Data augmentation is of great importance to alleviate the insufficiency of training samples, and further improve wildlife detection accuracy. However, current data augmentation methods tend to augment all kinds of samples equally, ignoring the problem of uneven distribution of the number and size of all kinds of samples in wildlife detection datasets, resulting in poor generalization of the model. To address this problem, this paper proposes a joint distribution and class-based data augmentation method for wildlife detection. In this method, diverse rather than universal data augmentation methods are introduced for different classes with a small proportion. This makes the distributions of different classes more balanced. Therefore, each class even with a small number of samples gets good training as well. To evaluate the effectiveness of the proposed method, extensive comparative experiments are conducted. Experimental results show the superiority of our proposed method. Specifically, the detection accuracy of Faster RCNN with Swin Transformer as the backbone network is improved by 0.8% to 96.2% after data augmentation with our method.
{"title":"Joint Distribution and Class-based Data Augmentation for Wildlife Detection","authors":"Yunhao Pan, Chenhong Sui, Fuhao Jiang, Guobin Yang, Ankang Zang, Shengwen Zhou","doi":"10.1109/ICARCE55724.2022.10046567","DOIUrl":"https://doi.org/10.1109/ICARCE55724.2022.10046567","url":null,"abstract":"Data augmentation is of great importance to alleviate the insufficiency of training samples, and further improve wildlife detection accuracy. However, current data augmentation methods tend to augment all kinds of samples equally, ignoring the problem of uneven distribution of the number and size of all kinds of samples in wildlife detection datasets, resulting in poor generalization of the model. To address this problem, this paper proposes a joint distribution and class-based data augmentation method for wildlife detection. In this method, diverse rather than universal data augmentation methods are introduced for different classes with a small proportion. This makes the distributions of different classes more balanced. Therefore, each class even with a small number of samples gets good training as well. To evaluate the effectiveness of the proposed method, extensive comparative experiments are conducted. Experimental results show the superiority of our proposed method. Specifically, the detection accuracy of Faster RCNN with Swin Transformer as the backbone network is improved by 0.8% to 96.2% after data augmentation with our method.","PeriodicalId":416305,"journal":{"name":"2022 International Conference on Automation, Robotics and Computer Engineering (ICARCE)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124269188","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 : 2022-12-16DOI: 10.1109/ICARCE55724.2022.10046460
Jian-Kun Lu, Yilin Wu, Dengxue Cao
Buck converter is a buck DC/DC converter. In this paper, an improved control algorithm based on traditional sliding mode control and ESO is proposed for Buck converter, which is affected by some uncertain external factors, such as sudden change of load and variation of reference voltage, in order to meet the needs of high-power voltage conversion occasions. The improved control algorithm firstly evaluates the state, input voltage and load of the whole system, and then uses sliding mode control to improve the overall performance of the system. Finally, the mutation of reference voltage and load is carried out on MATLAB/SIMULINK to verify the feasibility of the improved control algorithm. The simulation results show that compared with the traditional sliding mode control, the proposed control algorithm can increase the system responsiveness and improve the overall performance of the system compared with the traditional control methods.
{"title":"Sliding Mode Control of Buck Converter Using ESO","authors":"Jian-Kun Lu, Yilin Wu, Dengxue Cao","doi":"10.1109/ICARCE55724.2022.10046460","DOIUrl":"https://doi.org/10.1109/ICARCE55724.2022.10046460","url":null,"abstract":"Buck converter is a buck DC/DC converter. In this paper, an improved control algorithm based on traditional sliding mode control and ESO is proposed for Buck converter, which is affected by some uncertain external factors, such as sudden change of load and variation of reference voltage, in order to meet the needs of high-power voltage conversion occasions. The improved control algorithm firstly evaluates the state, input voltage and load of the whole system, and then uses sliding mode control to improve the overall performance of the system. Finally, the mutation of reference voltage and load is carried out on MATLAB/SIMULINK to verify the feasibility of the improved control algorithm. The simulation results show that compared with the traditional sliding mode control, the proposed control algorithm can increase the system responsiveness and improve the overall performance of the system compared with the traditional control methods.","PeriodicalId":416305,"journal":{"name":"2022 International Conference on Automation, Robotics and Computer Engineering (ICARCE)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126186718","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 : 2022-12-16DOI: 10.1109/ICARCE55724.2022.10046524
H. Cai, Gang Liu, Ziqi Zeng, Fangming Deng
The existing deep learning transmission line detection technology with cloud computing is faced with problems such as slow response speed, high communication cost, and difficult to obtain data scattered, as well as the huge amount of data, which causes huge pressure on cloud storage capacity and processing capacity. This paper proposes a transmission line defect detection technology based on adaptive federated learning (FL). Its advantage is that data does not need to be uploaded and shared, which not only reduces communication costs, but also improves data security. In this paper, an adaptive algorithm is added to the original FL algorithm, which can adaptively change the data volume of the next round of training according to the training effect of each round and the local training energy consumption, so as to achieve the optimal number of communication between the two, which greatly reduces the Improve training speed and reduce communication costs. Through experimental analysis, the model training efficiency of the adaptive FL proposed in this paper is 70% higher than that of the centralized cloud computing, and the computing cost is saved by about 40%.
