Yingli Gao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Yingli Gao has a total of 84 co-authors. Here are the top co-authors by number of shared publications:
Juncai Zhu
8 shared publicationsZuodong Cheng
6 shared publicationsBei He
School of Traffic and Transportation Engineering, Changsha University of Science and Technology, Changsha, Hunan, People’s Republic of China
5 shared publicationsWei Li
5 shared publications- Yuelin Li
Department of Civil Engineering, KU Leuven, Campus Bruges, 8200 Bruges, Belgium
5 shared publications Kairui Duan
School of Traffic and Transportation Engineering, Changsha University of Science and Technology, Changsha, China
4 shared publications
Latest Publications
- Open Access
Leakage-Controlled and Information-Bounded Evaluation of Multi-Task Learning for Crumb–Rubber Modified Asphalt: What Twenty Mix Designs Can and Cannot Support
Data-driven prediction of crumb–rubber modified asphalt (CRMA) binder properties is usually reported on datasets in which a small number of mix-designs is swept across several test temperatures, so that the row count greatly exceeds the number of independent experiments. This study asks what such a dataset can actually support. Using 221 laboratory measurements drawn from 20 independent CRMA mix designs, we evaluate a task-adaptive mixture-of-experts multi-task network (TA-MoE-MTL), a locked Huber-anchored hybrid extension, and fourteen deep and classical reference models for the joint prediction of penetration, softening point, ductility and rutting factor. Three methodological elements are introduced. First, model selection is made strictly nested and group-aware: the stopping epoch is chosen on an inner split of the training designs and the held-out designs are used once. Second, we bound what the recorded inputs can explain before any model is fitted: because the consistency targets are constant within a mix design and because eight designs share identical input vectors while their rutting factors differ, the attainable coefficient of determination for the rutting factor is 0.790 rather than unity. Third, performance is reported with each mix design weighted equally, so that high replication designs cannot dominate. The locked hybrid assigns 90% weight to a Huber-anchored robust expert branch and 10% to a freshly trained TA-MoE-MTL branch. It attains pooled out-of-fold coefficients of determination of 0.613/0.594/0.878/0.685 and ranks first of 16 models at a mean pooled R2 0.693, exceeding MLP–sklearn (0.656) by 0.037. Across three seeds, seeds 42/43/44 give mean pooled R2 values of 0.693/0.689/0.693 (mean 0.691 ± 0.002), and all three runs remain above the frozen MLP sklearn reference. A learning curve over the number of training designs is still rising at the largest size the data allow. The contribution of this work is an evaluation protocol for replicated mix design datasets, a way of bounding their information content, and a robust hybrid that exposes rather than hides the value of a simple small-sample anchor.
Open AccessLeakage-Controlled and Information-Bounded Evaluation of Multi-Task Learning for Crumb–Rubber Modified Asphalt: What Twenty Mix Designs Can and Cannot Support
Data-driven prediction of crumb–rubber modified asphalt (CRMA) binder properties is usually reported on datasets in which a small number of mix-designs is swept across several test temperatures, so that the row count greatly exceeds the number of independent experiments. This study asks what such a dataset can actually support. Using 221 laboratory measurements drawn from 20 independent CRMA mix designs, we evaluate a task-adaptive mixture-of-experts multi-task network (TA-MoE-MTL), a locked Huber-anchored hybrid extension, and fourteen deep and classical reference models for the joint prediction of penetration, softening point, ductility and rutting factor. Three methodological elements are introduced. First, model selection is made strictly nested and group-aware: the stopping epoch is chosen on an inner split of the training designs and the held-out designs are used once. Second, we bound what the recorded inputs can explain before any model is fitted: because the consistency targets are constant within a mix design and because eight designs share identical input vectors while their rutting factors differ, the attainable coefficient of determination for the rutting factor is 0.790 rather than unity. Third, performance is reported with each mix design weighted equally, so that high replication designs cannot dominate. The locked hybrid assigns 90% weight to a Huber-anchored robust expert branch and 10% to a freshly trained TA-MoE-MTL branch. It attains pooled out-of-fold coefficients of determination of 0.613/0.594/0.878/0.685 and ranks first of 16 models at a mean pooled R2 0.693, exceeding MLP–sklearn (0.656) by 0.037. Across three seeds, seeds 42/43/44 give mean pooled R2 values of 0.693/0.689/0.693 (mean 0.691 ± 0.002), and all three runs remain above the frozen MLP sklearn reference. A learning curve over the number of training designs is still rising at the largest size the data allow. The contribution of this work is an evaluation protocol for replicated mix design datasets, a way of bounding their information content, and a robust hybrid that exposes rather than hides the value of a simple small-sample anchor.
- 1 April 2026
- Published by Taylor & Francis in Journal of Sustainable Cement-Based Materials
- Vol. 15 (8) , 2695-2710
Silane-modified nonmetallic powders from waste printed circuit boards in fly ash-based geopolymers: mechanical performance and microstructural characteristics
1. Waste electrical and electronic equipment has increased rapidly in recent years, generating a large amount of waste printed circuit boards [1]. During conventional recycling processes, valuable ...
- 1 April 2026
- Published by Taylor & Francis in Journal of Sustainable Cement-Based Materials
- Vol. 15 (8) , 2695-2710
Silane-modified nonmetallic powders from waste printed circuit boards in fly ash-based geopolymers: mechanical performance and microstructural characteristics
1. Waste electrical and electronic equipment has increased rapidly in recent years, generating a large amount of waste printed circuit boards [1]. During conventional recycling processes, valuable ...
