|
Landslide hazards is a common natural disaster, not only which threats human life and safety, and property, environment, resources, and be destructive. How safely,effectivly and reliably analysize slope stability and assessment, prediction of the landslide is a far-reaching significance.Firstly, analysizing factors of slope stability and summary evaluation of slope stability methods, bringing the fuzzy C-means clustering algorithm and rough set theory into the slope stability analysis, study a variety of slope Factors of the samples deeply and evaluate the slope stability. Comparing the evaluation results with the field monitoring results to determine the validity and limitations of each model selection, and then presents a fuzzy C means clustering algorithm and rough set of coupling used in slope stability analysis. Fuzzy C-means clustering algorithm only gives the number of clusters, discrete the measured data, results can reflect the distribution of data more objectively and solve the limitations in discrete choice from standard continuous data. Rough set theory overcomes the traditional method to determine the weight on the deviation caused by the subjective experience. This coupling makes the slope stability evaluation method is more objective and also increased evaluation of the slope stability.In this thesis, combine theory with examples, combine project example of open pit slopes in Yunnan, the fuzzy C-means clustering algorithm and rough set theory coupled model of slope stability to analysize and validate. Analysize natural state of the slope stability, gave slope stability conditions of the open pit, the model is feasible, effective and reasonable. To further analysis and forecast slope stability status in the future rainfall, established fuzzy C-means clustering algorithm and rough set theory coupled model based on rainfall conditions, analysis Slope stability in the rainfall reaches its extreme (saturated) state, to predict the slope stability during rainfall conditions, it is beneficial for engineering to design and governance reasonably.
|