{"title":"Research on Transmission Line Defect Detection Based on Adaptive Federated Learning","authors":"H. Cai, Gang Liu, Ziqi Zeng, Fangming Deng","doi":"10.1109/ICARCE55724.2022.10046524","DOIUrl":"https://doi.org/10.1109/ICARCE55724.2022.10046524","url":null,"abstract":"The existing deep learning transmission line detection technology with cloud computing is faced with problems such as slow response speed, high communication cost, and difficult to obtain data scattered, as well as the huge amount of data, which causes huge pressure on cloud storage capacity and processing capacity. This paper proposes a transmission line defect detection technology based on adaptive federated learning (FL). Its advantage is that data does not need to be uploaded and shared, which not only reduces communication costs, but also improves data security. In this paper, an adaptive algorithm is added to the original FL algorithm, which can adaptively change the data volume of the next round of training according to the training effect of each round and the local training energy consumption, so as to achieve the optimal number of communication between the two, which greatly reduces the Improve training speed and reduce communication costs. Through experimental analysis, the model training efficiency of the adaptive FL proposed in this paper is 70% higher than that of the centralized cloud computing, and the computing cost is saved by about 40%.","PeriodicalId":416305,"journal":{"name":"2022 International Conference on Automation, Robotics and Computer Engineering (ICARCE)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116247254","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 : 2022-12-16DOI: 10.1109/ICARCE55724.2022.10046589
Xinhua Huang, Haocheng Liu, Qingkai Meng
The significance of astronomical site selection in site selection impacts the extent of utilization of astronomical science equipment and the accessibility of observation findings, which is related to both the progress of astronomical science and the expensive resource investment. In order to show the topography and important geographic factors of the Saishten Mountains and Xueshan Pastures potential areas. In this study, we undertake a 3D reconstruction of the astronomical site selection region using Chinese Gaofen satellite images, tailored high-resolution dem data, and a 3D platform, offering a precise visualization help for site planning. The study results show that (1) The 3D reconstruction is quite effective and useful for site selection, according to the results. (2) All candidate locations offer favorable topography, terrain, and elevation characteristics for the astronomical site selection, with the exception of the candidate site in the Saishten Mountains, which necessitates additional thought for the engineering and geological safety of the site selection. (3) Each candidate site's platform area is suitable for the deployment of astronomical research equipment, but the platform space at Snow Mountain Ranch is particularly well suited for the positioning of sizable astronomical scientific equipment.
{"title":"3D Reconstruction of Astronomical Site Selection Based on Multi-Source Remote Sensing","authors":"Xinhua Huang, Haocheng Liu, Qingkai Meng","doi":"10.1109/ICARCE55724.2022.10046589","DOIUrl":"https://doi.org/10.1109/ICARCE55724.2022.10046589","url":null,"abstract":"The significance of astronomical site selection in site selection impacts the extent of utilization of astronomical science equipment and the accessibility of observation findings, which is related to both the progress of astronomical science and the expensive resource investment. In order to show the topography and important geographic factors of the Saishten Mountains and Xueshan Pastures potential areas. In this study, we undertake a 3D reconstruction of the astronomical site selection region using Chinese Gaofen satellite images, tailored high-resolution dem data, and a 3D platform, offering a precise visualization help for site planning. The study results show that (1) The 3D reconstruction is quite effective and useful for site selection, according to the results. (2) All candidate locations offer favorable topography, terrain, and elevation characteristics for the astronomical site selection, with the exception of the candidate site in the Saishten Mountains, which necessitates additional thought for the engineering and geological safety of the site selection. (3) Each candidate site's platform area is suitable for the deployment of astronomical research equipment, but the platform space at Snow Mountain Ranch is particularly well suited for the positioning of sizable astronomical scientific equipment.","PeriodicalId":416305,"journal":{"name":"2022 International Conference on Automation, Robotics and Computer Engineering (ICARCE)","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114700521","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 : 2022-12-16DOI: 10.1109/ICARCE55724.2022.10046590
Chenyang Sun, Xiangjun Liu, Runjie Shen
When a dual manipulator robot performs handling operations in a complex space, it is important to plan a collision avoidance path to the target point quickly and accurately. In order to solve the problems of local minimum and high sampling randomness of traditional path planning algorithms, a new fusion algorithm is proposed for global path planning of a dual manipulator robot. First, an improved artificial potential field (IAPF) method is mentioned for path planning in the start and target point areas. Then, for the local minimum problem, an improved RRT algorithm (IRRT) based on the ε-greedy sampling target biasing strategy and repeated iterative update strategy is fused to reduce the randomness of random tree growth, and explore an optimal path that grows toward the target as much as possible for jumping out of the local minimum area. The URDF model file of the dual manipulator robot is created, and simulation experiments of path planning are conducted based on the Rviz visualization tool under Ros system, which proves the effectiveness of the fusion algorithm.
