Methods and Systems for Predicting Renormalization Groups and Evaluating Safety of Instability in Goaf Backfill
By combining renormalization group theory with equivalent blasting fatigue load, the problem that existing technologies cannot reflect multi-scale evolution processes is solved, enabling quantitative evaluation of the stability of backfill bodies and providing the maximum number of safe mining layers, which is applicable to on-site safety evaluation in mines.
Patent Information
- Application Number
- CN202611458027.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-18
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies cannot effectively reflect the multi-scale evolution process of local micro-damage, fracture clusters, and macroscopic connections caused by blasting fatigue loads. As a result, the stability evaluation results of the filling body can only provide binary conclusions or qualitative warnings, and cannot quantitatively determine the maximum number of safe mining layers in the footwall ore body.
By combining renormalization group theory with equivalent blasting fatigue load, and using the equivalent number of cycles as the external driving variable for renormalization group iteration, the cell-level micro-damage probability is coarsely recursively derived into the system-level instability probability, thereby constructing a continuously quantified safety index system and calculating the maximum safe mining layer layer by layer.
It realizes the inclusion of multi-scale self-similar processes from microscopic damage to macroscopic instability into the stability criterion, provides quantitative safety indicators, directly gives the maximum safe number of mining layers, and is suitable for routine safety assessment in mine sites.
Smart Images

Figure CN123115960A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of mine backfilling mining and rock mechanics, specifically to a method and system for predicting and assessing the renormalization group and safety of instability in goaf backfill bodies. Background Technology
[0002] In metal or non-metal mining, when the ore body has a large dip angle and there is a goaf above it, the common mining sequence is to first backfill the goaf in the hanging wall and then recover the ore body in the footwall. The core safety premise of this process is that the upper backfill body must remain stable throughout the subsequent mining of the footwall ore body. However, in actual engineering, two prominent problems are encountered: First, instability of the backfill body leads to a chain reaction of instability in the goaf. The shock waves and vibration loads generated by blasting operations during the mining of the footwall ore body repeatedly act on the upper backfill body. If the backfill body becomes unstable as a whole or collapses on a large scale, it will directly destroy the structural integrity of the upper goaf, triggering a chain of disasters such as rockfall and surface subsidence. Second, the mixing of backfill body with the ore exacerbates ore dilution. Once the backfill body cracks, falls off, or even collapses as a whole, its fragments will be mixed into the footwall ore pile being mined. Since the grade difference between the backfill body and the primary ore is significant, the mixing of backfill body with the ore body will directly lead to a decrease in ore grade, an increase in beneficiation costs, and a reduction in resource recovery efficiency. Therefore, determining the maximum number of mining layers in the footwall ore body after filling and treating the goaf in the upper hanging wall, while ensuring the stability of the filling body, has become a key technical bottleneck for the successful application of the pre-filling and post-mining process.
[0003] Currently, the evaluation methods for the stability of backfill bodies under the condition of pre-filling followed by mining, both domestically and internationally, mainly fall into the following categories: The empirical analogy method refers to similar mine cases and determines the number of safe mining layers based on experience. This method lacks quantitative basis and is prone to being conservative or risky. The limit equilibrium method treats the backfill body as a rigid body and calculates the anti-sliding safety factor according to static equilibrium conditions. This method does not consider the cyclic fatigue effect of dynamic blasting loads and cannot reflect the gradual instability process of microcrack initiation, propagation, and penetration. The numerical simulation method uses software such as finite difference and discrete element method to simulate the stress-strain response of the backfill body under blasting loads. Although this method can provide stress contour maps and displacement fields, it is difficult to capture the cross-scale abrupt changes from micro-damage to macro-instability, and it has a large computational load and strong parameter sensitivity, making it difficult to use for real-time decision-making. The single criterion method uses only the uniaxial compressive strength of the backfill body or the attenuation rate of acoustic wave velocity as stability indicators, ignoring the spatial non-uniformity and clustering effect of damage accumulation, and cannot predict the critical precursors before instability.
[0004] Renormalization group theory, originating from statistical physics, has been applied in fields such as earthquake prediction, material fracture, and rockburst early warning. Its core idea is to use scale transformation, i.e., coarsening, to recursively extrapolate the failure probability of micro-units to the instability probability of the macro-system, and to identify critical instability conditions at fixed points. In the mining field, some studies have also applied renormalization groups to instability early warning of pillar and backfill systems, optimization of boundary pillar thickness, and identification of roadway instability precursors. However, their technical objectives are respectively geared towards graded early warning, thickness optimization, and precursor identification, none of which involve solving for the specific number of mineable layers under layered mining conditions.
[0005] In summary, the common essential flaw of existing methods lies in their failure to incorporate the multi-scale evolution process of local micro-damage, crack clusters, and macroscopic continuity caused by blasting fatigue loads into the stability criterion. Summary of the Invention
[0006] To address the technical problems in the existing technologies, such as the overall criteria failing to reflect the multi-scale evolution of local damage, the evaluation results only providing binary conclusions of stability or instability or qualitative early warning levels, and the inability to quantitatively determine the maximum number of safe mining layers in the footwall ore body, this invention provides a method and system for predicting the reorganization group and evaluating the safety of instability of the goaf filling body. This method is applied to the pre-filling and post-mining process where the upper goaf is first filled and treated, and the footwall ore body is subsequently mined in layers.
[0007] This invention combines renormalization group theory with equivalent blasting fatigue load: using the equivalent number of cycles as the external driving variable for renormalization group iteration, the cell-level micro-damage probability is coarsely recursively derived into the system-level instability probability; and using the critical instability probability at the fixed point of the renormalization group as the physical threshold, a continuously quantifiable safety index system is constructed, and then the maximum safe mining layer number is given through layer-by-layer calculation, which can be directly used for mining design as a rigid constraint.
[0008] In one aspect, the present invention provides a method for predicting the renormalization group and assessing the safety of goaf backfill instability, the method comprising:
[0009] Determine the required strength of the backfill material for the treatment of the goaf in the upper hanging wall;
[0010] The filled goaf area is divided into a three-dimensional regular cell grid, and each cell is assigned attribute parameters including strength, elastic modulus, porosity, interfacial bonding coefficient and initial failure probability.
[0011] Based on the blasting design parameters for layered mining of the lower ore body, the peak vibration velocity of the particles transmitted to the upper filling body is calculated and converted into dynamic stress amplitude. According to the linear cumulative damage criterion, the blasting stress cycles of different amplitudes are converted into the equivalent number of cycles under the reference amplitude and accumulated layer by layer.
[0012] The fatigue damage and failure probability of each cell under the cumulative equivalent number of cycles are calculated based on the fatigue crack propagation equation and stiffness degradation rate.
[0013] The cellular grid is divided into primitive cells. A renormalization group transformation function is established with the through-connection path in the primitive cell as the failure criterion. The critical instability probability is obtained by solving the fixed point. The system instability probability is output through multiple iterations with the average failure probability of the cell as the initial value.
