检索结果-内蒙古大学图书馆

A fuzzy compensation-Koopman model predictive control design for pressure regulation in proten exchange membrane electrolyzer

引用

Chinese Journal of Chemical Engineering 2024年第12期76卷 251-263页

作者： Haokun Xiong Lei Xie Cheng Hu Hongye Su State Key Laboratory of Industrial Control Technology Zhejiang UniversityHangzhou 310027China

Proton exchange membrane(PEM)electrolyzer have attracted increasing attention from the industrial and researchers in recent years due to its excellent hydrogen production *** accurate models to predict their performance is crucial for promoting and accelerating the design and optimization of electrolysis *** work developed a Koopman model predictive control(MPC)method incorporating fuzzy compensation for regulating the anode and cathode pressures in a PEM electrolyzer.A PEM electrolyzer is then built to study pressure control and provide experimental data for the identification of the Koopman linear *** identified linear predictors are used to design the Koopman *** addition,the developed fuzzy compensator can effectively solve the Koopman MPC model mismatch *** effectiveness of the proposed method is verified through the hydrogen production process in PEM simulation.

关键词： Hydrogen production PEM electrolyzer Nonlinear control Model predictive control Koopman operator Fuzzy logic system

来源：评论

学校读者我要写书评

暂无评论

A Decoder-Free Reconstruction Method for Semi-Supervised Rail Surface Defect Detection

IEEE Transactions on Industrial Cyber-Physical Systems

引用

IEEE Transactions on industrial Cyber-Physical Systems 2025年 3卷 285-295页

作者： Liu, Chen Shi, Zhenyu He, Shibo Tang, Shunpu Yang, Qianqian The State Key Lab. of Industrial Control Technology Zhejiang University Hangzhou310027 China

Detecting defects on railway tracks is critical for the operation of high-speed trains. Despite a plethora of machine vision-based methods designed to tackle this problem, the majority adopt a supervised setting and demand considerable labeled training data, inclusive of defective samples, which is expensive and impractical. In this paper, we propose an Invertible Reconstruction neural Network (IRNet) for semi-supervised rail surface defect detection, where only normal images are accessible during training. Firstly, we devise an information-preserving feature encoder comprising several invertible blocks. This structure safeguards subtle visual patterns distinguishing normal and defective images from being obscured by background information, guaranteed by its mathematical reversibility property. Second, to overcome the overgeneralization issue of conventional autoencoders caused by imperfectly crafted decoders, we propose a novel decoder-free reconstruction workflow based on the invertible feature encoder. Specifically, we force one portion of extracted features to approach a predefined constant tensor during the training stage by minimizing their mean squared error. Next, we feed the remained features and the predefined constant tensor backward into the encoder to reconstruct the original images. During the testing phase, we formulate an anomaly score that consolidates the reconstruction error and mean squared error to spot defects. Extensive experiments are conducted on 4 real-world datasets. Our method consistently outperforms state-of-the-art techniques, demonstrating an average increase of 8.5% on the F1 score. © 2023 IEEE. All rights reserved.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

From Static and Dynamic Perspectives:A Survey on Historical Data Benchmarks of control Performance Monitoring

引用

IEEE/CAA Journal of Automatica Sinica 2025年第2期12卷 300-316页

作者： Pengyu Song Jie Wang Chunhui Zhao Biao Huang the State Key Laboratory of Industrial Control Technology College of Control Science and Engineering Zhejiang University IEEE the Department of Chemical and Materials Engineering University of Alberta

In recent decades, control performance monitoring(CPM) has experienced remarkable progress in research and industrial applications. While CPM research has been investigated using various benchmarks, the historical data benchmark(HIS) has garnered the most attention due to its practicality and effectiveness. However, existing CPM reviews usually focus on the theoretical benchmark, and there is a lack of an in-depth review that thoroughly explores HIS-based methods. In this article, a comprehensive overview of HIS-based CPM is provided. First, we provide a novel static-dynamic perspective on data-level manifestations of control performance underlying typical controller capacities including regulation and servo: static and dynamic properties. The static property portrays time-independent variability in system output, and the dynamic property describes temporal behavior driven by closed-loop feedback. Accordingly,existing HIS-based CPM approaches and their intrinsic motivations are classified and analyzed from these two ***, two mainstream solutions for CPM methods are summarized, including static analysis and dynamic analysis,which match data-driven techniques with actual controlling behavior. Furthermore, this paper also points out various opportunities and challenges faced in CPM for modern industry and provides promising directions in the context of artificial intelligence for inspiring future research.

