| Due to the influence of engineering geology,construction technology and other factors,the safety of diversion tunnel during the excavation has attracted much attention.During the construction and excavation of diversion tunnel,the surrounding rock of the tunnel is often deformed by the excavation effect,which will harm the safety and stability of the tunnel.The analysis and prediction of tunnel deformation monitoring is of great practical significance for ensuring the construction safety of diversion tunnel.In order to know the deformation law of surrounding rock of diversion tunnel during excavation and ensure the safe construction of tunnel project,this paper analyzes and predicts the deformation amount of tunnel excavation by establishing monitoring model.The deformation during tunnel excavation is affected by the excavation process.The overall deformation trend of the tunnel will be different with the excavation history and the degree of excavation.It is difficult for ordinary time functions to accurately capture the change law.Therefore,it is necessary to select the influencing factors which can reflect the excavation effect effectively to analyze the deformation of tunnel surrounding rock.In this paper,considering the mechanical characteristics of surrounding rock during tunnel excavation,starting from the genetic creep theory,based on the measured data and engineering characteristics,a set of influencing factors reflecting instantaneous deformation and creep deformation is constructed.In this paper,the cuckoo search algorithm(CS)with few parameters,simple operation and strong searching ability is used to optimize the BP neural network model.By combining the CS algorithm with the BP neural network,the CS-BP neural network model is established to train and predict the deformation of diversion tunnel excavation.The results show that the CS-BP model based on the genetic creep theory has a very good prediction effect on the deformation prediction of the diversion tunnel,which can accurately reflect the deformation trend of the surrounding rock of the tunnel,and it also can provides an effective method for the monitoring data analysis of the diversion tunnel during the excavation construction..The analysis of the monitoring data during the diversion tunnel provides an effective method.In order to verify the improvement effect of factor and CS algorithm based on genetic creep theory on the model,different influencing factors were substituted into the CS-BP model for effect comparison,and the same influencing factors were also substituted into the CS-BP and BP models for comparison.Through the model output results,it can be concluded that the genetic creep theory based on the subset can effectively improve the training and prediction effect of the model,and the CS algorithm has an obvious optimization effect on the performance of the BP neural network model.On the basis of constructing reasonable factors and optimizing the model,this paper studies and establishes the multi-point model of tunnel during excavation.Example analysis shows that the multi-measuring point model not only has good prediction accuracy,but also can comprehensively reflect the change law of surrounding rock of tunnel,has good integrity and anti-interference,and has deep application and promotion value in the research of diversion tunnel excavation deformation prediction. |
