Real-time monitoring and deep learning prediction modeling of rainfall infiltration effects in slope stability risk assessment

In this paper, from the mechanism of rainfall infiltration on slope stability, the effects of rainfall infiltration on the pore space, displacement field and velocity field of the slope were simulated under real-time monitoring conditions. After determining the influencing factors, a slope stability prediction model and an orthogonal test based on 9 factors and 4 levels were constructed to investigate the stability of the slope. Afterwards, the prediction model was optimized by combining BP neural network and alternating iteration global optimization algorithm, and the effect of the model on slope stability under rainfall infiltration conditions was examined. The results show that rainfall will cause the groundwater level of the slope geotechnical body to rise and reach saturation, resulting in a decrease in the rate of rainfall infiltration. After 15 hours of continuous rainfall, the slope is saturated, the surface runoff is no longer infiltrated, and the X-direction displacement of the slope tends to be stabilized. With the beginning of rainfall, the initial (0-15h) slope will appear settlement phenomenon, but 15 hours after the settlement area will eventually through, at this time the deformation rate of the slope X, Y direction tends to stabilize. Under the conditions of nine factors, such as daily rainfall, cohesion, etc., the test values are highly consistent with the actual conclusions, and the test accuracy of the three parallel networks is good (>99%), so it can be seen that the prediction model proposed in this paper can meet the needs of the application of slope stability risk assessment.