检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Li, Dongdong Dong, Jiuxiang College of Information Science and Engineering Northeastern University Shenyang110819 China State Key Laboratory of Synthetical Automation of Process Industries Northeastern University Shenyang110819 China Key Laboratory of Vibration and Control of Aero-Propulsion Systems Ministry of Education Northeastern University Shenyang110819 China

Policy iteration is one of the classical frameworks of reinforcement learning, which requires a known initial stabilizing control. However, finding the initial stabilizing control depends on the known system model. To relax this requirement and achieve model-free optimal control, in this paper, two different reinforcement learning algorithms based on policy iteration and variable damping coefficients are designed for unknown discrete-time linear systems. First, a stable artificial system is designed, and this system is gradually iterated to the original system by varying the damping coefficients. This allows the initial stabilizing control to be obtained in a finite number of iteration steps. Then, an off-policy iteration algorithm and an off-policy Q-learning algorithm are designed to select the appropriate damping coefficients and realize data-driven. In these two algorithms, the current estimates of optimal control gain are not applied to the system to re-collect data. Moreover, they are characterized by the fast convergence of the traditional policy iteration. Finally, the proposed algorithms are validated by simulation. Copyright © 2024, The Authors. All rights reserved.

关键词： Damping

来源：评论

学校读者我要写书评

暂无评论

Artificial Nose of Scalable Plasmonic Array Gas Sensor for Multi-Dimensional Sers Recognition of Volatile Organic Compounds

SSRN

引用

SSRN 2023年

作者： Qu, Cheng Fang, Hao Yu, Fanfan Chen, Jinai Su, Mengke Liu, Honglin China Light Industry Key Laboratory of Meat Microbial Control and Utilization School of Food and Biological Engineering Engineering Research Center of Bio-process Ministry of Education Hefei University of Technology Hefei230009 China

Gas sensing is one kind of promising analytical technique for high-flux screening and evaluation. Volatile organic compounds (VOCs) analysis by surface-enhanced Raman spectroscopy (SERS) is arising but usually focuses on a single species of VOCs due to the limitations of complexity, stability, and accuracy in practical applications. Canine animals have a sensitive sense of smell due to their large number of olfactory cells. Inspired by this, here we innovatively integrate scalable plasmonic arrays and multi-dimensional chemometrics for simultaneous SERS detection of multiple food-borne VOCs. Both direct and indirect SERS strategies are integrated by coating the array surface with MOFs or different monolayers to provide multi-dimensional recognition and readouts. Four species of VOCs including bacterial metabolites, hydrogen sulfide, aldehyde, and biogenic amine were selectively captured and produced unique SERS spectral profiles. The multi-dimensional outputs greatly increase the dimensionality of VOCs fingerprints, thereby significantly improve the sensitivity, reliability, and accuracy for freshness discrimination and forecasting in real food smell evaluation. By virtue of multi-dimensional SERS recognition and machine learning, the array gas sensor shows higher accuracy than traditional SERS analysis and is a promising platform for in situ, real-time and high-flux gas sensing and inspection. © 2023, The Authors. All rights reserved.

关键词： Volatile organic compounds

来源：评论

学校读者我要写书评

暂无评论

Catalytic enhancement of MOF-derived Fe@C catalysts in microwave-assisted co-pyrolysis: synergistic integration with bamboo-based pyrolytic char and microwave carbonization

引用

Energy 2025年 329卷

作者： Li, Pan Chen, Yasen Chen, Hai Chen, Wei Hu, Junhao Chen, Jiaheng Wang, Peiping Chang, Chun Pang, Shusheng School of Mechanical and Power Engineering State Key Laboratory of Biobased Transport Fuel Technology Zhengzhou University Zhengzhou China Key Laboratory of Energy Thermal Conversion and Control of Ministry of Education Southeast University Nanjing China Hunan Tai Tong Energy Management Co. LTD Changsha China Henan Key Laboratory of Green Manufacturing of Biobased Chemicals Puyang China Department of Chemical and Process Engineering University of Canterbury Christchurch New Zealand

Developing efficient catalysts for microwave-assisted co-pyrolysis of biomass and plastics (BP) is crucial for producing high-value products. This study introduces the design and synthesis of an enhanced Fe@C catalyst using MIL-101(Fe) as the precursor. During precursor preparation, different amounts of bamboo-based pyrolytic char (BC) were incorporated, and both conventional and advanced microwave carbonization techniques were employed to produce the Fe@C catalysts. Simultaneously, various characterization techniques were employed to analyze the physicochemical properties of the catalysts. The catalytic performance of the catalyst was evaluated in the microwave-assisted co-pyrolysis of bamboo and polyethylene, focusing on the distribution of pyrolysis gases and bio-oil products. The results showed that incorporating BC and applying microwave carbonization significantly enhanced the catalyst's selectivity for H2 and aromatic hydrocarbons. Notably, the Fe@C-8BC catalyst achieved the highest H2 content (55.65 vol%) and aromatic hydrocarbon content (28.07 %), with MAHs accounting for 51.8 % of total aromatics. Furthermore, the microwave-carbonized Fe@C catalyst exhibited superior BTEX selectivity, with a BTEX/aromatic ratio of 32.2 %. This study highlights the benefits of integrating BC and advanced microwave carbonization techniques in the preparation of MOF-derived catalysts, providing valuable insights for optimizing co-pyrolysis catalysts and advancing the synthesis of high-value products from waste materials. © 2025