{"title":"Research on Collision Avoidance Path Planning of Dual Manipulator Robot Based on Fusion Algorithm","authors":"Chenyang Sun, Xiangjun Liu, Runjie Shen","doi":"10.1109/ICARCE55724.2022.10046590","DOIUrl":"https://doi.org/10.1109/ICARCE55724.2022.10046590","url":null,"abstract":"When a dual manipulator robot performs handling operations in a complex space, it is important to plan a collision avoidance path to the target point quickly and accurately. In order to solve the problems of local minimum and high sampling randomness of traditional path planning algorithms, a new fusion algorithm is proposed for global path planning of a dual manipulator robot. First, an improved artificial potential field (IAPF) method is mentioned for path planning in the start and target point areas. Then, for the local minimum problem, an improved RRT algorithm (IRRT) based on the ε-greedy sampling target biasing strategy and repeated iterative update strategy is fused to reduce the randomness of random tree growth, and explore an optimal path that grows toward the target as much as possible for jumping out of the local minimum area. The URDF model file of the dual manipulator robot is created, and simulation experiments of path planning are conducted based on the Rviz visualization tool under Ros system, which proves the effectiveness of the fusion algorithm.","PeriodicalId":416305,"journal":{"name":"2022 International Conference on Automation, Robotics and Computer Engineering (ICARCE)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115837010","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 : 2022-12-16DOI: 10.1109/ICARCE55724.2022.10046579
Yimin Zhang, Chuxuan Chen, Ronglin Hu
Employing robots in badminton training contributes to a more accurate analysis of an athlete's movements and helps avoid injuries. Shuttlecock detection during the flying stage is a critical component of the badminton robot design. However, previous shuttlecock localization methods were unable to detect shuttlecock quickly and accurately in embedded device-based badminton robots, given scale variations, few extractable features, occlusion, and device limitation. In this paper, a deep learning-based shuttlecock localization method is proposed. First, an indoor shuttlecock dataset including 9548 shuttlecock images of various angles and scenes was constructed. Then a shuttlecock detection method YOLO-BTM is proposed, which is based on YOLOv4-Tiny. We proposed a new convolution block to replace the cross-stage partially block in the backbone, to improve the detection speed. To improve the network's ability to detect small objects, the efficient channel attention block is introduced in feature fusion. Finally, a comparative experiment on the accuracy of the method and the detection speed was conducted. The results show that the proposed YOLO-BTM has better performance in detection speed and accuracy compared to the existing state-of-the-art object detection methods on our own shuttlecock dataset. Our method enables real-time, accurate localization of shuttlecock and has the potential to be used in other embedded device based sports robots.
