aiming at the problem that the complex dynamic characteristics of the two-dimensional linear motor are difficult to carry out conventional mechanism modeling and other disturbances such as friction impedance during its movement a compensation scheme founded on the combination of tight format dynamic linearization model-free adaptive control and active disturbance rejection control technology is proposed, according to the data-driven control idea. The scheme provides an idea for solving the problem of friction disturbance of two-dimensional linear motors. After establishing the mathematical model of a two-dimensional linear motor, the scheme uses Matlab to simulate the algorithm. Then, owing to the influence of many adjustable parameters on the performance of the controller, and the problems of time-consuming and unsatisfactory optimization of many parameters, the controller parameters are optimized based on a genetic algorithm to improve the efficiency of parameter tuning.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125199360","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166179
Haorui Xu, Liang Cao
This paper studies the adaptive fault-tolerant quantized consensus control problem for a class of nonlinear multiagent systems with time-varying parameters and disturbances. With parameters compensation technique, a distributed adaptive consensus control scheme is developed, where the bound of the actuator fault parameters is estimated. Then a robust distributed adaptive quantized consensus tracking controller is designed to compensate the effect of unknown time-varying parameters and external disturbances. Based on Lyapunov stability theory, it is proven that the control strategy can guarantee the stability of the closed-loop systems, which is demonstrated by simulation results.
{"title":"Adaptive Quantized Consensus Control for Uncertain Nonlinear Multiagent Systems with Actuator Faults","authors":"Haorui Xu, Liang Cao","doi":"10.1109/DDCLS58216.2023.10166179","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166179","url":null,"abstract":"This paper studies the adaptive fault-tolerant quantized consensus control problem for a class of nonlinear multiagent systems with time-varying parameters and disturbances. With parameters compensation technique, a distributed adaptive consensus control scheme is developed, where the bound of the actuator fault parameters is estimated. Then a robust distributed adaptive quantized consensus tracking controller is designed to compensate the effect of unknown time-varying parameters and external disturbances. Based on Lyapunov stability theory, it is proven that the control strategy can guarantee the stability of the closed-loop systems, which is demonstrated by simulation results.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121223830","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10165863
Xin Zhang, Chunyang Wei, Cheng Zhang
In order to improve the reliability of the power generation system of ultra-supercritical units, a fault diagnosis algorithm based on the improved LeNet-5 network is proposed to address the problems of difficult feature extraction, low accuracy and reliance on manual experience of traditional fault diagnosis methods. Firstly, multi-scale convolutional kernels in parallel are used to extract more details of the fault features. By using the improved inception V2 network and residual neural network, more complete and accurate features can be extracted while avoiding the degradation of the model due to too deep layers. Then a combination of $1^{ast}1$ convolution and global average pooling is used instead of partial fully connected layers, which greatly reduces the parameters of the model and prevents model overfitting. The test shows that the fault identification rate of this method can be 98.42%.
{"title":"Improved LeNet-5 Network for Equipment Fault Diagnosis of Ultra-supercritical Units","authors":"Xin Zhang, Chunyang Wei, Cheng Zhang","doi":"10.1109/DDCLS58216.2023.10165863","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10165863","url":null,"abstract":"In order to improve the reliability of the power generation system of ultra-supercritical units, a fault diagnosis algorithm based on the improved LeNet-5 network is proposed to address the problems of difficult feature extraction, low accuracy and reliance on manual experience of traditional fault diagnosis methods. Firstly, multi-scale convolutional kernels in parallel are used to extract more details of the fault features. By using the improved inception V2 network and residual neural network, more complete and accurate features can be extracted while avoiding the degradation of the model due to too deep layers. Then a combination of $1^{ast}1$ convolution and global average pooling is used instead of partial fully connected layers, which greatly reduces the parameters of the model and prevents model overfitting. The test shows that the fault identification rate of this method can be 98.42%.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125776763","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166128
Guangbin Zhang, Junsheng Ren, Xiaowei Tan
Based on planar motion mechanism and overlapping mesh technique, the maneuverability hydrodynamic derivative of KVLCC2 ship model in viscous flow field is calculated. By numerical simulation of oblique shipping motion, pure sway motion and pure yaw motion, the calculated hydrodynamic force is compared with the experimental value under corresponding conditions. The calculated hydrodynamic derivative is in good agreement with the experimental value, and the accuracy of the calculated hydrodynamic derivative is high. On this basis, the trim is added to the ship to study the variation law of hydrodynamic derivative of ship maneuverability under the condition of trim.
