检索结果-内蒙古大学图书馆

Deep probabilistic graphical modeling for robust multivariate time series anomaly detection with missing data

RELIABILITY ENGINEERING & SYSTEM SAFETY 2023年第1期238卷

作者： Yang, Jingyu Yue, Zuogong Yuan, Ye Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automation Key Lab Image Proc & Intelligent Control Minist Educ Wuhan 430074 Peoples R China

Multivariate time series anomaly detection with missing data is one of the most pending issues for industrial monitoring. Due to scarcity of labeled anomalies, most advanced data-driven anomaly detection approaches fall in the unsupervised learning paradigm. As a premise in the presence of missing data, one needs to improve the data quality through data imputation with a separate model. Our concern lies in the consistency between data imputation and unsupervised learning for robust anomaly detection, regarding accurately discovering the spatiotemporal dependence among multiple variables over time. However, the existing practice tends to overlook this consistency and decouple the training process for these two closely linked tasks. This article novelly proposes a probabilistic multivariate time series anomaly detection framework that unifies data imputation and unsupervised learning. A deep probabilistic graphical model abbreviated SCNF is first devised for unsupervised density estimation. A tailored expectation maximization-based optimization scheme is then developed to achieve the joint training of data imputation and unsupervised learning with missing data. The efficacy is experimentally corroborated in several industrial applications, including chemical process, water treatment and network traffic. Briefly, the joint training framework enhances the AUROC of SCNF by averagely 6.34% for three applications under 50% data missing rate.

关键词： Multivariate time series Anomaly detection Missing data Probabilistic graphical model Expectation maximization

来源：评论

学校读者我要写书评

暂无评论

One-time-pads encryption in the tile assembly model

One-time-pads encryption in the tile assembly model

引用

作者： Chen, Zhihua Xu, Jin Key Laboratory of Image Processing and Intelligent Control Department of Control Science and Engineering Huazhong University of Science and Technology Wuhan 430074 China

Recent research has demonstrated that the ultra-scale computation by self-assembly DNA tiles can be implemented in the laboratory. One of the significant applications is the DNA-based cryptography systems. In this paper, we detail procedures for the DNA-based One-Time-Pads (OTP) cryptography, which is in principle unbreakable. In order to implement the whole encrypting and decrypting processes, we propose four tile systems: encrypting system, ciphertext extracting system, key extracting system and decrypting system. All the tile systems use θ(1) input tiles and compute in θ(n) steps according to the length of the input massage. The method of implementing the DNA OTP encryption extends the self-assemble tile models and achieves the real randomness in the DNA OTP. Copyright © 2010 American Scientific Publishers. All rights reserved.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

A Variant of P Machine: Splicing P Machine

引用

JOURNAL OF COMPUTATIONAL AND THEORETICAL NANOSCIENCE 2013年第6期10卷 1376-1384页

作者： Zhang, Xingyi Jiang, Yun Pan, Linqiang Anhui Univ Sch Comp Sci & Technol Minist Educ Key Lab Intelligent Comp & Signal Proc Hefei 230039 Peoples R China Huazhong Univ Sci & Technol Dept Control Sci & Engn Key Lab Image Proc & Intelligent Control Wuhan 430074 Peoples R China

Splicing P systems are a class of distributed parallel computing models in the framework of membrane computing, which are inspired by the recombination of DNA molecules under the influence of restriction enzymes. In this work, a variant of P machine, called splicing P machine, is presented, which can provide an automata approach to investigate splicing P systems. It is shown that such a device can do what the splicing P system does. Several examples are given to illustrate that splicing P machine can provide the same results as the corresponding splicing P system. This work provides an answer to an open problem formulated by G. Ciobaun and M. Gontineac.

关键词： Membrane Computing Splicing P System P Machine Automaton

来源：评论

学校读者我要写书评

暂无评论

Sliding mode control with fuzzy compensator of pneumatic muscle actuator

Sliding mode control with fuzzy compensator of pneumatic mus...

引用

作者： Wu, Jun Wang, Yongji Huang, Jian Xing, Kexin Key Laboratory of Image Processing and Intelligent Control Department of Control Science and Engineering HuaZhong University of Science and Technology Wuhan Hubei China

In this paper, the Pneumatic Muscle (PM) as actuator is investigated. The PM model is established by using a phenomenological model consisting of a contractile element, a spring element, and a damping element in parallel. Unlike the previous study, we fully consider the different characteristics of PM model in different pressure ranges, especially in low pressure range. In order to test the effectiveness of our PM modelling, the Sliding Mode control (SMC) is applied. For the purpose of improving the tracking accuracy, self-adaptation and robustness of PM control, Sliding Mode control with Fuzzy Compensator (FCSMC) is developed to make PM system track a desired trajectory within a guaranteed accuracy even though there are modelling uncertainties, friction, disturbance, and so on. Based on the Lyapunov theory, the stability criterion under FCSMC control strategy for PM system is established. The experimental results demonstrate the validity of PM modelling as well as indicate the effectiveness of the proposed FCSMC method. Copyright © 2012 Inderscience Enterprises Ltd.

