检索结果-内蒙古大学图书馆

IEEE Transactions on Artificial Intelligence 2021年第6期2卷 594-607页

作者： Dai, Wei Zhang, Rui Wang, Qianjin Cheng, Yuhu Wang, Xuesong The Engineering Research Center of Intelligent Control for Underground Space Ministry of Education China University of Mining and Technology Xuzhou221116 China The School of Information and Control Engineering China University of Mining and Technology Xuzhou221116 China The State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Shenyang110819 China The School of Electrical Engineering Yancheng Institute of Technology Yancheng221051 China

Recently, a novel learning algorithm called the broad learning system (BLS) has been established by using the theory of pseudoinverse and compressed sensing technology. As an alternative to deep learning, BLS has gained extensive attention in data analytics. Unlike the existing deep structures based learning tools (such as deep neural networks), it is able to rapidly achieve incremental learning, and does not require tedious retraining to remodel the system. However, given image data modeling tasks, 1-D BLS requires a conventional vectorization operation, the dimension of input will be quite large for high resolution image. This operation may result in huge computational complexity, thereby increasing the time of model construction. To resolve this issue, a 2-D version of BLS, namely 2-D-BLS, is first proposed to fast construct randomized learner models by using matrix inputs. Specifically, the proposed method employs left and right projecting vectors to replace the usual high dimensional input weight. Moreover, the theoretical analyses on the superiority of 2-D-BLS against BLS are presented in terms of the algorithm complexity and the difference of random parameter space. Empirical results on MNIST dataset, NORB 3-D object image recognition dataset, ORL dataset, YaleB dataset, and handwritten digit dataset demonstrate that the proposed 2-D-BLS outperforms the existing extensions of BLS and state-of-the-art 2-D machine learning methods in terms of the learning efficiency, showing good potential of 2-D-BLS for image data analytics. Impact Statement-as a novel learning tool, BLS has achieved great success in many fields. Image recognition is one of the most widely used fields of artificial intelligence. But, BLS is a 1-D model, the vectorization operation is necessary to convert matrix input data into a row or a column vector. This operation will lead to "explosion of dimensionality" and increase the time of model construction when it is coming to image data analytic

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Analysis of Operational Data and Self-learning control Rules for Mining Mill Equipment

Analysis of Operational Data and Self-learning Control Rules...

引用

Asian control Conference

作者： Guicheng Wang Liang Yuan Zhicheng Bai Guoliang Cai Ning Hua Xiaobin Li School of Electrical and Electronics Engineering Shanghai Institute of Technology Shanghai China State Key Laboratory of Intelligent Optimized Manufacturing in Mining & Metallurgy Process and Beijing Key Laboratory of Process Automation in Mining & Metallurgy Beijing China Zijin Intelligent Control (Xiamen) Technology Co. Ltd. Xiamen China

ISBN: (数字)9789887581598

ISBN: (纸本)9798331540845

The engineering application under industrial big-data analysis is of great significance for improving the optimization effect of enterprise production process control processes. By analyzing the operating principle of the semi-automatic mill and studying its control optimization strategy, based on the operational data of the SAG mill, industrial big-data analysis is adopted to self-learning its control optimization rules. Firstly, SPSS is used to preprocess the high-dimensional process data generated during the operation of the semi-automatic mill. Four parameters, namely power, axial pressure, grinding sound, and grinding concentration, are preliminarily selected. Then, PCA is used to screen and reduce the dimensionality of multidimensional data, and K-Means clustering algorithm is used to classify and identify the reduced dataset. After simulation research, the results show that PCA can reduce the redundancy of SAG mill operation data, and the control optimization rules formed by self-learning can serve as operational guidance for improving the efficiency and energy-saving of SAG mill operation. The effectiveness can be further verified in practical engineering applications in the later stage.