[0014] The safety index is composed of the renormalization group margin, the reduced safety factor, and the remaining lifetime ratio, calculated based on the system instability probability and the critical instability probability.
[0015] Starting from the first layer, the safety index is calculated layer by layer. When the safety index exceeds the limit, the calculation stops. The layer above the layer where the calculation stops is taken as the maximum number of safe mining layers and an early warning layer is output.
[0016] In the above method, cell discretization and attribute assignment provide a statistical basis and material parameters for multi-scale recursion; the conversion of equivalent cycle counts for blasting fatigue loads uniformly converts variable-amplitude and non-periodic blasting disturbances into an accumulative number of cycles, giving the renormalization group iteration a clear physical driving variable; the cell fatigue failure probability model transforms microcrack propagation and damage accumulation into calculable failure probabilities; the coarse-grained iteration of the renormalization group recursively converts the cell-level failure probability into the system-level instability probability through scale transformation, and provides a physically meaningful critical threshold at the fixed point; the safety index system maps the instability probability into a continuous quantitative index that can be directly understood and used by the engineering community; and the layer-by-layer calculation of the number of permissible mining layers directly binds the above probability evolution results with mining design parameters, outputting the number of executable mining permit layers. The above steps are sequentially linked to form a complete technical link from microscopic damage evolution to macroscopic mining permit determination.
[0017] In some preferred embodiments, the cell side length of the three-dimensional regular cellular mesh is between 0.5m and 2.0m, and the total number of cells is not less than 1000. The initial failure probability under static load is determined by the ratio of the relative strength deviation to the strength standard deviation using a standard normal distribution function.
[0018] This value ensures that the cell scale matches the microstructure scale of the filling body, while also ensuring that the total number of cells is sufficient to support the representativeness of the probability statistics. Using a normal distribution to characterize the intensity dispersion allows the initial failure probability to reflect the spatial non-uniformity of the filling body's strength, providing a reasonable initial state for subsequent fatigue accumulation.
[0019] In some preferred embodiments, the peak particle vibration velocity is calculated according to the Sadovsky formula from the maximum charge amount per segment, the center distance, and the site coefficient. The dynamic stress amplitude is obtained by multiplying the peak particle vibration velocity, the filling density, and the longitudinal wave velocity. The equivalent number of cycles for a single blast is taken as the power of the ratio of the dynamic stress amplitude to the reference stress amplitude.
[0020] This conversion link connects the design parameters that can be directly obtained in the blasting design with the dynamic response parameters of the filling body, so that the amplitude, frequency and attenuation law of each blast are uniformly converted into the standard cycle number, thereby transforming the non-periodic blasting disturbance into an accumulative damage driving quantity.
[0021] In some preferred embodiments, the fatigue damage degree of the cell is defined as the rate of elastic modulus degradation caused by crack propagation and is fitted by a power law form of the ratio of the number of cycles to the cell fatigue life. The cell failure probability is calculated as an exponential function that monotonically increases with fatigue damage degree, takes the value of zero when the damage degree is zero, and takes the value of less than one when the cell is completely damaged.
[0022] The power-law form allows fatigue damage to be directly obtained from the number of cycles and fatigue life, which is convenient for engineering applications. The exponential failure probability function does not fail when the damage is zero and still retains the possibility of not failing when the damage is complete, which is consistent with the physical fact that a cell can still maintain its load under geometric constraints. The scale parameter controls the steepness of the transition between damage and failure probabilities.
[0023] In some preferred embodiments, rainflow counting is performed on the measured blasting vibration waveform and the damage increment is accumulated successively, with the total damage amount replacing the fatigue damage degree calculated by constant amplitude cycles.
[0024] The actual blasting load is a variable amplitude cycle rather than a constant amplitude cycle. Rainflow counting can decompose the measured waveform into several stress cycles and accumulate them according to the linear cumulative damage criterion, so that the calculation of fatigue damage degree is closer to the actual load history on site.
[0025] In some preferred embodiments, the unit cell is a cubic unit cell composed of b×b×b units, the failure criterion is the existence of a failure unit cell connection path that runs through at least two opposite faces within the unit cell, and the renormalization group transformation function is established by enumerating the configurations of each failure unit cell.
[0026] Using the formation of a through-fracture surface as the criterion for unit cell failure directly corresponds to the physical fact in rock mechanics that "instability occurs as soon as a through-fracture surface is formed." The transformation function obtained through configuration enumeration gives a definite analytical expression to the recursion from the unit cell level failure probability to the unit cell level failure probability, which is convenient for fixed-point solution and iterative calculation.
[0027] In some preferred embodiments, the safety index includes the renormalization group margin, the reduced safety factor, and the remaining lifetime ratio, wherein the renormalization group margin is the difference between the critical instability probability and the system instability probability, the reduced safety factor is obtained by converting the ratio of the renormalization group margin to the critical instability probability through a mapping coefficient, and the remaining lifetime ratio is determined by the ratio of the current cumulative equivalent cycle count to the maximum allowable equivalent cycle count.
[0028] The three indicators characterize the same safety state from the perspectives of critical distance, engineering safety factor, and fatigue life, respectively. They not only retain the physical meaning of the critical threshold of the renormalization group, but also provide continuous quantitative results familiar to engineers, thus overcoming the problem of insufficient information granularity when only hierarchical early warning is used to represent the safety state.
[0029] In some preferred embodiments, the converted safety factor is set with three levels of engineering criteria: safe, critical, and dangerous, and a preset safety margin threshold is used as the stopping condition for calculating the number of layers to be mined layer by layer.
[0030] Clearly defined thresholds enable reproducible judgment rules for progressive calculations and allow evaluation results to be directly integrated with existing mine safety regulations and scheduling processes.
[0031] In some preferred embodiments, for layers determined to be warning layers, at least one of the following measures is taken before mining: reducing the maximum charge per section, increasing the inter-section delay, or reinforcing the upper backfill with grout. The safety degradation curve, margin curve, and maximum safe number of mining layers are then output.
[0032] The early warning layer and supporting measures enable the evaluation results to go beyond judgment and directly guide the adjustment of mining parameters and the formulation of reinforcement plans; the curve output provides mine scheduling with the safety status trend that changes with the number of layers.
[0033] On the other hand, the present invention also provides a renormalization group prediction and safety evaluation system for instability of goaf backfill, including a backfill strength determination module, a cell discretization module, a blasting load conversion module, a cell failure probability calculation module, a renormalization group iteration module, a safety degree calculation module, and a permitted mining layer determination module.
[0034] Each module of the system corresponds one-to-one with the steps of the aforementioned method, which facilitates the deployment of the method in the stability evaluation of mine backfill and the design of safe mining, and realizes the automated calculation from blasting design parameters to the number of allowable mining layers.
[0035] In another aspect, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for predicting and assessing the renormalization group instability of goaf filling bodies.
