中文摘要：

深基坑工程在设计、施工中存在一定的安全隐患，对其进行安全性评估具有重要意义。但是深基坑支护结构的安全性难以用一个准确的数值表述。本文选取侧压力，墙体侧斜，墙体弯矩，支撑轴力，立柱隆起及地表沉降作为评估指标；将深基坑的安全性等级划分为5级，并建立评估指标和安全等级的定量关系表；将模糊理论与神经网络技术结合起来，论证其在深基坑安全性评价的可行性，并建立用于深基坑安全性评估的模糊神经网络，并将训练好的网络用于某深基坑的安全性评估。结果表明：采用此方法对深基坑进行评估是深基坑安全评价的一个新的有效途径。

英文摘要：

To some extent , the design and construction progress of deep foundation pit project have poten-tial safety hazard , so that it is very significant to investigate the safety assessment of deep foundation pit .Howev-er , it is difficult to evaluate the safety grade of the structure of deep foundation pit with an accurate representation of numerical values.Lateral pressure, wall slope, wall bending -moment, axial force, column heaving -value and ground settlement were choosed as the indexes for safety evaluation of deep foundation pit .Here, a fuzzy theory and neural network technology we put together to build a fuzzy neural network for deep foundation pit safe -ty evaluation.The results show that：the use of fuzzy neural network on the deep foundation pit is a kind of high efficient, accurate, and feasible evaluation method in the safety assessment process .