中文摘要：

针对同一系列某花岗岩堆石料不同应力路径加载试验,进行了修正广义塑性模型与修正南水模型的应力路径适应性对比研究。研究结果表明：两类模型均可以很好地预测常规三轴加载试验结果;对于等p加载应力路径试验,二者可以很好地预测强度特征,但是预测的体变偏差较大,两类模型均低估了等p试验下堆石料的体缩特性;对于等应力比加载试验,南水模型预测结果优于广义塑性模型,对剪胀（缩）规律预测较好。广义塑性模型预测结果与试验值相差较大,主要原因在于常规三轴试验下的剪胀方程并不适用于等应力比试验,采用本文建议的修正剪胀方程后,广义塑性模型对等应力比路径试验预测精度明显提高。

英文摘要：

Taking tests of granite rockfill materials under various stress paths（ e. g.,triaxial test,constant p test and constant stress ratio test） as a case study,comparison between the generalized plasticity model and Shen＇s double yield surface model is made to study the adaptability of stress path. It is testified that the deformation of rockfill materials under triaxial stress paths can be well reflected by the two models. The strength prediction results of the two models are consistent with those of the constant p test,while prediction results of volumetric strain have larger deviations. Applying the two models to the constant p test may underestimate the shrinkage properties of rockfill materials. As for the constant stress ratio test,Shen＇s double yield surface model presents better results than the generalized plasticity model,well reflecting dilatancy and shrinkage properties of rockfill materials. However,the prediction results of the generalized plasticity model are quite different from experimental measurements,compared with Shen＇s double yield surface model. Therefore,the dilatancy equation of the conventional triaxial test is not suitable for the constant stress ratio test. The prediction precision of the generalized plasticity model for the constant stress ratio test is greatly improved by employing a modified dilatancy equation.