Method for evaluating environmental protection mining and filling efficiency risk in building dense area

Abstract

The application provides a kind of building dense area environmental protection mining filling efficiency risk assessment method, it is related to the field of intelligent mining, specifically includes: using finite element analysis software, the physical model of mining and filling of mining area is established;Strength development of filling body under different time and stress conditions is simulated by numerical simulation method;Using finite element post-processing function, stress and strain data of different parts of filling body are obtained;Combined with the flow, pressure data in the filling process, the compactness of the filling body is evaluated;The stability of the system composed of filling body and surrounding rock is analyzed by using digital twin model;The surface subsidence is calculated by using finite element model, and whether there is subsidence risk is determined;A multi-index comprehensive early warning model is established to generate suggestions for optimizing mining and filling parameters.The technical scheme of the application overcomes the problem that the existing technology cannot effectively control surface subsidence and reduce material waste, and realizes safe and efficient green mining in complex environment.

Description

Technical Field

[0001] This invention relates to the field of intelligent mining, specifically to a method for risk assessment of the environmental protection mining and backfilling efficiency in densely built-up areas. Background Technology

[0002] Against the backdrop of continuously rising global demand for mineral resources, mining operations in densely built-up areas are becoming increasingly common. To effectively ensure building safety and implement green and environmentally friendly mining principles, backfilling mining has become a widely used and important method. However, current methods for measuring backfilling effectiveness have significant shortcomings. Traditional measurement methods rely excessively on subjective human experience, and the monitoring data they are based on is extremely limited, making it difficult to reflect the actual effectiveness of the backfill in a real-time, comprehensive, and accurate manner throughout the entire mining process. For example, there is a lack of precise and effective quantitative methods for assessing the strength development process of the backfill; detailed detection of the spatial distribution of density within the backfill is impossible; and there are both theoretical and practical shortcomings in analyzing the synergistic mechanism between the backfill and the surrounding rock. These problems directly lead to difficulties for workers in adjusting mining parameters in a timely manner based on the true state of backfilling effectiveness during the mining process, thus significantly increasing the risk of surface subsidence, causing serious waste of backfill materials, and resulting in more severe damage to the surrounding ecological environment.

[0003] Therefore, there is a need for a risk assessment method for environmentally friendly mining and backfilling efficiency in densely built-up areas that can effectively control surface subsidence, reduce material waste, and achieve safe and efficient green mining in complex environments. Summary of the Invention

[0004] The main objective of this invention is to provide a method for risk assessment of the environmental protection mining and backfilling efficiency in densely built-up areas, in order to solve the problem that existing technologies cannot effectively control surface subsidence and reduce material waste, and achieve safe and efficient green mining in complex environments.

[0005] To achieve the above objectives, this invention provides a method for risk assessment of the effectiveness of environmentally friendly mining and backfilling in densely built-up areas, specifically including the following steps: S1 integrates ground-penetrating radar data, borehole camera image data, and rock mechanics parameter data.

[0006] S2 collects backfilling parameters and mining progress data. Backfilling parameters include: the energy change relationship of backfilling materials in the pipeline and the hydration reaction rate of backfilling materials.

[0007] S3. Using the geological data integrated and processed in step S1, a physical model of mining and backfilling in the mining area is established using the finite element analysis software ANSYS.

[0008] S4. The real-time data collected in step S2, including filling parameters and mining progress, is transmitted in real time to the physical model constructed in step S3 through the data interface. Using data interpolation and fitting algorithms, the discrete monitoring data is mapped to continuous physical model nodes to obtain a digital twin model after data mapping and dynamic updating.

[0009] S5, based on a digital twin model, uses numerical simulation to simulate the strength development of the filling material under different time and stress conditions.

[0010] S6 analyzes the stress distribution and deformation inside the filling body in the digital twin model, and uses the finite element post-processing function on the digital twin model to obtain stress and strain data of different parts of the filling body; combined with the flow rate and pressure data during the filling process, the compactness of the filling body is evaluated.

[0011] S7. Considering the interaction between the surrounding rock and the backfill, the stability of the system composed of the backfill and the surrounding rock is analyzed using a digital twin model at different mining stages.

[0012] S8 uses a digital twin model to calculate the surface subsidence and, combined with the building's allowable subsidence threshold, determines whether there is a risk of subsidence.

[0013] S9 establishes a multi-indicator comprehensive early warning model. Based on the simulation results of the digital twin model, combined with the genetic algorithm, suggestions for optimizing mining and backfilling parameters are generated.

[0014] Furthermore, step S1 specifically includes the following steps: S1.1, using ground-penetrating radar, the transmission frequency range is set to... Between MHz, the range of underground penetration depths is The system explores the stratigraphic structure and geological features; using borehole camera equipment, it takes 360-degree pictures of the inside of the borehole to obtain detailed information on rock texture and joints and fractures.

[0015] S1.2, rock mechanical parameters were obtained through triaxial compression tests; during the test, a servo-controlled loading system was used, with a loading rate range of [missing information]. Axial pressure was applied to the rock specimen at a speed of mm / min. and lateral pressure Calculate the elastic modulus of rock based on Hooke's Law : ; in, Let A be the stress on the rock, and A be the cross-sectional area of ​​the rock specimen. For rock strain, This is the axial deformation. This represents the axial deformation.

