Digital Identification Of Rock Mass Structure And Characterization Of Mechanical Parameter Variability And Its Engineering Application

A rock mass is a geological object consisting of rock blocks and discontinuities.As the basic elements of a rock mass,discontinuities control the physical and mechanical properties and failure boundaries of the rock mass.The randomness of the spatial distribution of rock mass discontinuities makes rock mass mechanical parameters uncertain.However,deterministic parameters are widely used in analyzing many rock masses engineering problems,making it difficult to obtain robust evaluation conclusions for related engineering design,mechanism research,and risk decisionmaking.Therefore,precisely quantitatively characterizing the structural characteristics of rock masses and the variability of their mechanical parameters has important theoretical significance and engineering value for solving complex engineering problems of rock masses.Given this,the research focuses on the key issues of digital identification of rock mass structure and characterization of mechanical parameter variability.It adopts multidisciplinary interdisciplinary research methods such as algorithm research and program development,theoretical and statistical analysis,numerical simulation,and engineering analysis.The proposed digital identification method of rock mass structural characteristics is organically connected with the rock mass mechanical parameter evaluation method,and then the digital rock mass structure and mechanical parameter evaluation software based on multi-source data is developed.Based on the Wuyue Pumped Storage Power Station Project,the analysis and engineering application of the various characteristics of rock mass structure and mechanical parameters were carried out,forming a closed loop between the research and engineering application of digital identification of rock mass structure and mechanical parameter variability characterization methods.The main findings of the study are as follows:(1)A method for identification of rock mass discontinuity based on the multirule region growth algorithm is proposed.The K d-tree algorithm is used to organize the point cloud data,the principal component analysis algorithm is used to calculate the normal vector and curvature of the point cloud,and the multinormal pair consistency voting algorithm is used to optimize the calculation result of the normal vector for the sharp point cloud.The discontinuity recognition strategy of "growth first and optimization" is adopted to merge the oversegmentation recognition results of the traditional region growth algorithm with the multirange constraint optimization strategy,which reduces the amount of discontinuity calculation and obtains a robust recognition result.The comparison test of several data shows that the proposed method shows good accuracy and computational efficiency.(2)An FCM-WOA clustering algorithm for global optimization of the orientation of rock mass discontinuity is constructed.A cluster mean multiple method for eliminating the orientation of noise data is proposed,which solves the problem that traditional methods are sensitive to noise data.The initial clustering center is determined by max-min-distance methods,and the fuzzy c-means(FCM)algorithm’s clustering findings are then used to find the global optimal solution by the whale optimization algorithm(WOA),which significantly increases the accuracy of the global clustering of discontinuity orientation.Combined with the elbow method,the automatic determination of the optimal clustering number is completed,and a new clustering method for discontinuity orientation of rock mass is formed.Multi-data testing shows that the proposed method shows great progress in noise data elimination,global optimization clustering,and automatic grouping.(3)The quantitative characterization method of the geometrical characteristics and development characteristics of the rock mass structure is improved.Based on the recognition of discontinuities and occurrence grouping,a quantitative characterization method of discontinuity spacing based on point cloud data is proposed.Based on the definition of equivalent discontinuity trace,the method of calculating the threedimensional trace length of rock outcrop discontinuity is perfected.A quantitative multi-angle roughness coefficient characterization method is proposed,considering discontinuous roughness anisotropy.Based on image recognition technology,a quantitative characterization method for the aperture of rock mass discontinuities is proposed.Then,the digital characterization and random feature analysis of rock mass structural features of specific engineering examples are realized.(4)Based on the quantification of structural properties,a method for empirically estimating rock physical parameters is devised.Based on the generalized Hoek?Brown empirical strength criterion and the digital characterization of rock mass structural characteristics,the calculation expression of the geological strength index(GSI)based on structural rating(SR)and digital joint characteristic factor JDC is established,and the GSI correction method of rock mass considering groundwater condition are proposed.An evaluation system of rock mass mechanical parameters based on GSI digital characterization is proposed,which realizes the effective connection between GSI quantification and digital characterization of rock mass structure.The proposed method is applied to evaluate rock mass mechanical parameters in an engineering example.Comparative analysis shows that the proposed process can effectively assess the mechanical parameters of the rock mass.(5)A digital evaluation system for rock mass structure and mechanical parameters for engineering applications is developed.A digital evaluation framework of rock mass structure and mechanical parameters is proposed,integrating field data collection,preprocessing,and digital analysis.It focuses on the requirements of convenient acquisition of multi-source rock mass outcrop data,rapid reconstruction of point cloud data,and programmatic evaluation of rock mass properties.The engineering deployment framework for 3D point cloud reconstruction and the lightweight capture equipment for multi-source rock outcrops data improved.The digital assessment software Rig RM for rock mass structure and mechanical parameters is created based on the MATLAB platform.This software enables the quantitative characterization of rock mass structure data and the programmed evaluation of empirical mechanical parameters,allowing the research findings to be applied to engineering.(6)The analytical framework of mechanical parameter variability based on the randomness of rock mass structure is formed,and the engineering application is carried out.A digital evaluation framework for mechanical parameters based on the randomness of rock mass structure is proposed.Combined with Rig RM software,the probability distribution function of the rock mass structure and mechanical parameters of the underground powerhouse of the Wuyue Pumped Storage Power Station are determined.The point estimation method is used to carry out the probability analysis of the stability of the main underground caverns,revealing the probability of the stability of the caverns under the condition of variation in rock mass mechanical parameters.According to the surrounding rock support pressure and the probability distribution of the plastic zone of the surrounding rock of the cavern,the anchoring support scheme of the underground cavern is determined;the numerical simulation analysis of the effectiveness of the selected support measures shows that the anchoring scheme determined based on the probability of the plastic zone can well enhance the stability of the underground cavern.