A simulation method for geological mineral resource reserves

Abstract

The present invention relates to the technical field of mineral resource reserve simulation, and in particular to a method for simulating geological mineral resource reserves. The method comprises the following steps: collecting multi-source heterogeneous data in a target geological area, and extracting comprehensive features to obtain a comprehensive data set; constructing an initial static geological model based on the comprehensive data set to obtain an initial static geological model; constructing a multi-physics field coupling model based on the initial static geological model to obtain a multi-physics field coupling model grid; simulating the migration and enrichment of ore-forming fluids based on the multi-physics field coupling model grid to obtain a simulation result of ore-forming fluid migration; updating the geometric morphology of the geological body of the initial static geological model based on the simulation result of ore-forming fluid migration to obtain a dynamic geological model; training a reserve prediction model based on the dynamic geological model to obtain a reserve prediction model. The present invention uses mineral resource reserve simulation technology to more accurately and dynamically evaluate mineral resource reserves.

Description

Technical Field

[0001] The present invention relates to the technical field of mineral resource reserve simulation, and in particular to a method for simulating geological mineral resource reserves. Background Art

[0002] Accurate assessment of mineral resource reserves is key to mine planning, development and operation. Its development process, from early experience-based estimates to modern computer-based complex models, reflects the improvement of geological understanding and technical capabilities. In the early days, it relied on the experience of geologists and manual calculations. The method was simple but low in precision and highly subjective. Subsequently, computer technology was introduced and geostatistical methods (such as Kriging) were used for reserve estimation, which improved the ability to handle spatial variability and became the mainstream method. Later, three-dimensional modeling technology and numerical simulation methods emerged, which made it possible to build more sophisticated models to study mineralization processes and predict mining impacts. Despite the continuous development of technology, existing methods still have the limitations of static models and do not consider the impact of mining disturbances:

[0003] Limitations of static models: Traditional geological modeling mainly constructs static models, which are like "snapshots" of mineral deposits and cannot reflect the dynamic formation, evolution and enrichment process of mineral deposits. This leads to insufficient understanding of the causes of mineral deposits and future change trends, and limited predictive capabilities.

[0004] Failure to consider the impact of mining disturbances: Traditional reserve assessments are usually completed before mining and are static assessments that ignore the impact of mining activities on the geological environment and remaining reserves, such as stress changes, rock deformation, groundwater flow, etc. This leads to inaccurate reserve assessments, difficulty in optimizing mining plans, and underestimated safety risks. Summary of the invention

[0005] Based on this, it is necessary to provide a simulation method for geological mineral resource reserves to solve at least one of the above technical problems.

[0006] To achieve the above purpose, a method for simulating geological mineral resource reserves comprises the following steps:

[0007] Step S1: collect multi-source heterogeneous data in the target geological area and perform comprehensive feature extraction to obtain a comprehensive data set;

[0008] Step S2: constructing an initial static geological model according to the comprehensive data set to obtain an initial static geological model; constructing a multi-physics field coupling model according to the initial static geological model to obtain a multi-physics field coupling model grid; simulating the migration and enrichment of ore-forming fluids according to the multi-physics field coupling model grid to obtain a simulation result of ore-forming fluid migration; updating the geometric morphology of the geological body of the initial static geological model according to the simulation result of ore-forming fluid migration to obtain a dynamic geological model;

[0009] Step S3: extracting geological features from the dynamic geological model, and training a reserve prediction model using machine learning to obtain a reserve prediction model; using the reserve prediction model to perform probabilistic resource prediction to obtain a probabilistic resource map;

[0010] Step S4: Obtain a mineral mining plan; apply mining plan boundary conditions to the dynamic geological model according to the mineral mining plan to obtain a mining boundary condition model; perform a mining disturbance multi-physical field coupling simulation based on the mining boundary condition model and the dynamic geological model to obtain a mining disturbance response field; calculate the reserves affected by the disturbance based on the mining disturbance response field, the dynamic geological model and the probabilistic resource map to obtain the reserves affected by the disturbance; perform a dynamic geological model update on the deformation update geological model to obtain an updated dynamic geological model; perform a prediction result error analysis based on the reserves affected by the disturbance and the updated dynamic geological model to obtain a model prediction error report; adjust the model parameters of the updated dynamic geological model based on the model prediction error report to obtain an adjusted dynamic geological model to achieve the simulation task of geological mineral resource reserves.

[0011] The present invention integrates multi-source heterogeneous data, including geospatial data, geophysical data, geochemical data, and historical data, and performs feature extraction and fusion to construct a comprehensive, multi-dimensional integrated data set. This effectively improves the information content and reliability of the data, and provides a solid data foundation for subsequent geological modeling and resource prediction. By constructing an initial static geological model, and combining geological historical constraints and multi-physical field coupling simulation, a dynamic simulation of geological evolution and mineralization process is realized. This can not only more accurately characterize the geological structure and ore body distribution of the mining area, but also reveal the migration law and enrichment mechanism of the ore-forming fluid, and provide a more reliable geological model for subsequent reserve prediction. Using machine learning technology, combined with geological features extracted by the dynamic geological model and historical exploration and mining data, the reserve prediction model is trained. This allows reserve prediction to not only utilize the information of the geological model, but also fully explore the value of historical data and improve prediction accuracy. At the same time, probabilistic resource prediction and uncertainty quantification provide more comprehensive information for resource assessment and risk management. Integrating the mining plan into the geological model realizes dynamic simulation and reserve prediction of mining disturbances. By simulating the changes in multiple physical fields during the mining process, and combining actual production data and monitoring data for model verification and correction, it is possible to more accurately predict remaining reserves and assess mining risks, providing more scientific decision support for mine production planning and sustainable development. Dynamically updating geological models and resource maps enables real-time tracking and management of mineral resources, making reserve assessments closer to actual mining conditions. Therefore, the present invention provides a simulation method for geological mineral resource reserves, which effectively solves the shortcomings of existing geological mineral resource reserve simulation methods in terms of static models and failure to consider the impact of mining disturbances. By simulating the temporal evolution of geological evolution and mining disturbances, incorporating the interactions of multiple fields such as geology, physics, and chemistry into the model, and closely combining mining engineering with geological models, it is possible to more accurately and dynamically assess mineral resource reserves, providing more reliable decision support for the sustainable development of mines.

[0012] Preferably, step S1 comprises the following steps:

[0013] Step S11: collecting geospatial data of the target geological area to obtain original geospatial data;

[0014] Step S12: unifying the spatial reference of the original geographic spatial data and performing geometric correction to obtain geometrically corrected data;

[0015] Step S13: standardizing the attribute data of the geometrically corrected data and performing semantic coordination to obtain standardized attribute data;

[0016] Step S14: performing spatial interpolation and resampling on the standardized attribute data to obtain spatial consistency data;

[0017] Step S15: fusing multi-source data on the spatial consistency data and extracting data features to obtain a feature fusion data set;

[0018] Step S16: construct a comprehensive data set for the feature fusion data set to obtain a comprehensive data set.