关键词： Dimensionality reduction Energy consumption Technological innovation Redundancy Force process control Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

Deformation Pattern Shape Sensing for Visuotactile Endoscope

Deformation Pattern Shape Sensing for Visuotactile Endoscope

引用

2024 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2024

作者： Yu, Yanzhe Tian, Yuan Cui, Shaowei Wang, Yupeng Wang, Rui Wang, Shuo Beijing University of Science and Technology Beijing100083 China The University of Sheffield Department of Automatic Control and Systems Engineering Sheffield United Kingdom Institute of Automation Chinese Academy of Sciences State Key Laboratory of Multimodal Artificial Intelligence Systems Beijing100190 China University of Chinese Academy of Sciences School of Artificial Intelligence Beijing100049 China Center for Excellence in Brain Science and Intelligence Technology Chinese Academy of Sciences Shanghai200031 China

ISBN: (纸本)9798350372601

The integration of visuotactile sensor technology with endoscopy enables the provision of tactile information for minimally invasive surgeries and other operations. This paper proposes a real-time sensing framework for contact deformation pattern shape, providing a foundation for precise perception in miniaturized visuotactile sensors. Specifically, we first construct a Deformation Pattern shape Descriptor (DPSD) and then track the key points in the descriptor using optical flow to obtain contact deformation shape information. Through extensive qualitative and quantitative experiments, our proposed algorithm achieves an average perception accuracy of around 83.81% for contact pattern shapes. It demonstrates consistent perception accuracy across various contact force magnitudes, directions, and scenarios, highlighting its robustness. Additionally, compared to traditional marker-level methods, it achieves a notable 46.5% improvement in displacement field perception accuracy. This research elevates tactile perception to the dimension of contact pattern shape sensing and has the potential for application in miniaturized visual tactile sensors. © 2024 IEEE.

关键词： Endoscopy

来源：评论

学校读者我要写书评

暂无评论

A Koopman Operator-Based Finite Impulse Response Filter for Nonlinear Systems

A Koopman Operator-Based Finite Impulse Response Filter for ...

引用

IEEE Conference on Decision and control

作者： Zhichao Pan Biao Huang Fei Liu Key Laboratory of Advanced Process Control for Light Industry (Ministry of Education) Institute of Automation Jiangnan University Wuxi China department of chemical and materials engineering University of Alberta Edmonton Alberta Canada

This paper proposes a novel Koopman operator-based finite impulse response (KFIR) filter for nonlinear dynamic systems. This filter is generalized from the minimum variance unbiased (MVU) FIR filter for linear systems by using a global linear approximation of the nonlinear dynamics obtained from Koopman operator theory and the extended dynamic mode decomposition (EDMD) algorithm. Based on the recursive linear model, a reduced-order FIR filtering structure is proposed, and the optimal gain is derived to minimize the trace of the estimation error covariance. Unlike traditional methods, the KFIR filter requires no prior knowledge of the initial state and fully utilizes the data of a moving horizon. Simulation results show that the proposed filter has excellent robustness against unexpected modeling uncertainties and inaccurate noise information, making it suitable for real applications.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Inspection trolley based on the distance detection obstacle

Inspection trolley based on the distance detection obstacle

引用

Chinese control and Decision Conference, CCDC

作者： Guicheng Wang Ao Sheng Xiaobin Li Min Zhang Changliang Guan Fengfeng Yin School of Electrical and Electronics Engineering Shanghai Institute of Technology Shanghai China Library Shanghai Institute of Technology Shanghai China Zijin Intelligent Control (Xiamen) Technology Co. Ltd. Xiamen China State Key Laboratory of Process Automation in Mining & Metallurgy and Beijing Key Laboratory of Process Automation in Mining & Metallurgy Beijing China