{"title":"YOLO-BTM: A Novel Shuttlecock Detection Method for Embedded Badminton Robots","authors":"Yimin Zhang, Chuxuan Chen, Ronglin Hu","doi":"10.1109/ICARCE55724.2022.10046579","DOIUrl":"https://doi.org/10.1109/ICARCE55724.2022.10046579","url":null,"abstract":"Employing robots in badminton training contributes to a more accurate analysis of an athlete's movements and helps avoid injuries. Shuttlecock detection during the flying stage is a critical component of the badminton robot design. However, previous shuttlecock localization methods were unable to detect shuttlecock quickly and accurately in embedded device-based badminton robots, given scale variations, few extractable features, occlusion, and device limitation. In this paper, a deep learning-based shuttlecock localization method is proposed. First, an indoor shuttlecock dataset including 9548 shuttlecock images of various angles and scenes was constructed. Then a shuttlecock detection method YOLO-BTM is proposed, which is based on YOLOv4-Tiny. We proposed a new convolution block to replace the cross-stage partially block in the backbone, to improve the detection speed. To improve the network's ability to detect small objects, the efficient channel attention block is introduced in feature fusion. Finally, a comparative experiment on the accuracy of the method and the detection speed was conducted. The results show that the proposed YOLO-BTM has better performance in detection speed and accuracy compared to the existing state-of-the-art object detection methods on our own shuttlecock dataset. Our method enables real-time, accurate localization of shuttlecock and has the potential to be used in other embedded device based sports robots.","PeriodicalId":416305,"journal":{"name":"2022 International Conference on Automation, Robotics and Computer Engineering (ICARCE)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134269291","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 : 2022-12-16DOI: 10.1109/ICARCE55724.2022.10046644
Qimeng Chen, Tong Zheng, Liu Liu, Longji Yu, Zhong Chen
The existing state-of-the-arts two-stage oriented object detectors have no significant improvement in the label assignment strategies, and the most widely-used one is the so-called Max IoU Assigner (MIA). In this paper, we first illustrate that MIA may cause matching conflicts in some cases, hinder the matching of ground-truth (GT) boxes with high-quality samples, which is extremely harmful to the training process. After that, we propose a Harmonized Label Assigner (HLA) for the oriented RPN, which can automatically harmonize the assignment priority of each GT box according to the corresponding number of candidate samples, solve the matching conflicts, and improve the detection accuracy of the two-stage oriented detectors. Finally, we implement the proposed HLA on Oriented R-CNN and conduct sufficient experiments on two public datasets (MAR20 and HRSC2016). Without tricks, our HLA significantly improves the detection accuracy of the detector to 83.97% mAP (on MAR20) and 90.42% mAP (on HRSC2016), respectively.
{"title":"HLA: Harmonized Label Assigner for Two-stage Oriented Object Detection","authors":"Qimeng Chen, Tong Zheng, Liu Liu, Longji Yu, Zhong Chen","doi":"10.1109/ICARCE55724.2022.10046644","DOIUrl":"https://doi.org/10.1109/ICARCE55724.2022.10046644","url":null,"abstract":"The existing state-of-the-arts two-stage oriented object detectors have no significant improvement in the label assignment strategies, and the most widely-used one is the so-called Max IoU Assigner (MIA). In this paper, we first illustrate that MIA may cause matching conflicts in some cases, hinder the matching of ground-truth (GT) boxes with high-quality samples, which is extremely harmful to the training process. After that, we propose a Harmonized Label Assigner (HLA) for the oriented RPN, which can automatically harmonize the assignment priority of each GT box according to the corresponding number of candidate samples, solve the matching conflicts, and improve the detection accuracy of the two-stage oriented detectors. Finally, we implement the proposed HLA on Oriented R-CNN and conduct sufficient experiments on two public datasets (MAR20 and HRSC2016). Without tricks, our HLA significantly improves the detection accuracy of the detector to 83.97% mAP (on MAR20) and 90.42% mAP (on HRSC2016), respectively.","PeriodicalId":416305,"journal":{"name":"2022 International Conference on Automation, Robotics and Computer Engineering (ICARCE)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131532314","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In view of the long-term operation and maintenance needs of the Five-hundred-meter Aperture Spherical radio Telescope (FAST) feed cabin, focusing on the practical problems of limited installation space of the feed cabin, numerous disassembly and assembly bolts, low disassembly and assembly efficiency of the feed cabin and high cost, an intelligent robot arm integrating positioning, identification and disassembly is designed. Through the stability analysis of the control system, the amplitude margin is 5.69dB and the phase margin is 49.1deg, meeting the system stability requirements. Through finite element static analysis, modal analysis, harmonic response analysis and test verification, the effective operation and maintenance radius of the robot arm is 1.5m, the maximum torque can reach 130N·m, and the movement space is ±170 °, which can realize the rapid disassembly and assembly of M8 bolts and M12 bolts, and provide technical support for the operation and maintenance of FAST feed cabin.
