A5055127380- Machine Fault Diagnosis Techniques
- Industrial Vision Systems and Defect Detection
- Advanced Measurement and Detection Methods
- Fault Detection and Control Systems
- Advanced machining processes and optimization
- Optical measurement and interference techniques
- Gear and Bearing Dynamics Analysis
- Engineering Diagnostics and Reliability
- Image Processing Techniques and Applications
- Manufacturing Process and Optimization
- Perovskite Materials and Applications
- Optical Systems and Laser Technology
- Advanced Measurement and Metrology Techniques
- Digital Transformation in Industry
- Advanced Machining and Optimization Techniques
- Structural Health Monitoring Techniques
- Organic Light-Emitting Diodes Research
- Forest Biomass Utilization and Management
- Spectroscopy and Chemometric Analyses
- Surface Roughness and Optical Measurements
- Conducting polymers and applications
- Anomaly Detection Techniques and Applications
- Cloud Computing and Resource Management
- Biofuel production and bioconversion
- Organic Electronics and Photovoltaics
Guilin University of Electronic Technology
2023-2025
Hengyang Normal University
2017-2025
China University of Petroleum, Beijing
2015-2024
Northwestern Polytechnical University
2024
Zhengzhou University of Light Industry
2024
China Mobile (China)
2024
Beijing University of Posts and Telecommunications
2023
Harbin Normal University
2016-2023
Guangdong Ocean University
2023
Henan University
2022-2023
In modern industries, machine health monitoring systems (MHMS) have been applied wildly with the goal of realizing predictive maintenance including failures tracking, downtime reduction, and assets preservation. era big machinery data, data-driven MHMS achieved remarkable results in detection faults after occurrence certain (diagnosis) prediction future working conditions remaining useful life (prognosis). The numerical representation for raw sensory data is key stone various successful...
In modern manufacturing systems and industries, more research efforts have been made in developing effective machine health monitoring systems. Among various approaches, data-driven methods are gaining popularity due to the development of advanced sensing data analytic techniques. However, considering noise, varying length irregular sampling behind sensory data, this kind sequential cannot be fed into classification regression models directly. Therefore, previous work focuses on feature...
With significant advancement in information technologies, Digital Twin has gained increasing attention as it offers an enabling tool to realise digitally-driven, cloud-enabled manufacturing. Given the nonlinear dynamics and uncertainty involved during process of machinery degradation, proper design adaptability a model remain challenge. This paper presents reference for rotating fault diagnosis. The requirements constructing are discussed, updating scheme based on parameter sensitivity...
As network techniques rapidly evolve, attacks are becoming increasingly sophisticated and threatening. Network intrusion detection has been widely accepted as an effective method to deal with threats. Many approaches have proposed, exploring different targeting types of traffic. Anomaly-based is important research development direction detection. Despite the extensive investigation anomaly-based techniques, there lacks a systematic literature review recent datasets. We follow methodology...
Effective machine health monitoring systems are critical to modern manufacturing and industries. Among various approaches, data-driven methods gaining in popularity due the development of advanced sensing data analytic techniques. However, sensory that is a kind sequential can not serve as direct meaningful representations for conditions its noise, varying length irregular sampling. A majority previous models focus on feature extraction/fusion involve expensive human labor high quality...
Digital twin introduces new opportunities for predictive maintenance of manufacturing machines which can consider the influence working condition on cutting tool and contribute to understanding application predicted results. This paper presents a data-driven model digital twin, together with hybrid prediction method based deep learning that creates technique enhanced machining prediction. First, five-dimensional is introduced highlights performance data analytics in construction. Next,...
This paper addresses the cross-domain feature extraction and fusion from time-domain frequency-domain with spectrum envelop preprocessing time domain synchronization average principle using Transfer Component Analysis (TCA) for gearbox fault diagnosis. Considering TCA is developed based on kernel methods, effects of different kernels including Gaussian kernel, Linear Polynomial PolyPlus performance are investigated evaluated in comprehensive experiments testbed under various operating...
Feature extraction is an important step in conventional vibration-based fault diagnosis methods. However, the features are usually empirically extracted, leading to inconsistent performance. This paper presents a new automatic and intelligent method based on convolution neural network. Firstly, vibration signal processed by wavelet transform into multi-scale spectrogram image manifest characteristics. Next, directly fed network learn invariant representation for recognize status diagnosis....
ABSTRACT Gearbox failure becomes a major concern for reliability of wind turbine because complex repair procedures, long downtime and high replacement costs. Prior studies showed that the majority gearbox failures were initiated from bearing failures. Because low signal‐to‐noise ratio (mixture defect signals gear meshing signals) transient nature signals, it poses significant difficulty diagnosis in at incipient stage. To address it, this paper presents an effective fault component...
With movement toward complication and automation, modern machinery equipment encounters the problems of diversity complex origination faults, incipient weak complicated monitoring systems, massive data, which are all challenging current fault diagnosis technologies. Co nventional machine learning techniques, such as support vector back propagation, have disadvantages in handling non-linear relationships structure data. Deep (DL) methods a greater capability to address heterogeneous signals,...