[0019] The present invention adopts a variety of data collection methods, such as unmanned aerial vehicle remote sensing, RTK-GPS and geophysical exploration, and collects and organizes existing geological data, so as to obtain comprehensive, multi-scale original geospatial data of the target area, and provide a rich data basis for subsequent geological modeling and analysis, thereby improving the reliability and accuracy of the model. All original geospatial data are unified under the same coordinate system and projection system, and the image data is geometrically corrected, eliminating the spatial deviation and geometric distortion that may exist between different data sources, ensuring the spatial consistency of the data, and laying the foundation for subsequent multi-source data fusion and analysis. By standardizing the attribute data of different data sources, such as unifying units, data types and encoding methods, and performing semantic coordination, the ambiguity and inconsistency of data attributes are solved, and the comparability and fusibility of data are improved. The discrete point data is spatially interpolated, and the raster data of different resolutions are resampled, and different types of data are converted into raster data with the same spatial resolution and range, which ensures the continuity and consistency of the data in space and creates conditions for multi-source data fusion. Through multi-source data fusion technology, such as principal component analysis (PCA), a variety of different types of data are integrated together to extract the most useful features for mineral resource prediction, such as terrain features, geophysical anomalies, geochemical anomalies, etc., effectively improving the signal-to-noise ratio and information content of the data, and providing high-quality input data for subsequent machine learning model training. All processed data, including raster data and vector data, are integrated into a unified geographic database, and topological relationships and metadata information between data are established, constructing a structured, complete, easy-to-access and use comprehensive data set, providing a convenient data platform for subsequent geological modeling and resource prediction.

[0020] Preferably, step S2 comprises the following steps:

[0021] Step S21: constructing an initial static geological model according to the comprehensive data set to obtain an initial static geological model;

[0022] Step S22: applying geological history constraints to the initial static geological model to obtain a geological evolution stage model;

[0023] Step S23: constructing a multi-physics field coupling model according to the geological evolution stage model to obtain a multi-physics field coupling model grid;

[0024] Step S24: performing ore-forming fluid migration and enrichment simulation according to the multi-physics field coupling model grid to obtain ore-forming fluid migration simulation results;

[0025] Step S25: updating the model attributes of the initial static geological model and updating the geometric shape of the geological body according to the ore-forming fluid migration simulation results to obtain a dynamic geological model.

[0026] The present invention constructs a three-dimensional initial static geological model containing strata, faults and grade information by utilizing multi-source data in a comprehensive data set, such as geological outcrops, drilling data, DEM data, etc. The model can intuitively display the initial geological structure and ore body distribution characteristics of the mining area, laying the foundation for subsequent dynamic simulation and reserve prediction. By imposing geological history constraints, such as tectonic stress fields and fault slip, the geological evolution process of the mining area, such as fold deformation, fault activity and magma intrusion, is simulated, making the geological model more consistent with the actual geological evolution history, and providing important geological background information for understanding the mineralization process. Based on the geological evolution stage model, a multi-physics field coupling model grid containing rock physical properties and boundary conditions is constructed, which provides a discretized calculation domain and necessary parameter input for subsequent numerical simulations, ensuring the accuracy and reliability of the simulation results. The grid encryption strategy improves the simulation accuracy of key areas. Through the coupled simulation of fluid flow, heat transfer and chemical reaction, the simulation results of ore-forming fluid migration and enrichment were obtained, including fluid pressure field, temperature field, chemical component concentration field and mineral precipitation distribution, revealing the migration path and enrichment law of ore-forming fluid in geological history, and providing key information for predicting the distribution of mineral resources. According to the simulation results of ore-forming fluid migration, the properties (such as temperature, pressure, chemical component concentration, mineral composition) and geometric morphology of the initial static geological model were updated, and a dynamic geological model containing geological evolution and mineralization process information was constructed. This model more accurately reflects the geological reality of the mining area and provides a more reliable model basis for subsequent reserve prediction and mining disturbance simulation.

[0027] Preferably, step S24 comprises the following steps:

[0028] Step S241: setting the initial fluid and rock chemical compositions according to the multi-physics field coupling model grid to obtain an initial chemical composition model;

[0029] Step S242: constructing a hydrothermal reaction network of the initial chemical composition model to obtain a hydrothermal reaction network model;

[0030] Step S243: constructing chemical reaction control equations according to the hydrothermal reaction network model, and configuring the chemical reaction control equations on the multi-physics field coupling model grid to obtain a group of chemical reaction control equations;

[0031] Step S244: numerically solving the chemical component migration and reaction of the chemical reaction control equations according to the multi-physics field coupling model grid and the initial chemical composition model to obtain chemical component concentration distribution data;

[0032] Step S245: performing mineral dissolution, mineral precipitation and mineral replacement simulation according to the chemical component concentration distribution data and the hydrothermal reaction network model to obtain mineral phase distribution change data;

[0033] Step S246: tracking the enrichment of ore-forming elements according to the chemical component concentration distribution data and the mineral phase distribution change data to obtain the enrichment distribution data of ore-forming elements;

[0034] Step S247: Visualize and analyze the chemical component concentration distribution data, mineral phase distribution change data, and ore-forming element enrichment distribution data to obtain ore-forming fluid migration simulation results.

[0035] The present invention establishes an initial chemical environment for simulating the migration of ore-forming fluids by setting the chemical composition of the initial fluid and rock, provides basic data for subsequent chemical reaction simulations, and makes the simulation results closer to the actual geological conditions. A hydrothermal reaction network model is constructed to define the chemical reactions that may occur between fluids and rocks, including parameters such as reaction type, equilibrium constant, and reaction rate, and a simulation framework for chemical reactions is established, which provides a key chemical reaction mechanism for simulating the migration and enrichment of ore-forming elements. The hydrothermal reaction network model is converted into a chemical reaction control equation, and configured into a multi-physical field coupling model grid, which realizes the coupling of chemical reactions with fluid flow and heat transfer, can more accurately simulate the chemical reaction dynamics in the ore-forming process, and takes into account the interaction between different physical fields. By numerically solving the chemical reaction control equation group, the chemical component concentration distribution data is obtained, which reveals the migration and distribution laws of ore-forming elements in the simulation area, and provides important input parameters for subsequent mineral dissolution, precipitation, and replacement simulations. Based on the chemical component concentration distribution data and the hydrothermal reaction network model, the mineral dissolution, precipitation and replacement processes were simulated, and the mineral phase distribution change data were obtained, which revealed the formation and evolution laws of minerals in the mineralization process and provided an important basis for predicting the location and morphology of ore bodies. The consideration of competitive precipitation simulation and solid phase diffusion makes the simulation results closer to the actual situation. By tracking the migration and enrichment process of ore-forming elements, the enrichment distribution data of ore-forming elements were obtained, and the enrichment area and degree of ore-forming elements were clarified, providing key information for evaluating the potential of mineral resources. Visualizing and analyzing the simulation results can intuitively display the migration of ore-forming fluids, chemical reactions and mineral evolution processes, which is helpful to deeply understand the ore-forming mechanism and predict the distribution of mineral resources, and provide guidance for subsequent exploration and development.

