<p>In this study, the fundamental mechanical properties of silt in Yellow River flooded area were analyzed, and the effects of various influencing factors on its mechanical behavior were considered. A fractional-order elasto-plastic constitutive model was established to reflect the state-dependent behavior and non-associated flow of the soil. Shear-dilation relationships based on the Riemann–Liouville (R–L) and Caputo fractional derivatives were derived. The gradient non-orthogonality of fractional differentials was utilized to describe the non-orthogonal relationship between the plastic flow direction and the yield surface, thereby avoiding the limitations of traditional elasto-plastic models that require separate definitions for loading and plastic potential surfaces. The proposed model contains ten parameters, each with a clear physical meaning and ease of determination. A parameter calibration method was introduced, and a sensitivity analysis was carried out to evaluate the influence of individual parameters on the model’s predictive accuracy. Finally, the predictive performances of the R–L and Caputo derivative forms were compared and analyzed. The results have demonstrated that the fractional-order elasto-plastic model is capable of accurately reproducing the key mechanical characteristics of silt in Yellow River flooded area, including compressibility, shear dilation, and state dependence. The R–L model and Caputo model have similar accuracy in simulating the strain-hardening law, but the R–L fractional constitutive model will underestimate the dilatancy of silt in Yellow River flooded area when predicting the development of volumetric strain.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Influence of different fractional calculus forms on the accuracy of elastic–plastic constitutive model for silt

  • Fengming Zhou,
  • Yuke Wang,
  • Xinqi Zhao,
  • Nengbo Cai,
  • Quanling Li,
  • Yu Li

摘要

In this study, the fundamental mechanical properties of silt in Yellow River flooded area were analyzed, and the effects of various influencing factors on its mechanical behavior were considered. A fractional-order elasto-plastic constitutive model was established to reflect the state-dependent behavior and non-associated flow of the soil. Shear-dilation relationships based on the Riemann–Liouville (R–L) and Caputo fractional derivatives were derived. The gradient non-orthogonality of fractional differentials was utilized to describe the non-orthogonal relationship between the plastic flow direction and the yield surface, thereby avoiding the limitations of traditional elasto-plastic models that require separate definitions for loading and plastic potential surfaces. The proposed model contains ten parameters, each with a clear physical meaning and ease of determination. A parameter calibration method was introduced, and a sensitivity analysis was carried out to evaluate the influence of individual parameters on the model’s predictive accuracy. Finally, the predictive performances of the R–L and Caputo derivative forms were compared and analyzed. The results have demonstrated that the fractional-order elasto-plastic model is capable of accurately reproducing the key mechanical characteristics of silt in Yellow River flooded area, including compressibility, shear dilation, and state dependence. The R–L model and Caputo model have similar accuracy in simulating the strain-hardening law, but the R–L fractional constitutive model will underestimate the dilatancy of silt in Yellow River flooded area when predicting the development of volumetric strain.