关键词： control performance monitoring (CPM) data-driven method historical data benchmark (HIS) industrial process performance index static and dynamic analysis

来源：评论

学校读者我要写书评

暂无评论

Toward Universal controller: Performance-Aware Self-Optimizing Reinforcement Learning for Discrete-Time Systems With Uncontrollable Factors

引用

IEEE Transactions on Systems, Man, and Cybernetics: Systems 2025年第5期55卷 3249-3260页

作者： Zhang, Jianfeng Zhang, Haoran Zhao, Chunhui Zhejiang University State Key Laboratory of Industrial Control Technology College of Control Science and Engineering Hangzhou310027 China

The industrial system usually contains not only controllable variables (CVs) but also uncontrollable variables (unCVs), e.g., weather conditions and friction. These unCVs have a direct impact on system control performance. Despite the success of current deep reinforcement learning (DRL) control algorithms, most of them neglect the impact of unCVs, which can cause the deterioration of control performance and instability of the system. To perceive and eliminate the impact of unCVs, a performance-aware self-optimizing universal controller (PASOUC) is designed in this article. The PASOUC aims at integrating the representation of unCVs and controller design to perceive and eliminate the impact of unCVs under different conditions, which goes beyond most existing control methods. Technically, a historical trajectory-inspired control performance perceptron is developed to perceive the impact of unCVs on system control performance under different conditions. Subsequently, a new performance-aware reward is designed to integrate the representation of unCVs and controller design while training the DRL controller. In addition, the domain randomization (DR) training strategy is employed to learn a universal control policy, which can access the approximate optimal trajectory under nonideal conditions. In this way, the impact of unCVs can be eliminated. To handle the low efficiency of the DR training, the policy improvement-policy proximal optimization (PI-PPO) is proposed to enhance the convergence speed of the DR training by performing explicit policy improvement. Finally, illustrative examples are presented to demonstrate the superiority of the proposed method. © 2013 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A blast furnace fault monitoring algorithm with low false alarm rate:Ensemble of greedy dynamic principal component analysis-Gaussian mixture model

引用

Chinese Journal of Chemical Engineering 2023年第5期57卷 151-161页

作者： Xiongzhuo Zhu Dali Gao Chong Yang Chunjie Yang State Key Laboratory of Industrial Control Technology College of Control Science and EngineeringZhejiang UniversityHangzhou 310027China

The large blast furnace is essential equipment in the process of iron and steel manufacturing. Due to the complex operation process and frequent fluctuations of variables, conventional monitoring methods often bring false alarms. To address the above problem, an ensemble of greedy dynamic principal component analysis-Gaussian mixture model(EGDPCA-GMM) is proposed in this paper. First, PCA-GMM is introduced to deal with the collinearity and the non-Gaussian distribution of blast furnace ***, in order to explain the dynamics of data, the greedy algorithm is used to determine the extended variables and their corresponding time lags, so as to avoid introducing unnecessary noise. Then the bagging ensemble is adopted to cooperate with greedy extension to eliminate the randomness brought by the greedy algorithm and further reduce the false alarm rate(FAR) of monitoring results. Finally, the algorithm is applied to the blast furnace of a large iron and steel group in South China to verify *** with the basic algorithms, the proposed method achieves lowest FAR, while keeping missed alarm rate(MAR) remain stable.

关键词： Chemical processes Principal component analysis Gaussian mixture model Process monitoring Ensemble Process control

来源：评论

学校读者我要写书评

暂无评论

An Enhanced Distributed Optimization Subject to Multiple Constraints: A Case Study on Entire Ethylene Separation Process

引用

IEEE Transactions on control of Network Systems 2025年第2期12卷 1227-1237页

作者： Liu, Bing Du, Wenli Li, Zhongmei East China University of Science and Technology State Key Laboratory of Industrial Control Technology Ministry of Education Shanghai200237 China Huzhou Institute of Industrial Control Technology Huzhou313099 China

This paper studies an enhanced distributed optimization algorithm in an undirected topology based on local communication and computation to optimize the sum of local objective functions under multiple constraints. In particular, the adapt-then-combine distributed inexact gradient tracking algorithm (ATC-DIGing-MC) is developed for smooth and strongly convex local functions with multiple constraints. By implementing the adapt-then-combine scheme and using the gradient tracking technique, rapidity and flexibility can be obtained by the ATC-DIGing-MC with uncoordinated step sizes. Subsequently, the parameter projection strategy is implemented to deal with realistic production limitations in the presence of multiple constraints. Meanwhile, rigorous theoretical proofs and convergence theorem are provided to verify the geometrical convergence rate of the ATC-DIGing-MC. Furthermore, the numerical experiments based on the entire ethylene separation process validate the superior performance of the ATC-DIGing-MC algorithm in terms of time and energy savings compared to established algorithms. © 2025 IEEE.

