Jinzhang Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Jinzhang Zhang has a total of 63 co-authors. Here are the top co-authors by number of shared publications:
- Dongming Zhang
Department of Geotechnical Engineering, Tongji University, Shanghai 200092, China
9 shared publications Dongming Zhang
7 shared publicationsHong-Wei Huang
6 shared publicationsHongwei Huang
6 shared publications- Hongwei Huang
Tongji University
6 shared publications Chong Tang
5 shared publications
Latest Publications
- 4 February 2026
- Published by Elsevier in Soil Dynamics and Earthquake Engineering
- Vol. 204
Probabilistic evaluation of seismic active earth pressure in spatially random and nonlinear soil masses
- 4 February 2026
- Published by Elsevier in Soil Dynamics and Earthquake Engineering
- Vol. 204
Probabilistic evaluation of seismic active earth pressure in spatially random and nonlinear soil masses
- Open Access
- 26 December 2025
- Published by MDPI AG in Applied Sciences
- Vol. 16 (1) , 280
Soil Classification from Cone Penetration Test Profiles Based on XGBoost
This study develops a machine-learning-based framework for multiclass soil classification using Cone Penetration Test (CPT) data, aiming to overcome the limitations of traditional empirical Soil Behavior Type (SBT) charts and improve the automation, continuity, robustness, and reliability of stratigraphic interpretation. A dataset of 340 CPT soundings from 26 sites in Shanghai is compiled, and a sliding-window feature engineering strategy is introduced to transform point measurements into local pattern descriptors. An XGBoost-based multiclass classifier is then constructed using fifteen engineered features, integrating second-order optimization, regularized tree structures, and probability-based decision functions. Results demonstrate that the proposed method achieves strong classification performance across nine soil categories, with an overall classification accuracy of approximately 92.6%, an average F1-score exceeding 0.905, and a mean Average Precision (mAP) of 0.954. The confusion matrix, P–R curves, and prediction probabilities show that soil types with distinctive CPT signatures are classified with near-perfect confidence, whereas transitional clay–silt facies exhibit moderate but geologically consistent misclassification. To evaluate depth-wise prediction reliability, an Accuracy Coverage Rate (ACR) metric is proposed. Analysis of all CPTs reveals a mean ACR of 0.924, and the ACR follows a Weibull distribution. Feature importance analysis indicates that depth-dependent variables and smoothed ps statistics are the dominant predictors governing soil behavior differentiation. The proposed XGBoost-based framework effectively captures nonlinear CPT–soil relationships, offering a practical and interpretable tool for high-resolution soil classification in subsurface investigations.
- 26 December 2025
- Published by MDPI AG in Applied Sciences
- Vol. 16 (1) , 280
Open AccessSoil Classification from Cone Penetration Test Profiles Based on XGBoost
This study develops a machine-learning-based framework for multiclass soil classification using Cone Penetration Test (CPT) data, aiming to overcome the limitations of traditional empirical Soil Behavior Type (SBT) charts and improve the automation, continuity, robustness, and reliability of stratigraphic interpretation. A dataset of 340 CPT soundings from 26 sites in Shanghai is compiled, and a sliding-window feature engineering strategy is introduced to transform point measurements into local pattern descriptors. An XGBoost-based multiclass classifier is then constructed using fifteen engineered features, integrating second-order optimization, regularized tree structures, and probability-based decision functions. Results demonstrate that the proposed method achieves strong classification performance across nine soil categories, with an overall classification accuracy of approximately 92.6%, an average F1-score exceeding 0.905, and a mean Average Precision (mAP) of 0.954. The confusion matrix, P–R curves, and prediction probabilities show that soil types with distinctive CPT signatures are classified with near-perfect confidence, whereas transitional clay–silt facies exhibit moderate but geologically consistent misclassification. To evaluate depth-wise prediction reliability, an Accuracy Coverage Rate (ACR) metric is proposed. Analysis of all CPTs reveals a mean ACR of 0.924, and the ACR follows a Weibull distribution. Feature importance analysis indicates that depth-dependent variables and smoothed ps statistics are the dominant predictors governing soil behavior differentiation. The proposed XGBoost-based framework effectively captures nonlinear CPT–soil relationships, offering a practical and interpretable tool for high-resolution soil classification in subsurface investigations.
- Top Cited Papers
- 7 February 2025
- Published by Elsevier in Computers and Geotechnics
- Vol. 181
Probabilistic analysis of ground surface settlement induced by super large diameter shield tunneling based on 3D random finite element method
- 7 February 2025
- Published by Elsevier in Computers and Geotechnics
- Vol. 181
Top Cited PapersProbabilistic analysis of ground surface settlement induced by super large diameter shield tunneling based on 3D random finite element method
- 1 November 2024
- Published by Canadian Science Publishing in Canadian Geotechnical Journal
- Vol. 61 (11) , 2417-2433
Displacement-based design of shield tunnel lining considering mobilized strength of undrained clay
This paper presents a novel displacement–performance design method for the shield tunnel based on the mobilized strength design (MSD) theory. First, the nonlinear relationship between the ground volume loss caused by tunnel excavation and the earth pressure was constructed to obtain the real earth pressure on the tunnel lining. Second, the nonlinear relationship between the convergence displacement of the lining and the ground resistance was established to acquire the subgrade reaction acting on the tunnel lining. Finally, the proposed method was verified by orthogonal numerical experiments and compared with the finite element method (FEM) results. Compared with the FEM, the average deviation of the bending moment and the axial force is about 9% and 1%. The average variation of the bending moment can be reduced to 1% after calibration. The results calculated by this developed method are better than the traditional beam–spring model. Its accuracy is close to the FEM based on the strata–structure concept. Therefore, the proposed method based on MSD theory in this paper has the characteristics of high efficiency and simplicity, which provides a new auxiliary design approach for the structural design of shield tunnels in clay areas.
