Research On Risk Management Of High Cutting Slopes In Caoguoshan Tunnel Shaft Project Of Manjin Expressway

China is in the key period of the gradual implementation of the 14 th Five-Year plan.The construction of high-cutting slope of railways and highways under complex terrain is increasing day by day.But the corresponding risk management is not in place and the research is not sufficient.In order to control the risk and reduce the loss in the process of high-cutting slope construction,firstly,this paper takes the risk management of high-cutting slope as the research object,adopts the basic theory of risk management,and analyzes the safety risk factors in the process of cutting high slope construction in detail.Secondly,the common basic methods of high-cutting slope risk assessment are compared and their adaptability is analyzed.When determining the weight of each risk factor,a neural network model suitable for the weight distribution of risk source evaluation of high-cutting slope engineering is trained by using the self-learning of neural network.Finally,the risk management of high-cutting slope in Caoguo Mountain tunnel shaft project of Wan-Jin expressway is studied,and the following conclusions are obtained:(1)By sorting out the construction procedures of slope engineering,combined with the analysis of slope instability factors of rock mechanics,the correlation between risk sources and risk events in the high-cutting slope construction is found and the main risk sources and formulate the evaluation standards of each risk source is analyzed.(2)Based on a large number of historical data collected by literature analysis,sorted according to the main risk sources as the data source,and based on the optimization and improvement of the neural network toolbox in MALAB,the parameters such as the number of hidden layer nodes and training functions of the neural network are determined by experimental method,and a neural network model with the highest and best fitting degree to the data source is trained.(3)By simultaneous interpreting the risk factors of high-cutting slope in the highway project of the Caoguo Mountain tunnel,the scoring values of each risk index are obtained.The neural network model is trained and compared with the risk assessment results obtained by the traditional methods.The feasibility of the neural network model is verified.Based on the scoring system,the risk control measures for this project are put forward,and the emergency plan for the construction risk of high cutting slope is established.The research provides a reference basis for slope project risk management under similar working conditions.