关键词： Sliding mode control

来源：评论

学校读者我要写书评

暂无评论

Fuzzy PID control of a wearable rehabilitation robotic hand driven by pneumatic muscles

Fuzzy PID control of a wearable rehabilitation robotic hand ...

引用

20th Anniversary MHS 2009 and Micro-Nano Global COE - 2009 International Symposium on Micro-NanoMechatronics and Human Science

作者： Wu, Jun Huang, Jian Wang, Yongji Xing, Kexin Xu, Qi Key Laboratory of Image Processing and Intelligent Control Department of Control Science and Engineering Huazhong University of Science and Technology 430074 Wuhan China

ISBN: (纸本)9781424450954

Intensive task-oriented repetitive physical therapies need be provided by individualized interaction between the patients and the rehabilitation specialists to improve hand motor performance for those survived from stoke and traumatic brain injury. The goal of this research is to develop a novel wearable device for robotic assisted hand repetitive therapy. We designed a pneumatic muscle (PM) driven therapeutic device that is wearable and provides assistive forces required for grasping and finger extension. The robot has two distinct degrees of freedom at the thumb and all other fingers. The embedded sensors can feedback position and force information for robot control and quantitative evaluation of task performance. It is potential of providing supplemental at-home therapy in addition to in the clinic treatment. To realize the trajectory tracking control, a fuzzy PID controller is designed for the proposed device. The experimental results show angle tracking control of the robotic hand using the fuzzy PID controller has better performance than that using a conventional PID controller. ©2009 IEEE.

关键词： Wearable technology

来源：评论

学校读者我要写书评

暂无评论

Echo state network-based internal model control for pneumatic muscle system

Echo state network-based internal model control for pneumati...

引用

2011 International Conference on Electrical, Information Engineering and Mechatronics, EIEM 2011

作者： Wu, Jun Wang, Yongji Huang, Jian Zhou, Hanying Cai, Hong Key Laboratory of Image Processing and Intelligent Control Department of Control Science and Engineering Huazhong University of Science and Technology 430074 Wuhan China

ISBN: (纸本)9781447124665

Pneumatic muscle (PM) has strong time varying characteristic. The complex nonlinear dynamics of PM system poses problems in achieving accurate modeling and control. To solve these challenges, we propose an echo state network (ESN)-based internal model control (IMC) for PM system in this paper. Here, ESN is employed for identifying the plant model and constructing controller. Recursive least squares (RLS) is used for the online update of ESN. The ESN-based IMC fully embodies the virtues of ESN and IMC. It can build high accurate plant model without detailed model information, as well as attain strong robustness by online self-tuning of controller and internal model. Experiment demonstrates the effectiveness of the proposed control algorithm. The results show that ESNBIMC achieves satisfactory tracking performance. © 2012 Springer-Verlag London Limited.

关键词： controllers

来源：评论

学校读者我要写书评

暂无评论

Signal Enhancement Method for Mechanical Fault Diagnosis in Flexible Drive-Train

引用

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS 2021年第3期68卷 2554-2563页

作者： Yao, Yuan Xie, Bin Lei, Li Li, Yesong Yin, Quan Huazhong Univ Sci & Technol Key Lab Minist Educ Image Proc & Intelligent Control Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China

Nowadays, mechanical fault detection techniques based on signals from motor drive are becoming popular. However, in electromechanical system with flexible drive-train, mechanical fault signature reflected in motor drive is suppressed by the transmission path. This signal attenuation reduces the effectiveness of fault detection. Therefore, research on the law of fault signature transmission from load to motor is necessary. In this article, a two-mass electromechanical system with a long shaft is concerned. First, the transmission path from load torque to electromagnetic torque is analyzed, and the concept of torque transmission efficiency is proposed. Then, the frequency characteristic method and transfer function method are presented to establish the signal transmission model. Finally, the model-based signal enhancement method is proposed to compensate the attenuation. Simulation and experimental results show that the proposed method can enhance the quality of fault signature transmission and improve the performance of fault classification.

