Research On Multi-point Earth Pressure Balance Optimization Control Of Shield Machine Based On Sparrow Algorithm

In recent years,China has made outstanding achievements in the field of scientific and technological innovation.Under the background of urbanization,the underground space is becoming smaller and smaller,and the development and utilization of underground space has become an urgent task.As a key tool of tunnel excavation,shield machine has been widely used in the field of tunnel construction.It plays a very important role in underground engineering such as resource exploitation,tunnel excavation and municipal construction,which is inseparable from its unique functions and many advantages,such as perfect terrain adaptation,high efficiency,high degree of automation,safety and environmental protection,small impact,etc.The earth pressure balance of shield enclosure has always been one of the key issues in the research.Because the imbalance of internal and external pressure will cause the collapse or uplift of the surface,it is necessary to accurately control the earth pressure balance.Therefore this paper first establishes the multi-point earth pressure prediction model of the sealed cabin based on the least square support vector machine（LS-SVM）,then optimizes the LS-SVM model with particle swarm optimization（PSO）algorithm to improve the prediction accuracy,Finally,the sparrow algorithm is used to optimize the Propulsion speed and screw conveyor speed of the shield machine,and the optimal control of the earth pressure balance of the shield machine is realized.Firstly,based on the analysis of the tunneling mechanism of shield machine,the relationship between the parameters and the earth pressure of the sealed cabin is deduced,and the input and output variables of the prediction model are obtained.Then,based on LS-SVM,a four point earth pressure prediction model is established.Because the selection of kernel function has a certain impact on the accuracy and effect of the prediction model,the radial basis function is selected by consulting and analyzing the relevant literature,Secondly,particle swarm optimization algorithm is used to optimize the prediction model based on least squares support vector machine,that is,to optimize the penalty factor and kernel parameters of radial basis function,and to feed back the optimal sum to LSSVM model to solve the regression function.Finally,the sparrow algorithm（SSA）is used to optimize the input variables of the prediction model,such as the propulsion speed and the screw conveyor speed.The optimization objective is that the predicted earth pressure values of LS-SVM model is infinitely close to the set value of earth pressure as far as possible,namely,the minimum value of the difference between the two.The optimal speed obtained is transmitted to the corresponding system to control the working state of the shield machine,then,the optimal control of earth pressure balance in shield tunnel is realized.According to the field construction data,the simulation experiment shows that the prediction effect is good and the error is controllable.Finally,the performance of sparrow search algorithm and particle swarm optimization algorithm is compared.The results show that the optimization performance of SSA is better than that of PSO,and the effectiveness of this method is verified.