Considerable uncertainties are included in ground properties used for tunnel designs due to the limited investigation and tests. In this study, a back analysis was performed to find optimal ground properties based on artificial neural network using both face mapping data and convergence measurement data. First of all, the rock class of a study tunnel is determined from face mapping data. Then the possible ranges of ground properties were selected for each rock class through a literature review on the previous studies and utilized to establish more precise learning data. To find an optimal training model, a sensitivity analysis was also conducted by varying the number of hidden layers and the number of nodes more minutely than the previous study. As a result of this study, more accurate ground properties could be obtained. Therefore it was confirmed that the accuracy of the results could be increased by making use of not only convergence measurement data but also face mapping data in tunnel back analyses using artificial neural network. In future, it is expected that the methodology suggested in this study can be used to estimate ground properties more precisely.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.