Study On The Crack Source Mechanism From Quantitative Inversion And Its Mechanical Behaviors During Coal And Rock Fracture

In the process of deep coal mining,rock burst disasters occur frequently,which seriously threatens the safe and efficient production of coal mines.The essence of rock burst is the sudden change of stress and structure in coal and rock mass.During the preparation and development of rock burst disaster,there usually occurs large-scale fracture induced by crack propagation.Therefore,it's of great significance,to develop an accurate inversion method and to characterize the crack dynamic propagation process,for revealing the mechanism of rock burst disaster and then realizing reliable early-warning.Influenced by the coupling of geological and mining conditions,the deformation process of coal and rock mass is complex and the fracture tpyes are various including tensile fracture,shear fracture and tensile-shear mixed fracture.The source parameter solutions from quantitative seismology theory can well describes the shear-slip fracture process.However,how to explain the source mechanism of tensile fracture and mixed-mode fracture is still a critical scientific problem.In this thesis,by the means of theoretical modeling,laboratory experiments,numerical simulation and field verification,the crack source model for different fracture types was constructed and a quantitative inversion method was proposed.Additionally,the spatial-temporal evolution of crack source parameters was studied,and the dynamic characteristics of the crack propagation and fracture criteria were also discussed.Finally,the filed verification tests were conducted.The main work and conclusions are as follows.To tackle the problems of black-box data results and cumbersome waveform post-processing,an acoustic emission(AE)test system was independently established,and the corresponding waveform analysis algorithms were developed,which can provide high-quality acoustic data for monitoring the whole fracture process of coal and rock.The established AE system can realize multi-channel synchronous triggering to acquire of original AE time histories fully and losslessly.The autoregressive Akaike information criterion(AR-AIC)for AE onset time picking was improced by adding the noise filering and quality evaluation functions,which greatly enhanced the probability of reliable picking for the waveform with low signal-to-noise ratio.Based on the variance-covariance matrix and error ellipsoid theory,the three-dimensional confidence intervals of the AE source coordinates were deterimined.And location error evaluation index was proposed to eliminate poor-quality location results effectively.Crack source model for different fracture types was constructed,and an absolute quantitative inversion method for obtaining crack source parameters was proposed.The amplitude scalar sensitive parameter was defined to recover the collected electrical signals to the disturbance displacements,which quantified the coupling quality of the sensors and laid the foundation for the absolute inversion for crack source parameters.The crack source model involving only displacement discontinuity(DD)behavior,i.e.,tensile or shear fracture,was established.The relationship between the DD tensor and the crack source parameters was deduced,and it was found that the intermediate eigenvalue of DD tensor must be zero,which provides a necessary constraint,in the proposed crack source model.Such a constrained optimization problem was solved,by combing the Lagrange multiplier and Levenberg-Marquardt iterative approach,to calculate the DD tensor components.The crack source parameters,such as crack type,crack volume,crack orientation and crack motion direction,can be obtained.The applicability of the crack source model and the correctness of the quantatitive inversion method were verified by quartz glass fracture experiments.Using the constructed crack source model,the source parameters of microcracking events were obtained by inversion during the coal and rock fracture.Then the temporal-spatial evolution laws,energy dissipation characteristics and generation mechanisms of microcracks were revealed.The results show that the microcracks are mainly distributed in the fracture process zone(FPZ).The existence of FPZ is the inherent reason for the occurrence of precursor information during coal and rock fractrue.The smaller the FPZ size,the less obvious the precursor features before peak stage and the more severe the dynamic failure during the post-peak stage.Based on fracture mechanics,the calculation formula of absolute energy for crack source was deduced,which reveals a quantitative relationship between the released energy and the new-formed crack size(i.e.,crack volume).In this deduced formula,the difficulty of quantifying the source energy caused by the elastic wave attenuation is solved effectively.The absolute magnitudes of crack events also obey the well-known Gutenberg-Richter relationship,where the orientations of the large-magnitude events are almost parallel to propagation path of macroscopic crack,and their motions are mainly along the directions of the maximum principal stress.Therefore,the orientation and motion characteristics of crack events can be used to prejudge the fracture extension direction and stress field distribution,which provide important information for early-warning of rock burst disasters.Through the secondary development of discrete element method(PFC^3D),the above-mentioned experimental results were verified by simulation,and the results are in good agreement.Considering the influence of FPZ,the dynamic characteristics of crack propagation and fracture criteria were systematically studied.From linear elastic fracture mechanics(LEFM),the relation between the unloading compliance and traction-free crack length was established;The displacement gradient threshold was innovatively proposed to quantify the effective crack length.So,the accurate measurement of the FPZ size,during crack propagation,was realized.Based on the FPZ effective crack model,the Ba(?)ant size effect law for coal and rock fracture was derived,which were applied to calculate the fracture parameters for large-scale filed coal and rock materials from laboratory tests.Using the size effect law,the non-linear R-curve criterion for coal and rock fracture was obtained,which can well explain the post-peak behavior transformation from class-I stable“ductile”failure(i.e.,snap-through)in small size beams to class-II unstable“brittle”failure(i.e.,snap-back)in large size beams.The R-curve criterion provides a theoretical basis for the stability analysis of crack propagation in coal and rock materials.According to AE data processing method,field microseismic(MS)monitoring data were analyzed at a fully mechanized mining face of a rock burst mine.The diaster source of a typical roof-fracture-induced rock burst was located,and the direct parameters of this disaster source,including fracture type,fracture azimuth,fracture movement direction and fracture dimension,were got by crack source model inversion.The results of location and inversion are in good agreement with the actual situation.The energy released from roof-fracture source was calculated by the deduced energy formula for crack source,which was close to the value of MS energy recorded from SOS system.Combined with the space structure of stope,the precursor response and dynamic evolution of this roof-fracture-induced rock burst were analyzed,from the aspects of time-varying characteristics,spatial distribution,and focal mechanism,which provides a reference for monitoring and early-warning of similar disasters.In this thesis,the crack source model and quantitative inversion method were constructed to reveal the crack propagation mechanism and dynamic fracture behavior of loaded coal and rock.It's of great theoretical significance and practical value for further study on the mechanism and prediction of rock burst disaster.The thesis contains 107 figures,17 tables,and 203 references.