检索结果-内蒙古大学图书馆

data-driven fault detection for Multimode Industrial Processes Based on Mixture of Autoencoders

ASIA-PACIFIC JOURNAL OF CHEMICAL ENGINEERING 2025年

作者： Cui, Guoqiang Li, Xu Wei, Chihang Shao, Weiming China Univ Petr East China Coll New Energy Qingdao Peoples R China Hangzhou Normal Univ Sch Informat Sci & Technol Hangzhou Peoples R China

data-driven fault detection technique has been widely applied to industrial processes to reduce production losses caused by faults. Due to complicated chemical reactions and phase changes, the relationships between process variables are nonlinear. In particular, many industrial processes often operate under multiple conditions, resulting in multimode characteristics in the collected data and further complicating the nonlinearities between process variables, rendering challenging difficulties for traditional fault detection methods. In view of this, this paper proposes a novel fault detection method for multimode industrial processes by developing a mixture of autoencoders (MixAE) model. The MixAE is first designed with Gaussian mixture model and multiple single-hidden layer autoencoders;then, an efficient learning algorithm, based on expectation-maximization algorithm and gradient ascent algorithm, is developed to train the MixAE. By using the confidence index provided by each autoencoder, a comprehensive evaluation statistic is formulated for detecting anomalies. Finally, the proposed fault detection method is validated with a numerical example and an experimental three-tank liquid level control system. The experimental results demonstrate that the proposed method achieves better fault detection performance for multimode industrial processes compared to traditional methods.

关键词： data-driven fault detection expectation-maximization gradient ascent mixture of autoencoders multimode industrial process

来源：评论

学校读者我要写书评

暂无评论

data-driven fault detection of heterogeneous multi-agent systems using combined hardware and temporal redundant information

引用

ISA TRANSACTIONS 2024年 147卷 90-100页

作者： Yang, Chen Li, Yan Chen, Qijun Tongji Univ Sch Elect & Informat Engn Shanghai Peoples R China 1239 Siping Rd Shanghai 200092 Peoples R China

This study addresses the fault detection (FD) problem in heterogeneous multi-agent systems (HMASs) with unknown system models. A novel data-driven FD scheme is proposed by properly combining hardware and temporal redundant information to accelerate the generation of fault detectors while ensuring detection accuracy. The computational burden associated with the FD scheme is alleviated by applying a two-step order reduction algorithm. Additionally, an optimization problem is formulated, simplified and solved to achieve a compromise between sensitivity to faults and robustness to disturbances, further enhancing the detection performance of agents. Through a series of examples and comparative experiments, the effectiveness and improvements of the proposed approach are demonstrated.

关键词： Heterogeneous multi-agent system data-driven fault detection Hardware redundant information Temporal redundant information Parity space

来源：评论

学校读者我要写书评

暂无评论

A data-driven fault detection and fault-tolerant control scheme for large-scale systems and its application on multi-area interconnected power systems

引用

IET CONTROL THEORY AND APPLICATIONS 2023年第4期17卷 400-418页

作者： Gao, Jingjing Yang, Xu Huang, Jian Peng, Kaixiang Univ Sci & Technol Beijing Sch Automat & Elect Engn Key Lab Knowledge Automat Ind Proc Minist Educ 30 Xueyuan Rd Beijing Peoples R China

Stimulated by the enhancing requirements for system safety as well as process reliability in complex industrial processes, this paper investigates a data-driven fault detection and fault-tolerant control scheme for large-scale systems. To this end, a pair-wise decomposition approach is first presented based on the inclusion principle, which expands the original large-scale system into a set of disjointed pair-wise subsystems. Then, for the pair-wise subsystems, the observer-based residual generators and observer-based state feedback controllers for fault detection and fault-tolerant control purpose are further developed in a decentralized way. On this basis, the fault-tolerant controller for the original large-scale system is obtained by the coordination and contraction of the decentralized observer-based state feedback controllers. Finally, the feasibility and effectiveness of the proposed scheme are demonstrated through the case study on a multi-area interconnected power system.

关键词： decentralised control fault-tolerant control purpose fault diagnosis Power system management, operation and economics large-scale systems power system faults Simulation, modelling and identification fault-tolerant control scheme decentralized observer-based state feedback controllers power system interconnection data-driven fault detection power system control control system synthesis complex industrial processes disjointed pair-wise subsystems observers large-scale system Power system control observer-based residual generators system safety multiarea interconnected power system Control of electric power systems Multivariable control systems pair-wise decomposition approach Control system analysis and synthesis methods fault tolerant control state feedback fault-tolerant controller Stability in control theory

来源：评论

学校读者我要写书评

暂无评论

Accuracy Analysis of Polynomial Model and Auto Regressive Model for data-driven fault detection 7

Accuracy Analysis of Polynomial Model and Auto Regressive Mo...

