土壤水力參數不確定性對未飽和邊坡穩定性影響之研究

FEffects of uncertainty of soil hydraulic parameters on stability analysis of unsaturated slope

黃獻廷(Tsien-Ting Huang) ； 葉信富(Hsin-Fu Yeh)

《中華防災學刊》 13卷2期 (2021/09) Pp. 149-162

摘要

土壤持水曲線（Soil Water Retention Curve, SWRC）是解決岩土工程中未飽和土壤問題的關鍵資料，如何確立土壤持水曲線參數，對於探討降雨入滲和邊坡穩定性是非常重要的，但現地測量獲取土壤數據相當昂貴且耗時，僅能獲得有限的數據資料，使得土壤持水曲線參數具有高度的不確定性。本研究旨在探討土壤持水曲線參數不確定性，對於降雨條件下未飽和土壤邊坡滲流和穩定性之影響。首先，採用貝氏更新（Bayesian updating）針對模型與現地土壤之土壤持水曲線參數進行後驗分布之更新，並以馬可夫鏈蒙地卡羅（Markov Chain Monte Carlo, MCMC）法產生隨機樣本，分析其參數不確定性。另外，建置不同信賴區間之土壤持水曲線參數模型，並使用水力耦合模型評估不同信賴區間之土壤持水曲線參數，在降雨條件下邊坡滲流和穩定性之影響。研究結果指出，信賴區間之百分位數越大時，影響土壤進氣值的參數“α”和土壤孔徑分布的參數n會增加，兩參數的變化導致土壤中壓力水頭和含水量受降雨影響程度增加，而當含水量增加時，土壤內部吸應力隨之增加，導致邊坡局部安全係數減少，以降低邊坡之穩定性。根據上述結果顯示，壓力水頭和含水量之預測會受土壤持水曲線參數不確定性所影響，並導致邊坡穩定性分析產生誤差。

關鍵字

土壤水力參數 ； 貝氏更新 ； 不確定性 ； 水力耦合模型

並列摘要

Soil-Water Retention Curve (SWRC) is considerably important to explore the infiltration of rainfall and slope stability. However, SWRCs exhibit high uncertainty since it's quite expensive and spending time to measure the behavior of soil sample in the field and only obtain limited data. This study aims to quantify the effects of uncertainty of SWRC parameters on seepage and stability of unsaturated soil slope under rainfall conditions. First, we adopt Bayesian updating to improve the posterior distribution of SWRC parameters between the model and field data. In order to analyze the uncertainty of SWRC parameters, we apply Markov Chain Monte Carlo (MCMC) method to produce random samples. Then, the different confidence intervals for the parameters of SWRC were constructed. We combine the hydraulic coupled model to estimate the seepage and stability analysis of the slope under rainfall conditions with the SWRC parameters of different confidence intervals. The results show that the higher the percentile value of confidence intervals, the larger the parameters of SWRC. The variation of SWRC parameters results in the effects of pressure head and water content of the soil under rainfall. The higher water content, the more suction stress in the soil. It leads to decreasing the local factor of safety (LFS) of the soil and the slope stability. According to the results, the estimation of pressure head and water content was influenced by the uncertainty of SWRC parameters and it results in generating the errors of slope stability.

並列關鍵字

soil hydraulic parameters ； Bayesian updating ； uncertainty ； hydraulic coupled model