Cement concrete is a critical pavement construction material. However, under prolonged exposure to heavy traffic loads and the combined effects of multiple factors, it frequently exhibits premature slab failure and concurrent multiple defects, severely limiting its service performance and lifespan. The dynamic behavior of cement concrete pavement under heavy-load conditions and the influence of subgrade geometry and pavement width on dynamic bearing performance remain insufficiently understood. To address this issue, this study employs finite element software Abaqus 2020 to construct a three-dimensional finite element model of heavy-duty cement concrete pavement under six typical conditions, including uncut and unfilled subgrade, low embankment, high embankment, cut slope, and different pavement widths. Utilizing an implicit dynamic algorithm, the model simulates and analyzes the acceleration response, dynamic stress distribution, and evolution of strain energy density within the pavement structure under vehicle dynamic loading. The results indicate that the peak acceleration is highest for the uncut subgrade (159.214 m/s2) and lowest under the cut condition (146.566 m/s2), demonstrating that the cutting structure can effectively suppress pavement vibration intensity. Among subgrade types, high embankments exhibit the greatest capacity for reducing strain energy concentration; at the slab corners, the baseline strain energy density of 8.882 J/m3 is reduced by 4.2%, 7.8%, and 5.0% under low embankment, high embankment, and cut conditions, respectively. Regarding pavement width, wider configurations reduce slab vibration intensities, stored strain energy, peak stresses, and stress concentrations, benefiting long-term service life, but concurrently elevate slab corner strain energy accumulation, increasing the risk of corner fracture and compromising load-bearing capacity. These findings provide scientific and technical support for the structural design and performance optimization of heavy-duty cement concrete pavements.
Qu et al. (Fri,) studied this question.