引用

IEEE 7th data driven Control and Learning Systems Conference (DDCLS)

作者： Sun, Bowen He, Zhangming Xu, Shuqing Zhou, Haiyin Wang, Jiongqi Natl Univ Def Technol Coll Liberal Arts & Sci Changsha 410072 Hunan Peoples R China China Acad Space Technol Beijing Inst Control Engn Beijing 100080 Peoples R China

ISBN: (纸本)9781538626184

The key of data-driven fault detection method lies in the full and effective understanding of the detected data, and the fitting for the detected data is an effective means to realize the parameterization of the data model. In this paper, the polynomial model and the autoregressive model are used to estimate and predict the non-stationary data and the stationary data respectively, so as to achieve the data-driven fault detection. The estimation accuracy of the parameter model is analyzed. The relationship between the prediction accuracy and the prediction duration, the polynomial fitting window, the fitting order are given theoretically. Finally, numerical simulation results are given, which can provide some support for data-driven fault detection to some extent.

关键词： data-driven fault detection Date Fitting Polynomial Model Auto Regressive Model Slide Window Prediction Accuracy

来源：评论

学校读者我要写书评

暂无评论

Online fault detection for transmission lines based on data-driven neural network using synchronized measurement

引用

JOURNAL OF THE CHINESE INSTITUTE OF ENGINEERS 2025年第3期48卷 205-213页

作者： Zhang, Tong Wang, Nan Shenyang Univ Technol Sch Artificial Intelligence Shenyang Peoples R China Shenyang Univ Coll Mech Engn Shenyang Peoples R China

A data-driven online fault detection method is proposed based on the characteristic reduction using the phase angle shift analysis. In this work, the shift feature in the amplitude-phasor signal is enhanced based on Kirchhoff law. Then, the fault characteristic entropy (FCE) is introduced to condense high-dimensional fault characteristics to low-dimensional eigenvectors, which achieves more distinct clustering. The FCE index measures the mutual dependence between the neighbor buses measured by phasor measurement units. Finally, the regularized radial basis neural network (RRBF) is utilized to implement fault section detection for low-dimensional eigenvectors. The proposed fault detection method based on FCE-RRBF is evaluated in IEEE-39 bus *** with CNN (Convolutional Neural Network), WNN (Convolutional Neural Network) and LSTM (Long Short Term Memory) methods, the FCE-RRBF method can accurately online detect the fault section. The FCE-RRBF method establishes the fundamental framework for the self-healing recovery in the transmission power system.

关键词： data-driven fault detection transmission network regularized radial basis neural network Shannon entropy

来源：评论

学校读者我要写书评

暂无评论

A data-driven subspace distributed fault detection strategy for linear heterogeneous multi-agent systems

引用

ISA TRANSACTIONS 2024年 146卷 186-194页

作者： Yazdanpanah, Nasim Farsangi, Malihe Maghfoori Seydnejad, Saeid R. Shahid Bahonar Univ Kerman Dept Elect Engn Kerman Iran

This paper presents a new data-driven subspace distributed fault detection strategy specifically designed for linear heterogeneous multi-agent systems (MASs). The proposed approach leverages the characteristics of heterogeneous MASs, where agents exhibit diverse dynamics and parameters. By utilizing subspace construction techniques, the proposed method captures the normal behavior of each agent and enables the detection of deviations that indicate the presence of faults. Unlike existing methods, the approach is completely data-driven and eliminating the need for centralized information or communication among the agents. Simulation results demonstrate the effectiveness and efficiency of the proposed approach in detecting simultaneous faults in different agents. Overall, the proposed approach represents a significant departure from existing methods and offers a powerful new tool for fault detection in heterogeneous multi-agent systems.

关键词： data-driven fault detection Heterogeneous Multi-agent systems Subspace Identification Method

来源：评论

学校读者我要写书评

暂无评论

A machine learning approach for particle accelerator errant beam prediction using spatial phase deviation

引用

NUCLEAR INSTRUMENTS & METHODS IN PHYSICS RESEARCH SECTION A-ACCELERATORS SPECTROMETERS DETECTORS AND ASSOCIATED EQUIPMENT 2024年 1063卷

作者： Yucesan, Yigit A. Blokland, Willem Ramuhalli, Pradeep Zhukov, Alexander Peters, Charles Brown, David Long, Cary Oak Ridge Natl Lab Oak Ridge TN 37830 USA

Particle accelerators are extremely complex systems that are expected to operate on high availability. Predicting impending failures only by utilizing data collected from diagnostic equipment already on board can help operators to avoid installing expensive sensors, unscheduled downtime and associated costs. For this purpose we explore the predictive power of Machine Learning algorithms to detect faulty beams prior to the failure. In this study, we propose a Machine Learning approach to model mapping from a pair of sensors located across the accelerator. While the model is trained to represent normal operation, we evaluate the predictive performance on known faulty beam pulses. We also investigate the model performance on unseen data through k -fold cross-validation. Then we recap the analysis with a neural architecture search and hyperparameter optimization study to fine tune our initial model. This paper will also introduce a sustainable framework that can standardize Machine Learning workflow applied to particle accelerators.

