A Soft Rock Tunnel Risk Assessment Model Based on the SSA-ENN Neural Network

In order to study the prediction method of surrounding rock deformation of soft rock tunnel, this paper constructs the Elman (SSA-ENN) prediction model optimized by sparrow search algorithm, and takes tianqiaoshan tunnel as the engineering support, and selects the monitoring data of vault settlement and horizontal convergence of surrounding rock deformation of soft rock tunnel as the training and test samples. Then, the measured results are compared with the prediction value of SSA-ENN soft rock tunnel surrounding rock deformation prediction model. Finally, taking DK110+605 section as an example, the SSA-ENN model is applied in engineering. In order to verify the effectiveness of SSA-ENN model, Elman neural network model and SSA-ENN model are predicted. The comparison results show that SSA-ENN model has the highest prediction accuracy, with R² value of 0.9965, RMSE value of 7.52 and MAE value of 0.24, which has high prediction accuracy and meets the requirements of guiding construction.