ZHU Chuanhua, HU Guangdao. Data Mining in Landslide Monitoring Based on Multidimensional Modeling[J]. Bulletin of Soiland Water Conservation, 2010, 29(4): 121-124.
ZHU Chuanhua, HU Guangdao. Data Mining in Landslide Monitoring Based on Multidimensional Modeling[J]. Bulletin of Soiland Water Conservation, 2010, 29(4): 121-124.DOI:
并在SQL Server 2005 Analysis Services软件中使用Microsoft时序算法实现数据挖掘
对滑坡位移进行预测分析
得到的预测数据与实测数据误差很小
表明Microsoft时序算法可用于滑坡监测数据的短期预测。
Abstract
Security monitoring is one of the most important ways to investigate and control landslides and displacement monitoring is the most common way to monitor landslides. The analysis of landslide displacement monitoring data is generally based on flat files. A different approach based on data warehouse is proposed
which analyzes in depth the time series data from landslide displacement monitoring and build a multidimensional model of landslide displacement monitoring using CASE tool Power Designer. By the approach
a multidimensional model is designed and applied in Baishuihe landslide in the Three Gorges Reservoir area. Data mining process is then implemented using the algorithm of Microsoft Time Series in SQL Server 2005 Analysis Services. Results show that the predicted value is approximately in accordance with the measured value and the algorithm is applicable in short-term prediction.