基于BP神经网络的隧道渗漏水等级评定模型研究

由于隧道渗漏水的普遍性和严重危害性,隧道渗漏水已成为隧道工程建设和维护的一大关键课题。针对我国隧道渗漏水的现状,通过综合分析影响隧道渗漏水的各种因素,找出影响隧道渗漏水的主要因素,提出包括隧道设计施工因素、防排水工程措施、围岩及地下水情况、自然地理情况4个一级指标共10个因素的评价指标集,构建基于BP神经网络的评价模型,并验证该模型的可靠性。该方法为进一步提高隧道渗漏水等级评价的精确度提供新的思路。

Study on Evaluation Model for Water Leakage Grades of Tunnels Based on BP Neural Network

Water leakage of tunnels is very popular and has serious adverse effect, therefore it has become one of the critical issues in the construction and maintenance of tunnels. Regarding the status of the water leakage of tunnels in China, the authors identify the main factors that influence the water leakage of tunnels by comprehensively analyzing the factors. The authors propose an evaluation index system consisting of 4 Grade I indexes (i.e., tunnel design and construction index, waterproofing and drainage index, surrounding rock mass and groundwater index and physical geography index), which include 10 factors. In addition, they establish an evaluation model based on BP neural network and verify the reliability of this model. This method offers a new way to improve the evaluation precision of the water leakage of tunnels.