检索结果-内蒙古大学图书馆

2024 International Annual Conference on Complex Systems and Intelligent Science, CSIS-IAC 2024

作者： Huang, Jianhang Jia, Lijie Zhou, Yitian Wang, Shuo Zhang, Xinliang Xu, Xiaozhuo School of Electrical Engineering and Automation Henan Polytechnic University Henan International Joint Laboratory of Direct Drive and Control of Intelligent Equipment Jiaozuo China Zhoushan Yangwangnaxin Technology Co. Zhoushan China School of Electrical Engineering and Automation Henan Polytechnic University Jiaozuo China

ISBN: (纸本)9798331504755

In deep network-based steel surface defect detection models, using only serial convolution operations can hinder the detection of small defects due to insufficient multi-scale detail information. To address this issue, a detection model based on joint multi-dimensional atrous convolution (JMDC) is proposed. By introducing atrous convolution with various kernel sizes into the feature extraction layer, a multi-channel parallel convolution chain is constructed to capture multi-scale defect features, enhancing the detail perception. Subsequently, a JMDC operator, generated from the residual network, fuses these features and establishes a channel constraint to improve the training stability and global feature extraction. Then by integrating the JMDC into the multi-level feature extraction layers, the improved YOLOv5 realizes the detection enhancement for the multi-scale defects. Correspondingly, the k-means algorithm is adopted to generate adaptive multi-scale anchor boxes with the reduction of the positional deviation. Experimental results show that the proposed approach has provided a detection accuracy of 81.7% on the NEU-DET dataset, yielding a significant improvement on the defect detection, especially for small targets. © 2024 IEEE.

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

Feedback Generative Adversarial Network With Channel-Space Attention for Image-Based Optimal Path Search Planning

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第8期5卷 4293-4307页

作者： Sun, Tao Li, Jian-Sheng Zhang, Yi-Fan Ru, Xin-Feng Wang, Ke Tsinghua University Department of Automation Beijing100084 China Dalian University of Technology School of Control Science and Engineering Dalian116024 China Dalian University of Technology Key Laboratory of Intelligent Control and Optimization for Industrial Equipment of Ministry of Education Dalian116024 China Tongji University School of Economics and Management Shanghai200092 China

Path planning and seeking is one of the most challenging and interesting problems in the field of artificial intelligence. In this article, we propose a new image-based path planning algorithm to overcome traditional design concepts based on complex topological networks. Specifically, a novel feedback generative adversarial network learning model with channel-space attention (CSA-FGAN) is developed to search and plan an optimal path. It is worth noting that the input and output of the proposed feedback learning model are RGB images with complex spatial obstacle environment maps and RGB images with feasible paths, respectively. In particular, in order to further enhance the feature extraction ability and path prediction accuracy of image data, the feedback attention mechanism controls and improves the depth of the network by forming a closed loop of the proposed learning model. Final, three application scenario image examples are conducted to verify the superiority of the proposed path planning policy. © 2020 IEEE.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

An Early Multi-Fault Diagnosis Method of Lithium-ion Battery Based on Data-Driven 41

An Early Multi-Fault Diagnosis Method of Lithium-ion Battery...

引用

第41届中国控制会议

作者： Xin Gu Yunlong Shang Chijun Li Yuhao Zhu Bin Duan Jinglun Li Wenyuan Zhao School of Control Science and Engineering Shandong University School of Automation Science and Electrical Engineering Beijing University of Aeronautics and Astronautics

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

Lithium-ion power battery is the "heart" of electric *** is significant to diagnose the battery fault quickly and accurately.A real-time data driven diagnosis approach for early battery faults based on improved principal component analysis is proposed in this *** technique rotates the battery pack voltage sequence into a new coordinate space through linear combination,while the detection metrics of square prediction errors and modified contribution plots are employed to achieve minor fault *** addition,the training sample of this method relies on the voltage sequence of the battery health state instead of the battery fault data which is difficult to *** experimental results demonstrate that this approach can not only locate the battery cell where the fault occurs,but also diagnose the open-circuit and short-circuit faults of the battery as well as the occurrence and duration of the fault in ***,the feasibility and robustness of the proposed method are verified by applying different experimental *** summary,the presented approach provides an easy-to-implement option that does not require accurate mathematical modeling,expert understanding,and complex computational processes.

关键词： Early fault diagnosis Lithium-ion battery pack Principal component analysis Data driven

来源：评论

学校读者我要写书评

暂无评论

Mechanical behavior of entangled metallic wire materials-polyurethane interpenetrating composites

引用

Defence Technology（防务技术） 2023年第2期20卷 120-136页

作者： Xiao-yuan Zheng Zhi-ying Ren Hong-bai Bai Zhang-bin Wu You-song Guo College of Mechanical Engineering and Automation Fuzhou UniversityFuzhou350116China Institute of Metal Rubber&Vibration Noise Fuzhou UniversityFuzhou350116China Gongbang Vibration Control Technology Co. LtdJiangsuJiangyin214431China