关键词： Aromatization

来源：评论

学校读者我要写书评

暂无评论

Dynamic Iterative Learning control for Relative Trajectory Tracking with Nonrepetitive Leader Path

Dynamic Iterative Learning Control for Relative Trajectory T...

引用

Data Driven control and Learning Systems (DDCLS)

作者： Yiming Wang Guina Wang Hongfeng Tao Yiyang Chen School of Mechanical and Electrical Engineering Soochow University Suzhou P. R. China Key Laboratory of Advanced Process Control for Light Industry of Ministry of Education Jiangnan University Wuxi P. R. China

ISBN: (数字)9798350361674

ISBN: (纸本)9798350361681

This paper proposes a collaborative iterative learning control scheme with optimal leader path planning strategy. Due to the demonds on repetitive tasks and static references under traditional ILC framework, it cannot fit the complex circumstances with multiple collaborative systems or satisfy certain control request on specific application. This paper introduces a dynamic norm-optimal iterative learning control (NOILC) framework based on an improved coordinate descent algorithm. Under the aforementioned framework, an optimal leader path can be found to fit certain performance requirements, i.e. energy reduction. Finally, the simulation results on a combined complex system verify the effectiveness of the proposed algorithm.

关键词： Learning systems Trajectory tracking Heuristic algorithms Simulation Collaboration control systems Task analysis

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

A Method for Constructing Optical Detection Model of Wheat Seed Purity Based on Sample Generation and Contrast Learning Strategy

SSRN

引用

SSRN 2024年

作者： Yan, Shengqi Zhao, Xin Zhu, Qibing Huang, Min Guo, Xinnian Key Laboratory of Advanced Process Control for Light Industry Ministry of Education Jiangnan University Wuxi214122 China Jiangsu Province Engineering Research Center of Smart Poultry Farming and Intelligent Equipment Suqian University Suqian223800 China

Seed purity is an essential indicator of seed quality, and use of optical techniques for seed purity testing has been extensively reported. Currently, most optical detection models for seed purity rely on the assumptions that the training set used to develop detection model contains all possible impurity varieties, and that the number of positive(the variety of interest) and negative samples(all possible impurities contained in that variety) in the training set is balanced. This assumption is often difficult to fulfil in practice due to the complexity of the seed production process. In order to solve the problem of unbalanced samples in the construction of the optical detection model and the problem of detection of unknown impurities (categories of impurities not contained in the training set) faced by the model in practical applications, an optical detection model construction method for wheat seed purity based on sample generation and contrast learning strategy was proposed in this study. Firstly, a linear data generation strategy is used to expand the dataset of impurity samples, enhance the diversity of impurity samples and improve the imbalance between the positive and negative samples, so as to improve the model's ability to identify impurity seeds. At the same time, a contrast learning strategy is introduced to strengthen the feature differences between the positive and negative samples and improve the model's recognition accuracy for unknown impurity samples. Hyperspectral images of six varieties with a total of 4200 wheat seeds were acquired, and a purity detection model for each variety was built based on the average spectra of the seeds. The purity detection model for each variety was constructed through four varieties (containing the variety itself as the positive sample, and three of the remaining five varieties as a small number of negative samples), while the test sample consisted of all six variety samples. The results showed that the recognition acc

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Weakly supervised object detection with interactive edge attentive collaboration

Weakly supervised object detection with interactive edge att...

引用

第34届中国控制与决策会议

作者： Wenlong Gao Ying Chen Yong Peng Key Laboratory of Advanced Process Control for Light Industry Ministry of EducationJiangnan University School of Internet of Things Engineering Jiangnan University

Weakly supervised object detectors based on image-level annotation tend to overfit in the discriminative regions while ignoring the integrity of the *** this paper,a novel weakly supervised object detection network with the interactive edge attentive collaboration module is proposed to alleviate the local optimal problem,in which edge attention is extracted as an object unity supervision for the detection,and a collaborative loss is introduced to enable VGG16 feature map with global attentive *** module can be detached from the network in the test period,which ensures the high efficiency of the detector without introducing any additional inference *** experiments are carried out on the PASCAL VOC 2007 and VOC 2012 datasets,which reach 52.3% mAP,67.7% CorLoc and 49.1%mAP,68.0% CorLoc respectively,outperforming state-of-the-arts.

