A5041454124- Fault Detection and Control Systems
- Machine Fault Diagnosis Techniques
- Industrial Vision Systems and Defect Detection
- Gear and Bearing Dynamics Analysis
- Advanced ceramic materials synthesis
- Solidification and crystal growth phenomena
- Advanced machining processes and optimization
- Advanced Statistical Process Monitoring
- Mineral Processing and Grinding
- Aluminum Alloy Microstructure Properties
- Advanced materials and composites
- Engineering Diagnostics and Reliability
- High Temperature Alloys and Creep
- Manufacturing Process and Optimization
- Metallurgical Processes and Thermodynamics
- Scheduling and Optimization Algorithms
- Aluminum Alloys Composites Properties
- Non-Destructive Testing Techniques
- Advanced Battery Technologies Research
- Spectroscopy and Chemometric Analyses
- Advanced Manufacturing and Logistics Optimization
- Integrated Circuits and Semiconductor Failure Analysis
- Intermetallics and Advanced Alloy Properties
- Structural Health Monitoring Techniques
- Ultrasonics and Acoustic Wave Propagation
Shanghai University
2015-2025
Tongji University
2016-2025
Tianjin Nankai Hospital
2023-2025
Tianjin Medical University
2023-2025
Guangdong Pharmaceutical University
2025
Beijing Anzhen Hospital
2023-2025
Capital Medical University
2023-2025
Huazhong University of Science and Technology
2020-2024
Luoyang Institute of Science and Technology
2024
Peking University Shenzhen Hospital
2024
This paper presents a similarity-based approach for estimating the Remaining Useful Life (RUL) in prognostics. The is especially suitable situations which abundant run-to-failure data an engineered system are available. Data from multiple units of same used to create library degradation patterns. When RUL test unit, it will be matched those patterns and actual life as basis estimation. tackle challenge problem defined by 2008 PHM Challenge Competition, which, unspecified provided set...
Surface defect detection of products is an important process to guarantee the quality industrial production. A task aims identify specific category and precise position in image. It hard take into account accuracy both, which makes it be challenging practice. In this study, a new deep neural network (DNN), RetinaNet with difference channel attention adaptively spatial feature fusion (DEA_RetinaNet), proposed for steel surface detection. First, differential evolution search-based anchor...
Vibration signals are generally utilized for machinery fault diagnosis to perform timely maintenance and then reduce losses. Thus, the feature extraction on one-dimensional vibration often determines accuracy of those models. These typical deep neural networks (DNNs), e.g., convolutional (CNNs), well in learning have been applied machine diagnosis. However, supervised CNN requires a large amount labeled images thus limits its wide applications. In this article, new DNN, residual autoencoder...
The sensitivity of various physical features that are characteristics bearing performance may vary significantly under different working conditions. Thus, it is critical to extract the most effective information from original generated vibration signals for defect classification and degradation assessment. This paper proposes a local nonlocal preserving projection (LNPP)-based feature extraction algorithm, which principal component analysis aims discover global structure Euclidean space...
In semiconductor manufacturing processes, defect detection and recognition in wafer maps have received increasing attention from industry. The various patterns provide crucial information for assisting engineers recognizing the root causes of fabrication problems solving them eventually. This paper develops a manifold learning-based map system. this system, joint local nonlocal linear discriminant analysis (JLNDA) is proposed to discover intrinsic that provides characteristics patterns. An...
Degradation parameter from normal to failure condition of machine part or system is needed as an object health monitoring in condition-based maintenance (CBM). This paper proposes a hidden Markov model (HMM) and contribution-analysis-based method assess the degradation. A dynamic principal component analysis (DPCA) used extract effective features vibration signals, where inherent signal autocorrelation considered. novel assessment indication, HMM-based Mahalanobis distance proposed provide...
Energy-efficient scheduling is highly necessary for energy-intensive industries, such as glass, mould or chemical production. Inspired by a real-world glass-ceramics production process, this paper investigates bi-criteria energy-efficient two-stage hybrid flow shop problem, in which parallel machines with eligibility are at stage 1 and batch machine 2. The performance measures considered makespan total energy consumption. Time-of-use (TOU) electricity prices different states of (working,...
The TianQin-1 satellite (TQ-1), which is the first technology demonstration for TianQin project, was launched on 20 December 2019. round of experiment had been carried out from 21 2019 until 1 April 2020. residual acceleration found to be about $1\times10^{-10}~{\rm m}/{\rm s}^{2}/{\rm Hz}^{1/2}$ at $0.1~{\rm Hz}\,$ and $5\times10^{-11}~{\rm $0.05~{\rm Hz}\,$, measured by an inertial sensor with a sensitivity $5\times10^{-12}~{\rm Hz}\,$. micro-Newton thrusters has demonstrated thrust...
Strip steel is an indispensable material in the manufacturing industry and defects of surface directly determine quality. Due to diversity complexity intraclass between interclass, a great deal manpower resources have been devoted defect detection. This article proposes new deep learning detection network, channel attention, bidirectional feature fusion on fully convolutional one-stage (CABF-FCOS) network achieve rapid effective strips. First, anchor-free FCOS proposed as framework eliminate...
Process signals show the characteristics of large scale, high dimension, and strong correlation in modern industrial processes, which brings a big challenge for process fault detection diagnosis. Due to powerful feature learning ability, deep has been widely used image visual processing. This article proposes new neural network (DNN), convolutional long short-term memory autoencoder (CLSTM-AE) from signals. The LSTM (ConvLSTM) is proposed describe distribution data learn effective features...
Lithium-ion batteries are the main energy source of devices, and estimation their state-of-health (SOH) has become a hot point in prognostics health management. However, many existing methods assume that training testing data follow same distribution. The model based on dataset under one working condition may be ineffective for another due to distribution discrepancy. Thus, this article proposes novel battery prognostic transfer learning. First, learning-based model, called deep domain...