Composite materials exhibit the impressive mechanical properties of high damping and stiffness,which cannot be attained by employing conventional single *** these lines,a novel material architecture is presented in this work in order to fabricate composites with enhanced mechanical *** specifically,entangled metallic wire materials were used as the active matrix,whereas polyurethane was employed as the reinforcement *** a result,an entangled metallic wire material-polyurethane composite with high damping and stiffness was prepared by enforcing the vacuum infiltration *** top of that,the mechanical properties(loss factor,energy consumption,and average stiffness)of the proposed composite materials were characterized by performing dynamic tests,and its fatigue characteristics were verified by the micro-interface bonding,as well as the macro-damage *** impact of the density,preloading spacing,loading amplitude,and exciting frequency on the mechanical properties of the composites were also thoroughly *** extracted results indicate that the mechanical properties of the composites were significantly enhanced than those of the pure materials due to the introduction of interface ***,the average stiffness of the composites was about 10 times the respective value of the entangled metallic wire ***,a rise in the loading period leads to some failure between the composite interfaces,which reduces the stiffness property but enhances the damping dissipation ***,a comprehensive dynamic mechanical model of the composites was established,while it was experimentally *** proposed composites possess higher damping features,i.e.,stiffness characteristics,and maintain better fatigue characteristics,which can broaden the application range of the *** addition,we provide a theoretical and experimental framework for the research and applications in the field of metal ma

关键词： Entangled metallic wire material Composites materials Damping property Stiffness Fatigue characteristics

来源：评论

学校读者我要写书评

暂无评论

An analysis towards integrated control of wastewater collecting and treatment systems 27

An analysis towards integrated control of wastewater collect...

引用

27th International Conference on System Theory, control and Computing, ICSTCC 2023

作者： Vasiliev, Iulian Luca, Laurentiu Barbu, Marian Vilanova, Ramon Caraman, Sergiu Department of Automation and Electrical Engineering Dunărea de Jos University of Galati Galati Romania Department of Automation Systems and Advanced Control Research Autonomous University of Barcelona Barcelona Spain

ISBN: (纸本)9798350337983

This paper deals with an analysis regarding the effect of Sewer Network operation on the performances of the integrated wastewater collecting and treatment system. The analysis is done by considering various control actions for the output storage tank of the sewer network, comparing a series of performance indicators defined for both subsystems with their "no control" value. Two cases have been considered regarding the control actions for the output storage tank of the sewer network: 1. keeping the output valve at a constant opening and 2. keeping the output valve at the maximum opening possible such as the outflow is limited to a constant value. While the second case did not prove advantageous, the first one showed that lowering the sewer network output valve opening would improve the integrated system performances. © 2023 IEEE.

关键词： Wastewater treatment

来源：评论

学校读者我要写书评

暂无评论

Harnessing Neural Unit Dynamics for Effective and Scalable Class-Incremental Learning 41

Harnessing Neural Unit Dynamics for Effective and Scalable C...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Li, Depeng Wang, Tianqi Chen, Junwei Dai, Wei Zeng, Zhigang School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China School of Information and Control Engineering China University of Mining and Technology Xuzhou China

Class-incremental learning (CIL) aims to train a model to learn new classes from non-stationary data streams without forgetting old ones. In this paper, we propose a new kind of connectionist model by tailoring neural unit dynamics that adapt the behavior of neural networks for CIL. In each training session, it introduces a supervisory mechanism to guide network expansion whose growth size is compactly commensurate with the intrinsic complexity of a newly arriving task. This constructs a near-minimal network while allowing the model to expand its capacity when cannot sufficiently hold new classes. At inference time, it automatically reactivates the required neural units to retrieve knowledge and leaves the remaining inactivated to prevent interference. We name our model AutoActivator, which is effective and scalable. To gain insights into the neural unit dynamics, we theoretically analyze the model's convergence property via a universal approximation theorem on learning sequential mappings, which is under-explored in the CIL community. Experiments show that our method achieves strong CIL performance in rehearsal-free and minimal-expansion settings with different backbones. Copyright 2024 by the author(s)

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

DUAL-COMPONENT INTER-DEPENDENT FAILURE MECHANISM ANALYSIS OF DIESEL ENGINE BASED ON COUNTERFACTUAL INFERENCE

DUAL-COMPONENT INTER-DEPENDENT FAILURE MECHANISM ANALYSIS OF...

引用

2024 Jiangsu Annual Conference on automation, JACA 2024

作者： Li, Heng Huang, Shoujin Ding, Yong Lu, Ningyun CRllege of Automation Engineering Nanjing University of Aeronautics and Astronautics Nanjing211106 China State Key Laboratory of Mechanics and Control of Mechanical Structures Nanjing211106 China

ISBN: (纸本)9781837242689

For diesel engines, the coupling between the high-pressure oil pump and injector complicates fault diagnosis, and relationships between faults and symptoms are complex. Without analyzing inter-dependent failure mechanisms, it is straightforward to fall into the trap of pseudo-relevance with big data and there is also no guarantee of the accuracy and robustness of the predictive model. Multi-facto accidents are characterized by strong concealment, complex factor association, irreversible state, etc. Analyzing inter-dependent failure mechanisms remains a challenging problem in the reliability and safety of diesel engines. To address these challenges, this paper constructs a dual-component inter-dependent failure mechanism analysis method. First, a fault causal network is constructed using domain knowledge. Next, the Leaky Noisy-Or assumption and hidden variable nodes are incorporated to establish a Structural Causal Model (SCM). In addition, counterfactual inference based on do operator intervention in twin networks is employed to calculate two metrics: expected sufficiency and expected disablement, measuring the strength of the causal explanation of faults on the symptom. Finally, the case study demonstrates that this method effectively provides a clear causal chain of failure and intuitively reveals the causal relationship between fault causes and fault symptoms. © The Institution of engineering & Technology 2024.

