检索结果-内蒙古大学图书馆

Multi-objective optimization for ladle tracking of aluminium tapping based on NSGA-II 1

学校读者我要写书评

暂无评论

12th International Conference on Bio-Inspired Computing: Theories and Applications, BIC-TA 2017

作者： Zhou, Kaibo Zou, Yutao Wang, Hongting Xu, Gaofeng Guo, Sihai School of Automation Huazhong University of Science and Technology Wuhan430074 China MOE Key Laboratory of Image Processing and Intelligence Control Wuhan430074 China School of Automation Wuhan University of Technology Wuhan430070 China

ISBN: (数字)9789811071799

ISBN: (纸本)9789811071782

In order to realize the optimization of ladle tracking of aluminium tapping, a mathematical model, which takes the grade of aluminium, the energy required for transportation and the optimum ratio of aluminium liquid into account, is established. The traditional method optimizes the impurity content and the transport distance based on a single objective optimization, but it requires the empirical values of the weight coefficients. The paper proposes a modified multi-objective optimization model with the elitist non-dominated sorting genetic algorithm (NSGA-II). In the crossover operator process and the mutation operator process, the separately improved methods are introduced based on the ladle tracking problem of aluminium tapping, which replaces the Simulated Binary Crossover (SBX) in the original NSGA-II algorithm into Partially Matched Crossover (PMX) based on natural number coding and uses exchange mutation (EM) operator. Finally, the practical production data of the aluminium factory is used to verify the validity and practicability of the method, and the results show that this method can obtain a feasible solution for the user to choose suitable solutions, and avoid the defects of selecting empirical weighting coefficient. © Springer Nature Singapore Pte Ltd 2017.

关键词： Multiobjective optimization

Fault diagnosis in aluminium electrolysis using a joint method based on kernel principal component analysis and support vector machines 12th

学校读者我要写书评

暂无评论

Fault diagnosis in aluminium electrolysis using a joint meth...

12th International Conference on Bio-Inspired Computing: Theories and Applications, BIC-TA 2017

作者： Zhou, Kaibo Xu, Gaofeng Wang, Hongting Guo, Sihai School of Automation Huazhong University of Science and Technology Wuhan430074 China MOE Key Laboratory of Image Processing and Intelligence Control Wuhan430074 China School of Automation Wuhan University of Technology Wuhan430070 China

ISBN: (纸本)9789811071782

As a key part of aluminium smelting, the operational conditions of aluminium electrolytic cells are of great significance for the stability of the aluminium electrolysis process. As a result, developing a effective process monitoring and multiple fault diagnosis model is essential. Traditional multi-classification methods such as neural networks and multiple support vector machines (multi-SVM) have good effects. However, the connatural limitations of these methods limit the prediction accuracies. To solve this problem, a hierarchical method for multiple fault diagnosis based on kernel principal component analysis (KPCA) and support vector machines (SVM) is proposed in this paper. Firstly, test statistics, such as the comprehensive index , the squared prediction error (SPE), and Hotellings T-squared (T2), are used for fault detection. To separate faults preliminarily, traditional K-means clustering as transition layer is applied to the principal component scores. Next, anode effect is recognized and classified by the established SVM prediction model. Compared with multi-SVM-based classification methods, the proposed hierarchical method can diagnosis different faults with a higher precision. The prediction accuracy can reach about 90%. © Springer Nature Singapore Pte Ltd 2017.

关键词： Support vector machines

A General End-to-end Diagnosis Framework for Manufacturing Systems

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Yuan, Ye Ma, Guijun Cheng, Cheng Zhou, Beitong Zhao, Huan Zhang, Hai-Tao Ding, Han State Key Lab of Digital Manufacturing Equipment and Technology Huazhong University of Science and Technology Wuhan430074 China School of Artificial Intelligence and Automation MOE Key Lab of Intelligent Control and Image Processing Huazhong University of Science and Technology Wuhan430074 China School of Mechanical Science and Engineering Huazhong University of Science and Technology Wuhan430074 China

The manufacturing sector is envisioned to be heavily influenced by artificial intelligence-based technologies with the extraordinary increases in computational power and data volumes1,2. A central challenge in manufacturing sector lies in the requirement of a general framework to ensure satisfied diagnosis and monitoring performances in different manufacturing applications. Here we propose a general data-driven, end-to-end framework for the monitoring of manufacturing systems. This framework, derived from deep learning techniques, evaluates fused sensory measurements to detect and even predict faults and wearing conditions. This work exploits the predictive power of deep learning to automatically extract hidden degradation features from noisy, time-course data. We have experimented the proposed framework on ten representative datasets drawn from a wide variety of manufacturing applications. Results reveal that the framework performs well in examined benchmark applications and can be applied in diverse contexts, indicating its potential use as a critical corner stone in smart manufacturing. Copyright © 2018, The Authors. All rights reserved.

