Safety Evaluate Compositive Analysis System On Rock Slope

Landslide has already became one of the three most geo-hazards, the others were earthquake and volcano. Because of slide, there are a lot of economy loss and personnel casualty. Recently, the western part of our country, there are a lot of rock slopes in there, so the safety research on stability analysis of slope is more important day after day. Because of that, developed safety estimate compositive system on rock slope in this paper. It not only can analyze safety of rock slope, but also can predict the displacement in short time and in long time, confirm the safety condition. The system has two parts: the safety analysis of rock slope and the displacement predicting of rock slope.There are lots of joints in rock slope, safety of rock slope is controlled by joints, but there are a few of joints whose direction is vertical, a majority of joints' direction is not vertical, so the Sarma is adopted to be the safety analysis method of rock slope. We didn't suppose several critical slip surfaces before confirming the safety factor, using the program which is edited by myself to search the critical slip surface. The safety analysis subsystem searches the critical slip surface based on the least shearing strength and A* algorithm. Before searching the critical slip surface, using 3-D Joint Network Simulated program to recurrence joint distribute in the slope and point joints. After the critical slip surface confirmed, the subsystem separates inclining the slope automatically and computes the slope safety factor.About the displacement predicting subsystem of slope, adopting neural network developed by simulated algorithm. Because that neural network has better nonlinear and parallel proceeding, it adapts to process displacement predicting. Neural network cant convergence at the least point and the network structure design is more subjective. In this paper, using simulated algorithm to overcome the disadvantage of BP neural network, and constitute the developed neural network model. In additional, using dynamic forecasting technique to reduce the influence of the prophase displacement, so it can achieve better forecast precision.We used the safety estimate compositive system on rock slope to analyze the "3.29 landslide" slope in Tian Huangping. The result is coincident to the real condition of the slope, so the safety estimate compositive system on rock slope is relatively viable.