本文選題：可靠度 + 無失效數(shù)據(jù)��； 參考：《湖南科技大學(xué)》2016年碩士論文

【摘要】：由于巖土工程可靠性問題中不確定因素較多,且檢測數(shù)據(jù)多為無失效數(shù)據(jù)(檢測試驗中沒有破壞或者失效的樣品數(shù)據(jù)),傳統(tǒng)的分析方法較難分析計算工程的可靠度。本文通過利用貝葉斯統(tǒng)計理論方法,將時間參數(shù)轉(zhuǎn)化為荷載參數(shù),把無失效數(shù)據(jù)可靠度模型應(yīng)用于樁基承載力可靠度的計算研究中。通過對單樁承載力可靠度的研究作為算例分析,驗證無失效數(shù)據(jù)可靠性模型在巖土工程中的可行性。并得到以下成果:1.針對巖土工程中的檢測數(shù)據(jù)失效率低,小子樣(樣本空間小),工程可靠度高,極易出現(xiàn)零失效數(shù)據(jù)情況這一特點。引入無失效數(shù)據(jù)可靠度理論,并對該理論研究方法進(jìn)行了一定的介紹。從理論的角度分析其在巖土工程可靠度方面的可行性。2.通過單參數(shù)指數(shù)分布無失效數(shù)據(jù)模型,建立關(guān)于樁基承載力的指數(shù)分布可靠度函數(shù),對實際工程單樁豎向抗壓靜載試驗數(shù)據(jù)進(jìn)行處理分析,發(fā)現(xiàn)該模型可以較好的使用于樁基無失效數(shù)據(jù)模型,能較好的與實際情況吻合。3.通過雙參數(shù)威布爾分布無失效數(shù)據(jù)模型,建立建立關(guān)于單樁承載力的指數(shù)分布可靠度函數(shù),對實際工程樁單樁豎向抗壓靜載試驗數(shù)據(jù)進(jìn)行處理分析,在加入驗前信息的同時引入小子樣概念,并對似然函數(shù)進(jìn)行修正。通過實際工程算例進(jìn)行分析計算,其結(jié)果能較好反映實際工程情況。4.由于樁基試驗數(shù)據(jù)較少,一般為小子樣數(shù)據(jù),而試驗過程中經(jīng)常難以獲取真實的數(shù)據(jù)。本文提出利用Bayes方法,針對小子樣數(shù)據(jù)進(jìn)行污染回歸處理。從而對用來計算分析可靠度的樣本數(shù)據(jù)進(jìn)行修正。
[Abstract]:Because there are many uncertain factors in the reliability of geotechnical engineering, and most of the test data are non-failure data (there is no damage or failure sample data in testing test), the traditional analysis method is difficult to analyze and calculate the reliability of engineering. In this paper, by using Bayesian statistical method, time parameters are transformed into load parameters, and the reliability model of no-failure data is applied to the calculation of bearing capacity reliability of pile foundation. The reliability of single pile bearing capacity is analyzed as an example to verify the feasibility of the reliability model of no-failure data in geotechnical engineering. And get the following results: 1. In view of the characteristics of low failure rate and small sample (small sample space and high engineering reliability) in geotechnical engineering, zero failure data is easy to appear. The reliability theory of non-failure data is introduced, and the research method of this theory is introduced. The feasibility of this method in geotechnical engineering reliability is analyzed from the angle of theory. Based on the single parameter exponential distribution data model, the reliability function of the exponential distribution of pile foundation bearing capacity is established, and the vertical compressive static load test data of practical single pile are processed and analyzed. It is found that the model can be applied to the pile foundation data model without failure, and it is in good agreement with the actual situation. 3. Based on the two-parameter Weibull distribution no-failure data model, the reliability function of exponential distribution on the bearing capacity of single pile is established, and the vertical compressive static load test data of practical single pile are processed and analyzed. At the same time, the concept of small sample is introduced and the likelihood function is modified. Through the analysis and calculation of practical engineering examples, the results can better reflect the actual engineering situation. 4. Because the pile foundation test data is few, it is usually small sample data, but it is often difficult to obtain the real data in the test process. In this paper, Bayes method is used to deal with the pollution regression of small sample data. Thus, the sample data used to calculate the reliability of the analysis are modified.
【學(xué)位授予單位】：湖南科技大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2016
【分類號】：TU473.1

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 茆詩松;羅朝斌;;無失效數(shù)據(jù)的可靠性分析[J];數(shù)理統(tǒng)計與應(yīng)用概率;1989年04期

，

本文編號：1965916