Landslides have an effect on lives and properties in many different countries [1], making control of the effects of natural hazards -especially landslides -an urgent problem [2]. Landslide can cause secondary damage, rockslide-dammed lakes [3][4], tsunamis [5][6][7], and block roads [8]. There are so many factors leading to landslides, such as hydrology [9], earthquakes, top loads, pedology, different layers, geomorphy [10], and geology-changing climate [11].A slope consists of soil and rock. Previous research treated geomaterials as homogeneous materials and did not separate soils and rock. Different geomaterials have different affects on slope stability. Mahmood and Kim [12] analyzed the effect of soil type on matric suction and stability of unsaturated slope, using the modified MohrCoulomb failure criterion and the saturated-unsaturated seepage model, and drew upon the fact that saturatedunsaturated hydraulic conductivity and matric suction head affect the matric suction results. Tiwari and Ajmera Pol. J. Environ. Stud. Vol. 26, No. 6 (2017), 2747-2757 AbstractRock-soil aggregate landslides are distributed all over the world and have done great harm to transportation networks, buildings, personal safety, and city construction. Although landslide studies usually focus on the slope of the single homogeneous material, few slopes are composed of or covered by various complicated geomaterials. This paper proposes a calculation model of the slope covered by rock-soil aggregate and analyzes Longhuguan landslide in Guangxi, China, considering rainfall. Fully considering the weak surface, a support design plan is given.The unbonded cable is taken to support the landslide, the concrete beams are adopted on the angle turning point of the surface, and the slide-resistant piles are used to jointly support the front edge of the slope. Top fissures are blocked and landslide drainage measures are taken. The displacement monitoring proves that the support is effective. This study can provide many references for landslide support and analysis covered by rock-soil aggregate.

At present, deep learning technology shows great market potential in broaching tool wear state recognition based on vibration signals. However, traditional single neural network structure is difficult to extract a variety of different features simultaneously and has low robustness, so the accuracy of wear status recognition is not high. In view of the above problems, a broaching tool wear recognition model based on ShuffleNet v2.3-StackedBiLSTM is proposed in this paper. The model integrates ShuffleNet v2.3, which has been channel shuffling, and StackedBiLSTM, a long and short-term memory network, to effectively extract spatial and temporal features for tool wear state recognition. Based on the innovative recognition model, the turbine disc fir-tree slot broaching experiment is designed, and the performance index system based on confusion matrix is adopted. The experimental research and results show that the model has outstanding accuracy, precision, recall, and F1 value, and the accuracy rate reaches 99.37%, which is significantly better than ShuffleNet v2.3 and StackedBiLSTM models. The recognition speed of a single sample was improved to 8.67 ms, which is 90.32% less than that of the StackedBiLSTM model.