[0036] Preferably, step S245 is specifically as follows:

[0037] According to the chemical component concentration distribution data, the instantaneous reaction rates of dissolution reaction, precipitation reaction and replacement reaction are calculated for the hydrothermal reaction network model to obtain the instantaneous reaction rate data;

[0038] Determine the potential precipitation minerals and supersaturation of the hydrothermal reaction network model, and obtain the potential precipitation mineral list and supersaturation data;

[0039] According to the instantaneous reaction rate data, the potential precipitation mineral list and supersaturation data, the competitive precipitation process is simulated to obtain the actual precipitation mineral increment data;

[0040] The mineral dissolution process is simulated based on the instantaneous reaction rate data to obtain the actual dissolved mineral reduction data;

[0041] According to the instantaneous reaction rate data and the actual dissolved mineral reduction data, the mineral replacement process and solid phase diffusion influence are simulated to obtain the mineral change data of the replacement reaction;

[0042] Extract the mineral composition of the previous time step from the initial chemical composition model to obtain the mineral composition of the previous time step; update the mineral composition of the grid unit according to the mineral composition of the previous time step, the actual precipitation mineral increment data, the actual dissolved mineral reduction data and the metasomatic reaction mineral change data to obtain the mineral composition data of the current time step;

[0043] The mineral phase distribution change is recorded for the mineral composition of the previous time step and the mineral composition data of the current time step to obtain the mineral phase distribution change data.

[0044] The present invention can accurately capture the dynamic changes of chemical reactions by calculating the instantaneous reaction rate of each reaction in each grid unit, and provide precise rate control for the subsequent simulation of mineral dissolution, precipitation and replacement processes, thereby improving the accuracy of the simulation. By determining the potential precipitation minerals and supersaturation, it is possible to predict which minerals may precipitate under specific conditions, and quantify the driving force of their precipitation, providing necessary information for the subsequent competitive precipitation simulation. The competitive relationship of multiple minerals precipitating at the same time is taken into account, the actual precipitation process is simulated, the changes in mineral combinations are more accurately reflected, the limitations of single mineral precipitation simulation are avoided, and the authenticity of the simulation is improved. By simulating the mineral dissolution process, the actual dissolution amount of each mineral can be calculated, providing data support for updating the mineral composition, and providing a material source for the replacement reaction. The mineral replacement process and the influence of solid phase diffusion are simulated, the complex interactions between minerals are more comprehensively considered, the simulation accuracy of the mineral evolution process is improved, and the characteristics of replacement alteration can be better reflected. The mineral composition information of the previous time step is extracted, which provides basic data for the update of the mineral composition of the current time step, ensuring the continuity of mineral evolution. Taking into account the mineral precipitation, dissolution and replacement processes, the mineral composition of each grid unit is updated, the dynamic changes of mineral content are tracked, and the evolution of mineral assemblages is more accurately reflected. Recording the changes in mineral phase distribution can track the evolution of mineral assemblages over time, providing important data support for analyzing the mineralization process and predicting the distribution of ore bodies.

[0045] Preferably, step S3 comprises the following steps:

[0046] Step S31: acquiring historical exploration data and historical mining data; extracting geological features from the dynamic geological model, historical exploration data and historical mining data, and preparing machine learning training data to obtain a machine learning training data set;

[0047] Step S32: training a reserve prediction model according to the machine learning training data set to obtain a reserve prediction model;

[0048] Step S33: inputting the physical field parameters output by the dynamic geological model into the reserve prediction model, performing probabilistic resource prediction, and performing uncertainty quantification to obtain probabilistic resource prediction results;

[0049] Step S34: construct a probability resource map according to the probability resource prediction results to obtain a probability resource map.

[0050] The present invention integrates the physical field parameters of the dynamic geological model, historical exploration data (drilling, geochemistry, geophysical exploration) and historical mining data, and performs feature extraction to construct a machine learning training data set containing multi-dimensional information. The data set not only contains traditional exploration and mining information, but also incorporates the simulation results of the dynamic geological model, providing a rich data basis for training a more accurate reserve prediction model. The reserve prediction model (such as a random forest regression model) is trained using the constructed machine learning training data set, so that the model can learn the complex nonlinear relationship between geological characteristics and ore grade, thereby improving the accuracy and reliability of reserve prediction. The training and testing process of the model ensures the generalization ability and prediction performance of the model. The physical field parameters output by the dynamic geological model are input into the trained reserve prediction model to realize the probabilistic resource prediction based on multi-physical field information. The uncertainty of the prediction results is quantified by the Bootstrap method, and not only the predicted value of the ore grade is obtained, but also the confidence interval of the prediction result is obtained, providing more comprehensive information for risk assessment and decision-making. Visualizing the probabilistic resource prediction results into a probabilistic resource map can intuitively display the spatial distribution, grade changes and prediction uncertainties of mineral resources, providing geologists and mining engineers with more intuitive and easy-to-understand mineral resource information, which helps to better carry out resource assessment and mining planning.

[0051] Preferably, step S4 comprises the following steps:

[0052] Step S41: obtaining a mineral mining plan; applying mining plan boundary conditions to the dynamic geological model according to the mineral mining plan to obtain a mining boundary condition model;

[0053] Step S42: performing a mining disturbance multi-physics field coupling simulation according to the mining boundary condition model and the dynamic geological model to obtain a mining disturbance response field;

[0054] Step S43: performing geological model deformation update on the dynamic geological model according to the mining disturbance response field to obtain a deformation updated geological model;

[0055] Step S44: performing a probability resource map disturbance adjustment on the probability resource map according to the mining disturbance response field to obtain a disturbance adjusted resource map;

[0056] Step S45: Calculate the reserves affected by the disturbance according to the deformation updated geological model, the disturbance adjusted resource map, and the mineral mining plan to obtain the reserves affected by the disturbance;

[0057] Step S46: dynamically updating the deformation updated geological model according to the disturbance adjusted resource map to obtain an updated dynamic geological model;

[0058] Step S47: collecting actual production data and monitoring data of the target geological area, and integrating the measured data to obtain a measured data set;

[0059] Step S48: performing prediction result comparison and error analysis on the reserves affected by the disturbance, the updated dynamic geological model and the measured data set to obtain a model prediction error report;

[0060] Step S49: adjusting the model parameters of the updated dynamic geological model according to the model prediction error report to obtain an adjusted dynamic geological model.

