Intelligent Prediction Of The Deep Rock Mechanics Under Coupling Effects Of Temperature And Pressure

Shallow surface resources are gradually exhausted,and deep resource exploitation is imperative.For deep underground engineering,the combined effect of high pressure and high temperature environment will have a significant impact on the mechanical behavior of deep rock masses.In order to explore the evolution of the mechanical behavior of deep underground rock masses under high temperature and high pressure conditions,this study relies on the indoor rock triaxial test data to discuss the two major external conditions factors of temperature and confining pressure on peak strength,peak strain and elastic modulus of rock mass Measure the influence of three typical mechanical parameters.Collect a large amount of rock triaxial compression test data at different temperatures and different confining pressures to build a parameter sample database,and analyze the law of temperature and confining pressure on rock mass mechanical behavior.It is believed that the influence of temperature and confining pressure on rock mass mechanical behavior appears Highly nonlinear characteristics are difficult to quantitatively characterize with a single mathematical function.Therefore,the BP(Back Propagation)neural network algorithm is used to construct an intelligent prediction model for the mechanical properties of rock mass considering the coupling effect of temperature and pressure,and the model library of mechanical parameters is used to train the model to obtain an accurate prediction model,and then to the high temperature and high confining pressure environment.The numerical simulation and prediction of the mechanical properties of the deep rock mass are used to obtain the characteristics of the evolution behavior of the rock mass mechanical behavior under different temperature and confining pressure conditions,which can provide a theoretical basis for deep underground engineering.The main research work of this article is as follows:(1)Clarify the variation law of rock mass temperature and confining pressure with depth,and combine the "three high and one disturbance" environment of deep underground engineering to explore the temperature and confining pressure range of deep rock mass.Define the parameter "Rock Mass Structure" to exclude the influence of variables other than temperature and confining pressure on the test results.It is calculated that the ambient temperature of the rock mass 5000 m below the surface can reach 1000 ?,and the confining pressure can reach 135 MPa.(2)The test data is classified and the peak strength,peak strain and elastic modulus of the test results are qualitatively and quantitatively analyzed by the single factors oftemperature and confining pressure.The results show that the influence of temperature and confining pressure on rock mechanical behavior exhibit significant nonlinear characteristics,which is difficult to be characterized by a single mathematical function.The interaction principle can be used to calculate the temperature-pressure coupling interaction coefficient.There is a coupling effect that cannot be ignored.(3)According to the operation mechanism of the neural network algorithm,an appropriate artificial neural network is selected,a specific BP neural network model is constructed,and a sample database is used for training and evaluation of the reliability of the model,and further research is carried out.The study found that the mean square error of the neural network after training using the sample database can reach 0.01 and the R value reaches 0.8.It can be considered that the artificial neural network algorithm can better characterize the nonlinear behavioral characteristics of the rock mass mechanical properties,and can be better Ground reaction temperature-pressure coupling effect.(4)The mechanical behavior of rock under high temperature and high confining pressure is predicted and analyzed by using the trained neural network.It is found that the confining pressure threshold value of peak strain with temperature is 90 MPa,the peak strength decreases with temperature rising,the peak strain generally increases,and the elastic modulus first increases,then decreases,and then increases;while the peak strength and peak strain increase with confining pressure,and the temperature threshold value of elastic modulus changes with confining pressure is 700 ?,when the temperature is lower than the threshold value,the elastic modulus tends to decrease,when the temperature is higher than the threshold value,the elastic modulus tends to increase.