Under symmetric and unsymmetric loading acoustic emission ch
Update time：2009-03-06 11:11 Viewed：次
Abstract: using RFPA system developed by rock failure process analysis, to study the rock specimens under symmetric and asymmetric load presented during the failure process and failure process of the acoustic emission characteristics. Numerical simulation results demonstrate two kinds of conditions by micro fracture induced damage evolution process and macro and micro fracture source of acoustic emission events related to the spatial distribution features and the sequence of events. Numerical simulation results, points out that compared with symmetrical load compression experiment, the biggest difference is that in the asymmetric load compression local shear failure occurred in the experiments, and a more localized distribution of acoustic emission event source. In the physical sense of asymmetric load experiment provides the brittle failure under scheduled rupture path method, from the aspects of acoustic emission research for asymmetric experiment than symmetric to better observe the acoustic emission event source localization.
Abstract: the localization; Acoustic emission; Micro burst; Asymmetric add
1 lead it
Rocks, ceramics and other complex nature damage, is in the macroscopic damage happened before (main rupture) micro fracture (tiny). Due to the earth's crust contains many faults and joints, so earthquake is considered also contains faults and joints such as stress concentration source of micro fracture in the earth's crust. Based on this view, the earthquake with micro fracture of composite material such as rock, ceramic has much in common. Rock in the case of minor damage, and earthquakes have very similar characteristics: produce a called acoustic emission of elastic wave. Existing research shows that the micro fracture rock acoustic emission and the earthquake in more than 8 in the earth's crust level obey statistical regularity of similar magnitude. Therefore focus on the macro damage caused by micro fracture research to find the probability of a laboratory sample, based on the unified specification and seismic rock bursts of mathematical model.
The analysis of the acoustic emission localization is a slightly used to study the rock rupture rock macroscopic damage of common tools. Most of the past research is on rock sample under symmetrical load break experiment for studying the acoustic emission. This paper, by using the numerical method in symmetric experiments at the same time, the asymmetric experiment was carried out. Asymmetric load to provide a new kind of used to study the rock to determine crack location method of the brittle failure. In symmetrical experiments, uniaxial and triaxial, crack location with some kind of fuzziness, asymmetric experiment is to eliminate the ambiguity. Another to determine crack location method of rock brittle failure under study was recently conducted with initial healing joint localization of rock damage research in advance. Asymmetric test or experiment with initial macroscopic defects, acoustic emission concentration of localization domain is predetermined, or by heterogeneity of stress or by material of macroscopic defects. However, in the asymmetric experiments destroy path can also be determined in advance, and contains natural healing in the experiments of rock failure joint is not always the case.
This paper by using Rock Failure Process Analysis system developed by RFPA Rock Failure Process Analysis simulation in symmetric and asymmetric two kinds of Rock Failure Process under different load conditions. According to analyzing the results of numerical simulation of stress evolution in both cases, the formation process of failure mode, and in the process of rock failure with acoustic emission characteristics and parameters, and the numerical simulation results and symmetric and asymmetric compression experiment of the formation and release of stress, X-ray tomography CT of the proceeds of the failure mode and comparing the experimental results.
Two numerical simulation methods and models
2.1 numerical simulation model
At a rock two-dimensional network system, the grid is divided into m * n area of a square unit, such as this kind of treatment may cause the stress concentration area calculation precision is not enough, but for a certain sample and the number of acoustic emission calculation behind is needed. If a further row grid, helps to improve the calculation accuracy. For the present general nonlinear finite element, the introduction of unit nonlinear macro nonlinear constitutive model to simulate the whole material, does not take into account the material unit parameters, such as rock heterogeneity and the characteristics of randomness. Research shows that 395, the cause of the formation rock material nonlinear macro may not be the medium of the mesoscopic unit nonlinear, but the medium of the mesoscopic unit heterogeneity and randomness. This heterogeneity makes rock material continuously in the process of carrying through the destruction of the mesoscopic element, of mesoscopic unit cell damage caused by macro media continuum damage, thus forming a macroscopic phenomenon of nonlinear deformation.
In view of the randomness of the rock materials, the unit cell and heterogeneity, this article take the Monte Carlo method and the combination of statistical description initialized on unit assignment. Mesoscopic unit cell parameters (strength, elastic modulus, etc.) the heterogeneity can be used to describe a statistical distribution function, combined with previous experience and the experimental results, this paper adopts Weibull statistical distribution form. All units in the network body strength and elastic modulus of the threshold O and E, respectively, each unit strength and elastic modulus of the threshold for O and E,
They should obey Weibull statistical distribution, namely
Type of m1, m2 for statistical distribution function of shape parameters, shape parameters show that the greater the material is uniform, while they tend to be infinite, becomes the ideal inhomogeneous brittle materials. The statistical distribution of unit cell of a sample space, even though they may be the same, the threshold of distribution function but not exactly the same, the spatial distribution of these element composition material on the macroscopic properties, such as rock may be broadly, the main difference lies in the details of the mesoscopic structure, reflects the mesoscopic disorder, this kind of mesoscopic disorder
Just embodies the discreteness of the unique characteristics of rock materials. General random distribution in the physical space disorder by Monte Carlo method to do this, this has not only statistical, and randomness of the parameter (disorder) unit method to satisfy the heterogeneity of network in the unit cell parameters and stochastic demands.