In Industry 4.0 background, unmanned full-process automated production has gradually become a hot topic, and the realization of intelligent and information factory requires the joint cooperation of multiple industries, in which the inspection device that replaces traditional manual work is an inescapable topic. At present, domestic and foreign research on intelligent inspection trolley is focused on obstacle avoidance function based on various sensors. In this context, it is introduced an algorithm based on monocular camera as a sensor, using Python language to access OpenCV library, and utilizing Gaussian filter function and Canny operator-based image contour extraction to complete the distance measurement. The algorithm can achieve the distance measurement and obstacle avoidance function at a lower cost. The test results show that the system can automatically identify the distance of obstacles in the inspection process, and according to the visual information captured by the camera in real time, it can independently generate steering action and collision reminder, and can complete the preliminary requirement of intelligent inspection.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Learning-based Adaptive Compliance Method for Symmetric Bi-manual Manipulation

arXiv

引用

arXiv 2023年

作者： Cao, Yuxue Zhao, Wenbo Wang, Shengjie Zheng, Xiang Ma, Wenke Wang, Zhaolei Zhang, Tao Beijing Institute of Control Engineering China Department of Automation Tsinghua University China Institute for Interdisciplinary Information Sciences Tsinghua University China Department of Computer Science City University of Hong Kong Hong Kong Qian Xuesen Laboratory of Space Technology China Beijing Aerospace Automatic Control Institute China

Symmetric bi-manual manipulation is an essential skill in on-orbit operationsdue to its potent load capacity. Previous works have applied compliant controlto maintain the stability of manipulations. However, traditional methods haveviewed motion planning and compliant control as two separate modules, which canlead to conflicts with the simultaneous change of the desired trajectory andimpedance parameters in the presence of external forces and ***, the joint usage of these two modules requires experts to manuallyadjust parameters. To achieve high efficiency while enhancing adaptability, wepropose a novel Learning-based Adaptive Compliance algorithm (LAC) thatimproves the efficiency and robustness of symmetric bi-manual ***, the algorithm framework integrates desired trajectory generationand impedance-parameter adjustment under a unified framework to mitigatecontradictions and improve efficiency. Second, we introduce a centralizedActor-Critic framework with LSTM networks preprocessing the force states,enhancing the synchronization of bi-manual manipulation. When evaluated indual-arm peg-in-hole assembly experiments, our method outperforms baselinealgorithms in terms of optimality and robustness. Copyright © 2023, The Authors. All rights reserved.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

A Norm Optimal Iterative Learning control Approach With Efficient Computation

A Norm Optimal Iterative Learning Control Approach With Effi...

引用

Data Driven control and Learning Systems (DDCLS)

作者： Luyuan Gao Zhihe Zhuang Hongfeng Tao Yiyang Chen Wojciech Paszke Key Laboratory of Advanced Process Control for Light Industry of Ministry of Education Jiangnan University Wuxi PR China School of Mechanical and Electrical Engineering Soochow University Suzhou PR China Institute of Automation University of Zielona Gora Zielona Gora Poland

ISBN: (数字)9798350361674

ISBN: (纸本)9798350361681

Iterative learning control (ILC) is excellently applied to output tracking for systems that perform repetitive tasks. Classical norm optimal ILC (NOILC) approach yields optimal control inputs by minimizing a quadratic cost function. However, this approach requires the introduction of supervectors, i.e., lifting technique, for system description, which raises a high computational load problem. The contribution of this paper is to propose a non-lifted NOILC (N-NOILC) approach by redesigning a cost function that does not involve the lifting technique, achieving the goal of efficient computation. Therefore, the proposed approach is particularly suitable for large tracking tasks with long trial duration and high sampling frequency. Moreover, considering the model uncertainty in practice, robust convergence is analyzed. Numerical simulations are shown to demonstrate the properties and advantages of the proposed approach.

关键词： Uncertain systems Uncertainty Optimal control Cost function Numerical simulation Computational efficiency Numerical models

来源：评论

学校读者我要写书评

暂无评论

Demand Response Optimization for the Enhancement of the Distribution System's Operation

Demand Response Optimization for the Enhancement of the Dist...

