Prediction of thermal conductivity of granitic rock: an application of arithmetic and salp swarm algorithms optimized ANN.
Saved in:
| Title: | Prediction of thermal conductivity of granitic rock: an application of arithmetic and salp swarm algorithms optimized ANN. |
|---|---|
| Authors: | Lawal, Abiodun Ismail1,2 (AUTHOR), Kwon, Sangki1 (AUTHOR) kwonsk@inha.ac.kr, Kim, Minju1 (AUTHOR), Aladejare, Adeyemi Emman3 (AUTHOR), Onifade, Moshood4 (AUTHOR) |
| Source: | Earth Science Informatics. Dec2022, Vol. 15 Issue 4, p2303-2317. 15p. |
| Subject Terms: | *Thermal conductivity, *Granite, *Arithmetic, *Rock properties, *Drill core analysis, *Nondestructive testing |
| Abstract: | Thermal conductivity (TC) is an important rock property as it determines its energy transfer potential. Compared with other rock properties like uniaxial comprehensive strength (UCS), it is rarely investigated. Hence, novel Arithmetic and Salp swarm optimized artificial neural network (ANN) models are used to predict the thermal conductivity of granitic rock based on the results of non-destructive tests. Fifty (50) core samples were obtained from the study location and tested in the laboratory. The results obtained from the laboratory investigations were used to perform the ordinary ANN and the optimized ANN models. The outcomes showed that the performances of the optimized ANN models are better than the ordinary ANN model. The results were also compared with the multiple linear regression model (MLR) although the predictive strength of the MLR model is extremely low. The proposed models were mathematically transformed into simple mathematical models, and a graphic user interface (GUI) prepared with the Visual basic programming language was developed. The proposed models can be practically implemented for TC prediction. [ABSTRACT FROM AUTHOR] |
| Database: | Energy & Power Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Thermal conductivity (TC) is an important rock property as it determines its energy transfer potential. Compared with other rock properties like uniaxial comprehensive strength (UCS), it is rarely investigated. Hence, novel Arithmetic and Salp swarm optimized artificial neural network (ANN) models are used to predict the thermal conductivity of granitic rock based on the results of non-destructive tests. Fifty (50) core samples were obtained from the study location and tested in the laboratory. The results obtained from the laboratory investigations were used to perform the ordinary ANN and the optimized ANN models. The outcomes showed that the performances of the optimized ANN models are better than the ordinary ANN model. The results were also compared with the multiple linear regression model (MLR) although the predictive strength of the MLR model is extremely low. The proposed models were mathematically transformed into simple mathematical models, and a graphic user interface (GUI) prepared with the Visual basic programming language was developed. The proposed models can be practically implemented for TC prediction. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 18650473 |
| DOI: | 10.1007/s12145-022-00880-x |
