| Autor: |
Yafei Hu, Keqing Li, Bo Zhang, Bin Han |
| Předmět: |
|
| Zdroj: |
Journal of Renewable Materials; 2023, Vol. 11 Issue 9, p3463-3484, 22p |
| Abstrakt: |
The super-fine particle size of tailings is its drawback as a recycled resource, which is reflected in the low strength of the new construction and industrial materials formed when it is mixed with cement and other cementitious materials. Therefore, it is crucial to study the effect of tailings particle size and cementitious material on the strength of tailings wet shotcrete (TWSC) and to investigate the optimal mix proportion. In this paper, a multivariate nonlinear response model was constructed by conducting central composite experiments to investigate the effect of different factors on the strength of TWSC. The strength prediction and mix proportion optimization of TWSC are carried out by machine learning techniques. The results show that the response model has R² > 0.94 and P < 0.01, which indicates that the model has high reliability. Moreover, the strength of TWSC increases with the increase of tailings fineness modulus and decrease of water-binder ratio, while it also increases and then decreases with the increase of replacement rate of slag powder to cement (SRC rate). The extreme learning machine (ELM) constructed in this paper predicts the strength of TWSC with an accuracy of more than 98% and achieves rapid prediction under multi-factor conditions. It is worth mentioning that the ELM combined with the genetic algorithm (ELM-GA) collaboratively solved to obtain the mix proportion for C15 and C20 strength grades of TWSC and the maximum error is verified by experiments to be less than 2%. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