[0061] The present invention constructs a mining boundary condition model by converting a mineral mining plan into model boundary conditions and applying these boundary conditions to a dynamic geological model, so as to prepare for the subsequent simulation of the impact of mining disturbance on the geological environment and make the simulation closer to the actual mining situation. Through multi-physical field coupling simulation, the comprehensive impact of mining activities on the geological environment, such as stress changes, groundwater flow and temperature changes, is taken into account, and a mining disturbance response field containing information on changes in various physical fields is obtained, which provides comprehensive data support for subsequent analysis and prediction. According to the displacement field data in the mining disturbance response field, the dynamic geological model is deformed and updated, so that the model can reflect the deformation of the geological body caused by mining activities, such as surface settlement, rock mass displacement, etc., and improves the accuracy and reliability of the model. According to the mining disturbance response field, the probability resource map is adjusted, and the impact of mining disturbance on the grade and mineability of the ore body, such as ore crushing or dilution, is taken into account, so that the resource assessment is more accurate and can reflect the changes brought about by mining activities. Taking mining plan, geological model deformation and resource map adjustment into consideration, the reserves affected by disturbance are calculated, and the dynamic evaluation of remaining mineral resources is realized, providing a more reliable basis for mine production planning and decision-making. The resource information after disturbance adjustment is updated to the deformation updated geological model, and a dynamic geological model containing the latest mining information and geological changes is obtained, which provides an updated model basis for the next round of simulation and prediction, and realizes the dynamic update and iterative optimization of the model. By collecting actual production data and monitoring data and integrating them, a measured data set is constructed, which provides important actual reference data for model verification and correction, ensuring the practicality and reliability of the model. Comparing the simulation results with the measured data and error analysis, the prediction accuracy of the model can be evaluated, and the deficiencies of the model can be found, providing direction for the adjustment and improvement of model parameters, so as to continuously improve the prediction ability of the model. According to the model prediction error report, the parameters of the dynamic geological model are adjusted, such as correcting the geological parameters or improving the simulation parameters, so as to realize the correction and optimization of the model, improve the prediction accuracy and reliability of the model, and enable the model to better adapt to the actual situation.

[0062] Preferably, step S42 includes the following steps:

[0063] Step S421: configuring the ground stress change caused by mining according to the mining boundary condition model and the dynamic geological model to obtain a ground stress change model;

[0064] Step S422: extracting geological model parameters from the dynamic geological model to obtain rock mechanics parameters, hydrogeological parameters, thermal parameters, fault mechanics parameters and fault geometry information;

[0065] Step S423: solving the deformation field of the rock and soil mass caused by mining according to the geostress variation model and rock mechanics parameters to obtain the deformation field data of the rock and soil mass;

[0066] Step S424: using the rock and soil deformation field data as input, updating the hydrogeological parameters to obtain updated hydrogeological parameters; simulating the impact of mining on the groundwater flow field based on the updated hydrogeological parameters to obtain groundwater flow field data;

[0067] Step S425: performing coupled simulation of temperature field changes on the mining boundary condition model according to thermal parameters, rock and soil deformation field data, and groundwater flow field data to obtain temperature field data;

[0068] Step S426: performing fault activation and surface settlement analysis according to the rock and soil deformation field data and fault mechanics parameters to obtain fault activation and surface settlement analysis results;

[0069] Step S427: integrating the rock and soil deformation field data, groundwater flow field data, temperature field data, and fault activation and surface settlement analysis results into a mining disturbance response field to obtain a mining disturbance response field.

[0070] The present invention accurately simulates the initial geostress changes caused by mining by setting free surfaces and releasing stress in the mining area, provides accurate boundary conditions for subsequent rock deformation simulation, and provides basic data for analyzing geological disasters such as rock bursts and roof falls. Various necessary parameters are extracted from the dynamic geological model, including rock mechanics parameters, hydrogeological parameters, thermal parameters, fault mechanics parameters, and fault geometry information, providing complete parameter input for subsequent multi-physics field coupling simulation, ensuring the accuracy and reliability of the simulation. Based on the geostress change model and rock mechanics parameters, the rock and soil deformation field caused by mining is solved, and the rock displacement information is obtained, which provides a data basis for subsequent analysis of surface settlement, tunnel deformation, etc., and can be used to evaluate the stability of the rock mass caused by mining. The influence of rock deformation on hydrogeological parameters, such as changes in permeability, is considered, and the updated parameters are used to simulate the influence of mining on the groundwater flow field, and a more accurate groundwater flow prediction result is obtained, which can be used to evaluate the risk of mine water inrush and optimize drainage schemes. By coupling the simulation of temperature field changes, the effects of mining, groundwater flow and rock deformation on the temperature field are taken into account, and more comprehensive simulation results are obtained, which can be used to analyze the effects of temperature changes on rock mechanical properties and mineral reactions during mining, such as thermal cracking or changes in mineral solubility. By analyzing rock deformation and fault mechanical parameters, the possibility of fault activation and the scope and amplitude of surface subsidence are evaluated, providing important reference information for mine safety production and environmental protection. The simulation results of various physical fields and the analysis results of fault activation and surface subsidence are integrated into the mining disturbance response field, forming a complete data set that fully reflects the impact of mining activities on the geological environment and provides a complete data basis for subsequent reserve calculations and model updates.

[0071] Preferably, step S421 is specifically:

[0072] According to the mining boundary condition model, the excavation unit of the dynamic geological model is identified, and the free surface is set to obtain the excavation unit identification result and the initial free surface model;

[0073] A step-by-step excavation simulation is performed according to the excavation unit identification results, and stress release is performed according to the initial free surface model to obtain a step-by-step excavation stress release model;

[0074] The constitutive model of the filling body is constructed according to the mining boundary condition model and the dynamic geological model to obtain the constitutive model parameters of the filling body; the contact between the filling body and the surrounding rock is simulated according to the constitutive model parameters of the filling body to obtain the contact model between the filling body and the surrounding rock;

[0075] Dynamically adjust the boundary conditions of complex mining processes according to the mining boundary condition model and the dynamic geological model to obtain a dynamic mining boundary condition sequence;

[0076] According to the mining boundary condition model and the dynamic geological model, the equivalent simulation of blasting disturbance is carried out to obtain the blasting equivalent load model;

[0077] The initial free surface model, gradual excavation stress release model, filling body and surrounding rock contact model, dynamic mining boundary condition sequence and blasting equivalent load model are integrated to generate the ground stress change model caused by mining.

[0078] The present invention accurately simulates the formation of underground cavities during the mining process by identifying excavation units and setting free surfaces, provides an accurate geometric model for subsequent stress release and deformation calculations, avoids the simplified treatment of the mining area as a homogeneous material, and improves the simulation accuracy. The gradual excavation simulation and stress release more realistically reflect the timing of the actual mining process, avoid the numerical error caused by instantaneous excavation, and can more accurately capture the dynamic evolution of the stress field during the mining process, such as analyzing stress concentration and redistribution phenomena. By constructing a constitutive model of the filling body and simulating the contact between the filling body and the surrounding rock, the impact of filling mining on the surrounding rock mass can be more accurately simulated, such as the supporting effect of the filling body on the surrounding rock and stress transfer, thereby improving the overall accuracy and reliability of the model. By dynamically adjusting the boundary conditions, complex mining processes such as layered mining and staged backfilling can be simulated, so that the model can adapt to different mining scenarios and accurately reflect the dynamic changes in the mining process, thereby improving the applicability and flexibility of the model. By equivalently simulating blasting disturbances, the complex effects of blasting are simplified into equivalent loads, avoiding complex blasting dynamics calculations, improving simulation efficiency, and capturing the main effects of blasting on the surrounding rock mass. The simulation results of various aspects are integrated into the ground stress change model, comprehensively considering the effects of multiple factors such as excavation, filling, and blasting on ground stress, and a more comprehensive and accurate ground stress change model is obtained, providing a reliable basis for subsequent deformation and stability analysis.