[0016] S1.3, a weighted fusion algorithm is used for feature fusion. For the first... Feature vectors of data The fused feature vector for: ; in, For the first The weights of each data point are calculated using the entropy weighting method: ; in, For the first The data point number The normalized values ​​of each feature. Furthermore, step S2 specifically includes the following steps: S2.1, Install pressure sensors and electromagnetic flow sensors on the filling pipeline, and analyze the energy change relationship of the filling material in the pipeline according to Bernoulli's equation: ; in, Density of the filling material; For flow rate, , To fill the cross-sectional area of ​​the pipe, For traffic; For height, , The pressure difference between two points It is the acceleration due to gravity. , , , , and The subscripts 1 and 2 represent any two cross sections inside the filling pipe.

[0017] S2.2, Multiple temperature and humidity sensors are installed at the filling site, and the hydration reaction rate of the filling material is quantified by combining the Arrhenius equation with a humidity correction coefficient: ; in, Pre-exponential factors; It is the activation energy; It is the gas constant; For temperature; This is the humidity correction function.

[0018] S2.3 uses a laser rangefinder to measure distance based on the principle of triangulation. : ; in, Baseline length; The focal length of the camera; The displacement of feature points in the image is calculated using an image matching algorithm. Image matching employs the ORB-SLAM algorithm, which extracts ORB feature points, calculates descriptors, and uses Hamming distance for feature matching. The matching error is... satisfy: ; in, and For the first The pixel coordinates of a pair of matching points The threshold value is used.

[0019] S2.4. Image recognition technology is used, employing a Convolutional Neural Network (CNN) model to identify mining areas. The CNN model is trained using labeled mining area image data. The CNN model uses a ResNet architecture, and the loss function is cross-entropy loss. : ; in, For the sample size, For the number of categories, For real labels, To predict probabilities.

[0020] Furthermore, step S3 specifically includes the following steps: S3.1, In the modeling process, the continuous rock mass and infill are discretized into a finite number of elements. Taking a two-dimensional plane stress problem as an example, the equilibrium equation of the element is expressed as: ; in, For a plane stress problem, the elements of the element stiffness matrix are... for: ; in, To perform integration on the volume of a single cell; The volume of a single finite element; and These are the units. The node, the first Each node corresponds to a sub-block of the strain matrix. for transpose, It is the elasticity matrix; The element node displacement vector. , , They are unit nodes exist , Displacement in direction; For the element nodal force vector, , , These are the effects applied to the unit nodes. Above , Force in a certain direction.

[0021] S3.2, Considering the mechanical properties of the rock, the Mohr-Coulomb strength criterion is used to describe the failure behavior of the rock. The yield condition of the Mohr-Coulomb strength criterion is: ; in, , Principal stress, For cohesion, The internal friction angle is used; the Burgers model is used to describe the rheological properties of the filling material, and the stress-strain relationship is... for: ; in, , It is the elastic modulus; , The viscosity coefficient; for Stress at any moment for Adaptability at all times This is the first derivative of strain with respect to time.

[0022] Furthermore, step S4 specifically includes the following steps: S4.1, using Lagrange interpolation, for a given... Data In the interval Interpolation function within for: , ; in, Interpolation interval The coordinates of any point to be solved within the interior.

[0023] S4.2, For dynamic data updates, the Kalman filter algorithm is used for data fusion. The state equation and observation equation are as follows: ; ; in, For state vectors, For the observation vector, refer to The state transition matrix at time t, refer to The observation matrix at time, and These are process noise and observation noise, respectively.

[0024] Furthermore, step S5 specifically includes the following steps: S5.1, During the simulation, the hydration reaction process of the filling material, the influence of temperature and humidity, and the degree of hydration reaction should be considered. Over time Changing models for: ; in, For the final degree of hydration, and These are the model parameters.

[0025] S5.2, Combining the filling material samples collected on-site, use ultrasonic testing equipment to measure the propagation speed of ultrasonic waves in the filling body. According to strength The strength of the empirical relationship is estimated using the empirical formula: ; in, , The coefficients related to the filling material were determined through fitting a large amount of experimental data. The least squares method was used for fitting, and the objective function was: ; in, To find the minimum value function, For the first The measured compressive strength values ​​in the experimental data set For the first The measured values ​​of ultrasonic propagation speed in the experimental data set.

[0026] Furthermore, step S6 specifically includes the following steps: S6.1, based on porosity The relationship between volumetric strain and volumetric strain was obtained through finite element simulation to obtain the volumetric strain of the filling material. Therefore, the porosity can be estimated. To assess density: , ; in, The volumetric strain of the filling material. This represents the volume change of the filling material. This represents the initial volume of the filling material.

[0027] S6.2, If stress concentration areas exist within the filling material, stress cloud diagram analysis should be used to determine if these areas have insufficient density; stress concentration factor. Defined as: ; in, For local maximum stress, This is the nominal stress.

