Structural Safety Evaluation of In-Service Tunnels Using an Adaptive Neuro-Fuzzy Inference System

Abstract

This study investigated structural safety evaluation of in-service tunnels. Considering the factors that affect the safety of in-service tunnels, the indices and grading standards for structural safety evaluations of concrete tunnels were first introduced. Subsequently, a safety evaluation system was established based on an adaptive neuro-fuzzy inference system (ANFIS). Taking a concrete tunnel as an example, according to structural characteristics of the tunnel and a Chinese specification, the indices and grading standards for the structural safety evaluation of the tunnel were determined, and a safety evaluation system for the tunnel was established. In the safety evaluation system, the effect of data size on the training results of the system was analyzed based on the theory of statistics and the Delphi method that are used for data inspection. The results show that the evaluation system has good learning and application ability. By using the field measured data of an in-service tunnel, after learning, the system can effectively imitate experts to do nonlinear fuzzy inference.