Research on a Multi-Dimensional Information Fusion Mechanical Wear Fault-Diagnosis Algorithm Based on Data Regeneration
Open Access
Abstract
Under laboratory conditions for recording a small amount of data, the characteristics of the phenomena distribution become a limitation of machine learning and advanced deep learning concepts for the diagnosis and localization of mechanical wear faults. In this paper, we adopt the combination of the diffusion model and TTT (test-time training), based on the sample distribution of feature data under the laboratory conditions, and we use the pre-trained decoder to decode the data into a continuous potential representation of natural language for sampling, to achieve data regeneration. Subsequently, the TTT algorithm becomes a model with weights in the hidden state itself. The gradient step on the self-supervised loss is selected as the update rule, which is trained synchronously during the testing time, adhering to the concept of migration learning, to construct a high-dimensional mapping relationship between the feature parameters and the failure modes of the mechanical wear. The final validation results show that the diagnosis accuracy reaches more than 95% for six types of typical aero-engine mechanical wear faults.Keywords
Funding Information
- National Science and Technology Major Program (Y2022-V-0002-0028, J2019-V-0003-0094, 2023A1515110319, 2025A04J3757, 242102210171)
- Guangdong Basic and Applied Basic Research Foundation (Y2022-V-0002-0028, J2019-V-0003-0094, 2023A1515110319, 2025A04J3757, 242102210171)
- Science and Technology Projects in Guangzhou (Y2022-V-0002-0028, J2019-V-0003-0094, 2023A1515110319, 2025A04J3757, 242102210171)
- key science and technology program of henan province (Y2022-V-0002-0028, J2019-V-0003-0094, 2023A1515110319, 2025A04J3757, 242102210171)
This publication has 65 references indexed in Scilit:
- Research on the prediction algorithm of aero engine lubricating oil consumption based on multi-feature information fusionApplied Intelligence, 2024
- Use of machine learning models in condition monitoring of abrasive belt in robotic arm grinding processJournal of Intelligent Manufacturing, 2024
- Digital Twin for wear degradation of sliding bearing based on PFENNAdvanced Engineering Informatics, 2024
- Bearing Fault Diagnosis Using Machine Learning Models2024
- Comparison of ML Algorithms and Neural Networks on Fault Diagnosis of a Worm GearJournal of Vibration Engineering & Technologies, 2024
- Image deep learning in fault diagnosis of mechanical equipmentJournal of Intelligent Manufacturing, 2023
- Intelligent fault diagnostic system for rotating machinery based on IoT with cloud computing and artificial intelligence techniques: a reviewSoft Computing, 2023
- Tool wear condition monitoring across machining processes based on feature transfer by deep adversarial domain confusion networkJournal of Intelligent Manufacturing, 2023
- A spatio-temporal fault diagnosis method based on STF-DBN for reciprocating compressorJournal of Intelligent Manufacturing, 2022
- Challenges in Engine Health Monitoring Instrumentation During Developmental Testing of Gas Turbine Engines2020