[0036] Compared with the prior art, the present invention has the following beneficial effects:
[0037] By equating blasting fatigue load to the number of cycles and using it as an external driving variable for renormalization group iteration, this invention can recursively transform the fatigue damage probability of micro-units into the instability probability of the macro-system through scale transformation. This incorporates the multi-scale self-similarity process of microcrack initiation, propagation, connection, and ultimately the formation of a through-fracture surface into the stability criterion, overcoming the shortcomings of the limit equilibrium method (which does not consider cyclic fatigue effects) and the single-criteria method (which ignores the spatial inhomogeneity of damage). By using the critical instability probability at the fixed point of the renormalization group as a physical threshold, and constructing a continuously quantified safety index system based on the renormalization group margin, the equivalent safety factor, and the ratio of remaining life, this invention can provide a quantitative answer to "how much safety is possible," rather than merely providing a binary conclusion of stability or instability or a qualitative warning level. By binding the blasting design parameters for layered mining of the footwall ore body with the equivalent number of cycles and calculating layer by layer, this invention can directly determine the maximum number of safe mining layers, answering the core engineering question of "how many layers can be mined at most," and providing a basis for refined mining permits based on layered blasting. By replacing iterative optimization with renormalization group transformation and iteration instead of repeatedly calling numerical models, this invention avoids the problems of large computational load, parameter sensitivity, and difficulty in real-time decision-making in numerical simulation methods. The entire evaluation process is based on analytical and iterative calculations of filling strength, blasting design parameters, and fatigue test parameters, without applying destructive loads to the filling, making it suitable for routine safety evaluations in mine sites. Attached Figure Description
[0038] Figure 1 This is a schematic diagram of the overall process of the method for predicting the renormalization group and evaluating the safety of instability of goaf filling bodies according to an embodiment of the present invention.
[0039] Figure 2 This is a schematic diagram of the safety degradation curve in an embodiment of the present invention;
[0040] Figure 3 This is a schematic diagram of the RG margin curve in an embodiment of the present invention;
[0041] Figure 4 This is a bar chart illustrating the instability probability evolution and safety factor in an embodiment of the present invention.
[0042] Figure 5(a) shows the relative positions of the structural models of goaf sections 4042 and 4043 in embodiment 4059.5 of the present invention;
[0043] Figure 5(b) shows the relative positions of the structural models of goaf sections 3041, 3042, and 3043 in Embodiment 4097 of the present invention;
[0044] Figure 5(c) shows the relative positions of the goaf, KT2-1 ore body and 3PD1 return airway structure model in an embodiment of the present invention;
[0045] Figure 5(d) shows the relative positions of the ore body structure model in the embodiment of the present invention;
[0046] Figure 5(e) shows the numerical calculation model of goaf filling and safe mining stability in an embodiment of the present invention;
[0047] Figure 6(a) is a schematic diagram of the distribution and evolution of the maximum principal stress during goaf filling in an embodiment of the present invention;
[0048] Figure 6(b) is a schematic diagram of the distribution and evolution of the maximum principal stress in the first, second and third layers of the KT2-1 section in the middle section of 4047 mining according to an embodiment of the present invention;
[0049] Figure 7(a) is a schematic diagram of the minimum principal stress distribution for filling the goaf in an embodiment of the present invention;
[0050] Figure 7(b) is a schematic diagram of the minimum principal stress distribution in the first, second, and third layers of the KT2-1 section in the middle section of section 4047 of the present invention in an embodiment of the present invention.
[0051] Figure 8(a) is a schematic diagram of vertical displacement analysis of goaf filling in an embodiment of the present invention;
[0052] Figure 8(b) is a schematic diagram of the vertical displacement analysis of the first, second and third layers of the KT2-1 section in the middle section of 4047 in an embodiment of the present invention;
[0053] Figure 9(a) is a schematic diagram of the layout of monitoring points for roof settlement in the goaf according to an embodiment of the present invention;
[0054] Figure 9(b) is a schematic diagram of the settlement curve of the goaf roof in an embodiment of the present invention. Detailed Implementation
[0055] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with embodiments. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.
[0056] This invention provides a method for predicting renormalization groups and evaluating safety in the case of instability of backfill in goaf areas. It is applied to a pre-backfilling and post-mining process where the upper goaf is first backfilled and treated, followed by layered mining of the lower ore body. This implementation includes a method implementation, a system implementation, and a computer-readable storage medium implementation. The method implementation mainly includes determining the strength of the backfill, cell discretization and attribute assignment, conversion of equivalent cycle number of blasting fatigue load, calculation of cell fatigue failure probability, coarse-grained iteration of the renormalization group, calculation of safety index, determination of the number of allowable mining layers, and numerical calculation verification. These are elaborated below.
[0057] Step S1: Treatment of goaf in the upper hanging wall and determination of the strength of the filling body.
[0058] This invention first determines the strength of the backfill material based on the scale and location of the upper goaf, using a combination of laboratory tests, engineering analogies, and theoretical formulas. The theoretical strength formulas can employ classic methods in the field of backfill strength determination, such as the Thomas method, Terzaghi method, and the Hawke-Brown strength criterion. Laboratory tests measure the uniaxial compressive strength, elastic modulus, and standard deviation of strength under different cement-sand ratios through backfill mix design experiments, providing statistical parameters for subsequent cellular attribute assignment. The output of this step is the backfill strength and its distribution characteristics required for subsequent cellular discretization.
[0059] Step S2: Discretization and attribute assignment of the cells of the goaf filling body.
[0060] The filled goaf area in the upper plate is divided into a three-dimensional regular cellular mesh. The cell side length is between 0.5m and 2.0m, and the total number of cells is no less than 1000 to ensure statistical representativeness. Each cell i is assigned the following initial attributes:
[0061] —The uniaxial compressive strength of cell i, MPa, was determined by the filler composition test;
[0062] —Elastic modulus, GPa;
[0063] —Initial porosity, dimensionless, determined by CT scan or MRI;
[0064] — Cohesion coefficient between the filling material and the surrounding rock, dimensionless, ranging from 0 to 1;
[0065] —The initial failure probability of a cell under static load is determined by the intensity distribution function:
[0066]
[0067] In the formula, It is the standard normal distribution function; The average uniaxial compressive strength of all cells, in MPa; The strength standard deviation is expressed in MPa.
[0068] Step S3: Calculation of equivalent cycle number of blasting fatigue load.
[0069] 3.1 Calculation of stress amplitude in a single blast.
[0070] Based on the blasting design parameters of the footwall ore body, the peak vibration velocity of the particles transmitted to the bottom of the hanging wall filling body is calculated using the Sadovsky formula:
[0071]
[0072] In the formula: The peak vibration velocity of the particle is given in cm / s. R is the maximum charge per section, in kg; R is the distance from the center of the blast source to the bottom of the filling material, in m. The field coefficient is determined by regression analysis of on-site blasting vibration tests.
[0073] Converting peak particle velocity to dynamic stress amplitude:
[0074]
[0075] In the formula: The density of the filling material is kg / m³. The longitudinal wave velocity of the filling body is m / s; The value is the dynamic stress amplitude, in Pa.
