TZTianqin Zeng

Tianqin Zeng

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Tianqin Zeng has a total of 3 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Bridging Heterogeneous Experimental Data and Soil Mechanics: An Interpretable Machine Learning Framework for Displacement-Dependent Earth Pressure

    by Tianqin Zeng, Zhe Zhang, Yongge Zeng

    Classical earth pressure theories often struggle to account for the complex coupling effects of wall displacement and spatial non-uniformity under non-limit states. This study presents an interpretable machine learning framework designed to extract universal mechanical laws from heterogeneous experimental datasets. Using a multi-source database of rigid retaining walls with sandy backfill, a three-stage feature refinement strategy is proposed that incorporates Recursive Feature Elimination, Collinearity Analysis, and Interpretability Comparison to identify a parsimonious set of five fundamental physical parameters. A SHapley Additive exPlanations-Categorical Boosting (CatBoost-SHAP) framework is established to predict the active earth pressure coefficient (K) and interpret the underlying mechanisms across various movement modes (RB, RT, and T). Results demonstrate that the model effectively captures the progressive evolution of shear bands and the soil arching effect. Specifically, a critical displacement threshold of Δ/H ≈ 0.006 is identified, marking the transition from mode-dominated stress non-uniformity to magnitude-driven limit states. Leave-One-Dataset-Out Cross-Validation (LODOCV) confirms the model’s ability to maintain physical consistency over purely statistical fitting despite significant inter-literature heterogeneity. Finally, a Graphical User Interface (GUI) is developed to facilitate rapid, displacement-based design in engineering practice. This research bridges the gap between empirical laboratory observations and generalized mechanical logic, providing a data-driven foundation for refined geotechnical design.

  • Open Access

    Precise Analysis and Prediction of Active Earth Pressure for Retaining Walls Based on Explainable Machine Learning

    by Tianqin Zeng, Zhe Zhang, Yongge Zeng

    The classical Rankine and Coulomb theories frequently encounter difficulties in accurately modeling the complex, nonlinear, and displacement-coupled behavior of earth pressure on retaining walls under non-limit states. The present st udy proposes a “key feature refinement strategy based on collinearity analysis” and employs the said strategy by applying it to model test data. The strategy identified an optimum set of five physical parameters , namely displacement mode (DM), relative displacement (Δ/H), relative depth (Z/H), unit weight (γ), and internal friction angle (φ). A machine learning (ML) model has been developed that integrates Categorical Boosting with SHapley Additive exPlanations (CatBoost-SHAP) . This model has been found to exhibit a marked enhancement in accuracy (R² = 0.917) when compared to classical theories , while concurrently offering the distinct advantage of explicit interpretability. SHAP analysis has been demonstrated to elucidate the nonlinear influence of each parameter . It is confirm ed that displacement mode is identified as the governing factor for spatial pressure distribution, and classical mechanisms such as top‑down stress relaxation in the rotation-about-the-base ( RB ) mode and soil arching in the rotation-about-the-base ( RT ) mode are visualized . Furthermore , a displacement‑dependent mechanical threshold (Δ/H ≈ 0.006) has been identified, which marks the transition from a mode‑dominated to displacement‑driven pressure evolution. In addition, t he proposed approach is integrated in to a graphical user interface (GUI ) that is designed to be user‑friendly, thereby furnishing practitioners with a precise tool for designing retaining walls . The validation of the model's performance against independent experimental results has demonstrated its superior agreement and practical utility under displacement-controlled conditions in comparison to conventional methods.

  • Open Access

    Explainable Prediction of UHPC Tensile Strength Using Machine Learning with Engineered Features and Multi-Algorithm Comparative Evaluation

    by Zhe Zhang, Tianqin Zeng, Yongge Zeng, Ping Zhu

    To explore a direct predictive model for the tensile strength of ultra-high-performance concrete (UHPC), machine learning (ML) algorithms are presented. Initially, a database comprising 178 samples of UHPC tensile strength with varying parameters is established. Then, feature engineering strategies are proposed to optimize the robustness of ML models under a small-sample condition. Further, the performance and efficiency of algorithms are compared under default hyperparameters and hyperparameter tuning, respectively. Moreover, the utilization of SHapley Additive exPlanations (SHAP) enables the analysis of the relationships between UHPC tensile strength and its influencing factors. The quantitative analysis results indicate that ensemble algorithms exhibit superior performance, indicated by R2 values of above 0.92, under default hyperparameters. After hyperparameter tuning, both conventional and ensemble models achieve R2 values exceeding 0.94. However, Bayesian ridge regression (BRR) consistently demonstrates a suboptimal performance, irrespective of hyperparameter tuning. Notably, Categorical Boosting (CatBoost) requires a substantial duration of 1208 s, which is notably more time-consuming than that of other algorithms. The most influential feature identified is fiber reinforcement index with a contribution of 37.5%, followed by the water-to-cement ratio, strain rate, and cross-sectional size. The nonlinear relationship between UHPC tensile strength and the top four factors is visualized, and the critical thresholds are identified.