- 1 November 2024
- Published by Canadian Science Publishing in Canadian Geotechnical Journal
- Vol. 61 (11) , 2417-2433
Displacement-based design of shield tunnel lining considering mobilized strength of undrained clay
This paper presents a novel displacement–performance design method for the shield tunnel based on the mobilized strength design (MSD) theory. First, the nonlinear relationship between the ground volume loss caused by tunnel excavation and the earth pressure was constructed to obtain the real earth pressure on the tunnel lining. Second, the nonlinear relationship between the convergence displacement of the lining and the ground resistance was established to acquire the subgrade reaction acting on the tunnel lining. Finally, the proposed method was verified by orthogonal numerical experiments and compared with the finite element method (FEM) results. Compared with the FEM, the average deviation of the bending moment and the axial force is about 9% and 1%. The average variation of the bending moment can be reduced to 1% after calibration. The results calculated by this developed method are better than the traditional beam–spring model. Its accuracy is close to the FEM based on the strata–structure concept. Therefore, the proposed method based on MSD theory in this paper has the characteristics of high efficiency and simplicity, which provides a new auxiliary design approach for the structural design of shield tunnels in clay areas.
- 10 July 2024
- Published by Springer Nature
- p. 281-287
Three-Dimensional Stratigraphic Reconstruction Based on Improved Coupled Markov Chain
- 10 July 2024
- Published by Springer Nature
- p. 281-287
Three-Dimensional Stratigraphic Reconstruction Based on Improved Coupled Markov Chain
- 10 July 2024
- Published by Springer Nature
- p. 1063-1069
Effect of Adjacent Excavation on Existing Tunnels in Spatially Variable Soil
- 10 July 2024
- Published by Springer Nature
- p. 1063-1069
Effect of Adjacent Excavation on Existing Tunnels in Spatially Variable Soil
- Open Access
Quantitative Study on the Impact of Surcharge on Nearby Foundations
Situated within the context of a soft ground foundation at an iron ore mining site, this study investigates the impact of substantial surcharges on the settlement of such foundations and the adjacent infrastructure. By employing the finite-difference numerical software FLAC3D 6.0, a series of three-dimensional simulations were conducted to assess the stress response and deformation of gallery pile foundations, shallow foundations, and mine shed pile foundations to step loading. This study integrates the analysis of soil strength augmentation under considerable stress and its attenuation characteristics under significant deformation. Various reinforcement measures, such as the implementation of stone columns, prefabricated vertical drain, and surcharge preloading techniques, were examined for their capacity to consolidate the foundation, reduce settlement, and mitigate impacts on adjacent structures. The results reveal that horizontal displacements in the pile and shallow foundations escalate progressively with additional surcharge throughout the operational period. The most pronounced horizontal deviation in the pile foundations is observed at the juncture between sand and silt strata. Stone columns act effectively as a barrier to the sliding surface, consequently reducing the influence of surcharge on the movement of the foundation.
Open AccessQuantitative Study on the Impact of Surcharge on Nearby Foundations
Situated within the context of a soft ground foundation at an iron ore mining site, this study investigates the impact of substantial surcharges on the settlement of such foundations and the adjacent infrastructure. By employing the finite-difference numerical software FLAC3D 6.0, a series of three-dimensional simulations were conducted to assess the stress response and deformation of gallery pile foundations, shallow foundations, and mine shed pile foundations to step loading. This study integrates the analysis of soil strength augmentation under considerable stress and its attenuation characteristics under significant deformation. Various reinforcement measures, such as the implementation of stone columns, prefabricated vertical drain, and surcharge preloading techniques, were examined for their capacity to consolidate the foundation, reduce settlement, and mitigate impacts on adjacent structures. The results reveal that horizontal displacements in the pile and shallow foundations escalate progressively with additional surcharge throughout the operational period. The most pronounced horizontal deviation in the pile foundations is observed at the juncture between sand and silt strata. Stone columns act effectively as a barrier to the sliding surface, consequently reducing the influence of surcharge on the movement of the foundation.
- 9 April 2024
- Published by Springer Nature in International Journal of Civil Engineering
- Vol. 22 (8) , 1339-1356
Experimental and Numerical Study of Braced Retaining Piles with Asymmetrical Excavation
- 9 April 2024
- Published by Springer Nature in International Journal of Civil Engineering
- Vol. 22 (8) , 1339-1356
Experimental and Numerical Study of Braced Retaining Piles with Asymmetrical Excavation