Prediction model for permanent deformation of railway subgrade using an artificial neural network; [MODELO DE PREVISÃO DA DEFORMAÇÃO PERMANENTE DE FUNDAÇÕES DE VIAS-FÉRREAS COM RECURSO A UMA REDE NEURONAL ARTIFICAL]

• Type: Article in International Scientific Journal
• Date: 2023

Authors

Ramos, A (Author) Other; Correia, AG (Author) Other; Rui Calçada (Author) FEUP

Journal

The Journal is awaiting validation by the Administrative Services.

Title: Geotecnia

Vol. 2023

Pages: 25-41

ISSN: 0379-9522

Indexing

Scopus - 0 Citations

Other information

Authenticus ID: P-00Z-F52

DOI: 10.14195/2184-8394_159_2

Abstract (EN): The prediction of the permanent deformation in the subgrade and its reliability is one of the main concerns of the Railway Infrastructure Managers, as it can influence the reduction of the maintenance costs of the track in service. This study proposes a novel methodology for predicting permanent deformation based on a parametric study performed using a hybrid approach that includes the short and long term performance. The conducted study allowed the construction of a robust database used in this study to forecast the permanent deformation. The database feeds the neural network model, whose performance was evaluated using different metrics: MAE, MSE, RMSE, standard deviation, and regression coefficient. The model was tested and validated based on experimental results. The obtained results demonstrate that the developed model is rapid and efficient in accurately predicting the permanent deformation induced by the passage of trains. The model has the potential to be implemented in a computational decision support system for railway track maintenance and management. © 2023 Sociedade Portuguesa de Geotecnia.

Language: Portuguese

Type (Professor's evaluation): Scientific

No. of pages: 16