[0079] Preferably, step S426 is specifically:

[0080] Constructing a fault geometry model based on fault geometry information and performing model mesh division to obtain a fault geometry model and a fault unit mesh;

[0081] Apply fault contact constraints according to the fault geometry model and fault mechanics parameters, and determine the fault constitutive model parameters to obtain the fault contact constraint model and fault constitutive model parameters;

[0082] According to the deformation field data of the rock and soil mass and the fault contact constraint model, the stress state of the fault plane is calculated for the fault unit grid to obtain the stress distribution data of the fault plane;

[0083] According to the fault contact constraint model, fault plane stress distribution data and fault constitutive model parameters, the fault stability is evaluated and the fault slip amount is calculated to obtain the fault stability evaluation results and fault slip amount data;

[0084] Extract surface settlement information from the rock and soil deformation field data to obtain surface settlement displacement data;

[0085] Conduct surface settlement analysis on the surface settlement displacement data and visualize the surface settlement to obtain the surface settlement analysis diagram and surface settlement curve;

[0086] The fault stability evaluation results, fault slip data, surface settlement displacement data, surface settlement analysis diagram and surface settlement curve are integrated to obtain the fault activation and surface settlement analysis results.

[0087] The present invention can more accurately express the spatial morphology and distribution characteristics of the fault by constructing a fine fault geometry model and grid, providing a geometric basis for subsequent fault mechanics analysis and improving the calculation accuracy, especially in the area near the fault. By applying fault contact constraints and determining fault constitutive model parameters, such as friction angle and cohesion, the interaction between the rock masses on both sides of the fault and the mechanical behavior of the fault can be simulated, providing necessary mechanical constraints and parameter inputs for fault stability assessment. Based on the rock and soil deformation field data and the fault contact constraint model, the stress state of the fault plane is calculated, and the stress distribution data on the fault plane is obtained, providing key stress information for evaluating the stability of the fault. By analyzing the stress state of the fault plane and the fault mechanics parameters, such as using the Coulomb friction criterion, the stability of the fault can be evaluated, and the possible slippage of the fault can be predicted, providing early warning information for mine safety production, and helping to formulate corresponding prevention and control measures. Surface settlement information is extracted from the rock and soil deformation field data, and surface settlement displacement data is obtained, providing basic data for analyzing the scope and amplitude of surface settlement. By analyzing and visualizing the surface settlement displacement data, we can intuitively understand the distribution law and change trend of surface settlement, such as the maximum settlement amount and settlement range, which provides a basis for evaluating the impact of mining on the surface environment and helps to formulate corresponding control measures. By integrating the fault stability assessment results, fault slip, surface settlement data, analysis diagrams and curves, we can obtain a complete fault activation and surface settlement analysis result, which can more comprehensively evaluate the impact of mining activities on the geological environment and provide more comprehensive reference information for mine safety production and environmental protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] Figure 1 A schematic diagram of the steps of a simulation method for geological mineral resource reserves;

[0089] Figure 2It is a schematic diagram of the detailed implementation steps of step S2 in the present invention.

[0090] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0091] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0092] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0093] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0094] To achieve this, please refer to Figure 1 to Figure 2 , a simulation method for geological mineral resource reserves, comprising the following steps:

[0095] Step S1: collect multi-source heterogeneous data in the target geological area and perform comprehensive feature extraction to obtain a comprehensive data set;

[0096] Step S2: constructing an initial static geological model according to the comprehensive data set to obtain an initial static geological model; constructing a multi-physics field coupling model according to the initial static geological model to obtain a multi-physics field coupling model grid; simulating the migration and enrichment of ore-forming fluids according to the multi-physics field coupling model grid to obtain a simulation result of ore-forming fluid migration; updating the geometric morphology of the geological body of the initial static geological model according to the simulation result of ore-forming fluid migration to obtain a dynamic geological model;

[0097] Step S3: extracting geological features from the dynamic geological model, and training a reserve prediction model using machine learning to obtain a reserve prediction model; using the reserve prediction model to perform probabilistic resource prediction to obtain a probabilistic resource map;

[0098] Step S4: Obtain a mineral mining plan; apply mining plan boundary conditions to the dynamic geological model according to the mineral mining plan to obtain a mining boundary condition model; perform a mining disturbance multi-physical field coupling simulation based on the mining boundary condition model and the dynamic geological model to obtain a mining disturbance response field; calculate the reserves affected by the disturbance based on the mining disturbance response field, the dynamic geological model and the probabilistic resource map to obtain the reserves affected by the disturbance; perform a dynamic geological model update on the deformation update geological model to obtain an updated dynamic geological model; perform a prediction result error analysis based on the reserves affected by the disturbance and the updated dynamic geological model to obtain a model prediction error report; adjust the model parameters of the updated dynamic geological model based on the model prediction error report to obtain an adjusted dynamic geological model to achieve the simulation task of geological mineral resource reserves.

[0099] In the embodiment of the present invention, reference Figure 1 FIG. 1 is a schematic diagram of the steps of the method for simulating geological mineral resource reserves of the present invention. In this example, the method for simulating geological mineral resource reserves includes the following steps:

[0100] Step S1: collect multi-source heterogeneous data in the target geological area and perform comprehensive feature extraction to obtain a comprehensive data set;

[0101] In the embodiment of the present invention, multi-source heterogeneous data of the target area are collected by various means, including drone remote sensing, RTK-GPS positioning, geophysical exploration, and collection and collation of existing geological data. Subsequently, these data are subjected to spatial benchmark unification, geometric correction, attribute data standardization, semantic coordination, spatial interpolation and resampling to ensure the consistency of the data in space and attributes. Finally, multi-source data fusion and feature extraction are performed, such as using the PCA method to reduce dimensionality and extract terrain features, to construct a comprehensive data set for subsequent modeling, and save it in HDF5 format.

[0102] Step S2: constructing an initial static geological model according to the comprehensive data set to obtain an initial static geological model; constructing a multi-physics field coupling model according to the initial static geological model to obtain a multi-physics field coupling model grid; simulating the migration and enrichment of ore-forming fluids according to the multi-physics field coupling model grid to obtain a simulation result of ore-forming fluid migration; updating the geometric morphology of the geological body of the initial static geological model according to the simulation result of ore-forming fluid migration to obtain a dynamic geological model;

[0103] In an embodiment of the present invention, an initial static geological model is constructed based on a comprehensive data set, including a stratigraphic model, a fault model, and a grade model. Then, according to geological historical records, tectonic stress fields and fault slips are applied to the initial model to simulate the geological evolution process and obtain a geological evolution stage model. Next, a multi-physics field coupling model grid is constructed based on the geological evolution stage model, and rock physical properties and boundary conditions are assigned to each grid unit. The grid is used to simulate the migration and enrichment of ore-forming fluids, simulate fluid flow, heat transfer, and chemical reactions, and obtain fluid pressure fields, temperature fields, chemical component concentration fields, and mineral precipitation distributions. Finally, the properties and geometric morphology of the initial static geological model are updated according to the simulation results to obtain a dynamic geological model containing geological evolution information.