[0028] Furthermore, step S7 specifically includes: The system stability is evaluated using the limit equilibrium method. Taking slope stability analysis as an example, the safety factor is... The calculation formula is: ; in, It is cohesive force; The length of the sliding surface; For normal force, , For the first Normal stress on the sliding surface For the corresponding area; It is the internal friction angle; For downward force, , For the first The total weight of the sliding body. For the first The inclination angle of the smooth surface, is the sine of the tilt angle.

[0029] Furthermore, step S8 specifically includes: Calculation of surface subsidence using the finite element model Combined with the building's allowable settlement threshold [ ],when At that time, it was determined that there was a risk of settlement; the settlement prediction model is expressed as: ; in, This represents the total number of strata in the overlying rock above the mining area. This is the settlement influence coefficient; Additional stress caused by mining; This corresponds to the thickness of the strata.

[0030] Furthermore, step S9 specifically includes the following steps: S9.1, the calculation model for early warning indicators is as follows: ; in, This is a comprehensive early warning value; For the first The weight of each indicator; For the first The dimensionless value of each indicator, ,in, The actual value of the indicator. and These are the minimum and maximum values ​​of the indicator.

[0031] S9.2 employs a genetic algorithm for optimization, outputting the optimized parameter combination and the corresponding objective function value. The optimization objective functions include minimizing infill cost, maximizing infill efficiency, and minimizing surface subsidence. Represented as: ; in, For filling costs, , For the first The unit price of the material For the first The amount of each material used; A comprehensive score for filling efficiency; This refers to surface subsidence. , , The weights of each objective.

[0032] Constraints include: Strength constraints: ,in, Minimum strength required for the filling material; stability constraints: ,in, Minimum safety factor required by the system; construction constraints: ,in, and These are the minimum and maximum values ​​of the filling flow rate, respectively.

[0033] The present invention has the following beneficial effects: This invention constructs a closed-loop architecture encompassing "physical entity - data acquisition - digital twin model - performance evaluation - intelligent decision-making - entity feedback." In practice, the data acquisition module first uses ground-penetrating radar and sensors to acquire geological parameters, filling parameters, and mining progress data. Then, the digital twin model construction module uses finite element analysis and Kalman filtering techniques to establish a dynamic virtual model that accurately maps to the physical entity. Subsequently, the filling performance evaluation module quantitatively evaluates the performance from dimensions of strength, density, and stability. Finally, the intelligent decision-making module generates tiered early warnings and parameter optimization suggestions based on the evaluation results, which then influence the mining site. In practical applications, this method can effectively control surface subsidence and reduce material waste, enabling safe, efficient, and green mining in complex environments. Attached Figure Description

[0034] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings: Figure 1 A flowchart of a method for risk assessment of environmentally friendly mining and backfilling efficiency in densely built-up areas according to the present invention is shown.

[0035] Figure 2 A graph showing the comparison of infill strength monitoring accuracy is presented.

[0036] Figure 3 A bar chart comparing the accuracy of multidimensional measurements is shown.

[0037] Figure 4 A comparison chart of the accuracy of infill strength monitoring is shown.

[0038] Figure 5 The diagram illustrates the convergence process of the multi-objective genetic algorithm.

[0039] Figure 6 A radar chart showing the overall system performance is displayed.

[0040] Figure 7 The graph shows the real-time monitoring fluctuation of the filling flow rate. Detailed Implementation

[0041] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] like Figure 1 The method for risk assessment of environmental protection mining and backfilling efficiency in densely built-up areas, as shown, specifically includes the following steps: S1 integrates ground-penetrating radar data, borehole camera image data, and rock mechanics parameter data.

[0043] S2 collects backfilling parameters and mining progress data. Backfilling parameters include: the energy change relationship of backfilling materials in the pipeline and the hydration reaction rate of backfilling materials.

[0044] S3. Using the geological data integrated and processed in step S1, a physical model of mining and backfilling in the mining area is established using the finite element analysis software ANSYS.

[0045] S4, Virtual Model Mapping: The real-time data collected in step S2, including filling parameters and mining progress, is transmitted in real time to the physical model constructed in step S3 through the data interface; using data interpolation and fitting algorithms, the discrete monitoring data is mapped to continuous physical model nodes to obtain a digital twin model after data mapping and dynamic updates.

[0046] S5, Strength Assessment: Based on the digital twin model, numerical simulation methods are used to simulate the strength development of the filling material under different time and stress conditions.

[0047] S6, Compaction Assessment: By analyzing the stress distribution and deformation inside the filling body in the digital twin model, and using the finite element post-processing function on the digital twin model, stress and strain data of different parts of the filling body are obtained; combined with the flow rate and pressure data during the filling process, the compactness of the filling body is assessed.

[0048] S7, Calculate the slope safety factor: Considering the interaction between the surrounding rock and the backfill, the stability of the system composed of the backfill and the surrounding rock is analyzed using a digital twin model at different mining stages.

[0049] S8, Surface Settlement Assessment: The finite element model mounted on the digital twin model is used to calculate the surface settlement, and combined with the building's allowable settlement threshold, it is determined whether there is a risk of settlement.

