Disaster effect of freez-thaw on the frozen wall of subway connection channel and its average temperature prediction
The ground freezing method has been increasingly applied to improve the properties of surrounding formations during the construction of tunnel connecting channels in soft soil areas,owing to the advantages of being easy to use,high reliability,good groundwater control,and small disturbance to the soil.The average temperature of the frozen wall,as one of the important parameters to measure the freezing quality,is quite difficult to be obtained due to its complicated solution procedure.In this paper,a chain effect mechanism of freez-thaw disasters on the frozen wall of the subway connection channel is established,including the fracture of frozen pipes,the convergent deformation of the frozen tunnel in the frozen area,the disaster of convergent deformation of the frozen curtain,the seepage of freezing curtain and the ground settlement during the melting of the ground frozen wall.Considering the layout of the freezing pipes,a complete data set is established by using the equivalent trapezoid method of the average temperature of frozen wall.Based on the highly nonlinear characteristics of the neural network,a general model for predicting the average temperature of the frozen wall is established when the freezing pipes are spaced in a single row,double rows,and multi-rows(n>3).The results show that the neural network prediction model based on the complete data set can effectively reflect the relationship between different freezing tube layout parameters,freezing wall thickness and freezing wall average temperature.The average freezing wall temperature decreased gradually with the increase of freezing tube row number and freezing wall thickness,but increased gradually with the increase of brine temperature,freezing tube spacing and row distance.
frozen wallaverage temperaturedisaster effect of freez-thawconnection channelsubway
NETL
NSTL
万方数据
维普
