A5047315829- Machine Fault Diagnosis Techniques
- Gear and Bearing Dynamics Analysis
- Engineering Diagnostics and Reliability
- Advanced Measurement and Detection Methods
- Structural Integrity and Reliability Analysis
- Software Engineering Research
- Regional Development and Environment
- Software System Performance and Reliability
- Anomaly Detection Techniques and Applications
- Remote Sensing and Land Use
- Mechanical Failure Analysis and Simulation
- Thermography and Photoacoustic Techniques
- Advanced Sensor and Control Systems
- Software Reliability and Analysis Research
Dalian Polytechnic University
2024
National Administration of Surveying, Mapping and Geoinformation of China
2024
China University of Mining and Technology
2024
Beihang University
2022-2023
Vibration signals collected in real industrial environments are usually limited and unlabeled. In this case, fault diagnosis methods based on deep learning tend to perform poorly. Previous work mainly used the unlabeled data of same diagnostic object improve accuracy, but it did not make full use easily available from different sources. study, a signal momentum contrast for unsupervised representation (SMoCo) contrastive algorithm—momentum visual Learning (MoCo)—is proposed. It can learn how...
To address the limited data problem in real-world fault diagnosis, previous studies have primarily focused on semi-supervised learning and transfer methods. However, these approaches often struggle to obtain necessary data, failing fully leverage potential of easily obtainable unlabeled from other devices. In light this, this paper proposes a novel network architecture, named Signal Bootstrap Your Own Latent (SBYOL), which utilizes vibration signals challenging issues variable working...
Abstract. Cultivated land is the basic resource and material condition for human survival, providing necessary basis agricultural development modernization (Tian Shi, 2024). For this reason, paper takes Beijing as an experimental area to study cultivated extraction method analyze distribution pattern of degree fragmentation. The results show that (1) are evaluated by using confusion matrix Kappa coefficient, coefficient obtained be 0.8358.(2) in mainly distributed southeast well northwest...
Although there are many software reliability growth models, they still face some challenges. Firstly, these models can be divided into two categories: parametric and non-parametric models. The often make assumptions that differ from the actual situation, which affects their generalizability performance. Secondly, although based on deep learning have shown impressive performance, mainly focus recurrent neural networks, prone to problems such as gradient explosion, disappearance long-term...