关键词： Flexible drive-train mechanical fault diagnosis motor drive signal enhancement torque transmission

来源：评论

学校读者我要写书评

暂无评论

Hard sample mining makes person re-identification more efficient and accurate

引用

NEUROCOMPUTING 2020年第0期382卷 259-267页

作者： Chen, Kezhou Chen, Yang Han, Chuchu Sang, Nong Gao, Changxin Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Key Lab Minist Educ Image Proc & Intelligent Control Wuhan 430074 Peoples R China

In recent years, the field of person re-identification has made significant advances riding on the wave of deep learning. However, owing to the fact that there are much more easy examples than those meaningful hard examples in a dataset, the training tends to stagnate quickly and the model may suffer from over-fitting, which leads to some error matching of models especially for some hard samples during the test process. Therefore, the hard sample mining method is fateful to optimize the model and improve the learning efficiency. In this paper, an Adaptive Hard Sample Mining algorithm is proposed for training a robust person re-identification model. No need for hand-picking the images in the batch or designing the loss function for both positive and negative pairs, we can briefly calculate the hard level by comparing the prediction result with the true label of the sample. Meanwhile, taking into account the change in the number of samples required for the model during training process, an adaptive threshold of hard level can make the algorithm not only stay in step with training process harmoniously but also alleviate the under-fitting and over-fitting problem simultaneously. Besides, the designed network to implement the approach is very efficient and has good generalization performance that can be combined with various existing models readily. Experimental results on Market-1501, DukeMTMC-reID and CUHK03 datasets clearly demonstrate the effectiveness of the proposed algorithm. (C) 2019 Elsevier B.V. All rights reserved.

关键词： Person re-identification Hard sample mining Deep learning

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning-Based Security-Constrained Battery Scheduling in Home Energy System

引用

IEEE TRANSACTIONS ON CONSUMER ELECTRONICS 2024年第1期70卷 3548-3561页

作者： Wang, Bo Zha, Zhongyi Zhang, Lijun Liu, Lei Fan, Huijin Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Key Lab Minist Educ Image Proc & Intelligent Control Wuhan 430074 Peoples R China

The home energy system today involves multiple renewable energy sources and battery energy storage systems, which can be considered as a microgrid. The battery energy storage system is a key component in the home energy system for the sake of filling the gap between the user demand and volatile energy supplies to maximize the techno and economic performances. However, the battery scheduling must suffer the stochastic nature of renewable energy resources and loads, which results in an intractable multi-period stochastic optimization problem with security constraints. An improved actor-critic-based reinforcement learning is proposed for this issue, where a distributional critic net is applied for faster and more accurate reward estimation under uncertainties, and a policy net incorporating protective secondary control is adopted to satisfy security constraints, preventing the unsafe state of batteries during the trial-and-error process. Numerical tests show that the proposed approach outperforms conventional reinforcement learning algorithms, as well as the rule-based battery scheduling approach while guaranteeing safe operation. The robustness and adaptability of the proposed method are also verified in case studies with different optimization tasks.

关键词： Batteries State of charge Security Degradation Trajectory Q-learning Monte Carlo methods Home energy system energy storage renewable energy resources security-constrained battery scheduling reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Different Set Domain Adaptation for Brain-Computer Interfaces: A label Alignment Approach

引用

IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING 2020年第5期28卷 1091-1108页

作者： He, He Wu, Dongrui Huazhong Univ Sci & Technol Minist Educ Key Lab Image Proc & Intelligent Control Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China

A brain-computer interface (BCI) system usually needs a long calibration session for each new subject/task to adjust its parameters, which impedes its transition from the laboratory to real-world applications. Domain adaptation, which leverages labeled data from auxiliary subjects/tasks (source domains), has demonstrated its effectiveness in reducing such calibration effort. Currently, most domain adaptation approaches require the source domains to have the same feature space and label space as the target domain, which limits their applications, as the auxiliary data may have different feature spaces and/or different label spaces. This paper considers different set domain adaptation for BCIs, i.e., the source and target domains have different label spaces. We introduce a practical setting of different label sets for BCIs, and propose a novel label alignment (LA) approach to align the source label space with the target label space. It has three desirable properties: 1) LA only needs as few as one labeled sample from each class of the target subject;2) LA can be used as a preprocessing step before different feature extraction and classification algorithms;and, 3) LA can be integrated with other domain adaptation approaches to achieve even better performance. Experiments on two motor imagery datasets demonstrated the effectiveness of LA.

关键词： Covariance matrices Electroencephalography Manifolds Feature extraction Calibration Silicon Pipelines Brain-computerinterface EG label alignment Riemannian geometry domain adaptation transfer learning

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：