ISSN 1003-8035 CN 11-2852/P
    冯谕,曾怀恩,涂鹏飞. 遗传算法下的滑坡蠕滑位移预测模型研究[J]. 中国地质灾害与防治学报,2024,35(1): 82-91. DOI: 10.16031/j.cnki.issn.1003-8035.202209038
    引用本文: 冯谕,曾怀恩,涂鹏飞. 遗传算法下的滑坡蠕滑位移预测模型研究[J]. 中国地质灾害与防治学报,2024,35(1): 82-91. DOI: 10.16031/j.cnki.issn.1003-8035.202209038
    FENG Yu,ZENG Huaien,TU Pengfei. Research on prediction model of landslide creep displacement on genetic algorithm[J]. The Chinese Journal of Geological Hazard and Control,2024,35(1): 82-91. DOI: 10.16031/j.cnki.issn.1003-8035.202209038
    Citation: FENG Yu,ZENG Huaien,TU Pengfei. Research on prediction model of landslide creep displacement on genetic algorithm[J]. The Chinese Journal of Geological Hazard and Control,2024,35(1): 82-91. DOI: 10.16031/j.cnki.issn.1003-8035.202209038

    遗传算法下的滑坡蠕滑位移预测模型研究

    Research on prediction model of landslide creep displacement on genetic algorithm

    • 摘要: 滑坡位移预测是预报滑坡灾害的重要依据,以往的滑坡位移预测模型多数为时间序列预测模型、BP神经网络预测模型、Gaussian拟合预测模型以及其他一些非线性预测模型。这些滑坡位移预测模型在建立上缺乏力学理论支撑,对不同力学特性产生的滑坡位移预测分析上没有针对性。文章针对力学特性为重力蠕变型滑坡位移的预测,提出一种基于遗传优化算法的滑坡蠕滑位移非线性预测模型。以鲁家坡滑坡东侧J05监测点的累计水平位移为例,划定测试区域与预测区域进行模型预测分析,并将新模型预测结果与Gaussian拟合预测模型、BP神经网络预测模型预测结果进行对比分析。结果表明,相较于传统预测模型,新模型的预测效果有所提升,有一定的工程价值与实践价值。

       

      Abstract: Landslide displacement prediction is an important basis of predicting landslide disasters. Most of the previous landslide displacement prediction models include time series prediction models, BP neural network prediction models, Gaussian fitting prediction models, and various other nonlinear prediction models. However, these landslide displacement prediction models lack the foundation of mechanical theory in the establishment and have no pertinence in the prediction and analysis of landslide displacement resulting from diverse mechanical properties. In this paper, a nonlinear prediction model of landslide creep displacement based on genetic optimization algorithm is proposed for the prediction of gravity creep landslide displacement. Using the cumulative horizontal displacement data from monitoring point J05 on the eastern side of the Lujiapo landslide as a case study, the test area and prediction area are delimited for model prediction analysis. The results of the new model are compared with those of Gaussian fitting model and BP neural network model. The results indicate that, in comparison to the traditional prediction models, the new model exhibits improved predictive performance, offering a certain engineering value and practical value.

       

    /

    返回文章
    返回