【摘要】 为提高三峡库区堆积层滑坡变形趋势判断的准确性,该文以R/S分析和混沌优化PSO-SVM模型为基础,构建了滑坡变形趋势判断模型和变形预测模型,判断和预测三峡库区堆积层滑坡的变形趋势。经实例检验表明:R/S分析模型能很好地评价滑坡的变形趋势,且累计变形序列和速率变形序列的Hurst指数均大于0.5,呈持续增加趋势,但累计变形序列的Hurst指数相对更大,趋势性也相对更强;在预测过程中,混沌优化PSO-SVM模型的平均相对误差均小于2%,最大相对误差也仅为1.83%,具较高预测精度,满足期望要求,且变形预测结果与趋势判断结果具有较好的一致性;通过两实例的综合应用,验证了本文模型的普遍适用性。通过该文研究,相互佐证了趋势判断模型和变形预测模型在滑坡变形规律研究中的有效性,为滑坡变形规律研究提供了一种新的思路。更多还原

【Abstract】 In order to improve the accuracy of judging the deformation trend of accumulative landslide in the Three Gorges Reservoir area, this paper constructs a landslide deformation trend judgment model and a deformation prediction model based on R/S analysis and chaotic optimization PSO-SVM model to judge and predict deformation trend of accumulated layer landslide in Three Gorges Reservoir area. Examples show that the R/S analysis model can evaluate the deformation trend of landslide very well, and the Hurst exponents of the cumulative deformation sequence and the rate deformation sequence are greater than 0.5, showing a continuous increasing trend, but the Hurst exponents of the cumulative deformation sequence are relatively larger and the trend is relatively stronger. In the prediction process, the average relative error of the chaotic optimization PSO-SVM model is less than 2%, and the maximum relative error is less than 2%. The difference is only 1.83%, which has a high prediction accuracy and meets the expectation requirements, and the deformation prediction results are in good agreement with the trend judgment results. Through the comprehensive application of two examples, the universal applicability of the model is verified. Through the research in this paper, the validity of trend judgment model and deformation prediction model in the study of landslide deformation law is proved, which provides a new way of thinking for the study of landslide deformation law.更多还原