| 应用贝叶斯动态模型的地基沉降概率分析与预测 |
| |
| 引用本文： | 魏冠军,党亚民,章传银.应用贝叶斯动态模型的地基沉降概率分析与预测[J].测绘科学,2012,37(2):52-53,90. |
| |
| 作者姓名： | 魏冠军 党亚民 章传银 |
| |
| 作者单位： | 1. 山东科技大学,山东青岛 266510;兰州交通大学,兰州730070
2. 山东科技大学,山东青岛 266510;中国测绘科学研究院,北京100039 |
| |
| 基金项目： | 国家863重点项目(2009AA121400) |
| |
| 摘 要： | 考虑到地基沉降预测模型中参数的时变特性及预测结果的可靠性,本文提出地基沉降概率预测方法:运用贝叶斯动态模型建立地基沉降过程的状态方程和观测方程,利用参数先验信息并结合含有噪声的前期沉降观测数据,对沉降状态参数进行Bayes后验概率推断,通过不断的"概率预测-修正"递推运算,获得最优沉降状态概率估计来预测地基沉降量。数值实例结果表明,与其他预测方法相比较,本文的方法是可行有效的。
|
| 关 键 词： | 贝叶斯估计 先验信息 卡尔漫滤波 极大似然估计 动态线性模型 沉降预测 |
Foundation settlement probability analysis and prediction based on Bayes dynamic linear model |
| |
| WEI Guan-jun
,
DANG Ya-min
,
ZHANG Chuan-ying.Foundation settlement probability analysis and prediction based on Bayes dynamic linear model[J].Science of Surveying and Mapping,2012,37(2):52-53,90. |
| |
| Authors: | WEI Guan-jun DANG Ya-min ZHANG Chuan-ying |
| |
| Institution: | g③(①Shangdong University of Science and Technology,Shandong Qingdao 266510,China;②Lanzhou Jiaotong University,Lanzhou 730070,China;③Chinese Academy of Surveying and Mapping,Beijing 100039,China) |
| |
| Abstract: | Taking into account the time-varying characteristics of parameters in foundation settlement prediction model and the reliability of prediction result,the paper proposed the probability method of foundation settlement predication.State equation and observation equation of foundation settlement were established by using Bayesian dynamic model.Combining parameters prior information with the early settlement observation data containing noise,the settlement state parameters were deduced with Bayes Posterior Probability.Optimal settlement state estimation used continuous Probability Forecast-Fixed recursion operator to predict the probability of foundation settlement.Numerical example showed that compared with other prediction methods,the method of Bayesian dynamic model was feasible and effective. |
| |
| Keywords: | Bayes estimation prior information Kalman filter maximum likelihood estimation dynamic linear model(DLM) settlement prediction |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