关键词： Object Detection Weakly supervised Learning Edge Feature Image processing

来源：评论

学校读者我要写书评

暂无评论

Catalytic Performance of Au-Ti Active Site on Mww Zeolite for Oxidative Cyclohexane to Cyclohexanol/One in Oxysteam Condition

SSRN

引用

SSRN 2024年

作者： Jin, Fang Liu, Sijia Wu, Guiying Shao, Xue Yang, Yuqi Zhan, Xin Hubei Novel Reactor & Green Chemical Technology Key Laboratory Key Laboratory for Green Chemical Process of Ministry of Education School of Chemical Engineering and Pharmacy Wuhan Institute of Technology Hubei Wuhan430205 China Hubei Key Laboratory of Mine Environmental Pollution Control & Remediation Hubei Polytechnic University China Wuhuan Engineering Co. LTD China

A mixture of cyclohexanol and cyclohexanone (KA oil) is the main raw material for caprolactam and adipic acid. The Ti and Au was sequentially introduced into the deboronated MWW zeolite by atom planting (AP) method and deposition-precipitation method as the catalyst for cyclohexane oxidation by molecular oxygen with higher selectivity of KA oil. The tetra-coordinated Ti species introduced by AP method in the zeolite framework can highly disperse the Au nano-particles as an active site for the cyclohexane oxidation. The flexible MWW zeolite layer structure facilities modification the pore structure of ERB-1 zeolite by introduction of mesopore and exposed the more silanol group on the external surface of the zeolite the exposed pore size structure can improve the cyclohexane conversion and cyclohexanol selectivity. The optimal Au nano-particle size for oxidation is around 2 nm. Moreover, the cyclohexane oxidation conditions can be optimised by varying the oxygen pressure and solvent polarity. © 2024, The Authors. All rights reserved.

关键词： Nanoparticles

来源：评论

学校读者我要写书评

暂无评论

Extended NNs-Based Adaptive Event-Triggered Asymptotic control for Nonlinear Systems Against Actuator Faults

Extended NNs-Based Adaptive Event-Triggered Asymptotic Contr...

引用

International Symposium on Autonomous Systems (ISAS)

作者： Shanlin Liu Yingwei Zhang Xudong Zhao Hesheng Wang College of Information Science and Engineering Northeastern University Shenyang China State Laboratory of Synthesis Automation of Process Industry Northeastern University Shenyang China Faculty of Electronic Information and Electrical Engineering Dalian University of Technology Dalian China Department of Automation Key Laboratory of System Control and Information Processing of Ministry of Education Key Laboratory of Marine Intelligent Equipment and System of Ministry of Education Shanghai Engineering Research Center of IntelligentControl and Management Shanghai Jiao Tong University Shanghai China

ISBN: (数字)9798350363173

ISBN: (纸本)9798350363180

In this article, an adaptive event-triggered asymp¬totically fault-tolerant control (FTC) issue for nonlinear systems with actuator faults is investigated. An extended neural networks (NNs) technique is introduced to estimate unknown functions of the system under consideration. Compared with traditional NNs that ignore approximation errors, the proposed extended NNs further processes approximation errors, reducing the conservatism of the proposed algorithm. Then, within the framework of backstepping method, a new adaptive control strategy is developed by combining event-triggered mechanism (ETM) and bound estimation (BE) approach, which ensures the asymptotic convergence of system states while also reducing the transmission of unnecessary information, thus improving the utilization of communication resources. At the same time, the designed controller eliminates the adverse effects of actuator failures and system nonlinear terms. Finally, a simulation example of a single-link robot system is provided to demonstrate that the developed control strategy can ensure that all signals are bounded and system states are asymptotically convergent.

关键词： Actuators Backstepping Fault tolerant systems process control Artificial neural networks control systems Approximation error

来源：评论

学校读者我要写书评

暂无评论

Optimal Iterative Learning control of Quantized Signals Based on Encoding-Decoding Method 10

Optimal Iterative Learning Control of Quantized Signals Base...

引用

10th IEEE Data Driven control and Learning Systems Conference, DDCLS 2021

作者： Huang, Yande Zhuang, Zhihe Tao, Hongfeng Chen, Yiyang Key Laboratory of Advanced Process Control for Light Industry of Ministry of Education Jiangnan University Wuxi214122 China School of Mechanical and Electrical Engineering Soochow University Suzhou215137 China

ISBN: (纸本)9781665424233

This paper considers optimal iterative learning control (ILC) to solve the trajectory tracking problem of systems using a network to transmit signals. Assuming that the relative quantization error of the logarithmic quantizer is white noise uniformly distributed in a known range, a cost function in the sense of mathematical expectation is established to deal with the relative quantization error, so as to obtain an iterative update law. By combining the logarithmic quantizer with the encoding and decoding method, the system can iteratively achieve high-precision tracking with limited communication capacity. Simulations based on an electro-mechanical control system model is provided to validate the proposed method. © 2021 IEEE.

关键词： Cost functions

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：