[0050] S9 establishes a multi-indicator comprehensive early warning model. Based on the simulation results of the digital twin model, combined with the genetic algorithm, suggestions for optimizing mining and backfilling parameters are generated.

[0051] Specifically, step S1 includes the following steps: S1.1, utilizing high-resolution ground-penetrating radar, the transmission frequency range is set at... Between MHz, the range of underground penetration depths is It can clearly detect the stratigraphic structure and geological formation; using borehole camera equipment, it can take 360-degree pictures of the inside of the borehole without blind spots, and obtain detailed information on rock texture and joints and fractures.

[0052] S1.2, rock mechanical parameters were obtained through triaxial compression tests; during the test, a servo-controlled loading system was used, with a loading rate range of [missing information]. Axial pressure was applied to the rock specimen at a speed of mm / min. and lateral pressure Calculate the elastic modulus of rock based on Hooke's Law : ; in, Let A be the stress on the rock (Pa) and A be the cross-sectional area of ​​the rock specimen (m²). For rock strain, The axial deformation is expressed in meters (m). The axial deformation is expressed in meters (m).

[0053] S1.3 utilizes multi-source data fusion technology to integrate ground-penetrating radar data, borehole camera image data, and rock mechanics parameter data. A feature layer fusion method is employed to extract and match data features from different data sources, constructing a high-precision geological model and laying a solid foundation for subsequent analysis. A weighted fusion algorithm is used for feature fusion, specifically for the first... Feature vectors of data The fused feature vector for: ; in, For the first The weights of each data point are calculated using the entropy weighting method: ; in, For the first The data point number The normalized values ​​of each feature. Specifically, step S2 includes the following steps: S2.1, Analyze the energy changes of the filling material inside the pipeline: Install a high-precision pressure sensor on the filling pipeline, and its measurement accuracy can reach... MPa, capable of real-time monitoring of filling material delivery pressure When paired with an electromagnetic flow sensor, the measurement accuracy is... Accurately acquire traffic (m³ / h).

[0054] The simplified form of Bernoulli's equation, which analyzes the energy change relationship of the filling material inside the pipe, is as follows: ; in, The density (kg / m³) of the filling material was determined accurately through laboratory testing. The velocity is (m / s). , The cross-sectional area (m²) of the filling pipe. For traffic; The height (m) is calculated using the relationship between pressure difference and density. , The pressure difference (Pa) between two points. The acceleration due to gravity (9.81 m / s²) , , , , and The subscripts 1 and 2 represent any two cross sections inside the filling pipe.

[0055] S2.2, Calculate the hydration reaction rate of the backfill material: Multiple temperature and humidity sensors are installed at the backfill site, with a temperature measurement accuracy of [insert accuracy here]. ℃, humidity measurement accuracy is %RH. Studies have shown that temperature and humidity The hydration reaction rate of the filling material It has a significant impact; the hydration reaction rate of the filling material is quantified by combining the Arrhenius equation with a humidity correction factor: ; in, The pre-exponential factor (s⁻¹) was obtained through experimental fitting. The activation energy (J / mol) was determined using thermal analysis experiments. The gas constant is 8.314 J / (mol・K); For temperature; The humidity correction function is determined based on an empirical model built from a large amount of experimental data, for example... ,in , , These are the coefficients obtained through experimental fitting.

[0056] S2.3, Mining progress data acquisition: A laser rangefinder is used, with a measurement accuracy of up to [missing information]. mm, distance measurement based on triangulation principle : ; in, The baseline length (m) is precisely calibrated. The focal length (m) of the camera is precisely calibrated at the factory. The displacement (m) of feature points in the image is calculated using an image matching algorithm. Image matching employs the ORB-SLAM algorithm, which extracts ORB feature points, calculates descriptors, and performs feature matching using Hamming distance. The matching error is... satisfy: ; in, and For the first The pixel coordinates of a pair of matching points The threshold value is used.

[0057] S2.4. Image recognition technology is used, employing a Convolutional Neural Network (CNN) model to identify mining areas. The CNN model is trained using labeled mining area image data. The CNN model uses a ResNet architecture, and the loss function is cross-entropy loss. : ; in, For the sample size, For the number of categories, For real labels, To predict probabilities, laser ranging data is combined to track the advance position and extraction volume of the mining face in real time, providing crucial data for subsequent backfilling efficiency analysis.

[0058] Specifically, step S3 includes the following steps: S3.1 During the modeling process, the continuous rock mass and filling body are discretized into a finite number of elements. The element type is selected according to the actual situation, such as tetrahedral elements, hexahedral elements, etc.

[0059] Taking a two-dimensional plane stress problem as an example, the equilibrium equations of the element are expressed as follows: ; in, The element stiffness matrix (Pa) is calculated using material properties and element geometry. For plane stress problems, the elements of the element stiffness matrix are... for: ; in, To perform integration on the volume of a single cell; The volume of a single finite element; and These are the units. The node, the first Each node corresponds to a sub-block of the strain matrix. for transpose, It is the elasticity matrix; The element node displacement vector. , , They are unit nodes exist , Displacement in direction; For the element nodal force vector, , , These are the effects applied to the unit nodes. Above , Force in a certain direction.

