… for solving the few sample problems in intelligent fault diagnosis of RMs. The … fault features
from input samples. The relation module calculates the relation score between the sample …

… of the equipment. Then, the frequency domain features reflecting the operating status of the
equipment are extracted as training samples. Finally, the fault diagnosis model based on the …

… To address the above-mentioned problems, a novel adver… fault diagnosis is proposed in
this paper, which requires only a few source samples and as low as one labeled target sample …

… fault diagnosis is carried out under the following assumptions: (1) There are n s labeled
samples and n t unlabeled samples in … The target samples only cover a subset of fault classes in …

… In this paper, a deep learning-based synthetic over-sampling method is proposed for machinery
fault diagnosis with imbalanced data. Two stages are included in the proposed method. …

… for machinery fault diagnosis and achieved good performance in the scenarios that sufficient
labeled fault samples … of labeled samples, while obtaining sufficient labeled fault samples in …

… diagnose rotating machinery fault. Although the above methods eliminate the error of manual
feature extraction and simplify the process of sample … fault diagnosis for small fault sample …

… require a large number of labeled samples for training. This greatly limits the application of
intelligent computing methods for the machinery fault diagnosis. Moreover, the adaptability of …

… generated fault samples are used as the attacks to diagnostic … training on the original samples
and new samples with slight … This allows the fault data for a specific fault category to live …

… [30] developed a new-type information fusion framework based on matrix space; the
comparison results confirmed its excellent performance in dealing with machinery fault diagnosis …