Highway subgrade bears millions of traffic loads over the years, and its strength, stiffness, and long-term stability gradually decline. In this paper, dynamic triaxial tests were carried out to study the time evolution and spatial distribution of strain and pore pressure of highway-subgrade soil under the action of traffic load. The influence of traffic load on subgrade deformation was analyzed. Furthermore, a numerical-calculation model of the subgrade was established. The deformation characteristics of subgrade under driving load were analyzed. The main conclusions can be drawn as follows: (1) With the increase in loading times, the cumulative strain and pore pressure can be roughly divided into three stages: rapid-growth stage, slow-growth stage, and equilibrium stage. (2) The influence of traffic load on the cumulative strain and pore-water pressure of subgrade soil decreases rapidly with the increase in depth. (3) The amplitude of traffic load has a tremendous influence on the strain and pore pressure of subgrade soil, especially for shallow subgrade. (4) As the distance from the subgrade surface increases, the maximum deformation appears at the edge of the subgrade.

Highway widening has become an effective way to improve the carrying capacity of existing expressways. However, assume that the differential settlement between the new and the old subgrades is not controlled. In that case, it will lead to longitudinal cracks on the pavement, which will seriously affect the safety of vehicles. In this study, the variations of elastic modulus and dynamic elastic modulus of new and old subgrades under traffic load were studied by tests. A hyperbolic model of dynamic elastic modulus increasing with strain is proposed. The numerical calculation model of new and old subgrades under traffic load is established. The deformation characteristics of new and old subgrades without treatment measures and setting up geogrid and geocell are studied. The main conclusions are as follows: (1) with the increase of traffic load and subgrade depth, the elastic modulus changes significantly in the early stage and small in the later stage. (2) Under the condition of no treatment measures, the vertical and horizontal deformation of the subgrade surface show a trend of increasing and then decreasing with the distance from the center line of the subgrade. (3) Setting geogrid and geocell can effectively reduce the differential settlement of the old and new subgrades. (4) The effect of setting geogrid to reduce the differential settlement of old and new subgrades is better than that of setting geogrid.

Key core technology innovation has become an important strategic path for countries to maintain industrial security amid intensifying global technological competition. As an important innovation paradigm, R&D collaboration is generally regarded as an effective way to achieve such innovation. However, the key issue of which collaborative mechanism is most effective at promoting key core technology innovation remains insufficiently explored. Therefore, systematically comparing the effectiveness of different mechanisms of collaborative innovation is of great strategic significance for achieving key core technology innovation and overcoming Western technological blockades. In this study, the R&D level and market share of key core technology were incorporated into an analytical framework and applied to a differential game focused on the innovation behaviors of leading enterprises, supporting enterprises, and academic research institutions under Nash non-collaborative, cost-sharing, and collaborative mechanisms. A simulation analysis was conducted using the MATLAB 2020a software. The results show that the optimal strategies for the key core technology innovation of innovation entities are negatively correlated with the cost coefficient, discount rate, technology, and market recession coefficient. Meanwhile, they are positively correlated with the sensitivity coefficient of technology R&D and market promotion. Furthermore, the R&D levels and market shares of key core technology are highest under the collaborative mechanism. In this scenario, the revenues of the innovation entity and the overall system reach Pareto optimality. Within a threshold range, the cost-sharing mechanism significantly improves innovative efforts, the R&D level, and the market share of key core technology, leading to a Pareto improvement for both the participants’ and overall system’s revenues compared to the non-collaborative mechanism. This study not only contributes to theoretical results of differential games but also provides valuable suggestions for policymakers and innovation entities to foster key core technology innovation from the perspective of collaboration.

With advantages in robustness to variations in ambient illumination and active sensing, light detection and ranging has become a primary modality for environmental perception in underground coal mines characterized by inadequate illumination and high dust concentrations. However, raw point clouds lack semantic attributes, which limits their direct application in scene understanding for digital mines. Although fully supervised deep learning methods for point cloud semantic segmentation achieve high performance, they rely heavily on labor-intensive point-wise annotations, incurring high costs and potential human error. To address this, this paper proposes a weakly supervised semantic segmentation framework specifically designed for coal mine roadways. The framework establishes a comprehensive workflow spanning from data acquisition to sparse-supervised training. By leveraging sparse labels to drive model optimization, the method maintains segmentation performance while significantly reducing annotation costs, achieving a trade-off between engineering precision and deployment overhead. Experimental results demonstrate that using only 0.1% sparse annotations, the proposed method achieves an overall accuracy of 91.64% and a mean intersection over union of 78.83%. Compared to fully supervised models, it reduces the annotation time by a factor of 2–3 times improvement in annotation efficiency with a precision loss of approximately 7%. The segmented structural point clouds are utilized for roadway deformation monitoring and parametric geometric reconstruction. This supports a comprehensive digital twin framework that enables high-fidelity modeling and early warning of centimeter-scale deformation. This research confirms that the proposed technology provides a cost-effective and efficient path for scene understanding in intelligent mines, demonstrating practical value and potential for practical deployment.