| GM—BP串联组合预测模型在滑坡变形监测中的应用 |
| |
| 引用本文： | 曹冬冬,徐秋艳.GM—BP串联组合预测模型在滑坡变形监测中的应用[J].中国煤炭地质,2012(2):47-49,60. |
| |
| 作者姓名： | 曹冬冬 徐秋艳 |
| |
| 作者单位： | [1]长安大学地质工程与测绘学院,陕西西安710054 [2]陕西省煤田地质局,陕西西安710054 |
| |
| 摘 要： | 为了提高滑坡的预测精度，通过对灰色GM（1，1）模型与BP神经网络模型各自优缺点及互补性的分析，建立了GM—BP串联组合预测模型。模型首先采用等维动态GM（1，1）模型进行初步预测，然后利用BP神经网络对初步预测的结果进行训练及仿真，通过数据的归一化处理，参数的判定选取，获得组合模型预测值。以茅坪滑坡为例，对位移进行了预测。通过数据的对比分析，发现GM—BP串联组合预测模型在短期预测精度上高于单一模型。
|
| 关 键 词： | 等维动态 归一化 GM—BP组合预测模型 滑坡预测 |
Application of GM-BP Tandem Combination Forecast Model to Landslide Deformation |
| |
| Cao Dongdong,Xu Qiuyan.Application of GM-BP Tandem Combination Forecast Model to Landslide Deformation[J].Coal Geology of China,2012(2):47-49,60. |
| |
| Authors: | Cao Dongdong Xu Qiuyan |
| |
| Institution: | 1. College of Geology Engineering and Geomatics, Changan University, Xian, Shaanxi 710054; 2. Shaanxi Bureau of Coal Geological Exploration, Xian, Shaanxi 710054) |
| |
| Abstract: | In order to improve the prediction precision of the landslide, the GM-BP tandem combination forecast model is established through analysis on pros and cons ＆ complementary of both grey model and BP neural network model. The model start with a preliminary prediction by employing the equivalent dimensionality dynamic GM （l,1） model, following a training and simulation with the results derived from the last step by utilizing BP neutral neural network. Then the combination model prediction value is obtained through data normalization and parameters selection. Take the case Maoping landslip for example, after the prediction to displacement and comparative analysis, it is demonstrated that the precision of GM-BP tandem combination forecast model is higher than single model. |
| |
| Keywords: | equivalent dimensionality dynamic normalization GM-BP combination forecast model landslip prediction |
| 本文献已被 维普 等数据库收录！ |
