检索结果-内蒙古大学图书馆

第40届中国控制会议

作者： Jun Chen Xiaoqiang Chen Linbo Luo Ganbei Wang School of Automation China University of Geosciences Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education

In view of the difficulty and low accuracy of small object detection for remote sensing images, this paper proposes a small object detection algorithm based on contextual information fusion to solve the problem of real-time detection accuracy of small object. In this paper, we use bottom-up VGG16 network to realize multi-scale feature extraction to deal with the problem of insufficient image feature extraction. To direct at the problem that the feature information of each feature layer is single, the shallow feature layer and the deep feature layer are fused through the feature fusion module, which achieves the purpose that some feature layers have more abundant fusion features in the structure level. Aiming at the problem that the detection objects in remote sensing images are mainly small and medium-sized objects, this paper proposes to use the multivariate information of four different scale feature layers for classification prediction and regression prediction, so as to reduce the complexity of network *** experimental results show that the proposed small object detection algorithm based on the fusion of four scale deep and shallow contextual information can obtain good accuracy and real-time performance on the NWPU VHR-10 dataset, improve the detection accuracy on the basis of ensuring the real-time detection, and perform well in the small object detection task of remote sensing images.

关键词： remote sensing deep learning small object detection SSD feature fusion

来源：评论

学校读者我要写书评

暂无评论

Detection of Sequential Alarm Patterns in complex Industrial Facilities Using ClaSP and Top-K Algorithms

Detection of Sequential Alarm Patterns in Complex Industrial...

引用

第40届中国控制会议

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

In modern complex industrial processes,due to poor rationalization of alarm systems and the complexity of process interconnections,alarm floods are commonly *** floods are also identified as the main causes of many industrial *** valid approach to deal with alarm floods is to mine meaningful alarm sequential patterns from alarm *** identified patterns can help to analyze root causes or to configure dynamic alarming *** this paper,a method based on the combination of ClaSP and Top-K is proposed to mine interesting alarm sequential patterns from historical alarm *** contributions of this study are twofold:1) A pattern mining approach is adapted to mine interesting patterns from alarm flood sequences;2) A pattern compression strategy is proposed to reduce pattern redundancy.A case study is presented to demonstrate the effectiveness of the proposed method.

关键词： Alarm system alarm flood sequential pattern mining process monitoring

来源：评论

学校读者我要写书评

暂无评论

Formation control of Nonlinear Multi-Agent systems with Actuator and Communication Attacks

Formation Control of Nonlinear Multi-Agent Systems with Actu...

引用

第40届中国控制会议

作者： Shihao Wang Shiqi Zheng Chenghao Zhao Haitao Jian Haiming Li School of Automation China University of Geosciences Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education

In this paper, we focus on the formation control problems of MAS over a directed graph with actuator and communication attacks. The considered system is composed of a leader, some followers and an attacked communication network. Firstly,a new distributed observer is proposed to estimate the leader information despite communication attacks. Then, for high-order nonlinear systems, we develop an adaptive control strategy to solve the actuator attack by using Nussbaum function and backstepping technique, so that the agent with actuator attacks can follow the leader’s trajectory. Finally, a simulation example is proposed to verify the results of this paper.

关键词： Multi-agent systems formation control distributed observer attack nussbaum function

来源：评论

学校读者我要写书评

暂无评论

Finite-Time State Estimation for Discrete-Time Nonlinear Singularly Perturbed complex Networks under New Event-Triggered Mechanism

Finite-Time State Estimation for Discrete-Time Nonlinear Sin...

引用

第40届中国控制会议

作者： Chao Yang Xiongbo Wan School of Automation China University of Geosciences Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education

This paper investigates the finite-time state estimation problem for a class of discrete-time nonlinear singularly perturbed complex networks under a new dynamic event-triggered mechanism(DETM). This new DETM is devised to adjust the date packet transmissions flexibly with hope to save network resources. By constructing a new Lyapunov function dependent on the information of the singular perturbation parameter(SPP) and DETM, a sufficient condition is derived which ensures that the error dynamics of state estimation is finite-time stable. The parameters of the state estimator are given by means of the solutions to several matrix inequalities and the upper bound of the SPP can be evaluated simultaneously. The effectiveness of the designed state estimator is demonstrated by a numerical example.

