Forecasting SYM‐H Index: A Comparison Between Long Short‐Term Memory and Convolutional Neural Networks
Abstract Forecasting geomagnetic indices represents a key point to develop warning systems for the mitigation of possible effects of severe geomagnetic storms on critical ground infrastructures. Here we focus on SYM‐H index, a proxy of the axially symmetric magnetic field disturbance at low and midd...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-02-01
|
| Series: | Space Weather |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2020SW002589 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841536403179044864 |
|---|---|
| author | F. Siciliano G. Consolini R. Tozzi M. Gentili F. Giannattasio P. De Michelis |
| author_facet | F. Siciliano G. Consolini R. Tozzi M. Gentili F. Giannattasio P. De Michelis |
| author_sort | F. Siciliano |
| collection | DOAJ |
| description | Abstract Forecasting geomagnetic indices represents a key point to develop warning systems for the mitigation of possible effects of severe geomagnetic storms on critical ground infrastructures. Here we focus on SYM‐H index, a proxy of the axially symmetric magnetic field disturbance at low and middle latitudes on the Earth's surface. To forecast SYM‐H, we built two artificial neural network (ANN) models and trained both of them on two different sets of input parameters including interplanetary magnetic field components and magnitude and differing for the presence or not of previous SYM‐H values. These ANN models differ in architecture being based on two conceptually different neural networks: the long short‐term memory (LSTM) and the convolutional neural network (CNN). Both networks are trained, validated, and tested on a total of 42 geomagnetic storms among the most intense that occurred between 1998 and 2018. Performance comparison of the two ANN models shows that (1) both are able to well forecast SYM‐H index 1 h in advance, with an accuracy of more than 95% in terms of the coefficient of determination R2; (2) the model based on LSTM is slightly more accurate than that based on CNN when including SYM‐H index at previous steps among the inputs; and (3) the model based on CNN has interesting potentialities being more accurate than that based on LSTM when not including SYM‐H index among the inputs. Predictions made including SYM‐H index among the inputs provide a root mean squared error on average 42% lower than that of predictions made without SYM‐H. |
| format | Article |
| id | doaj-art-5885dbba84a245dd8ce12b1e4710dd01 |
| institution | Kabale University |
| issn | 1542-7390 |
| language | English |
| publishDate | 2021-02-01 |
| publisher | Wiley |
| record_format | Article |
| series | Space Weather |
| spelling | doaj-art-5885dbba84a245dd8ce12b1e4710dd012025-01-14T16:30:33ZengWileySpace Weather1542-73902021-02-01192n/an/a10.1029/2020SW002589Forecasting SYM‐H Index: A Comparison Between Long Short‐Term Memory and Convolutional Neural NetworksF. Siciliano0G. Consolini1R. Tozzi2M. Gentili3F. Giannattasio4P. De Michelis5Department of Computer, Control and Management Engineering Antonio Ruberti Sapienza University of Rome Rome ItalyINAF‐Istituto di Astrofisica e Planetologia Spaziali Rome ItalyIstituto Nazionale di Geofisica e Vulcanologia Rome ItalyDepartment of Computer, Control and Management Engineering Antonio Ruberti Sapienza University of Rome Rome ItalyIstituto Nazionale di Geofisica e Vulcanologia Rome ItalyIstituto Nazionale di Geofisica e Vulcanologia Rome ItalyAbstract Forecasting geomagnetic indices represents a key point to develop warning systems for the mitigation of possible effects of severe geomagnetic storms on critical ground infrastructures. Here we focus on SYM‐H index, a proxy of the axially symmetric magnetic field disturbance at low and middle latitudes on the Earth's surface. To forecast SYM‐H, we built two artificial neural network (ANN) models and trained both of them on two different sets of input parameters including interplanetary magnetic field components and magnitude and differing for the presence or not of previous SYM‐H values. These ANN models differ in architecture being based on two conceptually different neural networks: the long short‐term memory (LSTM) and the convolutional neural network (CNN). Both networks are trained, validated, and tested on a total of 42 geomagnetic storms among the most intense that occurred between 1998 and 2018. Performance comparison of the two ANN models shows that (1) both are able to well forecast SYM‐H index 1 h in advance, with an accuracy of more than 95% in terms of the coefficient of determination R2; (2) the model based on LSTM is slightly more accurate than that based on CNN when including SYM‐H index at previous steps among the inputs; and (3) the model based on CNN has interesting potentialities being more accurate than that based on LSTM when not including SYM‐H index among the inputs. Predictions made including SYM‐H index among the inputs provide a root mean squared error on average 42% lower than that of predictions made without SYM‐H.https://doi.org/10.1029/2020SW002589forecastingneural networksgeomagnetic stormsSYM‐H index |
| spellingShingle | F. Siciliano G. Consolini R. Tozzi M. Gentili F. Giannattasio P. De Michelis Forecasting SYM‐H Index: A Comparison Between Long Short‐Term Memory and Convolutional Neural Networks Space Weather forecasting neural networks geomagnetic storms SYM‐H index |
| title | Forecasting SYM‐H Index: A Comparison Between Long Short‐Term Memory and Convolutional Neural Networks |
| title_full | Forecasting SYM‐H Index: A Comparison Between Long Short‐Term Memory and Convolutional Neural Networks |
| title_fullStr | Forecasting SYM‐H Index: A Comparison Between Long Short‐Term Memory and Convolutional Neural Networks |
| title_full_unstemmed | Forecasting SYM‐H Index: A Comparison Between Long Short‐Term Memory and Convolutional Neural Networks |
| title_short | Forecasting SYM‐H Index: A Comparison Between Long Short‐Term Memory and Convolutional Neural Networks |
| title_sort | forecasting sym h index a comparison between long short term memory and convolutional neural networks |
| topic | forecasting neural networks geomagnetic storms SYM‐H index |
| url | https://doi.org/10.1029/2020SW002589 |
| work_keys_str_mv | AT fsiciliano forecastingsymhindexacomparisonbetweenlongshorttermmemoryandconvolutionalneuralnetworks AT gconsolini forecastingsymhindexacomparisonbetweenlongshorttermmemoryandconvolutionalneuralnetworks AT rtozzi forecastingsymhindexacomparisonbetweenlongshorttermmemoryandconvolutionalneuralnetworks AT mgentili forecastingsymhindexacomparisonbetweenlongshorttermmemoryandconvolutionalneuralnetworks AT fgiannattasio forecastingsymhindexacomparisonbetweenlongshorttermmemoryandconvolutionalneuralnetworks AT pdemichelis forecastingsymhindexacomparisonbetweenlongshorttermmemoryandconvolutionalneuralnetworks |