[0060] S3.2, the physical model simulates the deformation process of the surrounding rock after ore body mining and the supporting effect of the infill material on the surrounding rock at different stages. Considering the mechanical properties of the rock, the Mohr-Coulomb strength criterion is used to describe the failure behavior of the rock. The yield condition of the Mohr-Coulomb strength criterion is: ; in, , Principal stress (Pa). Cohesion (Pa). The internal friction angle is given in rad. The Burgers model is used to describe the rheological properties of the filling material, and the stress-strain relationship is shown. for: ; in, , The elastic modulus (Pa) was determined experimentally. , The viscosity coefficient (Pa·s) was determined using rheological experiments. for Stress (Pa) at time t. for Adaptability at all times The first derivative of strain with respect to time is (s⁻¹).

[0061] Specifically, step S4 includes the following steps: S4.1, using Lagrange interpolation, for a given... Data In the interval Interpolation function within for: , ; in, Interpolation interval The coordinates of any point to be solved within the interior.

[0062] S4.2, by updating data in real time, the virtual model can accurately reflect the dynamic changes of the actual mining and backfilling process. For dynamic data updates, a Kalman filter algorithm is used for data fusion, and the state equation and observation equation are as follows: ; ; in, For state vectors, For the observation vector, refer to The state transition matrix at time t, refer to The observation matrix at time, and These are process noise and observation noise, respectively. and Let covariance be the variance. The Kalman filter recursive formula is: ; ; ; ; ; in, Based on The optimal estimate at time t, for The prior estimate of the system state at time t; refer to The covariance matrix of the prior estimate of the system state at time step; refer to The Kalman gain matrix at time t is the core parameter of the Kalman filter.

[0063] Specifically, step S5 includes the following steps: S5.1, During the simulation, the hydration reaction process of the filling material, the influence of temperature and humidity, and the degree of hydration reaction should be considered. Over time Changing models for: ; in, For the final degree of hydration, and These are model parameters, which are related to temperature and humidity.

[0064] S5.2, Combining the filling material samples collected on-site, ultrasonic testing equipment is used, with a frequency range of [missing information]. By measuring the propagation speed of ultrasound in the filling material (m / s), based on strength Intensity estimation is performed based on the empirical relationship (Pa), and the empirical formula is as follows: ; in, , The coefficients related to the filling material were determined through fitting a large amount of experimental data. The least squares method was used for fitting, and the objective function was: ; in, To find the minimum value function, For the first The measured compressive strength values ​​in the experimental data set For the first The measured values ​​of ultrasonic propagation speed in the experimental data set.

[0065] Meanwhile, core sampling was used to take samples of the filling material on-site and conduct compressive strength tests in the laboratory to verify the accuracy of the simulation results.

[0066] Specifically, step S6 includes the following steps: S6.1, based on porosity The relationship between volumetric strain and volumetric strain was obtained through finite element simulation to obtain the volumetric strain of the filling material. Therefore, the porosity can be estimated. To assess density: , ; in, The volumetric strain of the filling material. Volume change of the filling material (unit: m) 3 ), The initial volume of the filling material (unit: m) 3 ).

[0067] S6.2, If there are large stress concentration areas inside the filling material, stress cloud diagram analysis should be used to determine if the stress concentration areas have insufficient density; stress concentration factor. Defined as: ; in, For local maximum stress, This is the nominal stress.

[0068] Specifically, step S7 is as follows: The system stability is evaluated using the limit equilibrium method. Taking slope stability analysis as an example, the safety factor is... The calculation formula is: ; in, Cohesion (Pa) is determined through in-situ testing or laboratory experiments. The length of the slip surface (m) is calculated from the model. The normal force (N) is calculated based on the stress distribution. , For the first Normal stress (Pa) on the sliding surface. The corresponding area (m²); The internal friction angle (rad) was determined experimentally. The sliding force (N) is calculated from gravity and slope. , For the first Total weight of the sliding body (N). For the first The inclination angle (rad) of the smooth surface. The sine of the inclination angle is used to characterize the effect of the sliding surface slope on the sliding force (the steeper the slope, the greater the force). The larger the value, the greater the downward force. The larger (the larger).

[0069] Specifically, step S8 is as follows: Calculation of surface subsidence using the finite element model Combined with the building's allowable settlement threshold [ ],when At that time, it was determined that there was a risk of settlement; the settlement prediction model is expressed as: ; in, This represents the total number of strata in the overlying rock above the mining area. The settlement impact coefficient was determined through regression analysis of historical data. The additional stress (Pa) caused by mining is calculated using a digital twin model; This represents the corresponding stratum thickness (m).

[0070] Specifically, step S9 includes the following steps: S9.1 Establish a multi-indicator comprehensive early warning model, transforming assessment results such as intensity, density, and stability into early warning indicators. Set early warning thresholds; when an assessment indicator exceeds the threshold, different levels of early warning signals (e.g., yellow, orange, and red alerts) are triggered. The early warning indicator calculation model is as follows: ; in, This is a comprehensive early warning value; For the first The weights of each indicator are determined using the Analytic Hierarchy Process (AHP). For the first The dimensionless values ​​of each indicator are calculated using a linear normalization method. ,in, The actual value of the indicator. and These are the minimum and maximum values ​​of the indicator.