[0076] 3.2 Equivalent loop count conversion.
[0077] Using the Miner linear cumulative damage criterion, blasting stress cycles of different amplitudes are equivalent to the standard number of cycles under a baseline amplitude. Let the mining of the j-th layer contain... The stress amplitude of the k-th blast is... The equivalent number of cycles corresponding to this explosion is:
[0078]
[0079] In the formula: m is the reciprocal of the slope of the SN curve, reflecting the sensitivity of the fatigue life of the filling material to the stress amplitude; it is determined by the fatigue test of the filling material according to... The fitting results show that the cemented filler is a quasi-brittle porous composite material. Its microcrack propagation leads to a higher SN power law exponent than that of metallic materials. A large number of calculations show that this value is usually taken as 3~8. Specifically, it is determined by the mix proportion, curing age, uniaxial compressive strength and fatigue test under different stress levels. When there is no test data, m=5 is taken.
[0080] The reference stress amplitude is taken as 0.3 times the characteristic uniaxial compressive strength of the filling material, i.e. This level corresponds to a relatively representative medium-to-high dynamic stress amplitude in multiple downhole blasting disturbances, and is within the effective stress range commonly used in fatigue tests of filling bodies, making it convenient to generalize different blasting stress amplitudes to a unified benchmark.
[0081] After the first layer of mining is completed, the cumulative equivalent number of cycles is... The sum of the equivalent number of cycles for each layer and each explosion:
[0082]
[0083] Step S4: Cellular fatigue damage probability model.
[0084] 4.1 Fatigue damage evolution based on Paris formula.
[0085] This invention uses the Paris equation to describe the propagation behavior of microcracks in filled bodies under cyclic loading, and its basic form is as follows:
[0086]
[0087] In the formula: is the half-length of the crack, in mm; N is the number of cycles, dimensionless. This represents the stress intensity factor amplitude, in MPa·m. 1 / 2 C and n are material constants, determined by indoor fatigue crack propagation tests or equivalent fatigue life tests of the filling material. For tailings cemented filling materials with low cementitious material content and a 28-day uniaxial compressive strength of approximately 1-3 MPa, n can be obtained from tests or analogy with low-strength cementitious materials by approximately 2.5-3.5; for cemented filling materials with a strength of approximately 3-6 MPa, sufficient curing, and high density, n is approximately 3.2-4.5. This invention does not limit specific values; the actual filling material mix ratio and test results shall prevail.
[0088] For quasi-brittle materials such as infills, the relationship between the stress intensity factor amplitude and the stress amplitude and crack length is as follows:
[0089]
[0090] In the formula, Y is the geometric shape factor, which is dimensionless.
[0091] Substituting the above equation into the Paris formula and integrating, we obtain the result that the crack extends from the initial length a0 to the current length a. N Required number of loops:
[0092]
[0093] The expression for the crack half-length as a function of the number of cycles is obtained by solving the problem:
[0094]
[0095] 4.2 Definition of fatigue damage degree.
[0096] The fatigue damage degree of a cell is defined as the stiffness degradation rate caused by crack propagation:
[0097]
[0098] Based on the relationship between crack density and elastic modulus in damage mechanics (Mori-Tanaka method):
[0099]
[0100] In the formula: Let be the Poisson's ratio of cell i, which is dimensionless; The critical crack half-length is in mm, corresponding to the crack size when the cell completely fails.
[0101] Substituting the expression for crack half-length, we obtain an explicit expression for fatigue damage degree with respect to the number of cycles:
[0102]
[0103] To simplify engineering applications, the above equation is fitted into a power-law form:
[0104]
[0105] In the formula: The fatigue life of cell i, i.e., D i The number of loops when = 1; The damage nonlinearity exponent is determined by the Paris formula:
[0106]
[0107] 4.3 Consideration of the modification of blast load characteristics.
[0108] Actual blasting loads are not constant-amplitude cycles, but variable-amplitude cycles. This invention uses the rainflow counting method to decompose the measured blasting vibration waveform into several stress cycles, and performs cumulative damage calculation according to the Miner criterion. For the t-th blast, the measured vibration waveform, after rainflow counting, yields S. t There are 10 stress cycles, and the stress amplitude of each cycle is 10 ... The incremental damage caused by this blast is:
[0109]
[0110] The cumulative damage after T blasts is the sum of the damage increments from each blast:
[0111]
[0112] 4.4 Cell failure probability model.
[0113] The failure probability of cell i after the Nth equivalent cycle is defined as:
[0114]
[0115] In the formula: λ is a scaling parameter, which is calibrated by laboratory cyclic loading and unloading tests, and is generally taken as 0.15 to 0.35; The probability of cell failure is dimensionless and ranges from 0 to 1.
[0116] The physical meaning of this model is: when When it approaches 0, A value approaching 0 indicates that the device does not fail when there is no damage; when... When it approaches 1, It is still less than 1, which retains the physical meaning that the cell may not fail due to geometric constraints even in a completely damaged state; λ controls the steepness of the curve between damage and failure probability. The smaller λ is, the more sensitive the failure probability is to damage.
[0117] Step S5: Coarse-grained iteration of the renormalization group.
[0118] 5.1 Unit cell division and state determination.
[0119] The 3D cellular mesh is divided into primitive cells of size b×b×b (b is 2 or 3), and each primitive cell consists of b 3 The unit cell is composed of individual cells, and its state is determined by the failure modes of its internal cells. The failure state determination rule for a unit cell is defined as follows: when there is a failure cell path that penetrates at least two opposing surfaces, the unit cell is considered to have failed. This rule corresponds to the physical fact in rock mechanics that "the formation of a through-fracture surface signifies instability."
[0120] It should be noted that when b=3, the unit cell is 3×3×3, containing 27 units, and is still constructed according to the "opposite face failure connectivity" criterion. Determined by numerical enumeration or graph theory connectivity algorithm And solve for the fixed points. b=3 can improve the coarsening scale representation, but the configuration space is 2^27, and the computational cost is much greater than b=2. The two are equivalent under the same physical framework and convergence logic, differing only in scale and precision. In the example, b=2 is sufficient to determine the instability threshold of the filling body, the safety degradation, and the maximum safe number of mining layers L. safe Therefore, the following derivations and examples are given with b=2 as an example. It should be noted that b=3 is not excluded in the implementation of this invention, and the examples and core derivations are given with b=2, which is computationally efficient and applicable to engineering.
[0121] 5.2 Renormalization group transformation function.