关键词： complex networks dynamic event-triggered mechanism singularly perturbed systems state estimation

来源：评论

学校读者我要写书评

暂无评论

A Few-Shot Learning Approach for Fault Diagnosis Based on Event Sequences of Alarm Floods

A Few-Shot Learning Approach for Fault Diagnosis Based on Ev...

引用

第32届中国过程控制会议(CPCC2021)

作者： Guang Yang Wenkai Hu Weihua Cao Min Wu School of Automation China University of Geosciences Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education

Alarm systems are commonly deployed in modern industrial facilities for monitoring of process ***,due to the presence of nuisance alarms and alarm floods,the efficiency of many real alarm systems is much *** that problems,such as chattering alarms,redundant alarms,and false alarms,can be well solved,alarm floods are still difficult to *** an alarm flood situation,a number of alarms appear and overwhelm the plant operators;as a result,the operator may overlook the critical alarms and thus the situation may get *** paper studies the root cause analysis of alarm floods,and proposes a Few-Shot Learning(FSL) approach to diagnose faults under alarm flood situations based on alarm event sequences extracted from alarm and event(A&E) *** with the existing methods based on continuous-valued process data,the proposed method does not need to conduct feature selection or dimension reduction,and thus is more straightforward and computationally *** addition,the proposed method only requires a few shots of faulty data to train a fault diagnosis model,and thus shows better applicability in *** experimental results on the Vinyl Acetate Monomer(VAM) benchmark dataset showed the superiority of the proposed model against the state-of-the-art approaches.

关键词： Fault diagnosis alarm flood few-shot learning alarm and event data

来源：评论

学校读者我要写书评

暂无评论

Prediction Model of Oxygen-Enrichment-Related Gas Utilization Rate Based on Multiple Time Scales

Prediction Model of Oxygen-Enrichment-Related Gas Utilizatio...

引用

第33届中国控制与决策会议

作者： Jiajia Peng Jianqi An Xin Chen Min Wu School of Automation China University of Geosciences Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education

In the smelting process of a blast furnace(BF),the oxygen enrichment operation is an important operation to control the gas utilization rate(GUR).It affects the development of GUR by changing the BF states on multiple time *** at the problem that oxygen enrichment has multi-time-scale effects on GUR,this paper establishes a multi-time-scale prediction model based on mechanism analysis and data ***,the influence of oxygen enrichment on BF state is analyzed through the ironmaking mechanism,and three mechanism chains of oxygen enrichment affecting GUR are ***,the GUR data is decomposed to obtain the multi-time-scale components and the frequency distribution of each *** on the above analysis,the multi-time scale relationship between the three mechanism chains and the GUR components is ***,a multi-time-scale prediction model of oxygen enrichment on GUR based on BP neural network is established by combining the relationship between the mechanism chain and GUR.

关键词： Blast Furnace Oxygen Enrichment Gas Utilization Rate Multi-Time-Scale Analysis Prediction Model

来源：评论

学校读者我要写书评

暂无评论

A pulse position coding-based mud pulse signal identification algorithm for Measurement While Drilling(MWD) system

A pulse position coding-based mud pulse signal identificatio...

引用

第32届中国过程控制会议(CPCC2021)

作者： Ziliang Yu Weihua Cao Chao Gan Min Wu School of Automation China University of Geosciences Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education

Accurate identification of mud pulse signal is crucial for Measurement While Drilling(MWD) system due to its vital role in improving the drilling safety and *** this paper,a pulse position coding-based mud pulse signal identification algorithm is proposed for MWD system via two *** the signal preprocessing stage,wavelet filtering is introduced to reduce the noises in the raw mud pulse ***,a polynomial fitting-based detection method is used to remove the baseline drift in the *** the signal identification stage,a pulse signal position identification model is established to detect the pulse position,which does not need to set the detection threshold *** comparison results demonstrate that the proposed method has higher identification efficiency and accuracy than the conventional methods.

