Neural network model applied to electromagnetic shielding effectiveness of ultra-light Ni/Cu coated polyester fibrous materials

The purpose of effective electromagnetic interference (EMI) shielding is to prevent EMI from smartphone, wireless, and utilization of other electronic devices. The electrical conductivity of materials strongly influences on the EMI shielding properties. In this work, mainly focus to predict the EMI...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Scientific reports 2022-05, Vol.12 (1), p.8609-8609, Article 8609
Hauptverfasser: Periyasamy, Aravin Prince, Muthusamy, Lekha Priya, Militký, Jiri
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The purpose of effective electromagnetic interference (EMI) shielding is to prevent EMI from smartphone, wireless, and utilization of other electronic devices. The electrical conductivity of materials strongly influences on the EMI shielding properties. In this work, mainly focus to predict the EMI shielding effectiveness on the ultralight weight fibrous materials by artificial neural network (ANN). Prior to the ANN modelling, the ultra-lightweight fibrous materials were electroplated with different concentration of Ni/Cu and then coated with different silanes. This work utilizes the algorithm to provide accurate quantitative values of EMI shielding effectiveness (EM SE). To compare its performance, the experimental and the predicted EM SE values were validated by root-mean-square error (RMSE), mean absolute percentage error (MAPE) values and correlation coefficient ‘r’. The proposed ANN results accurately predict the experimental data with correlation coefficients of 0.991 and 0.997. Further due to its simplicity, reliability as well as its efficient computational capability the proposed ANN model permits relatively fast, cost effective and objective estimates to be made of serving in this industry.
ISSN:2045-2322
2045-2322
DOI:10.1038/s41598-022-12593-8