灰色模型改进的大坝变形分形几何监控模型

发布时间：2018-11-26 11:50

【摘要】：为了对大坝安全进行准确监控,利用分形几何理论预测大坝变形。针对一般常维分形分布不能很好分析大坝变形数据的问题,对监测数据进行N阶累计和变换,对变换后的数据利用分段变维分形模型计算各阶变形维数序列,再选择效果较好分形维数已知序列预测未知分形维数,最后反推大坝变形预测数据。针对传统变维分形预测模型分形维数预测方法的不确定性和所需监测数据量大的缺点,利用灰色模型预测分形维数,建立改进的大坝分形几何监控模型。结合工程实例,对比插值法预测分形参数的传统分形几何预测模型和灰色模型改进后的预测模型之间的预测精度,结果表明,改进分形模型不仅在预测精度上有所提高,而且更具稳定性和抗波动性。
[Abstract]:In order to monitor dam safety accurately, the fractal geometry theory is used to predict dam deformation. In order to solve the problem that the normal fractal distribution can not well analyze the dam deformation data, the N order cumulative and transformation of the monitoring data is carried out, and the transformed data is calculated by using the piecewise variable dimension fractal model to calculate the deformation dimension sequence of each order. The known series of known fractal dimensions are selected to predict the unknown fractal dimensions. Finally, the dam deformation prediction data are backfired. Aiming at the uncertainty of fractal dimension prediction method of traditional variable dimension fractal prediction model and the shortcoming of large amount of monitoring data, an improved dam fractal geometric monitoring model is established by using grey model to predict fractal dimension. Combined with engineering examples, the prediction accuracy between the traditional fractal geometric prediction model and the improved grey model is compared. The results show that the improved fractal model can not only improve the prediction accuracy, but also improve the prediction accuracy of the improved fractal model. And more stable and anti-volatility.
【作者单位】：河海大学水文水资源与水利工程科学国家重点实验室;河海大学水资源高效利用与工程安全国家工程研究中心;河海大学水利水电学院;大唐环境产业集团股份有限公司;
【基金】：国家自然科学基金资助项目(51579086,51479054,51379068,51139001) 江苏省杰出青年基金资助项目(BK20140039) 江苏高校优势学科建设工程资助项目(水利工程)(YS11001)
【分类号】：TV698.11

【相似文献】