Deep Learning Models for Estimation of the SuperDARN Cross Polar Cap Potential

Simple item page
dc.contributor.authorLiu, Erxiaoen
dc.contributor.authorHu, Hongqiaoen
dc.contributor.authorLiu, Jianjunen
dc.contributor.authorQiao, Leien
dc.contributor.departmentElectrical and Computer Engineeringen
dc.date.accessioned2020-10-08T12:49:08Zen
dc.date.available2020-10-08T12:49:08Zen
dc.date.issued2020-08en
dc.description.abstractWe present deep learning models for cross polar cap potential (CPCP) by applying multilayer perceptron (MLP) and long short-term memory (LSTM) networks to estimate CPCP based on Super Dual Auroral Radar Network (SuperDARN) measurements. Three statistical parameters are proposed, which are root-mean-square error (RMSE), mean absolute error and linear correlation coefficient (LC), to validate and test the models by measuring their performance on an independent data set that was withheld from the training data set. In addition, we compare the models with previous work. The results show that the deep learning models can successfully reproduce the CPCP values with much lower RMSE (8.41 kV for MLP and 7.20 kV for LSTM) and mean absolute error (7.22 kV for MLP and 6.28 kV for LSTM) and higher LC (0.84 for MLP and 0.90 for LSTM) than do the other models, which have RMSE larger than 10 kV and LC lower than 0.75. The deep learning models can also express the CPCP nonlinear properties (saturation effect) accurately. This study demonstrates that deep learning techniques can enhance the ability to predict CPCP.en
dc.description.notesThis work was supported in part by the National Key R&D Program of China (2018YFC1407300 and 2018YFC1407304), in part by the National Natural Science Foundation of China (Grants 41704154, 41431072, and 41674169) and in part by Opening Fund of SOA Key Laboratory for Polar Science (Grant KP201503). The authors acknowledge the use of SuperDARN data (URL: ), which are freely available through the SuperDARN website at Virginia Polytechnic Institute and State University (URL: ). SuperDARN is a collection of radars funded by the national scientific funding agencies of Australia, Canada, China, France, Italy, Japan, Norway, South Africa, the United Kingdom, and the United States of America. We also acknowledge the use of NASA/GSFC's Space Physics Data Facility's OMNI Web service and OMNI data (URL: ). We thank Professor John Michael Ruohoniemi at Virginia Tech for the fruitful discussions and comments on this research.en
dc.description.sponsorshipNational Key R&D Program of China [2018YFC1407300, 2018YFC1407304]; National Natural Science Foundation of ChinaNational Natural Science Foundation of China (NSFC) [41704154, 41431072, 41674169]; Opening Fund of SOA Key Laboratory for Polar Science [KP201503]; national scientific funding agency of Australia; national scientific funding agency of Canada; national scientific funding agency of China; national scientific funding agency of France; national scientific funding agency of Italy; national scientific funding agency of Japan; national scientific funding agency of Norway; national scientific funding agency of South Africa; national scientific funding agency of the United Kingdom; national scientific funding agency of United States of Americaen
dc.format.mimetypeapplication/pdfen
dc.identifier.doihttps://doi.org/10.1029/2020EA001219en
dc.identifier.eissn2333-5084en
dc.identifier.issue8en
dc.identifier.othere2020EA001219en
dc.identifier.urihttp://hdl.handle.net/10919/100311en
dc.identifier.volume7en
dc.language.isoenen
dc.rightsCreative Commons Attribution 4.0 Internationalen
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/en
dc.titleDeep Learning Models for Estimation of the SuperDARN Cross Polar Cap Potentialen
dc.title.serialEarth and Space Scienceen
dc.typeArticle - Refereeden
dc.type.dcmitypeTexten

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
2020EA001219.pdf
Size:
4.06 MB
Format:
Adobe Portable Document Format
Description:
Download

Collections

Scholarly Works, Electrical and Computer Engineering