检索结果-内蒙古大学图书馆

2024 China Automation Congress, CAC 2024

作者： Liu, Jianyang Gan, Chao Cao, Weihua School of Automation China University of Geosciences Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan430074 China

ISBN: (纸本)9798350368604

Geological exploration is essential for economic and social progress, with lithology identification being a key technology that improves exploration precision and resource development efficiency. This paper presents a real-time identification method of formation lithology in drilling process based on pre-training language model, which is divided into three stages. In this first stage, the drilling data undergoes rigorous cleaning, feature selection, and normalization to eliminate noise and anomalies. In the second stage, key model inputs such as weight on bit (WOB), rate of penetration (ROP), rotation speed (RPM), torque, and pump volume (P) are identified through detailed drilling mechanism analysis. In the last stage, a real-time identification method of formation lithology in drilling process based on pre-training language model is proposed, which realizes high-precision real-time identification in drilling process. The effectiveness of the proposed method is proved by comparison with common methods. © 2024 IEEE.

关键词： Lithology

来源：评论

学校读者我要写书评

暂无评论

Non-Convex Operating Zone Establishment and Alarm Limit Design for Multivariate Alarm systems Based on Grid Clustering 43

Non-Convex Operating Zone Establishment and Alarm Limit Desi...

引用

43rd Chinese control Conference, CCC 2024

作者： Wei, Yujie Hu, Wenkai Li, Yupeng Cao, Weihua China University of Geosciences School of Automation Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan430074 China

ISBN: (纸本)9789887581581

Alarm systems are core components in large complex industrial facilities to ensure safe and efficient operation. Traditional univariate alarm system design methods ignore the correlation of process variables and thus may generate massive false alarms and missed alarms. An effective way of designing a multivariate alarm system is to establish the normal operating zones (NOZs) based on historical data under normal conditions. These NOZs serve as the basis for setting alarm limits in multivariate alarm systems. Given that the actual correlations between related variables are frequently unknown, a robust modeling approach is necessary to effectively handle both convex and non-convex NOZs. Thus, this paper presents a systematic method for nonconvex operating zone establishment and alarm limit design for multivariate alarm systems. A multivariate non-convex NOZ is established based on grid clustering. Then, the multidimensional grid boundary is extracted and used to determine the alarm limits for real-time samples. A numerical example is used to demonstrate the effectiveness of the proposed method. © 2024 Technical Committee on control Theory, Chinese Association of Automation.

关键词： Alarm systems

来源：评论

学校读者我要写书评

暂无评论

A Review on 3D Hand Pose and Shape Reconstruction from Color Images 43

A Review on 3D Hand Pose and Shape Reconstruction from Color...

引用

43rd Chinese control Conference, CCC 2024

作者： Leng, Huajian Ge, Yuting Xu, Chi China University of Geosciences School of Automation Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Ministry of Education Engineering Research Center of Intelligent Technology for Geo-Exploration Wuhan430074 China

ISBN: (纸本)9789887581581

The task of hand reconstruction aims to estimate the shape and pose of hands from the input images. Deep learning algorithms have shown remarkable performance in the field of computer vision in recent years, promoting the development of related algorithms. This article reviews the 3D hand reconstruction technologies, introducing the research background, value, and directions, summarizing the characteristics of datasets and evaluation metrics, and discussing the types of parametric, non-parametric, and joint mesh representation methods;it then outlines research directions in single-hand, hand-object interaction, and two-hand reconstruction, and analyzes their progress. Finally, it summarizes both the hand datasets and the algorithms, and provides a perspective on the future development of hand reconstruction. © 2024 Technical Committee on control Theory, Chinese Association of Automation.

关键词： 3D reconstruction

来源：评论

学校读者我要写书评

暂无评论

A Novel Thermal Analysis Method for Tubular PM Linear Motors Based on Transfer Learning

引用

IEEE Transactions on Transportation Electrification 2025年第3期11卷 7379-7388页

作者： Wu, Tao Dai, Peipei Xue, Gang Guo, Youguang Lei, Gang Zhu, Jianguo Wang, Yifei China University of Geosciences School of Automation Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China University of Technology Sydney School of Electrical and Data Engineering NSW2007 Australia University of Sydney School of Electrical and Computer Engineering SydneyNSW2006 Australia