[0071] S9.2 employs a genetic algorithm for optimization, outputting the optimized parameter combination and the corresponding objective function value. The optimization objective functions include minimizing infill cost, maximizing infill efficiency, and minimizing surface subsidence. Taking multi-objective optimization as an example, the objective function... Represented as: ; in, The cost of filling (in yuan). , For the first The unit price of the material (yuan / ton). For the first Amount of each material used (tons); The comprehensive score for filling performance is calculated by weighting indicators such as strength, density, and stability. Surface subsidence (m); , , The weights of each objective are determined using an expert scoring method.

[0072] Constraints include: Strength constraints: ,in, Minimum strength (Pa) stability constraint required for the filling material: ,in, Minimum safety factor required by the system; construction constraints: ,in, and These are the minimum and maximum filling flow rates (m³ / h), respectively.

[0073] The main steps of using a genetic algorithm for optimization include: Encoding: The mining and backfilling parameters (such as backfilling material ratio, backfilling flow rate, mining step distance, etc.) are encoded in binary or real numbers.

[0074] Initialize the population: Randomly generate a certain number of initial solutions as the initial population.

[0075] Fitness calculation: Calculate the fitness value of each individual based on the objective function.

[0076] Selection: Selecting superior individuals from the population through selection operations such as roulette wheel selection and tournament selection.

[0077] Crossover: Perform a crossover operation on the selected individuals to generate new individuals.

[0078] Mutation: Performing mutation operations on individuals to increase population diversity.

[0079] Iteration: Repeated selection, crossover, and mutation operations until the termination condition is met (such as the maximum number of iterations, convergence of the objective function, etc.).

[0080] Output the optimized parameter combination and the corresponding objective function value.

[0081] The optimal mining and backfilling parameters output by the genetic algorithm are transformed into control commands by the intelligent decision-making module. After being fed back to the physical equipment for adjustment, the system collects new data through sensors and updates the digital twin model. The optimization effect is verified in a virtual environment and then evaluated a second time by the performance evaluation module. If the expected results are not achieved, the new data is input into the genetic algorithm to start a new round of optimization, forming a closed loop of "parameter optimization - implementation feedback - model update - performance evaluation - algorithm iteration" to ensure that mining under the structure is always in a safe and efficient state.

[0082] To verify the beneficial effects of the present invention, comparative tests were conducted. Table 1 shows the comparison data between the effects of the present invention and the traditional method.

[0083] Table 1 Comparison Data Table The above comparative tests are explained as follows: 1. All data are derived from parallel tests under the same geological conditions, building protection levels, and mining scale; 2. Calculation method for improvement / optimization: (value of traditional method - value of this invention) / value of traditional method × 100% (error index), (value of this invention - value of traditional method) / value of traditional method × 100% (benefit / safety index); 3. Allowable settlement threshold, minimum safety factor, etc. are all set with reference to the relevant standards of the "Technical Specification for Safety of Mining Under Buildings".

[0084] As shown in Table 1, the method proposed in this invention has high measurement accuracy, enhanced safety assurance, effectively controls surface subsidence and reduces material waste, and has high construction feasibility.

[0085] To verify the beneficial effects shown in Table 1 of this invention, a comparative experiment was conducted.

[0086] like Figure 2 The diagram shows the introduction of additive white Gaussian noise to simulate sensor interference in a mining environment, demonstrating the data fusion effect based on Kalman filtering. The red curve represents the traditional method, which is greatly affected by environmental noise fluctuations; the blue curve represents the method provided by this invention, which filters out noise through an algorithm and closely follows the black true value curve.

[0087] Regarding the measurement errors in the compaction of the backfill, surface settlement, and system stability safety factor monitored in the comparative experiments, such as... Figure 3 As shown, it is intuitively demonstrated that the error of traditional methods is between 15% and 22% in four dimensions, while the error of the present invention is reduced to less than 5%.

[0088] like Figure 4 As shown, the comparison of settlement prediction based on the finite element model is presented. The red curve shows that under the traditional working conditions, the settlement on the 15th day exceeded the safety threshold of 40mm, reaching 52mm. The green curve shows that after parameter optimization, the settlement curve was suppressed within the safe range and finally stabilized at around 28mm.

[0089] The comparative experiment used a Mohr-Coulomb constitutive model for the rock with fixed boundary conditions. The red curve represents the traditional parameters, which exceeded the allowable threshold on the 7th day of mining. The green curve represents the optimized model, which, by adjusting the early strength of the backfill and using a genetic algorithm to provide the optimal mix ratio, effectively supported the overburden and controlled the final settlement to 28 mm. This demonstrates the effectiveness of the method provided in this invention.

[0090] Regarding the calculation of economic benefit data, Table 1 shows that the unit filling cost decreased by 30.8%, from the original 185 yuan to 128 yuan. For example... Figure 5 The diagram shows the convergence process of the genetic algorithm, which is the solution result of the multi-objective optimization function.