[0122] Taking a cubic unit cell with b=2 as an example, the eight cells within the unit cell are numbered with coordinates (0,0,0), (1,0,0), (0,1,0), (1,1,0), (0,0,1), (1,0,1), (0,1,1), and (1,1,1). There are three possible penetration directions for failure: x, y, and z. Assuming each cell failure is an independent event with a probability p, then the unit cell failure probability p′ is equal to the sum of probabilities of all cell failure configurations leading to the formation of a penetration path. Enumeration is performed by classifying and listing the failures by the number of cells:
[0123] If all 8 cells fail: a through path will inevitably be formed, with a probability contribution of p. 8 ;
[0124] Seven cells fail: If any one cell survives, the other seven fail. Check if each case forms a through path. By enumeration, it is found that the survival of any one cell cannot block all three paths, resulting in C(8,1) = 8 configurations, with a probability contribution of 8p. 7 (1-p);
[0125] Six cells fail: Any two cells survive. We need to enumerate C(8,2) = 28 cases to find which surviving combinations can block all penetration paths. Geometric analysis shows that penetration in three directions can only be blocked when the two surviving cells are located at opposite ends of the diagonal of the same body (e.g., (0,0,0) and (1,1,1)). There are four such combinations. Therefore, C(8,2) - 4 = 24 configurations can form penetration paths, contributing 24p probabilities. 6 (1-p) 2 ;
[0126] Five cells fail: Any three cells survive. This requires enumerating C(8,3) = 56 cases. Analysis shows that when the three surviving cells form a triangle on the same face, they may block the penetration path. For example, if the surviving cells are (0,0,0), (1,0,0), or (0,1,0), they are on the bottom face but cannot block penetration in the z-direction (because there are still failed cells on the top face). In fact, penetration in that direction can only be blocked when three out of four cells on a face survive. Enumeration shows that 32 out of C(8,3) = 56 combinations form a penetration path, contributing a probability of 32p. 5 (1-p) 3 ;
[0127] Four cells fail: Any four cells survive. This requires enumerating C(8,4) = 70 cases. Analysis shows that a path through a given direction can only be blocked when the four surviving cells form a complete plane (e.g., all four cells on the bottom plane survive). There are six such planes (up, down, left, right, front, and back), corresponding to six survival combinations. Furthermore, other special configurations (e.g., cells surviving on two opposite edges) may also block the path. Enumeration shows that 16 out of the C(8,4) = 70 combinations form a path, contributing 16p probability. 4 (1-p) 4 ;
[0128] Three or fewer cells fail: a through path cannot be formed, and the probability contribution is 0.
[0129] The renormalization group transformation function for b = 2 cubic primitive cells is obtained by summing the results.
[0130] p′=R2(p=p 8 +8p 7 (1-p)+24p 6 (1-p) 2 +32p 5 (1-p) 3 +16p 4 (1-p) 4
[0131] 5.3 Solving for fixed points.
[0132] fixed point p c satisfy After rearranging and factoring out the common factor, p c =0 is a stable fixed point, and non-zero solutions are determined by the polynomial within the square brackets being zero. Using numerical methods such as Newton's iteration method, let:
[0133] f(p) = p 7 +8p 6 (1-p)+24p 5 (1-p) 2 +32p 4 (1-p) 3 +16p 3 (1-p) 4 -1
[0134] Iterate starting from the initial value p = 0.3 to obtain p. c ≈0.312.
[0135] 5.4 Iterative convergence and system instability probability.
[0136] Iterate over p′=R2(p): p (1) =R2(p), p(2) =R2(p (1) ), ..., p (n) =R2(p (n-1) And so on. The iterative convergence behavior is as follows:
[0137] When the initial p is less than p c When the results of each iteration monotonically decrease and tend to 0, it indicates that the system is macroscopically stable;
[0138] When the initial p is greater than p c When the value monotonically increases and tends to 1, it indicates macroscopic instability of the system;
[0139] When the initial p equals p c When the iteration result remains unchanged, the system is in a critical state.
[0140] After n iterations (usually convergence is achieved when n≥3), the probability of macroscopic instability of the system is... The average cell failure probability under the current equivalent number of cycles N After n-level renormalization group transformation, we get:
[0141]
[0142] The average value over all cells is obtained as follows:
[0143]
[0144] Where I represents the total number of cells.
[0145] Step S6: Construction of a security index system.
[0146] 6.1 Renormalization group margin.
[0147] The renormalization group margin is defined as the distance between the current instability probability of the system and the critical threshold.
[0148] M(N)=p c- P sys (N)
[0149] The system is in a stable state when M(N) is greater than 0; the system is in a critically unstable state when M(N) is equal to 0; and the system enters an unstable state when M(N) is less than 0.
[0150] 6.2 Calculate the safety factor.
[0151] Mapping the renormalization group margin to the safety factor form familiar to the engineering community:
[0152]
[0153] In the formula k fThe mapping coefficient is determined by the mine risk level and is generally taken as 2.5 to 3.0. When M(N) = p c At that time, F s (N)=1+k f This corresponds to the intact state; when M(N) = 0, F s (N) = 1, corresponding to the critical state; when M(N) is less than 0, F s (N) is less than 1, which corresponds to an unstable state.
[0154] Engineering Criterion Recommendation: F s (N)≥1.5 is considered safe, and normal mining is permitted; 1.2≤F s When (N) < 1.5, it is critical, requiring enhanced monitoring and appropriate reduction of blasting intensity; F s When (N) < 1.2, it is dangerous; mining should be stopped immediately and reinforcement measures should be taken.
[0155] The converted safety factor F s (N) is a continuously differentiable function with respect to the equivalent number of cycles N. Its continuous degradation with N characterizes the continuous depletion process of the safety reserve of the filling body, rather than a discrete division of the steady state.
[0156] 6.3 Remaining lifetime ratio.
[0157] The fatigue remaining life ratio of the filling material is defined as:
[0158]
[0159] In the formula The maximum permissible equivalent number of cycles for the filling material under the current blasting parameters is determined by... Confirmed, M th The safety margin threshold is typically set between 0.03 and 0.05. The value of η ranges from 0 to 1, with values closer to 0 indicating that the lifespan of the filling material is about to be exhausted.
[0160] Step S7: Determine the number of layers allowed to be extracted.
[0161] 7.1 Proceed with calculations layer by layer.
[0162] Assume the lower ore body is designed to be mined in L layers, and the calculation starts from the first layer and proceeds layer by layer. For the l-th layer (l = 1, 2, ..., L): based on the blasting design for this layer, calculate the cumulative equivalent cycle number N according to step S3. cum (l); N = N cum (l) Substitute into step S4 to calculate the failure probability p of each cell. i (N); Substitute into step S5 to perform renormalization group iteration, and obtain the system instability probability P. sys (l); Substitute into step S6 to calculate safety index M(l), F s(l), η(l).
[0163] 7.2 Determination of the maximum safe number of mining layers.
[0164] Starting from l=1, the determination is performed layer by layer. When M(l)≤M first appears... th or F s If (l) < 1.5, then the maximum safe number of mining layers is:
[0165] L safe =l-1
[0166] Simultaneously, the l-th layer is designated as the early warning layer. Before mining at this layer, one of the following measures must be taken: reduce the maximum charge per segment to decrease σ. a Increase inter-segment delay to reduce the equivalent number of cycles; reinforce the upper filling body with grout to improve σ. c,i and E i .
[0167] 7.3 Safety curve output.