Thermal modeling and analysis are critical for permanent magnet linear synchronous motors (PMLSMs), particularly in multi-physical analysis and motor design optimization. This paper proposes a new method for thermal steady state modeling and analysis based on transfer learning, which combines the thermal mechanism model with the data-driven model to achieve high-precision temperature analysis with small number of finite element analysis (FEA) samples. First, an equivalent thermal circuit (ETC) model is established for PMLSMs based on the structural parameters and heat transfer principles. Second, a hybrid model based on transfer learning is developed: the model is pre-trained by a large sample set generated from the ETC in the source field (mechanism analytical model), and then the trained model is transferred to the target field (FEA data-driven model). The parameters of the neural network in the target field are optimized by a few FEA sample sets with higher precision to improve the prediction accuracy of the target model. Experimental results demonstrate that the proposed method significantly boosts the performance of the deep neural network (DNN) model, particularly when training sets are small. The root mean square error (RMSE) of the DNN model decreases by 0.901 in the 5:5 (training/testing) dataset and 1.726 in the 2:8 dataset;and the corresponding R2 and maximal absolute error of motor performances are significantly decreased. The smaller the proportion of the training set, the better performance of the hybrid modeling will achieve. Furthermore, a temperature rise experiment conducted on a prototype motor validates the efficacy and precision of the proposed hybrid modeling methodology. © 2015 IEEE.

关键词： Permanent magnets

来源：评论

学校读者我要写书评

暂无评论

Multi user computation offloading decision in edge networks 43

Multi user computation offloading decision in edge networks

引用

43rd Chinese control Conference, CCC 2024

作者： Liu, Chang Yuan, Yan Cao, Weihua Hu, Yabin China University of Geosciences School of Automation Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan430074 China

ISBN: (纸本)9789887581581

Computing offload is a key technology in mobile edge computing (MEC). By offloading tasks to the edge servers, it solves the problem that terminal devices in the edge network cannot handle high computing and time sensitive tasks. However, in multi-user scenarios, the resources of edge servers are very limited, and without good offloading decisions, the performance of MEC systems cannot be guaranteed. In this paper, the computation offloading problem of multi-user MEC system is analyzed and modeled and a batch offloading method is proposed. And in response to the shortcomings of the existing algorithms, a neural network combined with genetic algorithm for offloading decision is proposed. Simulation results show that the proposed algorithm performs better than existing algorithms in terms of time complexity and accuracy. © 2024 Technical Committee on control Theory, Chinese Association of Automation.

关键词： Computation offloading

来源：评论

学校读者我要写书评

暂无评论

A Lightweight Steel Strip Defect Detection Method Based on YOLO 43

A Lightweight Steel Strip Defect Detection Method Based on Y...

引用

43rd Chinese control Conference, CCC 2024

作者： Qian, Hao Xu, Chi China University of Geosciences School of Automation Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Ministry of Education Engineering Research Center of Intelligent Technology for Geo-Exploration Wuhan430074 China

ISBN: (纸本)9789887581581

As an essential component of modern industry, steel strips play an indispensable role in various fields and serve as a crucial raw material in industrial production. However, due to various factors such as production technology, the surface of steel strips often exhibits unavoidable defects like scratches and crazings. Therefore, implementing online surface quality inspection for steel strips is one effective way to enhance product quality and production efficiency. This paper proposes a detection method for surface defects in steel strips based on YOLO. By optimizing the network structure of YOLOv5s, the performance on the NEU-DET dataset has improved by nearly 4 % compared to the original network's mAP value. Experimental results indicate that the detection performance of this method is significantly superior to the current mainstream lightweight detection networks. © 2024 Technical Committee on control Theory, Chinese Association of Automation.

关键词： Strip metal

来源：评论

学校读者我要写书评

暂无评论

Alarm Prediction with Variational Output Dimension for Alarm Floods Using Transformer 43

Alarm Prediction with Variational Output Dimension for Alarm...

引用

43rd Chinese control Conference, CCC 2024

作者： Pang, Long Jiang, Wenbin Hu, Wenkai Wang, Zhuang Cao, Weihua China University of Geosciences School of Automation Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Ministry of Education Engineering Research Center of Intelligent Technology for Geo-Exploration Wuhan430074 China

ISBN: (纸本)9789887581581

Industrial alarm systems play crucial roles in assessing the production status and ensuring process safety. However, the presence of alarm floods can make alarm systems partially or fully fail, and thus would cause serious consequences. Alarm prediction, as an advanced approach to address alarm floods, offers operators decision supports when an alarm flood occurs. Existing alarm prediction methods requires that the output alarm sequences are of a fixed dimension, making such methods unable to adapt to the changing alarm rate. Accordingly, this paper proposes an alarm prediction model based on Transformer. The proposed model can predict alarm sequences with variational dimensions and can also achieve remarkable accuracy. The effectiveness of the proposed method is demonstrated by a case study with data from the Vinyl Acetate Monomer (VAM) simulation model. © 2024 Technical Committee on control Theory, Chinese Association of Automation.

