http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114997084-A
Outgoing Links
Predicate | Object |
---|---|
assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_d6a6f422b091ba12ea61d4adbf1b0e8e |
classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F2113-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F2119-14 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/Y02E10-20 |
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F30-28 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F30-27 |
classificationIPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F113-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F119-14 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F30-28 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F30-27 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 |
filingDate | 2022-08-01-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_002c3a2467723232e8e114c6ef0b6d36 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_b9126d982681965cd84f99b1783a8980 |
publicationDate | 2022-09-02-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-114997084-A |
titleOfInvention | A method for optimizing the bucket blade profile of an impingement turbine |
abstract | The invention discloses a method for optimizing a bucket blade profile line of an impingement type water turbine, which belongs to the field of neural network prediction. The present invention applies data-driven machine learning to the design of water bucket molding lines, establishes a prediction model between molding line parameters and hydraulic efficiency based on computational fluid dynamics simulation sample data, and further combines the prediction model with a target optimization algorithm. The bucket profile parameters are optimized. Moreover, in the present invention, by performing segmental sampling of dense and sparse in the middle and sparse at the two ends within the range of the profile parameters, at the same time, the importance of the feature parameters is sorted by the method based on a posteriori knowledge, and the neighborhood range encryption is performed based on the preliminary optimal solution. The sampling method constructs training samples close to the global optimal solution, thus ensuring the prediction accuracy of the established prediction model for the parameter space where the global optimal solution is located without excessive consumption of computing resources. The invention can improve the efficiency of the optimal calculation of the bucket blade profile line of the impingement type water turbine. |
priorityDate | 2022-08-01-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 25.