[0168] The final output includes three results for mine scheduling decisions:
[0169] Safety degradation curve F s (l)~l, intuitively displaying the changing trend of the calculated safety factor with the number of mining layers;
[0170] The renormalization group margin curve M(l)~l shows the distance of the system from critical instability;
[0171] Maximum number of safe mining layers L safe This serves as a rigid constraint for mining design.
[0172] M th Originating from RG critical region residuals .in, The preset convergence tolerance coefficient is defined as pc, which represents the probability of the critical fixed point in the RG transform. This setting directly links the warning threshold to the "fixed point convergence accuracy," overcoming the blindness of existing technologies that use fixed empirical thresholds.
[0173] F s The third-level threshold is determined by M. th Through the same linear mapping Export (where k) f (The mapping coefficients are used). This indicates that the three-level warning (safe, critical, unstable) are not independent empirical values, but different scales of the same physical quantity, possessing mathematical consistency and derivability.
[0174] The progressive approach uses a recursive cumulative model: .in This is a newly added equivalent cycle for layer l. This mechanism ensures that fatigue damage increases monotonically with the layer sequence and does not reset, unlike existing methods that perform independent verification for each layer.
[0175] The judgment is based on priority and closed-loop recalculation (instability → grout replenishment → recalculation; criticality → load reduction → recalculation), forming a feedback control loop to ensure safe mining of the lower ore body.
[0176] Compared to traditional methods that only output discrete levels of stability or instability, this invention directly calculates the continuous variable values M(N) and F. s (N), η(N). The thresholds of these variables no longer originate from subjective experience, but from the fixed point pc of the renormalization group theory, possessing a clear physical critical meaning. Regarding the expression of safety reserves, traditional methods cannot quantify the reserve amount, only providing a qualitative judgment of the level; this invention, however, uses M(N) = p c- P sys (N) directly provides the probability distance of instability, realizing the numericalization of the safety reserve. Regarding the trend of change with the number of evolution steps N, traditional methods exhibit step-like jumps in early warning, are not differentiable, and struggle to capture continuous degradation of system performance; the index of this invention is continuous and differentiable, allowing for further calculation of the derivative dP. sys (N) / dN provides acceleration trend information for early warning. In terms of engineering output, traditional methods can only provide qualitative suggestions, while this invention can accurately output the safety reserve L. safe It can be directly used for quantitative engineering decisions such as blasting parameter allocation.
[0177] The following describes specific embodiments.
[0178] This invention uses the safe mining of KT2-1 copper mine in Shangri-La Hongniu as an example. The ore body dips at 80°, and there are 5 goafs in the hanging wall. The goafs need to be filled and treated before the ore body in the footwall can be mined. The volume of the goafs is approximately 120,000 m³.
[0179] The main parameters are as follows: cement-sand ratio of the filling material is 1:8, and the uniaxial compressive strength σ at 28 days is... c =1.54MPa, standard deviation s σ =0.48MPa; Filling density ρ = 2100kg / m³, Longitudinal wave velocity c p =2800m / s; the lower ore body is mined in 6 layers, each layer is 3m high, and the number of blasts per layer is K. j =6 times, maximum single-stage charge Q = 80kg; Sadovsky coefficient K = 173, α = 1.6, detonation center distance R varies with layer number: R1 = 20m, R2 = 25m, R3 = 30m, R4 = 35m, R5 = 40m, R6 = 45m; fatigue parameter m = 5, σ a,ref =0.96MPa, C f =10 -8λ=0.22, β i =0.85; renormalization group parameter b=2, p c =0.312, k f =2.8, M th =0.04.
[0180] The layer-by-layer calculation process and results are shown in Table 1 below:
[0181] Table 1
[0182] Floor number Explosion center distance R (m) <![CDATA[Dynamic stress amplitude σ a (MPa)]]> Equivalent number of cycles ΔN per cycle <![CDATA[Cumulative equivalent cycle number N cum > Cell failure probability p̄ <![CDATA[System instability probability P sys > Renormalization group margin M <![CDATA[Converted Safety Factor F s > 1 20 0.871 0.614 3.70 0.085 0.075 0.237 3.13 2 25 0.558 0.112 4.32 0.138 0.156 0.156 2.41 3 30 0.387 0.032 4.46 0.198 0.241 0.071 1.63 4 35 0.284 0.012 4.50 0.248 0.318 −0.006 0.94
[0183] The relationship between the number of mining layers and the equivalent safety factor, renormalization group margin, and system instability probability is shown in the figure. Figures 2-4 As shown.
[0184] Result: At the end of the 3rd layer, M(3) = 0.071 > M th =0.04, Fs(3)=1.63, about to enter the critical interval; at the end of the 4th layer, M(4)<0, Fs(4)<1.0, the system is unstable, therefore the maximum safe number of mining layers L safe =3 floors.
[0185] Engineering verification and numerical calculation verification.
[0186] In terms of engineering verification, the actual mine proceeded layer by layer according to the judgment logic of this invention (RG margin M and equivalent safety factor Fs). After the third layer was mined, the system determined that M was approaching the warning threshold M. th (Corresponding to Fs dropping to 1.63, rapidly approaching the critical range), so mining was suspended.
[0187] (1) Acoustic wave velocity quantification fit: Acoustic wave detection was performed on the filling material, and the measured wave velocity V3 decreased by 18.2% compared with the initial wave velocity V0. This attenuation magnitude is consistent with the critical damage degree predicted by the RG model of this invention (corresponding to M≈M). th The results show a high degree of agreement, verifying the physical consistency of wave velocity descent rate as a critical threshold.
[0188] (2) Microseismic nucleation trend is consistent: Monitoring shows that microseismic events are densely clustered at the bottom of the filling body, the frequency of events is significantly higher than the average of the first and second layers, and they show obvious nucleation trend, which is consistent with the phase transformation nucleation characteristics of the system before near instability predicted by RG theory.
[0189] (3) Grouting closed-loop verification: After grouting reinforcement (increasing σ) c(This caused M to rise), and the fourth layer was mined. After grouting, the measured wave velocity recovered to 92% of its initial value. During the advance, Fs remained ≥1.56 (safe zone), microseismic activity stabilized, and no backfill collapse occurred. This result quantitatively verifies the reliability and prediction accuracy of the "critical early warning-reinforcement-safe advance" closed-loop control of this invention.
[0190] In terms of numerical verification, a geomechanical model of the mine goaf filling was established to study the deformation, stress, and plastic zone distribution of the filling body within the goaf during mining, while monitoring the stope roof. A three-dimensional calculation model was established based on a comprehensive analysis of the occurrence characteristics, excavated orebody locations, excavation impact range, and surface topography of the shallow orebody in the Hongniu copper mine. The modeling process was as follows: First, the orebody outlines of the modeling locations circled in the horizontal CAD drawings of each segment were extracted to obtain the three-dimensional coordinates of the intersection points of the hanging wall and footwall boundaries along the strike of the orebody. Then, a box girder with preliminary design dimensions was established in the 3D mining software. The hanging wall and footwall surfaces were established using the obtained intersection coordinates. The box girder was further cut to obtain the orebody occurrence model, and the required excavated orebody portion was divided using the same process. Finally, Boolean operations were performed on each module, and the model node and unit information were input into auxiliary software to generate a finite difference model file.
