Research On TBM Tunneling Performance Analysis And Pediction For Ultra-long Tunnel

The tunneling performance of TBM mainly includes tunneling speed,construction speed,operation utilization rate and tool wear.The study of tunneling performance has always been an important topic in this field.Its research and prediction can provide theory for the construction period and cost estimation of TBM under different surrounding rock conditions.in accordance with.Based on the Xinjiang EH project,this paper analyzes and predicts the tunneling performance of TBM.First,through the collection,analysis and comparison of excavation data at the construction site,a comparative analysis of the utilization rate of TBM2 and TBM3 excavation operations,equipment integrity,footage,excavation speed and construction speed was carried out.The results showed that rock integrity affects the utilization rate of TBM The main factors,the TBM management level and maintenance level of the construction team have a greater impact on the utilization rate;the TBM maintenance level of the construction team is an important factor affecting the integrity of the equipment;the main factor affecting the TBM footage is the uniaxial compressive strength of the rock;The driving speed and construction speed of TBM2 and TBM3 show the same law,gradually decreasing from type II surrounding rock to type IV surrounding rock.Then,through the acquisition,analysis and comparison of tool consumption data at the construction site,the tool change times and wear amount of TBM2 were statistically analyzed for normal wear and abnormal damage of the tools.In the process of tunneling with a total length of 6 152.1 m,the cumulative wear of the center knife was 284 mm,all of which were abnormally damaged,and a total of 44 tools were changed;the cumulative wear of the face knives was 3 744 mm,and 180 tools were changed for normal wear and abnormal damage.57;the cumulative wear of side knives is 3 904 mm,262 knives are exchanged for normal wear,and 13 knives are abnormally damaged.Among them,the main forms of abnormal tool damage are the fracture of the cutter ring,the eccentric wear of the cutter ring,the damage of the bearing,and the falling off of the retaining ring.Secondly,the influence of tunneling parameters and surrounding rock excavability FPI on tool wear is analyzed.The results show that the surrounding rock excavability index FPI can better reflect the normal wear and abnormal damage of tools than the thrust and penetration.At the same time,this article gives the relationship between the type of surrounding rock under the condition of tuffaceous sandstone and the wear per linear meter of each tool.Finally,based on the on-site excavation data of two TBMs with a diameter of7.0m,the thrust,penetration,speed,torque,FPI and TPI of the stable section of the TBM2 in different surrounding rock categories are selected as the basic indicators,which are respectively correlated with the excavation speed Analyze,optimize the BP neural network with the improved PSO algorithm,and establish the TBM tunneling speed prediction model.At the same time,the BP neural network improved by the BP neural network and the standard particle swarm algorithm is used to predict the tunneling speed.The training process and prediction effects of the three algorithms are compared and analyzed.The improved PSO-BP neural network has high prediction results.The number of iteration steps is less,and the prediction performance is better than the other two algorithms,which verifies the advantages of the model in this paper.And with the data of Xinjiang EH Project TBM3,the reliability of the improved PSO-BP network tunneling speed prediction model was tested.The prediction accuracy of the model meets the actual engineering needs.It is suitable for real-time prediction of tunneling speed and early warning of bad geological conditions,as well as auxiliary TBM tunneling parameters.Optimization adjustment.The research in this paper provides references for the analysis and prediction of tunneling performance of other similar TBM projects,as well as the understanding of the law of tool wear.