http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113111820-B
Outgoing Links
Predicate | Object |
---|---|
classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F2218-08 |
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-214 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-00 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 |
filingDate | 2021-04-21-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
grantDate | 2022-05-13-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationDate | 2022-05-13-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-113111820-B |
titleOfInvention | Fault diagnosis method and device for rotating parts based on improved CNN and relational module |
abstract | The invention discloses a fault diagnosis method for rotating parts based on an improved CNN and a relationship module. A fault diagnosis metadata set is constructed and divided into a training set and a test set according to fault categories; fast Fourier transform is performed on samples of the original data set; A convolutional neural network diagnostic model consisting of an extraction module, a fusion module, and a relation module is established by three strategies: scaled convolution kernel, random pooling, and atrous convolution; the meta-learning method is used to train the model using the training set; the test set is used to pair The trained model is used for multi-class fault diagnosis of rotating parts with small samples. The invention can self-adaptively train a measure of the distance between samples, and utilize the characteristics of meta-learning to realize the rapid diagnosis of a new fault with only one marked sample, thereby solving the problem that the traditional method relies on large data volume and long-term training. It effectively solves the problem of cross-domain diagnosis of new faults under the condition of small samples. |
priorityDate | 2021-04-21-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 64.