关键词： Inference engines

来源：评论

学校读者我要写书评

暂无评论

A novel 3D-Space Vector Modulation for Three single-phase combined 400Hz Ground Power Unit 12

A novel 3D-Space Vector Modulation for Three single-phase co...

引用

12th IEEE International Conference on control, automation and Information Sciences, ICCAIS 2023

作者： Que, Son Tran Van Diep, Huynh Dich, Nguyen Quang Trung, Nguyen Kien Hanoi University of Science and Technology Institute for Control Engineering and Automation Ha Noi Viet Nam Hanoi University of Science and Technology Department of Industrial Automation Ha Noi Viet Nam

ISBN: (纸本)9798350328783

The three single-phase combined (TSC) topology has been ruling the power converter industries for many years. In this research, TSC topology with Lithium Battery DC source is proposed for 400Hz Ground Power Unit (GPU) due to its advantages particularly in increasing the system flexibility and reliability. Sinusoidal Pulse Width Modulation (SPWM) is one of the most viable PWM techniques for inverter, however, it is shown that SPWM has no good contribution on three phase control system, hence, Space Vector Modulation (SVM) technique is now more widely used. This paper proposes a novel effectual three-dimensional Space Vector Modulation (3D-SVM) technique for three single-phase combined inverter, which can provide up to 27 switching states, expanding the possible range to find the best decision boundary to compare with traditional SPWM. The effectiveness of the proposed method is verified through MATLAB simulations. © 2023 IEEE.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Simulation of non-inductive vector control of permanent magnet synchronous motor based on sliding mode observer 3

Simulation of non-inductive vector control of permanent magn...

引用

3rd IEEE International Conference on Electronic Technology, Communication and Information, ICETCI 2023

作者： Zhang, Caiyue Liu, Zipin Xu, Bowen Guilin University of Electronic Technology School of Electronic Engineering and Automation Guilin China Xi'An University of Science and Technology School of Electrical and Control Engineering Shaanxi Xi'an China Harbin Institute of Technology Shenzhen School of Mechanical Engineering and Automation Shenzhen China

ISBN: (纸本)9798350398410

Permanent magnet synchronous motors (PMSM) are widely used due to their numerous benefits. It is critical to get rotor position and speed information in order to operate the motor accurately. Sensorless control techniques have emerged as a popular study area both at home and overseas. The sliding mode observer (SMO) may indirectly detect rotor position and has the benefits of easy implementation and efficient algorithms. In this study, a mathematical model for sensorless control of a PMSM is developed using SMO, vector control, and other techniques. With a surface-mounted PMSM as the study object, a mathematical model for sensorless control of PMSM is developed. PMSM's sliding mode observer model is built in the matlab/simulink environment. Experiments demonstrate that the system can track the rotor position and speed of the motor precisely and fulfill the requirements of sensorless vector control of PMSM. © 2023 IEEE.

关键词： Sensorless control

来源：评论

学校读者我要写书评

暂无评论

A Combination of Soft Attention-aided CNN Models using Dempster-Shafer Theory for Skin Cancer Classification

A Combination of Soft Attention-aided CNN Models using Demp...

引用

International Conference on Current Problems of Applied Mathematics and Computer Systems, CPAMCS 2023

作者： Sarkar, Sujan Ray, Amartya Kaplun, Dmitrii Sarkar, Ram Department of Computer Science and Engineering Jadavpur University Kolkata India Department of Automation and Control Processes Saint Petersburg Electrotechnical University "LETI" Saint Petersburg Russia

ISBN: (纸本)9783031640094

Skin cancer is one of the most deadly forms of cancer in the world, causing hundreds of deaths every year. Researchers have been developing various computer-aided diagnosis (CAD) systems to help medical professionals classify the types of skin cancer using different image modalities. In this paper, we propose a model, which combines the outcomes of three separately trained attention-based convolutional neural networks (CNNs) using the Dempster-Shafer theory. The three base learners used in this paper are ResNet50, InceptionResNetV2 and Dense-Net201, each of which is aided by a soft attention module. For evaluating the proposed model, we have considered a publicly accessible skin cancer dataset, called HAM10000. This is a challenging dataset to work with as it is a class imbalanced dataset. Our proposed approach yields an accuracy of 0.932, which is better than many recently published methods found in the literature. The source code is available at: https://***/Cmatermedicalimageanalysis/DST_Combination © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

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：