Bayesian Network Based Expert System for Tunnel Surrounding Rockmass Classification
Hong Tao Xie · Applied Mechanics and Materials · 2013
Traditional surrounding rockmass classification methods have disadvantages of relative narrow scope of application, most of the time the classification result needs some modifications by geological expert and field situation. Based on the surrounding rockmass classification methods of BQ system, the Bayesian network and corresponding uncertainty reasoning principle has been introduced to develop an expert system for surrounding rockmass classification. By combining prior knowledge of domain experts with worksite data recorder, we get the posterior probability density of most nodes. The field practices proved that the expert system has good applicability.