[0104] Step S3: extracting geological features from the dynamic geological model, and training a reserve prediction model using machine learning to obtain a reserve prediction model; using the reserve prediction model to perform probabilistic resource prediction to obtain a probabilistic resource map;

[0105] In an embodiment of the present invention, geological features are extracted from a dynamic geological model, historical exploration data, and historical mining data to construct a machine learning training data set. Then, the data set is used to train a reserve prediction model, such as a random forest regression model, for predicting ore grade. The trained model is used to predict the physical field parameters output by the dynamic geological model to obtain the predicted ore grade value of each grid cell. The Bootstrap method is used to quantify the uncertainty of the prediction, and the mean and standard deviation of the grade prediction value of each grid cell are calculated. Finally, the prediction results are visualized to generate a probabilistic resource map containing the predicted mean and uncertainty indicators of the ore grade.

[0106] Step S4: obtaining a mineral mining plan; applying mining plan boundary conditions to the dynamic geological model according to the mineral mining plan to obtain a mining boundary condition model; performing mining disturbance multi-physical field coupling simulation according to the mining boundary condition model and the dynamic geological model to obtain a mining disturbance response field; calculating the reserves affected by the disturbance according to the mining disturbance response field, the dynamic geological model and the probabilistic resource map to obtain the reserves affected by the disturbance; performing a dynamic geological model update on the deformation update geological model to obtain an updated dynamic geological model; performing a prediction result error analysis according to the reserves affected by the disturbance and the updated dynamic geological model to obtain a model prediction error report; adjusting the model parameters of the updated dynamic geological model according to the model prediction error report to obtain an adjusted dynamic geological model to achieve the simulation task of geological mineral resource reserves;

[0107] In an embodiment of the present invention, a mineral mining plan is obtained and converted into the boundary conditions required for numerical simulation, which are applied to the dynamic geological model to obtain a mining boundary condition model. Then, a mining disturbance multi-physical field coupling simulation is performed to simulate the stress changes, groundwater flow and temperature changes during the mining process to obtain a mining disturbance response field. The geometric form of the dynamic geological model is updated according to the response field to obtain a deformation updated geological model, and the probability resource map is adjusted to obtain a disturbance adjusted resource map. Based on the updated geological model and resource map, the reserves affected by the disturbance are calculated. Actual production data and monitoring data are collected, compared with the simulation results, error analysis is performed, and a model prediction error report is generated. Finally, the parameters of the dynamic geological model are adjusted according to the error report to obtain an adjusted dynamic geological model, thereby realizing dynamic simulation and updating of geological mineral resource reserves.

[0108] Preferably, step S1 comprises the following steps:

[0109] Step S11: collecting geospatial data of the target geological area to obtain original geospatial data;

[0110] Step S12: unifying the spatial reference of the original geographic spatial data and performing geometric correction to obtain geometrically corrected data;

[0111] Step S13: standardizing the attribute data of the geometrically corrected data and performing semantic coordination to obtain standardized attribute data;

[0112] Step S14: performing spatial interpolation and resampling on the standardized attribute data to obtain spatial consistency data;

[0113] Step S15: fusing multi-source data on the spatial consistency data and extracting data features to obtain a feature fusion data set;

[0114] Step S16: construct a comprehensive data set for the feature fusion data set to obtain a comprehensive data set.

[0115] In an embodiment of the present invention, a variety of data collection methods are used to obtain geospatial data for a target geological area within a specific longitude and latitude range. A high-precision digital elevation model (DEM) and orthophoto of the target area are obtained by using an unmanned aerial vehicle equipped with a high-resolution camera and a lidar. RTK-GPS is used to accurately locate the geological outcrop and borehole position, and the occurrence, fault information, etc. are measured in combination with a geological compass, recorded in a field data collection app, and exported to a shapefile format. Geophysical exploration equipment, such as a gravimeter, a magnetometer, an electrical instrument, etc., are used to collect data along a pre-planned survey line or survey network, and the data is stored in a standard format, such as SEG-Y format (seismic data) and DAT format (gravity, magnetic data). Existing geological maps, geochemical analysis reports, mining records and other materials are collected and digitized, such as scanning them into images or entering them into a database.

[0116] All geospatial data are unified to the WGS84 coordinate system and UTM projection (select the appropriate UTM zone number according to the latitude and longitude of the target area). ArcGIS software is used to project and transform the geological outcrop and borehole location data in shapefile format. For orthophotos acquired by drones, ground control points (GCP) and ENVI software are used to perform geometric precision correction, eliminate image distortion, and perform orthorectification to generate orthophotos with real geographic coordinates. For geophysical exploration data, professional software is used for preprocessing, such as using Geosoft Oasis Montaj software to perform baseline correction, drift correction, and terrain correction on gravity data, and daily variation correction and magnetic anomaly separation on magnetic data.

[0117] Unify the units, data types, and encoding methods of attribute data in different data sources. For example, unify the length unit to meter, the mass unit to kilogram, and the concentration unit to ppm. Use a unified lithology classification standard, such as the International Stratigraphic Guide, to unify and encode the descriptions of lithology in different data sources. Establish a database, store all attribute data in the database, and define the data table structure and field types. Use SQL statements to clean and organize data, such as removing duplicate records, filling missing values, and handling outliers.

[0118] Use the inverse distance weighted interpolation method (IDW) to convert discrete drill hole grade data into continuous grade field raster data, and set appropriate search radius and power parameters. For surface elevation data DEM, use bilinear interpolation to resample it to the same resolution and range as the grade field raster data. For geophysical exploration data, such as magnetic anomaly data, use Kriging interpolation to convert it into raster data with the same resolution as other data, and select appropriate parameters based on the variogram.

[0119] Use principal component analysis (PCA) to extract features from multi-source raster data (such as DEM, grade field, magnetic anomaly field), reduce the dimensionality of multiple bands of data to a few principal components, and retain the main information. Calculate the distance from each grid cell to the nearest fault line, generate a distance map to the fault, and add it to the dataset as a new feature layer. Use the Sobel operator to detect edges on the DEM, extract terrain slope and aspect information, and add it to the dataset as a new feature layer.

[0120] Store all processed data, including raster data (such as grade fields, geophysical anomaly fields, terrain features) and vector data (such as geological outcrops, drill hole locations, fault lines) in a geographic database. Use a geographic database management system (such as ArcGIS) to establish topological relationships between data, such as defining the contact relationship between different geological units, the cutting relationship between faults and strata, etc. Establish metadata information for each data set, record data source, processing method, data accuracy and other information, to facilitate data management and use. Export the geographic database to a unified data format, such as HDF5 format, for subsequent model reading and processing.

