Research On The Parameter Estimate And The Reliability Analysis Of Engineering Slope Based On The Principle Of Maximum Entropy

The destabilization of slope, one of the three global geology damages to the world, which brings great lost to the possession and life of human beings, has attracted widely attention of the world. Safety coefficient is usually used to evaluate the slope stability in our country at present. In this method, engineers often fix the soil's parameters by experience based on sampling at random, without considering various uncertain factors. So the method has some subjectivity features.The analysis method for slopes based on the statistics and probability theory is researched in this paper. Some significant results in theory and practice are gained in the field of the soil's parameter estimate and reliability analysis.In this paper, several methods used for reliability analysis are introduced and contrasted first. Monte—Carlo stimulation method has a higher degree of assurance to satisfy the engineering needs. For the estimate of parameters, illuminated by the Principle of Maximum Entropy in thermodynamics, the method of parameter estimate based on the Principle of Maximum Entropy is researched. By deducing the criterion functions of the random back analysis by this method, the "natural" parameter was gained. The paper described the computational process of Monte—Carlo stimulation with the parameters estimated by the Principle of Maximum Entropy in detail as well. This process gets the probability of failure and the reliability index to evaluate the stability of slopes. The feasibility and correctness of the method is validated by its use in an example with quantities of statistics data at last.