Publicação
Validation of a rail temperature model with experimental measurements
| Resumo: | Rail temperature is a key factor when studying the effects of thermal buckling. Many models have been developed to simulate rail temperatures under various weather conditions. This work is based on the model developed by the Chungnam National University (CNU), which includes the shadow effect on the rail and the solar position to improve the temperature prediction during several periods of the day, validates it with experimental data, and compares it with a finite element model. Furthermore, a python library is developed based on the lumped thermal model with small adaptations, called railtemp. The python package has slightly better performance over the original CNU model, reaching a correlation factor R2 of 0.947 and a root mean square error of 2.6°C. Furthermore, a new proposal is presented to determine the temperatures on rail tracks based on air temperature. |
|---|---|
| Autores principais: | Piloto, P.A.G. |
| Outros Autores: | Frigeri, Ary Vinicius Nervis; Minhoto, Manuel; Silva, Dyorgge A. |
| Assunto: | Railway python Temperature Weather Finite element analysis Python Buckling |
| Ano: | 2022 |
| País: | Portugal |
| Tipo de documento: | artigo |
| Tipo de acesso: | acesso restrito |
| Instituição associada: | Instituto Politécnico de Bragança |
| Idioma: | inglês |
| Origem: | Biblioteca Digital do IPB |
| Resumo: | Rail temperature is a key factor when studying the effects of thermal buckling. Many models have been developed to simulate rail temperatures under various weather conditions. This work is based on the model developed by the Chungnam National University (CNU), which includes the shadow effect on the rail and the solar position to improve the temperature prediction during several periods of the day, validates it with experimental data, and compares it with a finite element model. Furthermore, a python library is developed based on the lumped thermal model with small adaptations, called railtemp. The python package has slightly better performance over the original CNU model, reaching a correlation factor R2 of 0.947 and a root mean square error of 2.6°C. Furthermore, a new proposal is presented to determine the temperatures on rail tracks based on air temperature. |
|---|