http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-110514366-B
Outgoing Links
Predicate | Object |
---|---|
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-2411 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-24155 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-253 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G01M3-2815 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/F17D5-02 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-214 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G01M3-28 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/F17D5-02 |
filingDate | 2019-08-22-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
grantDate | 2021-03-05-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationDate | 2021-03-05-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-110514366-B |
titleOfInvention | A weak leak detection method for pipelines under the condition of small samples |
abstract | The invention provides a method for detecting weak pipeline leakage under the condition of small sample, which relates to the technical field of pipeline leakage detection. The steps of the present invention are as follows: Step 1: obtain a real sample set, and generate a virtual sample set according to the real sample set; Step 2: perform combined feature extraction on the real sample set and the virtual sample set, and the combined feature extraction includes 7 kinds of statistical features and 1 set of symbolic transformation features; Step 3: According to the 7 statistical features and 1 set of symbolic transformation features, the Naive Bayesian method and the least squares support vector machine method are used to establish the naive Bayesian network pipeline small leakage identification model and the minimum leakage identification model respectively. The small leakage identification model of the pipeline by the square support vector machine is used to detect the small leakage of the pipeline. The method constructs a weak leak identification model from two aspects: increasing the number of weak leak samples and deeply mining the characteristics of weak leak samples, which greatly improves the detection accuracy of weak leaks in pipelines and ensures the safe operation of oil pipelines. |
priorityDate | 2019-08-22-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Predicate | Subject |
---|---|
isDiscussedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/compound/CID4095 http://rdf.ncbi.nlm.nih.gov/pubchem/substance/SID419507804 |
Total number of triples: 19.