加权函数组合预测边坡变形模型的研究  

PREDICTION OF SLOPE DEFORMATION MODES WITH WEIGHTED FUNCTION COMBINATION METHOD

作　　者：金海元[1] 徐卫亚[1] 

机构地区：[1]河海大学岩土工程研究所,南京210098

出　　处：《工程地质学报》2008年第4期518-521,共4页Journal of Engineering Geology

摘　　要：边坡变形监测是边坡监测的主要内容之一,其变形预测问题是边坡工程中主要技术难题之一。考虑边坡位移变形预测模型的局限性,如神经网络预测方法需要大量的实测数据作为学习样本,灰色系统模型要求原始数据序列必须满足指数规律,且数据序列变化速度不能太快等。建立了边坡变形反向传播神经网络预测模型,同时给出了灰色GM(1,1)边坡预测模型。提出边坡的神经网络与灰色系统加权函数组合预测模型,采用动态规划解法,将原模型转化为多阶段决策问题,使组合预测误差的平方和最小,得到组合权重,这样得到的变形预测结果的精度将大大提高,弥补了单一方法的局限性,满足工程预测的需要。通过边坡实例加以验证,加权函数组合预测模型的预测结果精度有一定提高,能够与实际监测数据相吻合,达到准确预测的目的。Deformation monitoring is one of the main contents in slope monitoring. Deformation prediction is one of the main technical problems. The current deformation prediction models have limitations. For example, the neural network predicting model needs massive data as study pieces and grey model requests that the original data must meet the exponent rule. A ＂back propagation neural network prediction model and grey system GM （ 1,1 ） model for slope deformation are built. In order to gain more accurate prediction results, a weighted function combining forecasting model of neural network and grey system on slope deformation is developed and a dynamic programming method is used to solve multi - step decision problem, which can minimize the square sum of the relative errors and gain the combination weight. So, the predicting result can greatly enhance the precision and make up the limitation of single method. Through calculating an actual slope the predicting results develop greatly and consistent with the monitoring data. The combining predicting model enriches the slope deformation prediction theory.

关 键 词：变形预测 神经网络 灰色系统 组合预测 动态规划 

分 类 号：TD824.7[矿业工程—煤矿开采]