Model Parameters Identification Procedure Of Chamber Pressure Control In Shield Tunneling

Earth pressure balance (EPB) shield machine is a kind of engineering machine used for underground excavation, which is primary used in shield tunnelling. With wide application of the shield in tunnel construction, ground settlement are received more and more attention. However, the decisive factor of ground settlement is the imbalanced earth pressure of the head chamber. The main research results are as followings:The controlling model with main tunneling parameters of the increment of earth pressure in the head chamber, thrust speed and the rotation speed of screw conveyor is established by considering the earth pressure balance. A shield tunneling engineering is taken as an example, a numerical simulation is presented on the process of earth pressure control in shield tunneling. The result shows when the thrust speed is constant, the earth pressure can be controlled within the range of set values by controlling rotation speed of screw conveyor. Finally, the earth pressure balance is realized.The tangent modulus of soil and volumetric efficiency of screw conveyor related to the state equations need to be determined. Least squares method, BFGS algorithm and genetic algorithm are introduced and applied to the parameter identification of controlling model.The numerical simulation analysis on parameter identification of shield machine is presented,in order to ensure the model for controlling earth pressure and parameter identification method are effective, a simulated experiment is studied on the laboratory test device of shield machine. The minimization of difference between observed chamber pressures in laboratory and computed ones from computational model is taken as objective function. The chamber pressure can change with adjusting rotation speed of screw conveyor because excavation speed is selected as constant.The three kinds of optimization algorithms are compared with each other, genetic algorithm can acquire the accurate identification parameters, BFGS algorithm takes second place, least squares method is the worst. Genetic algorithm could acquire valid identification result on the tangent modulus of soil and volumetric efficiency of screw conveyor, identification result of genetic algorithm is close to the parameter data of experiment, model parameters are provided for pressure control.