中国具有广大的季节冻土和多年冻土区,在冻土地区进行工程建设,就必须深入研究冻土的力学特性,以确保冻土地基上工程建筑物的稳定性。本文首先,简要介绍了我国冻土的分布状况和冻土区别于融土的基本特性,广义的冻土力学可分为冻融作用和已冻土力学性质两方面,冻胀、融沉和冻融循环引起的土力学性质的变化属于冻融作用的范畴。对于冻胀的研究较为深入,人们先后提出了多个理论来解释冻胀产生的机制,有的应用于计算分析中。对融沉的研究尽管具有较长的历史,但是多数停留在经验方法上,融化固结理论目前还有较大的局限性,因此提出一方面可以用人工神经网络法提高经验方法的精度和适用范围,另一方面应当发展融化固结大变形理论;冻融循环可以改变土的力学性质,介绍了作者的最近研究进展。针对已冻土的力学特性,从3方面进行了分析。冻土的强度主要沿用融土的强度理论,很难反映高应力下的压融现象;冻土动力学特性主要针对温度对动力学参数的影响,近年来冻土层对场地动力响应的影响越来越受到重视;冻土的本构关系多集中在蠕变研究,以经验公式法为主。最后,分析了多年冻土地区工程变形所涉及的物理力学过程。
China has a large area with seasonally and perennially frozen soils. It is necessary to study mechanical properties of frozen soils for the sake of constructions in cold regions, so as to guarantee their stability. In this paper, the distribution of frozen soils in China and the main features of frozen soils differing from unfrozen soils, are briefly described firstly. Generally speaking, frozen soil mechanics consists of two main aspects, i.e. freeze-thaw related problems and mechanical properties of the already-frozen soils. Frost heave, thaw settlement and freeze-thaw induced changes in mechanical properties are freeze-thaw related problems. Frost heave has been extensively studied. Various theories have been put forward to describe the mechanism of ice segregation which is considered as the main cause of frost heave; and some theories have even been applied to quantitative analysis. The study of thaw settlement has a long history, while most of the previous studies still stay in empirical stage. Thaw consolidation has been studied; but it has considerable limitations so far. Therefore, we suggest that the artificial neural network could be used to improve the precision and applicability in estimating thaw settlement on the one side, large strain theory could be used to improve thaw consolidation theory. For the already-frozen soils, this paper summarizes the previous studies on strength, stress-strain relationship and dynamic properties. Study of the strength of frozen soil borrows strength theories for unfrozen soils; however, they can hardly reflect stress melting of frozen soils under high stresses. The previous studies of stress-strain relationship of frozen soils have been focusing on creep; and the empirical relationship directly obtained from testing data is the mainstream. Influence of temperature on dynamic parameters of frozen soils seems to be the only task in frozen soil dynamics for a long time; while influence of frozen layer in dynamic response of a site has attracted attentions in recent years