岩土力学参数的概率分布类型直接影响岩土工程可靠性分析的计算结果和精度。在现场有限个小样本和多个先验分布的条件下,探讨如何综合利用已有的经验资料确定岩土参数的概率分布,提出了多源信息融合方法。通过分析各先验分布和现场试验信息的差异程度,确定各先验分布的权重,进而进行整体融合,实现统计意义上概率分布类型的优化。
The probabilistic model of rock and soil mechanics parameters affects the results and precision of reliability analysis and design. Based on testing data and some groups of prior distribution data (existing in history) in the same zone, this paper discusses the problem of how to estimate the distribution of rock and soil mechanical parameters for a small sample in the case of insufficient data. A multiple sources information fusion method based on Bayesian statistics theory is presented. Confirming the weight of every prior distribution by calculating the difference between its prior data and testing data, this paper brings about the optimum fitting of probabilistic model of rock and soil mechanical parameters.