Rotating machinery plays an important role in industrial systems, and faults in the machinery may damage the system health. A novel image-based diagnosis method using improved deep convolutional generative adversarial networks (DCGAN) is proposed for the feature recognition and fault classification of rotating machinery. First, vibration signal data from the rotating machinery is transformed into time-frequency feature 2-D image data by a continuous wavelet transform and used for fault classification with the neural network method. The adaptive deep convolution neural network (ADCNN) is then combined with the generative adversarial networks (GANs) to improve the performance of the feature self-learning ability from input data. Compared with different fault diagnosis methods, the proposed method has better performance for image feature classification in rotating machinery.
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC9570832 | PMC |
| http://dx.doi.org/10.3390/s22197534 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Analysis of the interaction mechanism between straw soil rotary tillage blade during rotary tillage process.
Sci Rep
January 2025
College of Mechanical and Electrical Engineering, Shihezi University, Shihezi, 832003, China.
In response to the rotary ploughing equipment in the stubble land to implement protective operations, the stubble is large in number and strong in toughness, not easy to crush, resulting in rotary ploughing equipment to produce entanglement and increased resistance to rotary ploughing and other issues. In this study, researchers designed a bionic rotary tillage blade (B-RTB) based on the bionic structural equations of the Marmota claw. A straw-soil complex shear performance test was conducted to investigate the effect of straw on soil shear strength.
View Article and Find Full Text PDFFault diagnosis of rotating parts integrating transfer learning and ConvNeXt model.
Sci Rep
January 2025
Department of Power Engineering, Naval University of Engineering, Wuhan, 430033, Hubei, China.
This paper proposes a fault diagnosis method for rotating machinery that integrates transfer learning with the ConvNeXt model (TL-CoCNN), addressing challenges such as small sample sizes and varying operating conditions. To meet the input requirements of the model while minimizing feature loss, an alternative approach to visualizing vibration data is introduced. Specifically, RGB images are synthesized from time-domain, frequency-domain, and time-frequency domain representations of the original signal, which are subsequently used as the input dataset.
View Article and Find Full Text PDFA fault diagnosis method for rolling bearing based on gram matrix and multiscale convolutional neural network.
Sci Rep
December 2024
College of Information Engineering, SuQian University, SuQian, 223800, China.
The safety and reliability of rotating machinery hinge significantly on the proper functioning of rolling bearings. In the last few years, there have been significant advances in the algorithms for intelligent fault diagnosis of bearings. However, the vibration signals collected by machines are inevitably affected by irrelevant noise because of the complex working environments of bearings.
View Article and Find Full Text PDFA zero-shot attribute-embedded model with a feature difference mapping sigmoid function for compound fault diagnosis of rotating machinery.
ISA Trans
December 2024
State Key Laboratory of Mechanical Transmission for Advanced Equipment, Chongqing University, Chongqing 400044, PR China. Electronic address:
Article Synopsis
- Current methods for detecting machinery compound faults struggle due to the lack of available training data, as collecting sufficient compound fault samples is often impractical in engineering.
- The paper introduces a zero-shot attribute-embedded model (ZSAECFD), which allows for diagnosing unseen compound faults using only single fault data by constructing attribute prototypes and utilizing a new activation function, F-sigmoid.
- The model demonstrates high diagnostic accuracy—81.82% for bearing faults and 88.17% for gear faults—showing its effectiveness compared to traditional methods, even without training on compound fault data.
Design and validation of a real-time cassava planter seed quality monitoring system based on optical fiber sensors and rotary encoders.
Front Plant Sci
December 2024
Agricultural Machinery Research Institute, Chinese Academy of Tropical Agricultural Sciences, Zhanjiang, Guangdong, China.
Targeting the issues of seed leakage and cutting segment adhesion due to poor seed feeding and cutting in real-time seed-cutting cassava planters, this study developed a seeding quality monitoring system. Based on the structure and working principle of the seed cutting and discharging device, the installation methods of the matrix fiber optic sensor and rotary encoder were determined. By combining the operational characteristics of the planter's ground wheel drive with seed cutting and seed dropping, a monitoring model correlating the sowing parameters with seed dropping time was established; a monitoring window was created by extracting and processing the rotary encoder pulse signal, and the number of seeds sown after each opposing cutter's operation was calculated based on the pulse width information within the monitoring window.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!