A5101659590- Advanced machining processes and optimization
- Advanced Machining and Optimization Techniques
- Industrial Vision Systems and Defect Detection
- Machine Fault Diagnosis Techniques
- Machine Learning and ELM
- Non-Destructive Testing Techniques
- Metal Alloys Wear and Properties
- Engineering Diagnostics and Reliability
- Spectroscopy and Chemometric Analyses
- Semiconductor materials and devices
- Electromagnetic Simulation and Numerical Methods
- Mineral Processing and Grinding
- Advanced Algorithms and Applications
- Semiconductor materials and interfaces
- Advanced Measurement and Detection Methods
- Surface and Thin Film Phenomena
- Graphene research and applications
- Integrated Circuits and Semiconductor Failure Analysis
- Tunneling and Rock Mechanics
- Advanced Sensor and Control Systems
- Particle accelerators and beam dynamics
- Lubricants and Their Additives
- Fault Detection and Control Systems
- NMR spectroscopy and applications
- Carbon Nanotubes in Composites
Wenzhou University
2014-2023
North China Electric Power University
2013
Institute of Microelectronics
2012
Chevron (Netherlands)
2002-2004
Recent advances in artificial intelligence (AI) technology have led to increasing interest the development of AI-based tool condition monitoring (TCM) methods. However, achieving good performance using these methods relies heavily on large training samples, which are both expensive and difficult obtain practical TCM applications. This article addresses this issue by employing a much smaller sample composed non-exhaustive sampling experimentally measured cutting force signals conjunction with...
Tool condition monitoring (TCM) is of great importance for improving the manufacturing efficiency and surface quality workpieces. Data-driven machine learning methods are widely used in TCM have achieved many good results. However, actual industrial scenes, labeled data not available time target domain that significantly affect performance data-driven methods. To overcome this problem, a new method combining Markov transition field (MTF) deep adaptation network (DDAN) proposed. A few...
In this paper, a new approach is proposed based on data fusing with vibration signals using time-frequency parameters, probabilistic principal component analysis (PPCA) and statistical inference, for improving the accuracy visibility of damage identification numerical control (NC) machine tools. Time-frequency feature components are put forward, which extracted from eight dimensionless parameters statistically in time frequency domains by PPCA. The Chi-2 statistic established according to...
Rolling bearings are crucial mechanical components in the industry. Timely intervention and diagnosis of system faults essential for reducing economic losses ensuring product productivity. To further enhance exploration unlabeled time-series data conduct a more comprehensive analysis rolling bearing fault information, this paper proposes technique based on graph node-level information extracted from 1D vibration signals. In technique, 10 categories signals sampled using sliding window...
This paper focuses on the fault diagnosis for NC machine tools and puts forward a method based kernel principal component analysis (KPCA) and<mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" id="M2"><mml:mrow><mml:mi>k</mml:mi></mml:mrow></mml:math>-nearest neighbor (<mml:math id="M3"><mml:mrow><mml:mi>k</mml:mi></mml:mrow></mml:math>NN). A data-dependent KPCA covariance matrix of sample data is designed to overcome subjectivity in parameter selection function used transform original...
In this paper, the flame pyrolysis method to synthesize self-oriented carbon nanotubes (CNTs) with Fe/Mo/Al2O3 as catalyst is studied. The outcome arrays are well aligned and relatively good in morphology. Furthermore, after electric field being applied synthesis, become straighter longer significantly, higher yield.
Timely and effective identification monitoring of tool wear is important for the milling process. However, traditional methods estimation have run into difficulties due to under small samples with less prior knowledge. This article addresses this issue by employing a multisensor method based on blind source separation technology. Stationary subspace analysis (SSA) technology applied transform signals stationary nonstationary sources without information signals. Ten dimensionless...
Transfer learning (TL) has been successfully implemented in tool condition monitoring (TCM) to address the lack of labeled data real industrial scenarios. In current TL models, domain offset joint distribution input feature and output label still exists after two domains is aligned, resulting performance degradation. A multiple spatial alignment (MSDA) method proposed, Including Correlation for deep adaptation (Deep CORAL) Joint maximum mean difference (JMMD). Deep CORAL employed learn...
Recent advances in artificial intelligence (AI) technology have led to increasing interest the development of AI-based tool wear condition monitoring methods, heavily relying on large training samples. However, high cost experiment and uncertainty change machining process lead problems sample missing insufficiency model stage, which seriously affects identification accuracy many AI models. In this paper, a novel method based finite-element modeling (FEM) synthetic minority oversampling...
Timely and accurate recognition of tool wear condition could reduce the machining cost greatly. However, problem sample missing imbalance affects classification accuracy AI models seriously. A novel strategy based on finite element method (FEM) support vector machine (SVM) is proposed to overcome above problem. Firstly, several experiments are carried out obtain experimental samples. Then, a FEM model established verified through Based model, simulated imbalanced samples in be supplemented...
Tool wear condition monitoring (TCM) is an important part of machining automation. In recent years, deep learning (DL)-based TCM methods have been widely researched. However, almost all DL-based need sufficient samples to obtain good accuracy, which hard for in terms cost and time. order enhance the recognition accuracy under small samples, this paper proposed a new improved multi-scale edge-labelling graph neural network (MEGNN). Firstly, signal cutting force sensor expanded...