[0091] The blue line shows that as the number of iterations increased, the cost dropped rapidly from 185 yuan and stabilized at 128 yuan; the red line shows that the performance score rose from 62 points to 91 points.

[0092] Figure 6 This presents a comparative score between traditional methods and the present invention in terms of safety assurance and economic benefits.

[0093] This invention also includes real-time monitoring and compliance analysis of filling flow rate. Figure 7 This represents a comparison of the filling flow rate monitored in real time.

[0094] Of course, the above description is not intended to limit the present invention, and the present invention is not limited to the examples given above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A method for risk assessment of the effectiveness of environmentally friendly mining and backfilling in densely built-up areas, characterized in that, Specifically, the steps include the following: S1 integrates ground-penetrating radar data, borehole camera image data, and rock mechanics parameter data; S2, collect backfilling parameters and mining progress data. The backfilling parameters include: the energy change relationship of the backfilling material in the pipeline and the hydration reaction rate of the backfilling material. S3. Using the geological data integrated and processed in step S1, a physical model of mining and backfilling in the mining area is established using the finite element analysis software ANSYS. S4. The real-time data collected in step S2, including filling parameters and mining progress, is transmitted in real time to the physical model constructed in step S3 through the data interface. Using data interpolation and fitting algorithms, the discrete monitoring data is mapped to continuous physical model nodes to obtain a digital twin model after data mapping and dynamic updating. S5, based on the digital twin model, uses numerical simulation to simulate the strength development of the filling body under different time and stress conditions; S6 analyzes the stress distribution and deformation inside the filling body in the digital twin model, and uses the finite element post-processing function on the digital twin model to obtain stress and strain data of different parts of the filling body; combined with the flow rate and pressure data during the filling process, the compactness of the filling body is evaluated. S7. Considering the interaction between the surrounding rock and the backfill, the stability of the system composed of the backfill and the surrounding rock is analyzed using a digital twin model at different mining stages. S8 uses a digital twin model to calculate the surface settlement using a finite element model, and combines this with the building's allowable settlement threshold to determine whether there is a risk of settlement. S9. Establish a multi-indicator comprehensive early warning model. Based on the simulation results of the digital twin model, combined with the genetic algorithm, generate suggestions for optimizing mining and backfilling parameters. Step S9 specifically includes the following steps: S9.1, the calculation model for early warning indicators is as follows: ； in, This is a comprehensive early warning value; For the first The weight of each indicator; For the first The dimensionless value of each indicator, ,in, The actual value of the indicator. and These are the minimum and maximum values ​​of the indicator; S9.2 employs a genetic algorithm for optimization, outputting the optimized parameter combination and the corresponding objective function value. The optimization objective functions include minimizing infill cost, maximizing infill efficiency, and minimizing surface subsidence. Represented as: ； in, For filling costs, , For the first The unit price of the material For the first The amount of each material used; A comprehensive score for filling efficiency; This refers to surface subsidence. , , The weights of each objective; Constraints include: Strength constraints: ,in, Minimum strength required for the filling material; stability constraints: ,in, Minimum safety factor required by the system; construction constraints: ,in, and These are the minimum and maximum values ​​of the filling flow rate, respectively.

2. The method for risk assessment of environmental protection mining and backfilling efficiency in densely built-up areas according to claim 1, characterized in that, Step S1 specifically includes the following steps: S1.1, using ground-penetrating radar, the transmission frequency range is set to... Between MHz, the range of underground penetration depths is To explore the stratigraphic structure and geological features; to use borehole camera equipment to take 360-degree pictures of the inside of the borehole and obtain detailed information on rock texture and joints and fractures; S1.2, rock mechanical parameters were obtained through triaxial compression tests; during the test, a servo-controlled loading system was used, with a loading rate range of [missing information]. Axial pressure was applied to the rock specimen at a speed of mm / min. and lateral pressure Calculate the elastic modulus of rock based on Hooke's Law : ； in, Let A be the stress on the rock, and A be the cross-sectional area of ​​the rock specimen. For rock strain, This is the axial deformation. The initial axial length of the rock specimen; S1.3, a weighted fusion algorithm is used for feature fusion. For the first... Feature vectors of data The fused feature vector for: ； in, For the first The weights of each data point are calculated using the entropy weighting method: ； in, For the first The data point number The normalized values ​​of each feature.

3. The method for risk assessment of environmental protection mining and backfilling efficiency in densely built-up areas according to claim 1, characterized in that, Step S2 specifically includes the following steps: S2.1, Install pressure sensors and electromagnetic flow sensors on the filling pipeline, and analyze the energy change relationship of the filling material in the pipeline according to Bernoulli's equation: ； in, Density of the filling material; For flow rate, , To fill the cross-sectional area of ​​the pipe, For traffic; For height, , The pressure difference between two points It is the acceleration due to gravity. , , , , and The subscripts 1 and 2 represent any two cross sections inside the filling pipe; S2.2, Multiple temperature and humidity sensors are installed at the filling site, and the hydration reaction rate of the filling material is quantified by combining the Arrhenius equation with a humidity correction coefficient: ； in, Pre-exponential factors; Activation energy; It is the gas constant; For temperature; This is a humidity correction function; S2.3 uses a laser rangefinder to measure distance based on the principle of triangulation. : ； in, Baseline length; The focal length of the camera; The displacement of feature points in the image is calculated using an image matching algorithm. Image matching employs the ORB-SLAM algorithm, which extracts ORB feature points, calculates descriptors, and uses Hamming distance for feature matching. The matching error is... satisfy: ； in, and For the first The pixel coordinates of a pair of matching points For threshold; S2.

