摘要
本次研究以河西走廊典型粗颗粒硫酸盐渍土地基为对象,通过室内重塑土样的溶陷特性试验,揭示粗颗粒盐渍土渗透系数及溶滤变形系数两个指标的影响因素及影响规律。试验模拟了150k Pa压力下上述两个指标在"含盐量—孔隙比"坐标平面上变化的神经网络曲面及其在平面内投影的等高线图。试验结果表明:经过冻融循环之后,盐分对周围土颗粒起到胶结作用,试样渗透系数降低,抗渗性能提高;孔隙比范围在0.43~0.52之间时试样的溶滤变形系数最大;当荷载超过150k Pa时,应力对溶滤变形系数的影响较为显著,而荷载小于150k Pa时,含盐量是溶滤变形系数的主控因素。
Taking the typical coarse grain sulfate saline soil in the Hexi Corridor as the study example,and through a series of melt sinking tests for the remodeling soil samples in laboratory,the influence of various factors on the permeability coefficient and the coefficient of deformation due to the leaching are studied. The experiment has simulated the neural network surface and the contour map of the two coefficients on the salt content-porosity ratio coordinate plane under the pressure 150 k Pa. The results show that the salt plays a role in cementation to the surrounding soil particles after the freezing and thawing cycles,and the permeability resistance is improved. When the porosity ratio is in the range of 0. 43 -0. 52,the coefficient of leaching deformation is maximum. When the load is more than 150 k Pa,the stress has a significant influence on the coefficient of leaching deformation,while the load is less than 150 k Pa,the salt content is the main influence factor for the coefficient.
出处
《工程勘察》
2017年第3期22-28,共7页
Geotechnical Investigation & Surveying
关键词
粗颗粒
盐渍土
溶陷
神经网络
coarse grain
saline soil
dissolution
neural network