关键词： Particle accelerators data-driven fault detection Neural architecture search Sustainable machine learning framework

来源：评论

学校读者我要写书评

暂无评论

A fault detection Approach for Nonlinear Systems Based on Deep Learning-Aided Kernel Representations

引用

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2025年

作者： Wu, Shimeng Zhu, Yimin Luo, Hao Wang, Hao Zhang, Jiusi Qiao, Xinyu Tian, Jilun Harbin Inst Technol Dept Control Harbin 150001 Peoples R China Harbin Inst Technol Simulat Ctr Harbin 150001 Peoples R China Harbin Inst Technol Natl Key Lab Complex Syst Control & Intelligent Ag Harbin 150001 Peoples R China Harbin Inst Technol Harbin 150001 Peoples R China Univ Elect Sci & Technol China Sch Automat Engn Chengdu 611731 Peoples R China

This article focuses on utilizing process data to detect faults in nonlinear systems. To accomplish this, stable image/kernel representation is learned for nonlinear systems using deep neural networks, which serve as the basis for residual generators and fault detection. First, the closed-loop image representation of nonlinear systems is identified using gate recurrent units and fully connected neural networks. The involved network topology is designed to learn the nonlinear mapping in the form of linear time-varying state space, allowing the extension of existing linear methods to nonlinear systems. Then, with the identified image representation, the data-driven realization of kernel representation is derived. Finally, the residual generator is developed utilizing the system's kernel representation to enable precise fault detection in nonlinear systems. The effectiveness of our study is demonstrated through a numerical benchmark study and an actual experiment on a real Mecanum-wheeled vehicle platform.

关键词： Nonlinear systems fault detection Image representation Generators fault diagnosis Deep learning Linear systems Computational modeling Training Knowledge based systems data-driven fault detection kernel representation neural networks nonlinear systems

来源：评论

学校读者我要写书评

暂无评论

A Novel Two-Stage fault-detection Method Based on Constrained RVM and Integrating LDA With Minimax Probability Machine

引用

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2023年第3期19卷 3198-3207页

作者： Yang, Chen Li, Yan Chen, Qijun Tongji Univ Sch Elect & Informat Engn Shanghai 201804 Peoples R China

This article presents a novel two-stage fault-detection (FD) method composed of a preclassifier and a reclassifier for complex industrial processes, where the preclassifier is developed by combining linear discriminant analysis and minimax probability machine to reduce dimensionality and classify fully separable data with low computation time. For overlapping data that cannot be separated by the preclassifier, a reclassifier is designed by constructing a constrained relevance vector machine (RVM), according to Neyman-Pearson principle, to decrease the missed alarm rate. The reclassifier has a lower computational load than traditional RVM due to the amount and dimensionality of reclassified data reduced by the first stage, thereby a balance between detection accuracy and computational burden of the whole FD method can be achieved. Finally, an industrial benchmark of Tennessee-Eastman process is utilized to verify the effectiveness of the proposed FD method.

关键词： fault detection Support vector machines Principal component analysis Informatics Task analysis Feature extraction Probabilistic logic Classification data-driven fault detection fault detection (FD) linear discriminant analysis (LDA) minimax probability machine (MPM) missed alarm rate (MAR) relevance vector machine (RVM)

来源：评论

学校读者我要写书评

暂无评论

Attack and Defense: Adversarial Security of data-driven FDC Systems

引用

IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS 2023年第1期19卷 5-19页

作者： Zhuo, Yue Yin, Zhenqin Ge, Zhiqiang Zhejiang Univ Coll Control Sci & Engn State Key Lab Ind Control Technol Hangzhou 310027 Peoples R China

In modern industries, data-driven fault detection and classification (FDC) systems can efficiently maintain industrial security and stability, while the security of the data-driven FDC system itself is rarely or even never considered. The security problem named adversarial vulnerability is the intrinsic of data-driven machine learning models, which will give incorrect predictions under the maliciously perturbed input data. This paper presents a work on this new security topic of the data-driven FDC systems, by 1) summarizing and comparing various recent and typical adversarial attack and defense methods for fault classifiers;2) proposing novel attack and defense techniques for unsupervised fault detectors;3) constructing a novel industrial adversarial security benchmark on FDC systems in the Tennessee-Eastman process (TEP) dataset;4) exploring and discussing which attack is most potentially threatening for FDC systems and which defense technique is most applicable to mitigate attacks. The results reveal unique security properties of FDC systems, mainly including 1) for fault classifiers, black-box attack is close to the attack strength of white-box FGSM and the universal transferable attack is not significantly stronger than random noise;2) weak adversarial training is excellent with high adversarial accuracy improvement and negligible clean accuracy decrease;3) fault detectors are intrinsically more robust, and can be well protected by strong adversarial training. More intriguing properties and profound insights are demonstrated in the paper. This pioneering work could guide researchers and practitioners in discovering and navigating the field of FDC system adversarial robustness, outlining the research directions and open problems.

关键词： Security Benchmark testing fault detection Robustness data models Training Image reconstruction Adversarial attack adversarial defense data-driven fault classification data-driven fault detection industrial intelligent systems security

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

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

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：