关键词： Optimization algorithms

来源：评论

学校读者我要写书评

暂无评论

An Accurate and Interpretable Framework for Trustworthy Process Monitoring

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第5期5卷 2241-2252页

作者： Wang, Hao Wang, Zhiyu Niu, Yunlong Liu, Zhaoran Li, Haozhe Liao, Yilin Huang, Yuxin Liu, Xinggao Zhejiang University State Key Laboratory of Industrial Control Technology College of Control Science and Engineering Hangzhou310027 China

Trustworthy process monitoring seeks to build an accurate and interpretable monitoring framework, which is critical for ensuring the safety of energy conversion plant (ECP) that operates under extreme working conditions, such as high pressure and temperature. Contemporary self-attentive models, however, fall short in this domain for two main reasons. First, they rely on stepwise correlations that fail to involve physically meaningful semantics in ECP logs, resulting in suboptimal accuracy and interpretability. Second, attention matrices are frequently cluttered with spurious correlations that obscure physically meaningful ones, further impeding effective interpretation. To overcome these issues, we propose AttentionMixer, a framework aimed at improving both accuracy and interpretability of the existing methods, and establish a trustworthy ECP monitoring framework. Specifically, to tackle the first issue, we employ a spatial adaptive message passing block to capture variatewise correlations. This block is coupled with a temporal adaptive message passing block through a mixing operator, yielding a multifaceted representation of ECP logs accounting for both the stepwise and variatewise correlations. Concurrently, to tackle the second issue, we employ a sparse message passing regularizer to filter out spurious correlations. We validate the efficacy of AttentionMixer using two real-world datasets from the radiation monitoring network for Chinese nuclear power plants. © 2020 IEEE.

关键词： Accident prevention

来源：评论

学校读者我要写书评

暂无评论

Pyramid Dynamic Bayesian Networks for key Performance Indicator Prediction in Long Time-Delay industrial Processes

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第2期5卷 661-671页

作者： Zeng, Lingquan Ge, Zhiqiang Zhejiang University State Key Laboratory of Industrial Control Technology College of Control Science and Engineering Hangzhou310027 China

Building dynamic Bayesian networks (DBNs) for time-delay industrial processes has always been tough, since the structure learning of the DBN is a NP-hard problem. In this article, a pyramid DBN (PDBN) framework is proposed to speed up modeling and improve feature learning for industrial processes with large time delays. In the PDBN framework, a sequence of small-sized DBNs are established, each of which is a progressively simpler representation of the previous layer, and the feature information learned by the DBN sequence corresponds to the layers in the pyramid. With information fusion of the feature pyramid and a further feature filtering with hill climbing based on Bayesian information criterion, we can restore the feature information of the time-delay industrial processes. Regression models are built based on the features learned by the PDBN framework for further key performance indicator estimation. Advantages of the method have been effectively validated on two actual industrial cases. With multilevel feature learning and filtering, both modeling speed and prediction accuracy have been greatly improved. The proposed method outperforms other similar works on identifying important features in industrial processes and existing DBN-based soft sensors. © 2023 IEEE.

关键词： Hidden Markov models

来源：评论

学校读者我要写书评

暂无评论

{1012}twin–twin intersection-induced lattice rotation and dynamic recrystallization in Mg–6Al–3Sn–2Zn alloy

引用

Journal of Magnesium and Alloys 2024年第4期12卷 1529-1539页

作者： Bin-Jiang Lv Sen Wang Fu-Hao Gao Ning Cui Yi-Nan Li Tie-Wei Xu Feng Guo School of Mechanical and Automotive Engineering Qingdao University of Technology266520China Key Lab of industrial Fluid Energy Conservation and Pollution Control Qingdao University of Technology266520China

This study investigated the formation mechanism of new grains due to twin–twin intersections in a coarse-grained Mg–6Al–3Sn–2Zn alloy during different strain rates of an isothermal *** results of electron backscattered diffraction investigations showed that the activated twins were primarily{1012}tension twins,and 60°<1010>boundaries formed due to twin–twin intersections under different strain *** twin variants with 60°<1010>boundaries transformed into new grains through lattice rotations at a low strain rate(0.01 s^(−1)).At a high strain rate(10 s^(−1)),the regions surrounded by subgrain boundaries through high-density dislocation arrangement and the 60°<1010>boundaries transformed into new grains via dynamic recrystallization.

关键词： Mg alloy Twin-twin intersection Lattice rotation Dynamic recrystallization

来源：评论

学校读者我要写书评

暂无评论

Reduced-Order Observer-Based Resilient control for MASs with Time-Varying Delay Against DoS Attacks

IEEE Transactions on Industrial Cyber-Physical Systems

引用

IEEE Transactions on industrial Cyber-Physical Systems 2024年 2卷 51-59页

作者： Qiu, Xiaojie Meng, Wenchao Xu, Jinming Yang, Qinmin Zhejiang University State Key Laboratory of Industrial Control Technology College of Control Science and Engineering Hangzhou310027 China

This paper concentrates on the resilient control issue in multi-agent systems (MASs) with the time-varying delay subjected to denial of service (DoS) attacks. To eliminate the impact of DoS attacks, a novel resilient control framework is constructed for MASs, where a variable following the Bernoulli distribution process is utilized to denote the behaviors of the attacker. Besides, we design a novel reduced-order (RO) observer with internal time-varying delay to estimate the unknown output of the system and compensate for lost data in the channels. Different from the general observers, the designed observer can ensure both the reduced-order observation errors and the full-order ones can converge to zero. Furthermore, an RO observer-based distributed protocol is devised such that the consensus errors of the closed-loop systems asymptotically converge to zero. Simulation comparisons are provided to validate the efficiency and superiority of our approach. © 2023 IEEE.

关键词： Multi agent systems

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：