引用

International Conference on Smart Energy Systems and Technologies (SEST)

作者： Maria Fotopoulou George Tsekouras Dimitrios Rakopoulos Vassiliki Kontargyri Department of Electrical and Electronics Engineering University of West Attica Athens Greece Chemical Process and Energy Resources Institute Centre for Research and Technology Hellas Athens Greece Control Systems and Industrial Automation Laboratory National and Kapodistrian University of Athens Athens Greece

ISBN: (数字)9798350386493

ISBN: (纸本)9798350386509

Following the latest energy market trends, active distribution systems incorporate an ascending mix of flexible loads, which can participate in Demand Response (DR) strategies. The purpose of this research paper is to present a DR strategy for the Distribution System Operator (DSO) in order to enhance the system's operation and mitigate power quality issues, such as voltage drops, and congestion. The proposed algorithm aims at smoothing the load curve, taking into consideration the available flexible load and the limitations of the system. The algorithm is demonstrated on a model of the distribution system of Skiathos island, Greece, in the context of ENFLATE, a Horizon Europe project, and future scenarios with photovoltaic (PV) penetration are studied as well. According to the results, the proposed DR algorithm manages to reduce the voltage drop, eliminate the number of events where the voltage drop exceeds the limits and reduce the losses of the distribution system up to 8%-9%, depending on the PV penetration.

关键词： Electric potential Smoothing methods Power quality Europe Voltage Artificial neural networks Transformers Demand response Load modeling Context modeling

来源：评论

学校读者我要写书评

暂无评论

Augmented Yield Analysis for Refined Elementary Mode Set of Penicillin Metabolic Network

Augmented Yield Analysis for Refined Elementary Mode Set of ...

引用

第33届中国过程控制会议

作者： Chenxi Tao Quan Li Zhonggai Zhao Fei Liu School of Internet of Things Engineering Jiangnan University Key Laboratory of Advanced Control of Light Industry Process (Ministry of Education) Institute of AutomationJiangnan University

Penicillin is usually produced by microbial growth and metabolism in a fermentation *** analysis and understanding of metabolic behavior are the keys to improving penicillin *** refined elementary model set(REMS) is a subset of the Elementary Models(EMs) necessary to describe specific metabolic *** plays an important role in mechanistic modeling and optimization of penicillin fermentation *** conventional yield analysis(YA) method can easily derive REMS using yield information,but the metabolic behavior described by the results may seriously deviate from *** this paper,we propose an augmented yield analysis(AYA) method to address this ***,basic metabolic pathways and metabolites characterizing their activities are identified;second,the production fluxes of the identified metabolites are combined with yield information to form an augmented yield space,into which the EMS is recoded;finally,the REMS is formed by a minimal set of EMs which can meet the preset requirements for metabolic *** REMS of the penicillin metabolic network is determined by AYA,and it can describe the actual metabolic behavior of penicillin,thereby verifying the effectiveness of AYA.

关键词： Penicillin metabolic network elementary mode analysis refined elementary mode set metabolic information augmented yield analysis

来源：评论

学校读者我要写书评

暂无评论

Enhanced Soft Sensor with Qualified Augmented Data Using Centroid Measurement Criterion

Enhanced Soft Sensor with Qualified Augmented Data Using Cen...

引用

2021 International Conference on Information, Cybernetics, and Computational Social Systems, ICCSS 2021

作者： Dai, Yun Yu, Qing Yang, Tao Yao, Yuan Liu, Yi Zhejiang University of Technology Institute of Process Equipment and Control Engineering Hangzhou310023 China Northeastern University State Key Laboratory of Synthetical Automation for Process Industries Shenyang110819 China National Tsing Hua University Department of Chemical Engineering Hsinchu30013 Taiwan

ISBN: (纸本)9781665402453

Development of reliable soft sensors using limited labeled samples is not an easy task in industrial processes. A selective generative adversarial network (SGAN)-based support vector regression (SGAN-SVR) soft sensor is proposed for quality prediction using limited labeled training data. Specifically, SVR is considered as a base prediction model. The Wasserstein GAN (WGAN) is adopted to capture the distribution of available labeled data and generate virtual candidates. Subsequently, using a proposed similarity measurement strategy, those synthetic data with more information are selected and introduced into the training set. Using the designed data augmentation approach, the SGAN-SVR model can achieve better prediction performance compared with the SVR soft sensor. The quality prediction results on an industrial polyethylene process demonstrate the effectiveness and advantages of the proposed method. © 2021 IEEE.

关键词： Generative adversarial 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：