关键词： Manufacture

Fast signal recovery from saturated measurements by linear loss and nonconvex penalties

学校读者我要写书评

暂无评论

arXiv 2018年

作者： He, Fan Huang, Xiaolin Liu, Yipeng Yan, Ming Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University The MOE Key Laboratory of System Control and Information Processing Shanghai200240 China School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu611731 China The Department of Computational Mathematics Science and Engineering Michigan State University MI United States

Sign information is the key to overcoming the inevitable saturation error in compressive sensing systems, which causes information loss and results in bias. For sparse signal recovery from saturation, we propose to use a linear loss to improve the effectiveness from existing methods that utilize hard constraints/hinge loss for sign consistency. Due to the use of linear loss, an analytical solution in the update progress is obtained, and some nonconvex penalties are applicable, e.g., the minimax concave penalty, the 0 norm, and the sorted 1 norm. Theoretical analysis reveals that the estimation error can still be bounded. Generally, with linear loss and nonconvex penalties, the recovery performance is significantly improved, and the computational time is largely saved, which is verified by the numerical experiments. Copyright © 2018, The Authors. All rights reserved.

关键词： Compressed sensing

Machine discovery of partial differential equations from spatiotemporal data

学校读者我要写书评

暂无评论

arXiv 2019年

作者： Yuan, Ye Li, Junlin Li, Liang Jiang, Frank Tang, Xiuchuan Zhang, Fumin Liu, Sheng Goncalves, Jorge Voss, Henning U. Li, Xiuting Kurths, Jürgen Ding, Han School of Artificial Intelligence and Automation State Key Laboratory of Digital Manufacturing Equipments and Technology Huazhong University of Science and Technology Wuhan430074 China School of Artificial Intelligence and Automation Key Laboratory of Image Processing and Intelligent Control Huazhong University of Science and Technology Wuhan430074 China School of Electrical Engineering and Computer Science KTH Royal Institute of Technology Stockholm11428 Sweden School of Mechanical Science and Engineering Huazhong University of Science and Technology Wuhan430074 China School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta30332 United States School of Power and Mechanical Engineering Wuhan University Wuhan430074 China Department of Engineering University of Cambridge CambridgeCB2 1PZ United Kingdom Luxembourg Centre for Systems Biomedicine University of Luxembourg BelvauxL-4367 Luxembourg Citigroup Biomedical Imaging Center Weill Cornell Medical College New York10021 United States Research Domain IV - Transdisciplinary Concepts & Methods Potsdam Institute for Climate Impact Research Potsdam14473 Germany School of Mechanical Science and Engineering State Key Laboratory of Digital Manufacturing Equipments and Technology Huazhong University of Science and Technology Wuhan430074 China

The study presents a general framework for discovering underlying Partial Differential Equations (PDEs) using measured spatiotemporal data. The method, called Sparse Spatiotemporal System Discovery (S3d), decides which physical terms are necessary and which can be removed (because they are physically negligible in the sense that they do not affect the dynamics too much) from a pool of candidate functions. The method is built on the recent development of Sparse Bayesian Learning;which enforces the sparsity in the to-be-identified PDEs, and therefore can balance the model complexity and fitting error with theoretical guarantees. Without leveraging prior knowledge or assumptions in the discovery process, we use an automated approach to discover ten types of PDEs, including the famous Navier-Stokes and sine-Gordon equations, from simulation data alone. Moreover, we demonstrate our data-driven discovery process with the Complex Ginzburg-Landau Equation (CGLE) using data measured from a traveling-wave convection experiment. Our machine discovery approach presents solutions that has the potential to inspire, support and assist physicists for the establishment of physical laws from measured spatiotemporal data, especially in notorious fields that are often too complex to allow a straightforward establishment of physical law, such as biophysics, fluid dynamics, neuroscience or nonlinear optics. Copyright © 2019, The Authors. All rights reserved.

关键词： Nonlinear optics

A deep learning-based remaining useful life prediction approach for bearings

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Cheng, Cheng Ma, Guijun Zhang, Yong Sun, Mingyang Teng, Fei Ding, Han Yuan, Ye The Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China The State Key Laboratory of Digital Manufacturing Equipment and Technology Huazhong University of Science and Technology Wuhan430074 China The School of Mechanical Science and Engineering Huazhong University of Science and Technology Wuhan430074 China School of Information Science and Engineering Wuhan University of Science and Technology Wuhan430081 China The College of Control Science and Engineering Zhejiang University Hangzhou310007 China The Department of Electrical & Electronic Engineering Imperial College London LondonSW7 2AZ United Kingdom

In industrial applications, nearly half the failures of motors are caused by the degradation of rolling element bearings (REBs). Therefore, accurately estimating the remaining useful life (RUL) for REBs is of crucial importance to ensure the reliability and safety of mechanical systems. To tackle this challenge, model-based approaches are limited by the complexity of mathematical modeling. Conventional data-driven approaches, on the other hand, require massive efforts to extract the degradation features and construct the health index. In this paper, a novel data-driven framework is proposed to exploit the adoption of deep convolutional neural networks (CNN) in predicting the RULs of bearings. More concretely, raw vibrations of training bearings are first processed using the Hilbert-Huang transform to construct a novel nonlinear degradation energy indicator which can be used as the training label. The CNN is then employed to identify the hidden pattern between the extracted degradation energy indicator and the raw vibrations of training bearings, which makes it possible to estimate the degradation of the test bearings automatically. Finally, testing bearings’ RULs are predicted through using a ǫ-support vector regression model. The superior performance of the proposed RUL estimation framework, compared with the state-of-the-art approaches, is demonstrated through the experimental results. The generality of the proposed CNN model is also validated by performance test on other bearings undergoing different operating conditions. Copyright © 2018, The Authors. All rights reserved.

