Zero-Shot Compound Fault Diagnosis Method Using Semantic Construction and Embedding
- 30 September 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-13
Abstract
Compound fault diagnosis has been a challenge in rotating machinery health management. Current compound fault diagnosis methods rely on a substantial number of labeled samples, which are impractical to obtain for every type of compound fault and pose safety risks when collected in industrial settings. To address these issues, we have built a bearing fault dataset that includes compound faults under multiple conditions. We propose a zero-shot compound fault identification model, which can be established solely using vibration data from single faults under no-load conditions and is suitable for multiple unknown load conditions. Firstly, we design a compound fault logical semantic (CFLS) construction method based on statistical indicators from vibration data of single faults under no-load conditions. Secondly, we develop a multi-head self-attention based feature extraction approach that extracts features from vibration data under various load conditions. Thirdly, we train a semantic embedding model using the semantics and features derived from single faults. By inputting the extracted features of compound faults into the model, we can get the compound fault dynamic semantic (CFDS) under multiple load conditions. Finally, the compound faults are identified by using semantics regression to compare the Euclidean distance between CFLS and CFDS. This method has been validated on two cases using both self-built HDU dataset and the Paderborn dataset. The results show that it can recognize compound faults across multiple unknown domains and the identification accuracy has reached 78.17%.Keywords
Funding Information
- National Natural Science Foundation of China (U22A2047)
- Science and Technology Key Project of Zhejiang Province of China (2021C01141)
- Joint Funds of the Zhejiang Provincial Natural Science Foundation of China (LHZY24A010006)
This publication has 33 references indexed in Scilit:
- Machinery cross domain degradation prognostics considering compound domain shiftsReliability Engineering & System Safety, 2023
- A zero-shot fault semantics learning model for compound fault diagnosisExpert Systems with Applications, 2023
- A variational local weighted deep sub-domain adaptation network for remaining useful life prediction facing cross-domain conditionReliability Engineering & System Safety, 2022
- Zero-shot learning for compound fault diagnosis of bearingsExpert Systems with Applications, 2022
- Actual bearing compound fault diagnosis based on active learning and decoupling attentional residual networkMeasurement, 2021
- Deep transductive network for generalized zero shot learningPattern Recognition, 2020
- A review of failure modes, condition monitoring and fault diagnosis methods for large-scale wind turbine bearingsMeasurement, 2020
- A comparison of random forest variable selection methods for classification prediction modelingExpert Systems with Applications, 2019
- Mechanical model development of rolling bearing-rotor systems: A reviewMechanical Systems and Signal Processing, 2018
- Gaussian kernel optimization for pattern classificationPattern Recognition, 2009