{"title":"Estimation of Ship Hydrodynamic Derivatives using Numerical PMM Test with Trim Conditions","authors":"Guangbin Zhang, Junsheng Ren, Xiaowei Tan","doi":"10.1109/DDCLS58216.2023.10166128","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166128","url":null,"abstract":"Based on planar motion mechanism and overlapping mesh technique, the maneuverability hydrodynamic derivative of KVLCC2 ship model in viscous flow field is calculated. By numerical simulation of oblique shipping motion, pure sway motion and pure yaw motion, the calculated hydrodynamic force is compared with the experimental value under corresponding conditions. The calculated hydrodynamic derivative is in good agreement with the experimental value, and the accuracy of the calculated hydrodynamic derivative is high. On this basis, the trim is added to the ship to study the variation law of hydrodynamic derivative of ship maneuverability under the condition of trim.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125901024","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166757
Peiyuan Li, Panshuo Li
The batch process is a typical manufacturing mode in industry. In this article, an adaptive ILC method is proposed for the batch process with time-varying and unknown parameters. The proposed method involves merging an adaptive updating law that utilizes the steepest descent method to estimate unknown parameters with a controller that adjusts the estimated system. The proposed condition ensures that the estimated parameter error remains bounded and that the estimated state error is stabilized. The controller utilizes the estimated results to steer the estimated system to track the reference trajectory. A numerical experiment is presented to demonstrate the efficiency of the proposed method.
{"title":"Adaptive Iterative Learning Control for Industry Batch Process with Time-Varying and Unknown Parameters","authors":"Peiyuan Li, Panshuo Li","doi":"10.1109/DDCLS58216.2023.10166757","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166757","url":null,"abstract":"The batch process is a typical manufacturing mode in industry. In this article, an adaptive ILC method is proposed for the batch process with time-varying and unknown parameters. The proposed method involves merging an adaptive updating law that utilizes the steepest descent method to estimate unknown parameters with a controller that adjusts the estimated system. The proposed condition ensures that the estimated parameter error remains bounded and that the estimated state error is stabilized. The controller utilizes the estimated results to steer the estimated system to track the reference trajectory. A numerical experiment is presented to demonstrate the efficiency of the proposed method.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"41 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126028591","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166074
Kehao Zhang, Huaiping Jin, Huaikang Jin, Bin Wang, Wangyang Yu
Wind energy has become an important part of national power systems due to its wide distribution, low cost, and non-polluting characteristics. However, the intermittence, randomness, and fluctuating of wind energy make it extremely difficult to connect wind power to the grid, which in turn affects the normal dispatch of power resources. Therefore, accurate wind power forecasting is crucial for power systems. Deep neural networks (DNNs) can efficiently capture high-dimensional nonlinear spatiotemporal features and are employed. The architectures of state-of-the-art DNNs are usually hand-designed by users with extensive expertise. In this paper, a gated recurrent unit neural networks for wind power forecasting approach based on surrogate-assisted evolutionary neural architecture search (SA-ENAS) is proposed. Firstly, SA-ENAS uses gated recurrent unit neural networks (GRU) to capture high-dimensional nonlinear spatiotemporal features, while incorporating delay variables into ENAS. Secondly, the GRU architecture is jointly encoded with delay variables. Then, the architecture search and delay variable selection are achieved using a surrogate model based ENAS approach. Finally, the effectiveness and superiority of the proposed method are verified through the case study of an actual wind farm dataset.
{"title":"Gated Recurrent Unit Neural Networks for Wind Power Forecasting based on Surrogate-Assisted Evolutionary Neural Architecture Search","authors":"Kehao Zhang, Huaiping Jin, Huaikang Jin, Bin Wang, Wangyang Yu","doi":"10.1109/DDCLS58216.2023.10166074","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166074","url":null,"abstract":"Wind energy has become an important part of national power systems due to its wide distribution, low cost, and non-polluting characteristics. However, the intermittence, randomness, and fluctuating of wind energy make it extremely difficult to connect wind power to the grid, which in turn affects the normal dispatch of power resources. Therefore, accurate wind power forecasting is crucial for power systems. Deep neural networks (DNNs) can efficiently capture high-dimensional nonlinear spatiotemporal features and are employed. The architectures of state-of-the-art DNNs are usually hand-designed by users with extensive expertise. In this paper, a gated recurrent unit neural networks for wind power forecasting approach based on surrogate-assisted evolutionary neural architecture search (SA-ENAS) is proposed. Firstly, SA-ENAS uses gated recurrent unit neural networks (GRU) to capture high-dimensional nonlinear spatiotemporal features, while incorporating delay variables into ENAS. Secondly, the GRU architecture is jointly encoded with delay variables. Then, the architecture search and delay variable selection are achieved using a surrogate model based ENAS approach. Finally, the effectiveness and superiority of the proposed method are verified through the case study of an actual wind farm dataset.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126048592","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166525
Chi Xu, Zhenhua Wang
To handle the interference of attitude maneuver and measurement noise in gyroscope fault detection, a data-driven time series model based on long short-term memory (LSTM) with residual smoothing is proposed. First, a LSTM network is used to build a time series model, which achieves effective mining of attitude system data and tracking gyroscope output. And a sliding window mechanism is involved for better prediction. Then, the residuals between estimation data and real data are smoothed by exponentially weighted moving average (EWMA) to reduce the effect of measurement noise on fault detection. Finally, the fault is determined by comparing the smoothed residuals with the threshold. Simulation results show that the model proposed in this paper is effective in both fault scenarios of gyroscopes and has higher accuracy than traditional fault detection models such as BP and RBF neural networks.