2.2 numerical simulation method
RFPA is used in this article the rock failure process analysis system between linear elastic finite element and nonlinear finite element of a "quasi nonlinear finite element program, 345. Will first imposed by external load is divided into tiny step load, load step size should be divided into can describe the actual loading process in a certain approximation degree of continuity, each load step is actually a linear elastic calculation, stress, strain and displacement are preserved, but the next step load is in the previous step calculation on the basis of stress and strain level. For the first load step, I have
The next step load calculation; If a unit is broken, and then based on the acoustic emission of rock and rock damage consistent assumption, that the ratio of acoustic emission is proportional to the unit area of the rock damage and calculate each acoustic emission events of energy:
After the strain energy, according to its size in the acoustic emission in the circle mark in the list
Unit burst, in the next step calculation assumes that the broken unit continue to exist, but a sharp drop in the elastic modulus and strength. This course of weakening rupture units, and satisfactory results are obtained in the finite element calculation. Calculated with the method of iterative, grow in certain boundary conditions and rules (unit cell damage rules), should be determined by the finite element stress rupture of unit cell. Unit burst, the boundary conditions change, the need to solve the stress field equations, so repeated solving until the steady state. Slowly changing outer boundary conditions, repeat the above process at a time, until materials macroscopic fracture.
3 the results of numerical simulation and analysis
3.1 micro fracture collapse induced macroscopic damage process
Figure 3 shows the rock micro fracture under symmetrical loading induced macroscopic damage evolution process. In the initial loading phase micro fracture is mainly characterized by randomness and disorder, the effects of single micro fracture range is small, the link between the micro fracture effect is very weak, the statistical independence between micro fracture. Increased with the increase of the displacement of the rock damage increasing, micro fracture more reinforcement effect of the correlation between the performance of micro short-range correlation, and within the scope of the local small scale cascade, macro crack appeared in the central part of the sample, but still maintain the overall stability of the whole system. In high damage
Phase, cascade micro fracture from small scale to large scale cascade, statistical fluctuation is more and more intense, micro fracture mutual correlation function is more and more big, finally through the whole sample, form the shear fracture mode. According to different properties of the plate and sample, of course, high ratio of differences may also form X conjugate macroscopic failure mode.
Figure 4 shows the micro fracture rock under asymmetrical loading induced macroscopic damage evolution process. Similar and symmetric load condition, in the initial stages of loading micro fracture characterized by randomness and disorder. However, because under the condition of asymmetric load, the sample of the boundary of the load and load not exist on the surface of a high stress, high, so the micro burst sooner cascade at local scope, so as to make the sample in the high stress, high produce macroscopic crack much earlier. And then with the increase of load displacement, crack along the stress concentration of the stress distribution as determined surface extending from the top of the sample to the low end, but the sample still not lose bearing capacity, only
The whole stress-strain curve at this time to create a larger fluctuation. As the displacement of the load continues to increase, the broken area is transferred to other area of the sample, until the whole sample lose bearing capacity. Figure "for the asymmetric load conditions observed by X-ray tomography CT) granite failure mode. The experimental results were compared with the results of numerical simulation it can be seen that the failure modes of both very similar, from the analysis can also see that behind the final failure mode are nearly the same, not only the whole space evolution of the acoustic emission event source is almost the same.
3.2 acoustic emission event source space sequence
Figure 6 is the space evolution of the acoustic emission event source under symmetrical load. In low damage phase, due to the heterogeneity of rock materials, acoustic emission presented a random distribution of the disorder. With the increase of load displacement, stress, though still lower than the compressive strength of sample, but the acoustic emission event source presents the localization, the distribution of acoustic emission events usually source of stress in the evolution and the sample of clusters are consistent, in the sample at the beginning of the middle and lower cascade. In the latter stages of the destruction of acoustic emission events to expand to the side of the sample
The formation of the world at this time, eventually fracture surface. Then the largest loss of acoustic emission events and present a more even distribution. Less on specimen near the end of acoustic emission events may be due to the end friction.
Figure 7 under asymmetrical loading the space evolution of the acoustic emission event source. And symmetric load, at the initial stage, the acoustic emission events took on a random distribution of the disorder. With the increase of load displacement, near the border with sample loading and loading of acoustic emission is very active. Then acoustic emission events shear fracture surface of tracing the propagation of the crack formation, transfer to the bottom from the top of the sample. After shear failure surface formation, active region of acoustic emission is transferred to other regions of the sample. In contrast,
Figure 8 shows a granite rock sample under asymmetrical loading source of acoustic emission events in the loading process of evolution, distribution of tracking the shear failure surface in the process of from top to bottom to form the outline of acoustic emission events. After shear failure form, the acoustic emission active region is transferred to the rest of the area of the sample. Figure 8 just shows the current acoustic emission source, had accumulated acoustic emission source, figure 7 shows not only the current acoustic emission source also shows the cumulative source of acoustic emission events before. Through the contrast can be seen that the numerical simulation shows the evolution process of the spatial distribution of acoustic emission event source and experimental results shows the evolution of the process is very similar.
Blog