{"title":"Design and Analysis of Intelligent Robot Arm for FAST Operation and Maintenance","authors":"Hongxi Ren, Ligang Qiang, Lin Li, Taokang Xiao, Dingan Song, Zhiyuan Zhang","doi":"10.1109/ICARCE55724.2022.10046453","DOIUrl":"https://doi.org/10.1109/ICARCE55724.2022.10046453","url":null,"abstract":"In view of the long-term operation and maintenance needs of the Five-hundred-meter Aperture Spherical radio Telescope (FAST) feed cabin, focusing on the practical problems of limited installation space of the feed cabin, numerous disassembly and assembly bolts, low disassembly and assembly efficiency of the feed cabin and high cost, an intelligent robot arm integrating positioning, identification and disassembly is designed. Through the stability analysis of the control system, the amplitude margin is 5.69dB and the phase margin is 49.1deg, meeting the system stability requirements. Through finite element static analysis, modal analysis, harmonic response analysis and test verification, the effective operation and maintenance radius of the robot arm is 1.5m, the maximum torque can reach 130N·m, and the movement space is ±170 °, which can realize the rapid disassembly and assembly of M8 bolts and M12 bolts, and provide technical support for the operation and maintenance of FAST feed cabin.","PeriodicalId":416305,"journal":{"name":"2022 International Conference on Automation, Robotics and Computer Engineering (ICARCE)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131797701","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 : 2022-12-16DOI: 10.1109/icarce55724.2022.10046627
{"title":"Directory of ICARCE 2022 Conference Proceedings","authors":"","doi":"10.1109/icarce55724.2022.10046627","DOIUrl":"https://doi.org/10.1109/icarce55724.2022.10046627","url":null,"abstract":"","PeriodicalId":416305,"journal":{"name":"2022 International Conference on Automation, Robotics and Computer Engineering (ICARCE)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123520052","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 : 2022-12-16DOI: 10.1109/ICARCE55724.2022.10046584
Dominik Polke, Florian Diepers, Elmar Ahle, D. Söffker
The chemical process industry is currently undergoing a transformation to Chemistry 4.0, where digitalization, modularization, sustainability, and the circular economy are coming into focus. A growing interest in the use of process data with the aim of gaining a better understanding of the production process and conserving resources can be observed. Data-driven modeling is used in chemical industry when the production process is too complex to be described by chemical laws. Gaining knowledge of the chemical relationships can lead to resource-conserving production. In this paper, a framework to optimize the process of data-driven modeling in an industrial environment is presented. For generating data-driven models of industrial processes, many manual and time-consuming steps have to be carried out. This leads to delay in information acquisition and process optimization. Therefore, the presented framework automates these steps to accelerate the process of data-driven modeling. The steps are to extract the data from a process control system (PCS), make the data available for data-driven modeling, train the model, and deploy the model for predicting the process. To achieve high availability of the data and generate data-driven models, cloud services are used. The framework of this paper is applied to a high-throughput formulation system (HTFS) for coatings. In this paper, Gaussian processes are used for data-driven modeling. The evaluation of the framework shows the usefulness in this domain, but also the flexibility and scalability of this framework.
{"title":"Development of a Framework for Data-Driven Modeling with Cloud Services in the Process Industry","authors":"Dominik Polke, Florian Diepers, Elmar Ahle, D. Söffker","doi":"10.1109/ICARCE55724.2022.10046584","DOIUrl":"https://doi.org/10.1109/ICARCE55724.2022.10046584","url":null,"abstract":"The chemical process industry is currently undergoing a transformation to Chemistry 4.0, where digitalization, modularization, sustainability, and the circular economy are coming into focus. A growing interest in the use of process data with the aim of gaining a better understanding of the production process and conserving resources can be observed. Data-driven modeling is used in chemical industry when the production process is too complex to be described by chemical laws. Gaining knowledge of the chemical relationships can lead to resource-conserving production. In this paper, a framework to optimize the process of data-driven modeling in an industrial environment is presented. For generating data-driven models of industrial processes, many manual and time-consuming steps have to be carried out. This leads to delay in information acquisition and process optimization. Therefore, the presented framework automates these steps to accelerate the process of data-driven modeling. The steps are to extract the data from a process control system (PCS), make the data available for data-driven modeling, train the model, and deploy the model for predicting the process. To achieve high availability of the data and generate data-driven models, cloud services are used. The framework of this paper is applied to a high-throughput formulation system (HTFS) for coatings. In this paper, Gaussian processes are used for data-driven modeling. The evaluation of the framework shows the usefulness in this domain, but also the flexibility and scalability of this framework.","PeriodicalId":416305,"journal":{"name":"2022 International Conference on Automation, Robotics and Computer Engineering (ICARCE)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-12-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117198050","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：