A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Prediction of Rock Mass Parameters Based on PCA and Random Forest Method
https://orcid.org/0000-0003-1665-2778
Abstract In this paper, a rock mass parameter prediction model is established based on principal component analysis (PCA) and random forest algorithm (Random Forest). The model is verified by the field data of Gaoligongshan tunnel project of Yunnan Darui Railway. First of all, the variables with the highest weight among the four principal components are selected as the overall input characteristics through PCA dimensionality reduction, that is, total thrust (Th), tunneling speed (PR), penetration (P) and horizontal support cylinder pressure (HP) are selected as the input parameters for modeling. Then, the quartz content, uniaxial saturated compressive strength, integrity coefficient and the number of joints per unit volume of rock mass are predicted respectively. Finally, in order to verify the accuracy of the prediction model proposed in this paper, the prediction results are tested based on mean absolute percentage error (MAPE) and root mean square error (RMSE). It is calculated that the MAPE values are 0.929%, 1.972%, 9.628% and 10.865%, respectively, and the RMSE values are 0.576, 0.889, 0.062 and 1.868, respectively. To sum up, the prediction model based on principal component analysis (PCA) and random forest algorithm proposed in this paper has a prediction accuracy of more than 89% for four key rock mass parameters. The proposed prediction model has been successfully applied in practical engineering, which is of great significance for guiding the TBM construction of Gaoligong Mountain Tunnel.
Prediction of Rock Mass Parameters Based on PCA and Random Forest Method
https://orcid.org/0000-0003-1665-2778
Abstract In this paper, a rock mass parameter prediction model is established based on principal component analysis (PCA) and random forest algorithm (Random Forest). The model is verified by the field data of Gaoligongshan tunnel project of Yunnan Darui Railway. First of all, the variables with the highest weight among the four principal components are selected as the overall input characteristics through PCA dimensionality reduction, that is, total thrust (Th), tunneling speed (PR), penetration (P) and horizontal support cylinder pressure (HP) are selected as the input parameters for modeling. Then, the quartz content, uniaxial saturated compressive strength, integrity coefficient and the number of joints per unit volume of rock mass are predicted respectively. Finally, in order to verify the accuracy of the prediction model proposed in this paper, the prediction results are tested based on mean absolute percentage error (MAPE) and root mean square error (RMSE). It is calculated that the MAPE values are 0.929%, 1.972%, 9.628% and 10.865%, respectively, and the RMSE values are 0.576, 0.889, 0.062 and 1.868, respectively. To sum up, the prediction model based on principal component analysis (PCA) and random forest algorithm proposed in this paper has a prediction accuracy of more than 89% for four key rock mass parameters. The proposed prediction model has been successfully applied in practical engineering, which is of great significance for guiding the TBM construction of Gaoligong Mountain Tunnel.
Prediction of Rock Mass Parameters Based on PCA and Random Forest Method
https://orcid.org/0000-0003-1665-2778
Abstract In this paper, a rock mass parameter prediction model is established based on principal component analysis (PCA) and random forest algorithm (Random Forest). The model is verified by the field data of Gaoligongshan tunnel project of Yunnan Darui Railway. First of all, the variables with the highest weight among the four principal components are selected as the overall input characteristics through PCA dimensionality reduction, that is, total thrust (Th), tunneling speed (PR), penetration (P) and horizontal support cylinder pressure (HP) are selected as the input parameters for modeling. Then, the quartz content, uniaxial saturated compressive strength, integrity coefficient and the number of joints per unit volume of rock mass are predicted respectively. Finally, in order to verify the accuracy of the prediction model proposed in this paper, the prediction results are tested based on mean absolute percentage error (MAPE) and root mean square error (RMSE). It is calculated that the MAPE values are 0.929%, 1.972%, 9.628% and 10.865%, respectively, and the RMSE values are 0.576, 0.889, 0.062 and 1.868, respectively. To sum up, the prediction model based on principal component analysis (PCA) and random forest algorithm proposed in this paper has a prediction accuracy of more than 89% for four key rock mass parameters. The proposed prediction model has been successfully applied in practical engineering, which is of great significance for guiding the TBM construction of Gaoligong Mountain Tunnel.
Prediction of Rock Mass Parameters Based on PCA and Random Forest Method
Hu, Jingjun (author) / Song, Zhicheng (author) / Si, Jingzhao (author) / Cao, Guicai (author) / Nie, Lichao (author) / Chen, Andong (author)
2023
Article (Journal)
Electronic Resource
English
BKL:
57.00$jBergbau: Allgemeines
/
38.58
Geomechanik
/
57.00
Bergbau: Allgemeines
/
56.20
Ingenieurgeologie, Bodenmechanik
/
38.58$jGeomechanik
/
56.20$jIngenieurgeologie$jBodenmechanik
Research on strength prediction of crack rock mass based on random forest algorithm
| Springer Verlag | 2024
Roughness coefficient of rock discontinuities based on random forest regression analyses
| DOAJ | 2023
Rock mass parameters for design
| British Library Conference Proceedings | 2008