http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-108896492-B
Outgoing Links
Predicate | Object |
---|---|
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G01N21-25 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-084 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G01N21-25 |
filingDate | 2018-08-07-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
grantDate | 2020-12-29-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationDate | 2020-12-29-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-108896492-B |
titleOfInvention | PSO-BP neural network model training method, storage medium and terminal |
abstract | The invention discloses a training method, a storage medium and terminal equipment of a PSO-BP neural network model for heavy metal content prediction, wherein the training method comprises the following steps: carrying out heavy metal content measurement treatment on the sample soil to obtain heavy metal content data of the sample soil; carrying out spectral reflectivity processing on the sample soil to obtain a spectral reflectivity curve of the processed sample soil; performing characteristic wave band selection processing according to the heavy metal content data of the sample soil and the spectral reflectivity curve of the processed sample soil to obtain a characteristic wave band of the sample soil; constructing a PSO-BP neural network model; and inputting the constructed PSO-BP neural network model by adopting the characteristic wave band of the sample soil for training and learning until the PSO-BP neural network model converges. In the embodiment of the invention, the accuracy of the estimation and prediction of the heavy metal content in the soil by the model trained by the embodiment of the invention is greatly improved. |
priorityDate | 2018-08-07-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 24.