Modeling the weight and length changes of the concrete exposed to sulfate using artificial neural network
Abstract: This paper presents the modeling of an experimental investigation carried out to evaluate some mechanical and durability properties of concrete mixtures in which cement was partially replaced with Class C and Class F fly ash. A total of 39 mixtures with different mix designs were prepared....
Guardado en:
| Autores principales: | , , , |
|---|---|
| Lenguaje: | English |
| Publicado: |
Escuela de Construcción Civil, Pontificia Universidad Católica de Chile
2018
|
| Materias: | |
| Acceso en línea: | http://www.scielo.cl/scielo.php?script=sci_arttext&pid=S0718-915X2018000300337 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| Sumario: | Abstract: This paper presents the modeling of an experimental investigation carried out to evaluate some mechanical and durability properties of concrete mixtures in which cement was partially replaced with Class C and Class F fly ash. A total of 39 mixtures with different mix designs were prepared. After compressive strength testing, the mixtures containing Class F and Class C fly ashes which had similar compressive strength values to control mixtures at 28 d for each series were used for sulfate resistance tests. The degree of sulfate attack was evaluated using expansion and weight loss. The test results indicated that Class C fly ash showed higher compressive strength than Class F fly ash. Moreover, the addition of fly ash significantly increased the resistance to sulfate attack when each amount of fly ash addition regardless of fly ash types was employed. In this paper, the Artificial Neural Network (ANNs) techniques were used to model the relative change in the weight and length of the concrete exposed to sulfate. The best algorithm for length changes of concrete exposed to sulfate is BFGS quasi-Newton backpropagation algorithm while the best algorithm for weight changes of concrete exposed to sulfate is the Levenberg-Marquardt backpropagation algorithm. |
|---|