关键词： Mud pulse signal Wavelet analysis Polynomial fitting Pulse position identification model

来源：评论

学校读者我要写书评

暂无评论

Data Room Management System Based on Target Detection Algorithm

Data Room Management System Based on Target Detection Algori...

引用

第40届中国控制会议

作者： Zhicheng Zhao Weihua Cao Kang Li Yitao Chen School of Automation China University of Geosciences Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education

In recent years,streaming media services have been widely used in data room management *** order to record the user operations,host machine alarms and reminder in real time,target detection algorithms for streaming media have played an important *** target detection methods,limited by detection speed,are not suitable for such real-time ***,in this paper,we applies SSD（Single Shot MultiBox Detector） algorithm to data room management ***,we constructs a host operating image data set,and then perform data amplification by adding noise and ***,the SSD model is used to train a ***,after obtaining the target detection results,the NMS（Non-Maximum Suppression）algorithm is used to avoid the redundant detection *** the same time,some improvement measures are put forward to solve the problem of small target,leading to the difficulty of feature extraction and low detection *** experimental results demonstrate that the model can detect and track the target better in the video and meet the requirements of the real-time performance and accuracy of the system.

关键词： target detection data room management system streaming media SSD

来源：评论

学校读者我要写书评

暂无评论

Parallel control for Continuous-Time Linear systems:A Case Study

引用

IEEE/CAA Journal of Automatica Sinica 2020年第4期7卷 919-928页

作者： Qinglai Wei Hongyang Li Fei-Yue Wang the State Key Laboratory of Management and Control for Complex Systems Institute of AutomationChinese Academy of SciencesBeijing 100190 the University of Chinese Academy of Sciences Beijing 100049 Qingdao Academy of Intelligent Industries Qingdao 266109China the Institute of Systems Engineering Macao University of Science and Technology

In this paper,a new parallel controller is developed for continuous-time linear *** main contribution of the method is to establish a new parallel control law,where both state and control are considered as the *** structure of the parallel control is provided,and the relationship between the parallel control and traditional feedback controls is *** the situations that the systems are controllable and incompletely controllable,the properties of the parallel control law are *** parallel controller design algorithms are given under the conditions that the systems are controllable and incompletely ***,numerical simulations are carried out to demonstrate the effectiveness and applicability of the present *** Terms-Continuous-time linear systems,digital twin,parallel controller,parallel intelligence,parallel systems.

关键词： Index Terms—Continuous-time linear systems digital twin parallel controller parallel intelligence parallel systems

来源：评论

学校读者我要写书评

暂无评论

Landslide Sensitivity Mapping Based on One-Dimensional Residual Convolutional Neural Networks

Landslide Sensitivity Mapping Based on One-Dimensional Resid...

引用

Chinese automation Congress (CAC)

作者： Yougang Ke Chengda Lu Min Wu Wanke Yu Ruiqing Niu School of Automation China University of Geosciences Wuhan China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan China Ministry of Education Engineering Research Center of Intelligent Technology for Geo-Exploration Wuhan China Institute of Geophysics and Geomatics China University of Geosciences Wuhan China

Landslide is a common geological disaster. Landslide sensitivity mapping (LSM) is the key technology for landslide monitoring, early warning and risk assessment. Deep learning shows good performance in feature extraction. This paper proposes a one-dimensional residual convolution neural network (1DRCNN), which takes Yunyang County, Chongqing City, China, located in the Three Gorges Reservoir area as the research area. Extracting 12 evaluation factors from multi-source remote sensing data and building a training set. The spatial probability of landslide occurrence is quantitatively predicted by the proposed model, and finally, the landslide sensitivity mapping is generated. Compared with the common machine learning models SVM and logistic regression, the results show that the AUC (area under curve) and accuracy of 1D-RCNN are 0.9860 and 93.38%, respectively, which proving that this method is effective and can provide a reference for disaster prevention and reduction.

关键词： Support vector machines Training Sensitivity Urban areas Feature extraction Reservoirs Terrain factors

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：