{"title":"Fault Detection for Satellite Gyroscope Using LSTM Networks","authors":"Chi Xu, Zhenhua Wang","doi":"10.1109/DDCLS58216.2023.10166525","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166525","url":null,"abstract":"To handle the interference of attitude maneuver and measurement noise in gyroscope fault detection, a data-driven time series model based on long short-term memory (LSTM) with residual smoothing is proposed. First, a LSTM network is used to build a time series model, which achieves effective mining of attitude system data and tracking gyroscope output. And a sliding window mechanism is involved for better prediction. Then, the residuals between estimation data and real data are smoothed by exponentially weighted moving average (EWMA) to reduce the effect of measurement noise on fault detection. Finally, the fault is determined by comparing the smoothed residuals with the threshold. Simulation results show that the model proposed in this paper is effective in both fault scenarios of gyroscopes and has higher accuracy than traditional fault detection models such as BP and RBF neural networks.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123517294","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 : 2023-05-12DOI: 10.1109/DDCLS58216.2023.10166508
Shoupeng Gao, Yueyang Li, Dong Zhao
Recently, the field of intelligent fault diagnosis has made great breakthroughs and achievements since feature extraction has a powerful ability to learn data. However, in non-Euclidean spaces, the types of bearing fault relationships are complex and the number of relationships is inconsistent, resulting in traditional deep learning methods that cannot accurately mine the potential relationships between fault information. To solve this problem, we propose a fault diagnosis method for rolling bearings based on a novel visibility graph (VG) and a new graph convolution neural (GCN) network. Specifically, a novel weighted visibility graph (WVG) method which can convert time series data into graph data is proposed. It can superiorly reflect the complex relationship between each factor in bearing fault diagnosis. In order to achieve fault diagnosis in the way of graph classification, we propose a new method SGIN+. It combines GraphSAGE and an improved graph isomorphic network (GIN), so that it can accurately learn the graph structure in large-scale classification tasks. The effectiveness of both WVG and SGIN+ is verified by a real bearing dataset.
{"title":"Fault Diagnosis for Rolling Bearings Based on Novel Visibility Graph and GCN Scheme","authors":"Shoupeng Gao, Yueyang Li, Dong Zhao","doi":"10.1109/DDCLS58216.2023.10166508","DOIUrl":"https://doi.org/10.1109/DDCLS58216.2023.10166508","url":null,"abstract":"Recently, the field of intelligent fault diagnosis has made great breakthroughs and achievements since feature extraction has a powerful ability to learn data. However, in non-Euclidean spaces, the types of bearing fault relationships are complex and the number of relationships is inconsistent, resulting in traditional deep learning methods that cannot accurately mine the potential relationships between fault information. To solve this problem, we propose a fault diagnosis method for rolling bearings based on a novel visibility graph (VG) and a new graph convolution neural (GCN) network. Specifically, a novel weighted visibility graph (WVG) method which can convert time series data into graph data is proposed. It can superiorly reflect the complex relationship between each factor in bearing fault diagnosis. In order to achieve fault diagnosis in the way of graph classification, we propose a new method SGIN+. It combines GraphSAGE and an improved graph isomorphic network (GIN), so that it can accurately learn the graph structure in large-scale classification tasks. The effectiveness of both WVG and SGIN+ is verified by a real bearing dataset.","PeriodicalId":415532,"journal":{"name":"2023 IEEE 12th Data Driven Control and Learning Systems Conference (DDCLS)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-05-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126636679","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：