[0191] The model measures 2100m x 1900m, with an elevation ranging from 3100m to 4300m. The ore body is located between elevations of 3821m and 4122m. The numerical model comprises two parts: the ore body and the surrounding rock, and is divided into 3,403,361 elements and 587,150 nodes. The computational model is as follows: Figures 5(a) to 5(e) As shown in Table 2, the parameter values for various materials in the model are as follows.
[0192] Table 2
[0193] Mineral name Bulk modulus K / GPa Shear modulus G / GPa Density ρ / kg·m⁻³ Cohesion c / MPa Internal friction angle φ / ° <![CDATA[Tensile strength σ t / MPa]]> Upper marble 3.22 2.11 2710 1.65 35.00 0.38 Lower hornfels 2.98 3.22 2660 1.60 34.00 0.37 Ore body 5.28 3.96 2920 2.03 43.43 0.24 filling body 1.53 0.98 1980 0.45 31.00 0.15 weathering layer 0.51 0.38 1850 0.05 23.20 0.02
[0194] Analysis was conducted on goaf profiles 3041, 3042, 3043, 4042, and 4043. The distribution of maximum principal stress is as follows: Figures 6(a) to 6(b) As shown, the goaf filling body is generally under low pressure, with no large-scale high pressure concentration. There are almost no tensile stress areas inside the filling body, indicating that the filling body as a whole will not experience tensile cracking. The compressive stress of most of the filling body is around 1.0 MPa. A small part of the filling body in goaf No. 4043 shows a compressive stress concentration area, with a compressive stress of approximately 4.0 MPa. The whole body is stable, and there are no areas with tensile stress exceeding 0.5 MPa. The filling body will not experience tensile failure, and its self-supporting stability meets the requirements.
[0195] Minimum principal stress distribution as follows Figures 7(a) to 7(b) As shown, the compressive stress of the filling body is generally in the range of 0.5MPa to 1.0MPa, with low pressure bearing capacity and good self-stability.
[0196] Reference Figures 8(a) to 8(b) Vertical displacement analysis shows that the settlement value of the roof of the filled goaf is generally less than 2.0 mm, with only a small settlement area appearing at the edge of the roof in some areas, and the maximum settlement in the whole area is about 4.0 mm. The deformation inside the filling body is uniform, and there are no large areas of high settlement zones, indicating that the filling body effectively supports the overlying rock strata and the roof has not undergone obvious bending and subsidence. The floor of the goaf has almost no uplift deformation, and the overall deformation is extremely small. The self-supporting effect of the filling is good, and the deformation of the surrounding rock is in the elastic stage with no risk of deformation instability.
[0197] In this simulation, one monitoring point was set up on the roof of each of the five goaf areas studied and the three surrounding goaf areas, for a total of eight monitoring points, as shown in Figure 9(a). The settlement curve of the goaf roof is shown in Figure 9(b). As the mining of the footwall continued, the settlement of the goaf roof increased continuously, and then tended to stabilize after mining was completed. The maximum settlement of the roof was 3.0 mm, and the minimum settlement was 1.6 mm, indicating that the goaf roof was stable.
[0198] Numerical calculation results show that when the five goaf areas are filled and treated, the stability of the filling body and the stope is good when the lower ore body is mined in layers. This is consistent with the reorganization group prediction results of this invention, proving the reliability of the method proposed in this invention.
[0199] This invention also provides a renormalization group prediction and safety evaluation system for instability of goaf backfill, used to implement the above-mentioned method. The system includes a backfill strength determination module, a cell discretization module, a blasting load conversion module, a cell failure probability calculation module, a renormalization group iteration module, a safety factor calculation module, and a permissible mining layer determination module. Each module corresponds to a specific function in the steps of the above method, facilitating the deployment of the method in the mine backfill stability evaluation and safe mining design process, and enabling automated calculation from blasting design parameters to the permissible mining layer.
[0200] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-mentioned method for predicting and assessing the renormalization group instability of the goaf filling body.
[0201] Through the above embodiments, the present invention can quantitatively provide the instability probability of the upper backfill under multiple blasting fatigue disturbances, the continuously quantified safety index, and the maximum number of safe mining layers of the lower ore body, based on the strength of the backfill, blasting design parameters, and fatigue test parameters, without applying destructive loads to the backfill, thus providing a basis for the safety evaluation and refined mining permitting of the pre-filling and post-mining process.
[0202] The above embodiments are merely illustrative of the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any equivalent substitutions or improvements made based on the technical concept of the present invention should be considered as falling within the scope of protection of the present invention.
Claims
1. A method for predicting and safety evaluation of goaf filling body instability by renormalization group, characterized in that, include: The step of determining the strength of the backfill is to determine the strength of the backfill required for goaf treatment by comprehensively using indoor tests, engineering analogies and theoretical strength formulas, based on the scale and location of the goaf in the hanging wall. The cell discretization and attribute assignment steps divide the filled upper goaf into a three-dimensional regular cell grid and assign attribute parameters to each cell. The attribute parameters include the cell's uniaxial compressive strength, elastic modulus, initial porosity, interfacial bonding coefficient between the filling body and the surrounding rock, and initial failure probability under static load. The steps for converting the equivalent cycle number of blasting fatigue load are as follows: Based on the blasting design parameters of each layer of the lower ore body, the peak vibration velocity of the particles transmitted to the upper filling body is calculated, the peak vibration velocity of the particles is converted into the dynamic stress amplitude, and the blasting stress cycles of different amplitudes are converted into the equivalent cycle number under the reference stress amplitude according to the linear cumulative damage criterion. The cumulative equivalent cycle number after the end of each layer of mining is obtained. The cell fatigue failure probability calculation steps are as follows: the propagation behavior of microcracks in the cell under cyclic loading is described based on the fatigue crack propagation equation; the fatigue damage degree of the cell is defined by the stiffness degradation rate caused by crack propagation; and the failure probability of each cell under the cumulative equivalent number of cycles is calculated based on the fatigue damage degree. The renormalization group coarsening iteration step divides the three-dimensional cellular mesh into primitive cells composed of several cells. The existence of a failure cell connection path that penetrates at least two opposite faces within the primitive cell is used as the failure criterion for the primitive cell. A renormalization group transformation function is established to recursively derive the primitive cell failure probability from the cell failure probability. The fixed point of the renormalization group transformation function is solved to obtain the critical instability probability. The average cell failure probability under the current cumulative equivalent cycle number is used as the initial value for multi-level iteration, and the system instability probability of the upper plate filling body is output. The safety index calculation steps are as follows: calculate the renormalization group margin based on the system instability probability and the critical instability probability, map the renormalization group margin to a converted safety factor, and calculate the remaining lifetime ratio based on the current cumulative equivalent cycle count and the maximum allowable equivalent cycle count, wherein the maximum allowable equivalent cycle count is determined by the condition that the renormalization group margin is equal to a preset safety margin threshold. The steps for determining the number of allowable mining layers are as follows: starting from the first layer of the footwall ore body, the calculation is performed layer by layer: based on the cumulative equivalent number of blasting designs for that layer, the cell failure probability, system instability probability, and safety index of that layer are calculated sequentially until a certain layer meets the condition that the renormalization group margin is less than or equal to the safety margin threshold, or the converted safety factor is less than the safety threshold. The layer above the stopped layer is taken as the maximum number of safe mining layers, and the stopped layer is output as the warning layer.
