准确预测打入桩的承载特性是一个重要的工程课题，特别是在近海工程中，如大型跨江海桥梁、风力涡轮机、港口和近海工程结构物平台等，这些重要结构物都将打入桩作为主要的基础形式，然而通过现场静载试验确定这类桩的承载力却十分有限。目前，国内外存在众多关于黏土中打入桩的竖向承载力计算方法，其中较为常用的有Furgo-96、NGI-99、ICP-05和UWA-13这4种设计方法。但将这些设计方法直接应用于工程设计和实践却仍然较少，其主要原因在于缺乏高质量的静载试验数据来评价这些设计方法的可靠性。因此，有必要搜集一个具有广泛代表性的桩基承载力静载试验数据库来客观评价上述设计方法的可靠性，用以帮助工程设计人员选择最适合的设计方法。基于这个目的，本研究在国际上一些常用数据库的基础上进行扩充整理，最终得到一个包含89根高质量打入桩静载荷试验数据的新数据库，称之为ZJU-ICL Clay数据库。采用上述4种设计方法预测ZJU-ICL Clay数据库的桩基承载力，并将预测结果与实测数据进行对比，应用统计方法，客观评价这4种方法的优缺点及可靠性。

Predicting the behavior of driven piles is a key issue in practical engineering, particularly in major bridge, wind turbines, harbor and offshore engineering applications where the piled foundations are widely used to underpin the superstructures and the pile capacity is seldom determined from pile load tests. Numerous methods exist for calculation of axial pile capacity in clay, including four design methods, namely Furgo-96, NGI-99, ICP-05 and UWA-13. However, application of the latter in mainstream civil engineering sector has been relatively slow, possibly because of the small number of high quality, full scale, case histories available to demonstrate their potential benefits. Thus, it is necessary to collect a more comprehensive database to evaluate the performance of the four design methods objectively and to assist designers in the selection of the most appropriate method. A new ZJU-ICL Clay database with 89 high-quality pile load tests is formed by expanding the major databases with new available pile load tests. Database assessment of these four methods and evaluation of the predictive performance by using mean ? and coefficient of variation COV are presented.