[0121] Preferably, step S2 comprises the following steps:

[0122] Step S21: constructing an initial static geological model according to the comprehensive data set to obtain an initial static geological model;

[0123] Step S22: applying geological history constraints to the initial static geological model to obtain a geological evolution stage model;

[0124] Step S23: constructing a multi-physics field coupling model according to the geological evolution stage model to obtain a multi-physics field coupling model grid;

[0125] Step S24: performing ore-forming fluid migration and enrichment simulation according to the multi-physics field coupling model grid to obtain ore-forming fluid migration simulation results;

[0126] Step S25: updating the model attributes of the initial static geological model and updating the geometric shape of the geological body according to the ore-forming fluid migration simulation results to obtain a dynamic geological model.

[0127] As an example of the present invention, refer to Figure 2 As shown, in this example, step S2 includes:

[0128] Step S21: constructing an initial static geological model according to the comprehensive data set to obtain an initial static geological model;

[0129] In an embodiment of the present invention, geological outcrops, borehole lithology data, fault line data, and surface DEM data in a comprehensive data set are read. A three-dimensional stratigraphic model is constructed using geostatistical methods, such as Kriging interpolation, to define the spatial distribution and contact relationship of different strata. A fault model is constructed based on the spatial position and occurrence information of the fault line, and the fault model is cut with the stratigraphic model so that the stratigraphic model conforms to the fault structure. Based on the borehole grade data, a three-dimensional grade model is constructed using distance inverse weighted interpolation or ordinary Kriging interpolation method to describe the grade distribution of the ore body. The constructed stratigraphic model, fault model, and grade model are integrated into an initial static geological model and saved in VTK format.

[0130] Step S22: applying geological history constraints to the initial static geological model to obtain a geological evolution stage model;

[0131] In an embodiment of the present invention, the time and sequence of occurrence of major tectonic events, such as fold deformation, fault activity, magma intrusion, etc., are determined based on regional geological data and isotope dating data. A tectonic stress field is applied to the initial static geological model using numerical simulation software, such as Abaqus, to simulate the fold deformation process. According to geological historical records, the time and displacement of the fault slip are set, and the slip fault model is used to simulate the impact of fault activity on the stratum. The simulation results of different geological evolution stages are saved as separate model files, for example, geological models of different time steps are saved in VTK format, and arranged in chronological order to form a geological evolution stage model.

[0132] Step S23: constructing a multi-physics field coupling model according to the geological evolution stage model to obtain a multi-physics field coupling model grid;

[0133] In an embodiment of the present invention, a finite element mesh for numerical simulation is generated based on the final form of the model of the geological evolution stage. The geological body is discretized, divided into tetrahedral or hexahedral units, and mesh encryption is performed at important locations such as faults and lithological interfaces to improve calculation accuracy. Each grid unit is assigned corresponding rock physical properties, such as density, porosity, permeability, thermal conductivity, specific heat capacity, Young's modulus, Poisson's ratio, etc. These parameters can be obtained from laboratory measurement data or literature. Define the boundary conditions of the multi-physics field coupling model, such as setting the heat flow boundary conditions at the bottom of the model, the fixed displacement boundary conditions on the side of the model, etc. The generated grid and boundary condition information are saved in a format readable by COMSOL Multiphysics, such as mphbin format.

[0134] Step S24: performing ore-forming fluid migration and enrichment simulation according to the multi-physics field coupling model grid to obtain ore-forming fluid migration simulation results;

[0135] In an embodiment of the present invention, a multi-physics field coupling model grid is imported into COMSOL Multiphysics. An appropriate physical field interface is selected, such as Darcy's law module to simulate fluid flow, heat transfer module to simulate heat transfer, and chemical reaction engineering module to simulate chemical reaction. Initial conditions and boundary conditions are set, such as setting initial fluid pressure, temperature, chemical component concentration, and setting boundary conditions for fluid inlet and outlet. According to mineralization theory and geochemical data, chemical reaction equations and reaction rate constants are defined, such as mineral dissolution and precipitation reactions. Numerical simulation is run to solve the coupling equations of fluid flow, heat transfer and chemical reaction, and fluid pressure field, temperature field, chemical component concentration field and mineral precipitation distribution at different time steps are obtained. The simulation results are output in VTK format, such as the physical field distribution and mineral precipitation amount of each time step.

[0136] Step S25: updating the model attributes of the initial static geological model and updating the geometric morphology of the geological body according to the ore-forming fluid migration simulation results to obtain a dynamic geological model;

[0137] In an embodiment of the present invention, the results of the ore-forming fluid migration simulation are mapped to the initial static geological model. The attribute values ​​of each grid cell in the initial static geological model, such as temperature, pressure, chemical component concentration, mineral composition, etc., are updated. According to the results of the mineral precipitation simulation, the geometric shape and grade distribution of the ore body in the initial static geological model are adjusted. If the simulation results show that the tectonic stress field has changed, it is necessary to update the geometric shape of the geological body according to the stress change, such as simulating fault sliding or fold deformation. The updated geological model, physical field distribution, and mineral distribution information are integrated into a dynamic geological model and saved in VTK format.

[0138] Preferably, step S24 comprises the following steps:

[0139] Step S241: setting the initial fluid and rock chemical compositions according to the multi-physics field coupling model grid to obtain an initial chemical composition model;

[0140] Step S242: constructing a hydrothermal reaction network of the initial chemical composition model to obtain a hydrothermal reaction network model;

[0141] Step S243: constructing chemical reaction control equations according to the hydrothermal reaction network model, and configuring the chemical reaction control equations on the multi-physics field coupling model grid to obtain a group of chemical reaction control equations;

[0142] Step S244: numerically solving the chemical component migration and reaction of the chemical reaction control equations according to the multi-physics field coupling model grid and the initial chemical composition model to obtain chemical component concentration distribution data;

[0143] Step S245: performing mineral dissolution, mineral precipitation and mineral replacement simulation according to the chemical component concentration distribution data and the hydrothermal reaction network model to obtain mineral phase distribution change data;

[0144] Step S246: tracking the enrichment of ore-forming elements according to the chemical component concentration distribution data and the mineral phase distribution change data to obtain the enrichment distribution data of ore-forming elements;

[0145] Step S247: Visualize and analyze the chemical component concentration distribution data, mineral phase distribution change data, and ore-forming element enrichment distribution data to obtain ore-forming fluid migration simulation results.

[0146] In an embodiment of the present invention, the chemical composition of the initial fluid in the simulation area is set based on the existing geochemical analysis data and regional hydrogeological data. For example, the concentration of the main ions in the initial fluid, such as Na+, K+, Ca2+, Mg2+, Cl-, SO42-, HCO3-, etc., and the concentration of ore-forming elements, such as Cu, Pb, Zn, Au, etc., are set. These concentration values ​​are assigned to the fluid units in the multi-physics coupling model grid. According to the rock type and mineral composition data, the chemical composition of the rock is set. For example, the content of the main minerals in different lithological units, such as quartz, feldspar, mica, calcite, etc., and the content of trace elements are set. These content values ​​are assigned to the rock units in the multi-physics coupling model grid. The fluid chemical composition and rock chemical composition data are integrated into an initial chemical composition model and saved in HDF5 format.