关键词： Polyvinyl acetates

来源：评论

学校读者我要写书评

暂无评论

A Detector-Free Image Matching Method based on Enhanced Feature Fusion 43

A Detector-Free Image Matching Method based on Enhanced Feat...

引用

43rd Chinese control Conference, CCC 2024

作者： Chen, Ziyang Xu, Chi China University of Geosciences School of Automation Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Ministry of Eduction Engineering Research Center of Intelligent Technology for Geo-Exploration Wuhan430074 China

ISBN: (纸本)9789887581581

Image feature matching is an important research field in computer vision that can be widely applied to advanced vision tasks, such as 3D reconstruction and visual tracking. Considering the low matching accuracy and poor robustness of traditional detector-based matching methods on weakly textured images, we propose a detector-free image matching method based on enhanced feature fusion. Given that weak texture images consist of a large number of similar or irregular pixels, we introduce a new model that enhances feature fusion to recover the underlying texture information of the image. First, we propose a bidirectional fusion model to extract multi-scale feature maps and add an initialization module to correlate the feature maps and enhance the representation ability of the backbone. Then, we perform feature transformation using the quadtree attention model and finally establish dense pixel-by-pixel matching at both coarse and fine scales. The quadtree attention model employs a hierarchical computational attention method, accurately capturing image details and remote dependencies while ensuring linear complexity. Experimental results demonstrate that the proposed method achieves higher matching accuracy and better generalization on large datasets such as MegaDepth and ScanNet, making it suitable for subsequent work. © 2024 Technical Committee on control Theory, Chinese Association of Automation.

关键词： Image matching

来源：评论

学校读者我要写书评

暂无评论

Distributed Power Source Planning Based on Improved Particle Swarm optimization 3

Distributed Power Source Planning Based on Improved Particle...

引用

3rd Asia Power and Electrical Technology Conference, APET 2024

作者： Wang, Jiahui Tan, Zhili School of Automation China University of Geosciences Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan430074 China

ISBN: (纸本)9798350367690

The multitude of advantages offered by distributed power generation units are recognized as crucial factors in enhancing the security of distribution networks. Ensuring the optimal size and placement of distributed generation (DG) units not only enhances system security but also increases profitability. Haphazard placement of DG units may lead to adverse impacts on the power system, resulting in increased power losses and diminished voltage distribution. Therefore, to ensure the optimal size and placement of DG units, this paper proposes a method to optimize DG scheduling and operation, thereby improving voltage distribution, reducing power losses, and enhancing the overall reliability of the entire grid. To reduce power loss (PL) and improve voltage distribution (VP), an adaptive chaotic particle swarm optimization (ACPSO) algorithm is employed to determine the optimal locations and sizes of DG units. By comparing this algorithm with traditional approaches, its efficacy in enhancing performance can be discerned. © 2024 IEEE.

关键词： Power distribution planning

来源：评论

学校读者我要写书评

暂无评论

Short-Term Load Forecasting Method Based on Multi-Branch Convolutional Network and Bayesian optimization 43

Short-Term Load Forecasting Method Based on Multi-Branch Con...

引用

43rd Chinese control Conference, CCC 2024

作者： Lv, Ji Ding, Min Li, Quanhui Zhou, Sibei Fang, Zhijian China University of Geosciences School of Automation Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan430074 China

ISBN: (纸本)9789887581581

Accurate short-term power load forecasting is important to ensure the reliability and economy of the power grid. Convolutional neural networks are widely used, but they are difficult to capture the multi-scale temporal characteristics of load data. To address this problem, this study proposes a short-term load forecasting method based on a multi-branch CNN architecture and Bayesian optimisation. By using different sizes of convolutional kernels in different branches, different time-scale features of power load data are captured to improve the identification of rapid changes and potential long-term trends. In addition, the hyperparameters of the forecasting model are automatically adjusted by a Bayesian optimisation algorithm to achieve the best forecasting performance. In forecasting experiments on real load data, the method in this paper significantly improves the forecasting accuracy compared to baseline models. © 2024 Technical Committee on control Theory, Chinese Association of Automation.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：