Research On Several Problems Of Microseismic Technology In Rock Mass Engineering

With the further implementation of "Western Development Strategy" and "One Belt and One Road Initiative",it is inevitable to encounter a lot of rock masses engineering problems with feature of "large buried depth and high stress".So,it is great important to carry out safety monitoring of engineering rock masses.Microseismic monitoring technology has been widely used in mining,hydropower,tunnel,energy storage and other fields in recent years,and has became one of the main means of rock mass engineering safety monitoring.However,this technology is a direct monitoring technology for physical parameters(such as amplitude,frequency,phase,etc.)rather than mechanical parameters(such as stress,strain,etc.).There are still many basic theoretical problems in the actual application process.Among them,the microseismic location method,the pick up of onset time of the microseismic signal P wave,and the focal mechanism based on the moment ten sor theory are the core theoretical issues of microseismic technology,and are also the current research hotspots of microseismic technology.It is of great theoretical and practical significance for promoting the application of microseismic technology and the effect of disaster warning in engineering to carry out relevant research on the above core theoretical issues.We combine theoretical analysis with practical engineering projects to carry out researchs and draw following conclusions:(1)Based on the focal mechanism of rock mass fracture,this paper firstly carried out the research of microseismic source location,summarized the advantages and disadvantages of the existing location methods,an d analyzed the main factors affecting the location accuracy.According to the factors that affect the location accuracy,the source location method and algorithm for picking P-wave onset time are improved respectively.Geiger method with problems of iterative initial value and not consider distance which contribute to informations missing is one of the most popular location method.Utilize Gaussian kernel function to define the distance weighting matrix,and use weighted least square method to solve the source parameter space.We called this as improved Geiger method.At the same time,combine L2 norm optimization meth od and improved Geiger method.In detail,we use L2 norm optimization method as iterative initial value of improved Geiger to solve the convergence problem due to initial value selection.In view of the existing algorithms for picking P-wave onset time,the accuracy is vulnerable to the influence of prior parameters.This paper firstly present a new method to pick onset time of P-wave based on dissipated power of damping energy.The seismic signal is transformed into response of a linear elastic fixed base single degree of freedom with viscous damping oscillator,then through the histogram method and AIC method to calculate the change of damping energy dissipation power to pick up the onset time of P wave.Compared with the classical STA/LTA and AIC methods,the proposed method has the advantages of not requiring the setting of prior parameters such as threshold value and time window,and can identify the noise waveform.Numerical test shows that the algorithm for picking P-wave onset time based on damping dissipated energy power has higher noise resistance and accuracy than STA/LTA and AIC methods.(2)On the basis of the source location,the focal mechanism solution of rock failure based on moment tensor is studied.Firstly,the integral expression of elastodynamics and the displacement field generated by the dislocation model are derived according to the mutuality of the displacement field.The selection of the far-field P-wave displacement time function under the point source approximation is explained in detail.The P-wave initial displacement time function is approximated by the sine function,which explain the misunderstanding caused by the insufficient explanation of the time function in existing domestic literatures.The expression of Green’s function for moment tensor inversion under point source approximation of uniaxial sensor is derived.In view of the shortcomings of the ISO,DC and CLVD components of moment tensor decomposition proposed by predecessors on the basis of laboratory hydraulic fracturing tests,the ISO,DC and CLVD components with compression components are further derived in this paper,and the source type is determined by DC-CLVD diamond diagram.(3)Based on Mohr-Coulomb criterion,the concept of main fault is defined by fault instability,which can determine the focal mechanism of main fault.As the moment tensor amplitude inversion is sensitive to noise,a moment tensor inversion method based on PCA(Principle Component Analysis)is adopted.The characteristic waveform of all effective sensors were solved by PCA.The correlation coefficient between the waveform received by the sensor and the characteristic waveform was taken as the weight and the PCA amplitude was taken as the effective amplitude of the moment tensor inversion.Eventually,the moment tensor was solved by the weighted least square method.Data test shows that this method can effectively reduce the sensitivity of the inversion process to noise compared with the traditional amplitude inversion method.To the inadequacy of using dislocation model to explain the non-shear component in engineering,the theory of generalized dislocation model method and its relationship with the moment tensor are discussed.Based on Ohtsu failure type criterion of rock masses,the ISO,CLVD and DC components expression are derived based on generalized dislocation model,and a new failure type criterion is established by the angle betw een the dislocation vector and normal vector of fault.This method does not need to decompose the moment tensor and solve the DC component,and the included angle can be directly solved by the eigenvalue of the moment tensor.At the same time,this new rock failure type crieterion is compared with the Ohtsu rock failure type criterion.The result shows this method can determine the fault type of compression(compression,compression-shear),which improves the existing methods for determining the fracture type of rock mass.(4)In view of the microseismic monitoring problems in practical engineering,based on the above theoretical methods,a valuable microseismic data from seismogenic process to seismogenic process in Xianglu Mountain.Jiangxi Province,which was detected on October 27,2018,was studied and analyzed.The temporal and spatial distribution of microseismicity from seismogenic process to seismic process is analyzed on the basis of the focal location.Based on the focal mechanism solution and the instability coefficient,the time evolution law of the fault is analyzed,and the occurrence of the main fracture surface is obtained.The rock mass failure types of the 17 location events are pressure-shear failure and tension-shear failure,which are in good agreement with the actual field results,indicating that it is feasible to apply the moment tensor theory used to study the large-scale rock mass failure in earthquake to the failure analysis of the mesoscale engineering rock mass.