Multi-objective Optimization Method Of TBM Boring Parameters Under Complex Geological Conditions And Software Development

As the focus of our country’s resource development and transportation infrastructure construction projects shifts to the west,the construction of deep and long tunnels is indispensable,and the TBM construction method is the first choice.At present,the selection and adjustment of tunneling parameters during the construction of TBM tunnels basically relies on the experience of the driver.Under complex geological conditions,the tunneling parameters cannot achieve the optimal solution for multiple goals such as tunneling efficiency and tunneling cost,resulting in reduced tunneling efficiency and tunneling costs increase,resulting in unnecessary waste of resources.Therefore,it is necessary to study the multi-objective optimization method of TBM tunneling parameters in complex geological conditions to provide guidance for actual engineering.On the basis of extensive research on TBM tunnelling parameters,geological parameters,tunneling efficiency and cost,and multi-objective optimization methods,this thesis relies on the research project of "Key Technology Research on TBM Tunneling of Ultra-Extra Long Tunnels",mainly using theoretical analysis and numerical calculation methods,Research on the multi-objective optimization method of TBM tunneling parameters in complex geological conditions.The main research results include:(1)The pros and cons of mainstream multi-objective optimization algorithms are analyzed,and NSGA-Ⅱ,MOPSO,and NSGA-Ⅲ algorithms are selected.Research the core code,test problems,and performance indicators of the algorithm,realize and test and compare and analyze it in Mat Lab platform programming.The results show that the NSGA-Ⅲ algorithm is suitable for high-dimensional multi-objective optimization,and the distribution of the MOPSO algorithm is not as good as that of the NSGA-Ⅱ algorithm,but the overall convergence and convergence speed are better.(2)Summarized the TBM engineering technology and tunneling parameters,geological parameter acquisition and screening methods,and proposed a real-time acquisition method of TBM geological data;defined the wear coefficient to simplify the measurement of the number of tool wear;established two TBM tunneling parameters effective data and stable data Screening the set;analyzing the filtered data,it is concluded that the advancing speed in the stable section tends to the actual speed,the actual speed change in the adjustment section lags behind the advancing speed change,and the penetration is the true value only in the stable section.(3)Through statistical analysis,it is concluded that the thrust speed and the cutter head torque are strongly correlated;the regression expression of the tunneling control parameters(cutter head speed,advance speed)and the tunneling load(total thrust,cutter head torque)under specific geological conditions is established;The traditional regression model of the tunneling target according to the specific geological conditions of the Kashuang tunnel is used;the support vector machine and decision tree algorithm are used to identify the two-class lithology of the tunneling parameters,and the accuracy is above 80%;the BP neural network is introduced to predict the tunneling efficiency and accuracy is 86%,and then the NSGAⅡ-BP multi-objective optimization algorithm is proposed.In the case verification,the optimized data is better than 92% of the original data.(4)Established a multi-objective optimization mathematical model of TBM tunnelling parameters.Using Visual Studio platform,SQL SERVER database and Visual C# computer language to develop "TBM tunneling parameter multi-objective optimization analysis software",the optimization effect was improved by more than 15% in case verification.