关键词： Hilbert-Huang transform

Nonconvex penalties with analytical solutions for one-bit compressive sensing

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Huang, Xiaolin Yan, Ming Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University MOE Key Laboratory of System Control and Information Processing Shanghai200240 China Department of Computational Mathematics Science and Engineering Department of Mathematics Michigan State University East LansingMI48824 United States

One-bit measurements widely exist in the real world and can be used to recover sparse signals. This task is known as one-bit compressive sensing (1bit-CS). In this paper, we propose novel algorithms based on both convex and nonconvex sparsity-inducing penalties for robust 1bit-CS. We consider the dual problem, which has only one variable and provides a sufficient condition to verify whether a solution is globally optimal or not. For positive homogeneous penalties, a globally optimal solution can be obtained in two steps: a proximal operator and a normalization step. For other penalties, we solve the dual problem, and it needs to evaluate the proximal operators for many times. Then we provide fast algorithms for finding analytical solutions for three penalties: minimax concave penalty (MCP), 0norm, and sorted 1penalty. Specifically, our algorithm is more than 200 times faster than the existing algorithm for MCP. Its efficiency is comparable to the algorithm for the 1penalty in time, while its performance is much better than 1. Among these penalties, sorted 1is most robust to noise in different settings. Copyright © 2017, The Authors. All rights reserved.

关键词： Compressed sensing

Atorvastatin effect evaluation based on feature combination of three-dimension ultrasound images

学校读者我要写书评

暂无评论

Atorvastatin effect evaluation based on feature combination ...

Medical Imaging 2016: Computer-Aided Diagnosis

作者： Yongkang, Luo Mingyue, Ding Department of Bio-medical Engineering College of Life Science and Technology Image Processing and Intelligence Control Key Laboratory of Education Ministry of China Huazhong University of Science and Technology Wuhan Hubei430074 China

ISBN: (纸本)9781510600201

In the past decades, stroke has become the worldwide common cause of death and disability. It is well known that ischemic stroke is mainly caused by carotid atherosclerosis. As an inexpensive, convenient and fast means of detection, ultrasound technology is applied widely in the prevention and treatment of carotid atherosclerosis. Recently, many studies have focused on how to quantitatively evaluate local arterial effects of medicine treatment for carotid diseases. So the evaluation method based on feature combination was proposed to detect potential changes in the carotid arteries after atorvastatin treatment. And the support vector machine (SVM) and 10-fold cross-validation protocol were utilized on a database of 5533 carotid ultrasound images of 38 patients (17 atorvastatin groups and 21 placebo groups) at baseline and after 3 months of the treatment. With combination optimization of many features (including morphological and texture features), the evaluation results of single feature and different combined features were compared. The experimental results showed that the performance of single feature is poor and the best feature combination have good recognition ability, with the accuracy 92.81%, sensitivity 80.95%, specificity 95.52%, positive predictive value 80.47%, negative predictive value 95.65%, Matthew's correlation coefficient 76.27%, and Youden's index 76.48%. And the receiver operating characteristic (ROC) curve was also performed well with 0.9663 of the area under the ROC curve (AUC), which is better than all the features with 0.9423 of the AUC. Thus, it is proved that this novel method can reliably and accurately evaluate the effect of atorvastatin treatment. © 2016 SPIE.

关键词： Feature extraction

Small universal simple spiking neural P systems with weights

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2014年第9期57卷 19-29页

作者： ZENG XiangXiang PAN LinQiang PREZ-JIMNEZ Mario J. Key Laboratory of Image Processing and Intelligent Control School of AutomationHuazhong University of Science and Technology Department of Computer Science and Artificial Intelligence University of SevillaAvda. Reina Mercedes s/nSevilla 41012Spain

Spiking neural P systems with weights(WSN P systems,for short)are a new variant of spiking neural P systems,where the rules of a neuron are enabled when the potential of that neuron equals a given *** is known that WSN P systems are universal by simulating register ***,in these universal systems,no bound is considered on the number of neurons and *** this work,a restricted variant of WSN P systems is considered,called simple WSN P systems,where each neuron has only one *** complexity parameter,the number of neurons,to construct a universal simple WSN P system is *** is proved that there is a universal simple WSN P system with 48 neurons for computing functions;as generator of sets of numbers,there is an almost simple(that is,each neuron has only one rule except that one neuron has two rules)and universal WSN P system with 45 neurons.

关键词： bio-inspired computing membrane computing P system spiking neural P system universal computing device