4. Image recognition technology is used, employing a Convolutional Neural Network (CNN) model to identify mining areas. The CNN model is trained using labeled mining area image data. The CNN model uses a ResNet architecture, and the loss function is cross-entropy loss. : ； in, For the sample size, For the number of categories, For real labels, To predict probabilities.

4. The method for risk assessment of environmental protection mining and backfilling efficiency in densely built-up areas according to claim 1, characterized in that, Step S3 specifically includes the following steps: S3.1, In the modeling process, the continuous rock mass and infill are discretized into a finite number of elements. For a two-dimensional plane stress problem, the equilibrium equations of the elements are expressed as: ； in, For a plane stress problem, the elements of the element stiffness matrix are... for: ； in, To perform integration on the volume of a single cell; The volume of a single finite element; and These are the units. The node, the first Each node corresponds to a sub-block of the strain matrix. for transpose, It is the elasticity matrix; The element node displacement vector. , , They are unit nodes exist , Displacement in direction; For the element nodal force vector, , , These are the effects applied to the unit nodes. Above , Directional force; S3.2, Considering the mechanical properties of the rock, the Mohr-Coulomb strength criterion is used to describe the failure behavior of the rock. The yield condition of the Mohr-Coulomb strength criterion is: ； in, , Principal stress, For cohesion, The internal friction angle is used; the Burgers model is used to describe the rheological properties of the filling material, and the stress-strain relationship is... for: ； in, , It is the elastic modulus; , The viscosity coefficient; for Stress at any moment for Adaptability at all times This is the first derivative of strain with respect to time.

5. The method for risk assessment of environmental protection mining and backfilling efficiency in densely built-up areas according to claim 1, characterized in that, Step S4 specifically includes the following steps: S4.1, using Lagrange interpolation, for a given... Data In the interval Interpolation function within for: ， ； in, Interpolation interval The coordinates of any point to be solved within the interior; S4.2, For dynamic data updates, the Kalman filter algorithm is used for data fusion. The state equation and observation equation are as follows: ； ； in, For state vectors, For the observation vector, refer to The state transition matrix at time t, refer to The observation matrix at time, and These are process noise and observation noise, respectively.

6. The method for risk assessment of environmental protection mining and backfilling efficiency in densely built-up areas according to claim 1, characterized in that, Step S5 specifically includes the following steps: S5.1, During the simulation, the hydration reaction process of the filling material, the influence of temperature and humidity, and the degree of hydration reaction should be considered. Over time Changing models for: ； in, For the final degree of hydration, and These are model parameters; S5.2, Combining the filling material samples collected on-site, use ultrasonic testing equipment to measure the propagation speed of ultrasonic waves in the filling body. According to strength The strength of the empirical relationship is estimated using the empirical formula: ； in, , The coefficients related to the filling material were determined through fitting a large amount of experimental data. The least squares method was used for fitting, and the objective function was: ； in, To find the minimum value function, For the first The measured compressive strength values ​​in the experimental data set For the first The measured values ​​of ultrasonic propagation speed in the experimental data.

7. The method for risk assessment of environmental protection mining and backfilling efficiency in densely built-up areas according to claim 1, characterized in that, Step S6 specifically includes the following steps: S6.1, based on porosity The relationship between volumetric strain and volumetric strain was obtained through finite element simulation to obtain the volumetric strain of the filling material. Therefore, the porosity can be estimated. To assess density: ， ； in, The volumetric strain of the filling material. This represents the volume change of the filling material. This represents the initial volume of the filling material. S6.2, If stress concentration areas exist within the filling material, stress cloud diagram analysis should be used to determine if these areas have insufficient density; stress concentration factor. Defined as: ； in, For local maximum stress, This is the nominal stress.

8. The method for risk assessment of environmental protection mining and backfilling efficiency in densely built-up areas according to claim 1, characterized in that, Step S7 is as follows: The system stability is evaluated using the limit equilibrium method. For slope stability problems, the safety factor is... The calculation formula is: ； in, It is cohesive force; The length of the sliding surface; For normal force, , For the first Normal stress on the sliding surface, For the corresponding area; It is the internal friction angle; For downward force, , For the first The total weight of the sliding body, For the first The inclination angle of the smooth surface, is the sine of the inclination angle.

9. The method for risk assessment of environmental protection mining and backfilling efficiency in densely built-up areas according to claim 1, characterized in that, Step S8 is as follows: Calculation of surface subsidence using finite element model Combined with the building's allowable settlement threshold [ ],when At that time, it was determined that there was a risk of settlement; the settlement prediction model is expressed as: ； in, This represents the total number of strata in the overlying rock above the mining area. This is the settlement influence coefficient; Additional stress caused by mining; This corresponds to the thickness of the strata.

Images