2. The method of claim 1, wherein, The cell side length of the three-dimensional regular cellular mesh is between 0.5m and 2.0m, and the total number of cells is not less than 1000; the initial failure probability under static load is determined by the ratio of the deviation of the uniaxial compressive strength relative to the average uniaxial compressive strength of all cells to the standard deviation of the strength using the standard normal distribution function.
3. The method of claim 1, wherein, The steps for converting the equivalent cycle number of blasting fatigue load specifically include: The peak vibration velocity of the mass point is calculated using the Sadovsky formula, based on the maximum charge amount per segment, the distance from the blast source center to the bottom of the filling body, and the site coefficient determined by regression from on-site blasting vibration tests. The peak vibration velocity of the particles is multiplied by the density of the filling material and the longitudinal wave velocity of the filling material to obtain the dynamic stress amplitude; Based on the linear cumulative damage criterion, the m-th power of the ratio of the dynamic stress amplitude of a single blast to the reference stress amplitude is taken as the equivalent number of cycles for that blast, where m is the reciprocal of the slope of the material's SN curve, obtained by fitting the fatigue test of the filling body. The equivalent number of cycles for each blast within each layer is accumulated layer by layer to obtain the cumulative equivalent number of cycles after the mining of each layer is completed.
4. The method of claim 1, wherein, The specific steps for calculating the cell fatigue failure probability include: The crack propagation rate of half-length with the number of cycles is described by the fatigue crack propagation equation, where the stress intensity factor amplitude is determined by the geometric shape factor, the dynamic stress amplitude, and the crack half-length. Integrating the fatigue crack propagation equation yields the number of cycles required for the crack to propagate from the initial half-length to the current half-length. The fatigue damage degree of the cell is defined by the rate of elastic modulus degradation caused by crack propagation, and the fatigue damage degree is determined by the relationship between crack density and elastic modulus in damage mechanics. The fatigue damage degree is fitted as a power law form of the ratio of the number of cycles to the cell fatigue life. The failure probability of the cell is calculated using an exponential function that increases monotonically with fatigue damage degree, takes zero when the damage degree is zero, and takes a failure probability value less than 1 when the fatigue damage degree is 1, based on the fatigue damage degree and the scale parameters calibrated by laboratory cyclic loading and unloading tests.
5. The method of claim 4, wherein, The method also includes variable amplitude load correction: rainflow counting is performed on the measured blasting vibration waveform, which is decomposed into several stress cycles. Each stress cycle is converted into a damage increment according to the linear cumulative damage criterion and accumulated successively. The total damage obtained by accumulation is used to replace the fatigue damage degree calculated according to the constant amplitude cycle.
6. The method according to claim 1, characterized in that, In the renormalization group coarsening iteration step, the unit cell is a cubic unit cell composed of b×b×b units, where b is 2 or 3; when b is 2, the renormalization group transformation function is: p' = R2(p) = p 8 + 8p 7 (1 - p) + 24p 6 (1 - p) 2 + 32p 5 (1 - p) 3 + 16p 4 (1 - p) 4 Where p is the cell failure probability and p′ is the unit cell failure probability; solve the equation p=R2(p) to obtain a non-zero fixed point, and use this non-zero fixed point as the critical instability probability; in the multi-level iteration, when the average cell failure probability is less than the critical instability probability, the system instability probability monotonically decreases and tends to zero, and when it is greater than the critical instability probability, it monotonically increases and tends to one. After the iteration reaches convergence, the system instability probability is output.
7. The method according to claim 1, characterized in that, In the calculation steps of the security index: The renormalization group margin is the difference between the critical instability probability and the system instability probability; a value greater than zero indicates stability, a value equal to zero indicates critical instability, and a value less than zero indicates instability. The calculated safety factor is determined by the ratio of the mapping coefficient to the renormalization group margin to the critical instability probability, and the mapping coefficient is determined by the mine risk level. The remaining life ratio is determined by subtracting the current cumulative equivalent cycle count from one to the maximum allowable equivalent cycle count. Its value ranges from zero to one, and the closer it is to zero, the closer the fatigue life of the filling body is to being exhausted.
8. The method according to claim 7, characterized in that, The mapping coefficient is between 2.5 and 3.0; the safety margin threshold is between 0.03 and 0.
05. The engineering criterion for calculating the safety factor is: When the converted safety factor is greater than or equal to 1.5, it is considered safe and normal mining is allowed; when it is greater than or equal to 1.2 but less than 1.5, it is considered critical and requires enhanced monitoring or reduction of blasting intensity; when it is less than 1.2, it is considered dangerous and mining is stopped.
9. The method according to claim 1, characterized in that, The method further includes: for the warning layer, before mining, taking at least one of the following measures: reducing the maximum charge amount per segment to reduce the dynamic stress amplitude, increasing the inter-segment delay to reduce the equivalent cycle number, or reinforcing the upper filling body with grout to improve the strength and elastic modulus of the filling body; and outputting the safety factor degradation curve with the number of mining layers, the reorganization group margin margin curve with the number of mining layers, and the maximum safe number of mining layers.
10. A system for predicting and assessing the renormalization group and safety of instability in goaf backfill, characterized in that, include: The backfill strength determination module is used to determine the required backfill strength for backfill treatment based on the scale and location of the goaf in the hanging wall. The cell discretization module is used to divide the filled upper goaf into a three-dimensional regular cell grid and assign attribute parameters to each cell; The blasting load conversion module is used to calculate the peak vibration velocity and dynamic stress amplitude of the mass points based on the blasting design parameters for layered mining of the lower ore body, and to convert the cumulative equivalent number of cycles according to the linear cumulative damage criterion. The cell failure probability calculation module is used to calculate the failure probability of each cell under the cumulative equivalent number of cycles based on the fatigue crack propagation equation and stiffness degradation rate. The renormalization group iteration module is used to establish a renormalization group transformation function with the intracellular interconnected path as the failure criterion, solve the fixed point to obtain the critical instability probability, and perform multi-level iterations to output the system instability probability. The safety factor calculation module is used to calculate the renormalization group margin, the reduced safety factor, and the ratio of remaining lifetime. The allowed mining layer determination module is used to calculate the safety index layer by layer starting from the first layer. When the safety index exceeds the limit, it stops and outputs the maximum safe mining layer and the warning layer.
