Abstract: In order to explore the simulation method of layered soft rock in model experiment, relying on the indoor test results of samples selected from a layered soft rock tunnel of Jiuzhaigou-Mianyang Expressway, the optimal ratio of barite powder, quartz sand, gypsum powder, talc powder and water was determined by many tests to simulate soft rock matrix. The film with holes was used to simulate the adhesion of weak surface of bedding, and the porosity was determined by direct shear test. And direct shear and uniaxial and triaxial tests were carried out on the samples with different bedding angles and thickness to reflect the anisotropy. The optimal ratio of soft rock matrix simulation is 0.55:0.15:0.07:0.06:0.17. Barite powder plays a decisive role in strength and failure deformation. Barite content is too low for easy crushing and too high for producing top-to-bottom penetration cracks. The film with 30% porosity has the best effect on simulating bedding. The strength of soil sample changes in U-shape with the layer angle, and the strength decreases with the decrease of the thickness of layers (no less than 2 cm). In the direct shear test, the shear plane of 45º bedding is inclined to the direction of bedding, while the shear plane of 90º bedding leads to cracks and fractures on both sides of the shear plane. In uniaxial and triaxial results, for 0º bedding, small angle inclined cracks generate, for 45º bedding inclined cracks perpendicular to the bedding plane and secondary cracks arise, and for 90º bedding vertical splitting along bedding plane emerges. The optimal layer thickness is determined to be 3 cm after compared with the field results. The direct shear test of layer angles and thickness is consistent with the results of uniaxial and triaxial tests, revealing the process of pore compaction closure-elastic-plastic expansion crack failure-creep of soil samples.

Key words: layered soft rock simulation, bedding anisotropy, uniaxial and triaxial test, direct shear test, bedding number