世界各地的學者各自提出不同的液化評估法，這些評估法主要分為三大類，分別根據SPT-N、CPT-qc、震測VS三種現地試驗資料為發展基礎，而各種評估法分析的結果皆不盡相同，發展背景亦有所不同，各評估法是否適用於台灣之地震與地質構造有待研究探討。 本研究以1999年台灣發生的921地震液化案例資料庫為主，此次地震的案例資料庫相當豐富，計有303組的SPT-N案例、135組的CPT-qc案例以及75組的震測VS案例，非常適合做為分析與研究之用；本研究結合了「可靠度分析」與「貝氏定理」兩種機率方法，求得各液化評估法之不確定性，此法之優點在於可將各種液化評估法的不確定性與變異性加以量化，以瞭解各評估方法準確性的差異，並可以模式修正因子將各方法調整至更一致的標準。本研究分析之評估方法包括世界上至要使用之SPT-N法有四種、CPT-qc法有三種、VS法有兩種。 本研究亦試著研究本土學者所發展之雙曲線液化評估法(HBF法)之準確姓，此方法還具有明確的物理意義而非單純的數據統計迴歸；分析結果顯示， SPT-N、CPT-qc與VS三種的HBF法均不遜於世界上現有的至要評估法，建議可供國內工程界參考使用。最後針對HBF法之公式將之修正得更簡單並保持相同的準確性，這樣可使HBF評估法更為方便使用。 Many simpilified methods for assessing liquefaction potential have been proposed by scholars around the world. These simpilified methods are based on SPT-N, CPT-qc and VS parameters from different in-situ tests. The result of analysis and the development background among these simpilified methods are different. Thus, it’s very important to understand if these simpilified methods are suitable to the earthquake enviroment and geology condition in Taiwan. This research collects a lot of case data from the 1999 Chi-Chi earthquake in Taiwan. There are totally 303 sets of SPT-N data, 135 sets of CPT-qc data and 75 sets of VS data on liquefaction and non-liquefaction case histories. This study combines reliability analysis and Bayes’ theorem to study the accurancy and characteristics of these different methods. The advantage of this analysis method is that it can quantify the uncertainty and accurancy of these simpilified methods by probability concept. In this research, a simpilified method (HBF method) is proposed owing to its formula having clear physical meaning and simple parameters. The analysis results show that the HBF method has high accurancy and easy to use. Thus, the HBF method is recommended to be used by engineers in Taiwan.