- Open Access
Influence of Polyurethane-Modified Polycarboxylate on Volume Deformation of Hydraulic Concrete
This study investigated the effect of a polyurethane-modified polycarboxylate superplasticizer (P-PCE) on the volume deformation of hydraulic concrete. Macroscopically, the autogenous and drying shrinkage of concrete incorporating different types and dosages of PCEs were measured to analyze their influence. Microscopically, scanning electron microscopy (SEM) was employed to observe the hydration product morphology at 7 and 28 days. Low-field nuclear magnetic resonance (NMR) was utilized to quantify the pore structure, and a fractal dimension model was applied to correlate the microstructural characteristics with the macroscopic deformation. The results demonstrated that, compared to conventional PCEs, the laboratory-synthesized P-PCE (40% solid content) significantly reduced shrinkage and improved pore structure, thereby enhancing the volumetric stability of hydraulic concrete. The experimental results showed that, compared to ordinary PCE, P-PCE reduced the 60-day autogenous-shrinkage strain by 8.8% and the drying-shrinkage strain by 8.4%. Additionally, it decreased the total porosity by 19.46%, while also optimizing the pore structure distribution, thereby significantly improving the volume stability of hydraulic concrete.
Open AccessInfluence of Polyurethane-Modified Polycarboxylate on Volume Deformation of Hydraulic Concrete
This study investigated the effect of a polyurethane-modified polycarboxylate superplasticizer (P-PCE) on the volume deformation of hydraulic concrete. Macroscopically, the autogenous and drying shrinkage of concrete incorporating different types and dosages of PCEs were measured to analyze their influence. Microscopically, scanning electron microscopy (SEM) was employed to observe the hydration product morphology at 7 and 28 days. Low-field nuclear magnetic resonance (NMR) was utilized to quantify the pore structure, and a fractal dimension model was applied to correlate the microstructural characteristics with the macroscopic deformation. The results demonstrated that, compared to conventional PCEs, the laboratory-synthesized P-PCE (40% solid content) significantly reduced shrinkage and improved pore structure, thereby enhancing the volumetric stability of hydraulic concrete. The experimental results showed that, compared to ordinary PCE, P-PCE reduced the 60-day autogenous-shrinkage strain by 8.8% and the drying-shrinkage strain by 8.4%. Additionally, it decreased the total porosity by 19.46%, while also optimizing the pore structure distribution, thereby significantly improving the volume stability of hydraulic concrete.
- Open Access
- 10 January 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (2) , 739
Numerical Investigation of Micromechanical Failure Evolution in Rocky High Slopes Under Multistage Excavation
High rock slopes are extensively distributed in areas of major engineering constructions, such as transportation infrastructure, hydraulic projects, and mining operations. The stability and failure evolution mechanism during their multi-stage excavation process have consistently been a crucial research topic in geotechnical engineering. In this paper, a series of two-dimensional rock slope models, incorporating various combinations of slope height and slope angle, were established utilizing the Discrete Element Method (DEM) software PFC2D. This systematic investigation delves into the meso-mechanical response of the slopes during multi-stage excavation. The Parallel Bond Model (PBM) was employed to simulate the contact and fracture behavior between particles. Parameter calibration was performed to ensure that the simulation results align with the actual mechanical properties of the rock mass. The research primarily focuses on analyzing the evolution of displacement, the failure modes, and the changing characteristics of the force chain structure under different geometric conditions. The results indicate that as both the slope height and slope angle increase, the inter-particle deformation of the slope intensifies significantly, and the shear band progressively extends deeper into the slope mass. The failure mode transitions from shallow localized sliding to deep-seated overall failure. Prior to instability, the force chain system exhibits an evolutionary pattern characterized by “bundling–reconfiguration–fracturing,” serving as a critical indicator for characterizing the micro-scale failure mechanism of the slope body.
- 10 January 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (2) , 739
Open AccessNumerical Investigation of Micromechanical Failure Evolution in Rocky High Slopes Under Multistage Excavation
High rock slopes are extensively distributed in areas of major engineering constructions, such as transportation infrastructure, hydraulic projects, and mining operations. The stability and failure evolution mechanism during their multi-stage excavation process have consistently been a crucial research topic in geotechnical engineering. In this paper, a series of two-dimensional rock slope models, incorporating various combinations of slope height and slope angle, were established utilizing the Discrete Element Method (DEM) software PFC2D. This systematic investigation delves into the meso-mechanical response of the slopes during multi-stage excavation. The Parallel Bond Model (PBM) was employed to simulate the contact and fracture behavior between particles. Parameter calibration was performed to ensure that the simulation results align with the actual mechanical properties of the rock mass. The research primarily focuses on analyzing the evolution of displacement, the failure modes, and the changing characteristics of the force chain structure under different geometric conditions. The results indicate that as both the slope height and slope angle increase, the inter-particle deformation of the slope intensifies significantly, and the shear band progressively extends deeper into the slope mass. The failure mode transitions from shallow localized sliding to deep-seated overall failure. Prior to instability, the force chain system exhibits an evolutionary pattern characterized by “bundling–reconfiguration–fracturing,” serving as a critical indicator for characterizing the micro-scale failure mechanism of the slope body.