Research on Geological Information Updating and Prediction of Tunnel Surrounding Rocks Based on Machine Learning

Surrounding rock geological information is the basic index for tunnel disaster prediction and forecasting, so dynamic updating and prediction of surrounding rock geological information in an efficient and intelligent manner can provide an important support for disaster prediction and forecasting. In this paper, a surrounding rock geological information updating and prediction model based on RNN (RNN) is constructed to carry out research on the prediction of the four geological parameter indexes of rock mass integrity, rock hardness, rock weathering degree, and water abundance of tunnel surrounding rocks. The various index data of excavated tunnel sections are collected, the original data is normalized, a time series prediction model for tunnel surrounding rock geological information is established based on the training set samples, and finally the updated prediction of surrounding rock information is achieved. Preliminary application shows that the updating prediction accuracy of surrounding rock geological information of three indexes including rock mass integrity is up to 87.5%, and the prediction accuracy of rock weathering degree reaches 75%. This is an approach with relatively high engineering practical value.