[0147] Based on the existing geochemical reaction database and thermodynamic data, such as the SUPCRT92 database, a hydrothermal reaction network model is constructed. This model describes the chemical reactions that may occur between fluids and rocks, such as mineral dissolution, precipitation, ion exchange, redox reactions, etc. According to the geological background and mineralization type of the simulation area, key chemical reactions related to the mineralization process are selected. For example, for hydrothermal gold deposits, gold complexation reactions and sulfide precipitation reactions can be considered. Parameters such as the equilibrium constant, reaction rate constant, and activation energy of each chemical reaction are determined. These reaction equations and parameters are integrated into the hydrothermal reaction network model and saved in XML format.

[0148] Based on the chemical reactions defined in the hydrothermal reaction network model, the chemical reaction control equations are constructed. For example, for mineral dissolution and precipitation reactions, the mass action law can be used to describe the reaction rate. The chemical reaction control equations are coupled with the fluid flow equations and the heat transfer equations to form a set of chemical reaction control equations. For example, fluid flow affects the transport of chemical components, and temperature affects the reaction rate. The chemical reaction control equations are configured into the multiphysics coupling model grid, and the corresponding control equations and parameters are specified for each grid cell.

[0149] Solve the set of governing equations for the chemical reaction using numerical methods, such as the finite element method or the finite difference method. Set the parameters for the numerical solution, such as time step, convergence accuracy, etc. Use the chemical component concentrations in the initial chemical composition model as initial conditions and the boundary conditions in the multiphysics coupling model grid as boundary conditions. Run the numerical simulation and calculate the concentration of each chemical component in each grid cell at each time step. Save the calculation results in HDF5 format, which contains the concentration values ​​of each chemical component in each grid cell at each time step.

[0150] According to the chemical component concentration distribution data, the reaction rate of each chemical reaction in each grid cell is calculated. According to the mineral dissolution, precipitation and replacement reactions defined in the hydrothermal reaction network model, the dissolution, precipitation and replacement of minerals in each grid cell are calculated. For example, if the dissolution rate of a mineral in a grid cell is greater than the precipitation rate, the mineral will dissolve in the cell. Update the content of the mineral in each grid cell. For example, if a mineral precipitates, increase the content of the mineral in the cell. Record the changes in the mineral content in each grid cell at each time step to obtain the mineral phase distribution change data and save it in HDF5 format.

[0151] According to the chemical component concentration distribution data and mineral phase distribution change data, the migration and enrichment process of ore-forming elements in the simulation area can be tracked. For example, the dissolution, transportation and precipitation of gold in the fluid can be tracked. The enrichment coefficient of the ore-forming element in each grid cell is calculated, that is, the ratio of the concentration of the ore-forming element in the cell to its concentration in the initial fluid. The enrichment coefficient of the ore-forming element in each grid cell at each time step is recorded to obtain the enrichment distribution data of the ore-forming element and save it in HDF5 format.

[0152] Use visualization software such as ParaView or VisIt to visualize the chemical component concentration distribution data, mineral phase distribution change data, and ore-forming element enrichment distribution data in three dimensions. For example, the chemical component concentration field, mineral distribution, and ore-forming element enrichment at different time steps can be displayed using isosurfaces, slices, streamlines, etc. Analyze the simulation results, such as studying the enrichment rules of ore-forming elements, the migration path of ore-forming fluids, and the key factors of mineralization. Organize the visualization results and analysis results into a report as the ore-forming fluid migration simulation results.

[0153] Preferably, step S245 is specifically as follows:

[0154] According to the chemical component concentration distribution data, the instantaneous reaction rates of dissolution reaction, precipitation reaction and replacement reaction are calculated for the hydrothermal reaction network model to obtain the instantaneous reaction rate data;

[0155] Determine the potential precipitation minerals and supersaturation of the hydrothermal reaction network model, and obtain the potential precipitation mineral list and supersaturation data;

[0156] According to the instantaneous reaction rate data, the potential precipitation mineral list and supersaturation data, the competitive precipitation process is simulated to obtain the actual precipitation mineral increment data;

[0157] The mineral dissolution process is simulated based on the instantaneous reaction rate data to obtain the actual dissolved mineral reduction data;

[0158] According to the instantaneous reaction rate data and the actual dissolved mineral reduction data, the mineral replacement process and solid phase diffusion influence are simulated to obtain the mineral change data of the replacement reaction;

[0159] Extract the mineral composition of the previous time step from the initial chemical composition model to obtain the mineral composition of the previous time step; update the mineral composition of the grid unit according to the mineral composition of the previous time step, the actual precipitation mineral increment data, the actual dissolved mineral reduction data and the metasomatic reaction mineral change data to obtain the mineral composition data of the current time step;

[0160] The mineral phase distribution change is recorded for the mineral composition of the previous time step and the mineral composition data of the current time step to obtain the mineral phase distribution change data.

[0161] In an embodiment of the present invention, the chemical component concentration distribution data of the current time step and the hydrothermal reaction network model are read. For each grid cell, the instantaneous reaction rate of each reaction is calculated based on the chemical component concentration in the cell and the reaction rate constant defined in the hydrothermal reaction network model. For example, for mineral dissolution reactions, according to the law of mass action, the dissolution rate is proportional to the activity of related ions in the fluid and the surface area of ​​the mineral. For mineral precipitation reactions, the precipitation rate is proportional to the activity of related ions in the fluid and the degree of supersaturation. The instantaneous reaction rate of each reaction in each grid cell is recorded in the instantaneous reaction rate data and saved in HDF5 format.

[0162] Traverse all mineral precipitation reactions in the hydrothermal reaction network model. For each grid cell, calculate the saturation index of each mineral based on the concentration of the chemical components in the cell. The saturation index is equal to the logarithm of the ratio of the ion activity product to the solubility product. If the saturation index is greater than 0, the mineral is in a supersaturated state and is a potential precipitation mineral. Record all potential precipitation minerals in each grid cell and their corresponding supersaturations in the potential precipitation mineral list and supersaturation data, and save them in HDF5 format.

[0163] For each grid cell, the possible precipitation reactions are determined based on the list of potential precipitation minerals and supersaturation data. If there are multiple potential precipitation minerals, competitive precipitation simulation is performed based on their supersaturation and instantaneous reaction rate. For example, the precipitation amount of each reaction can be weighted averaged according to the supersaturation. The actual precipitation amount of each potential precipitation mineral is calculated and the results are recorded in the actual precipitation mineral increment data and saved in HDF5 format.

[0164] For each grid cell, the amount of each mineral dissolved is calculated based on the instantaneous reaction rate data. If the dissolution rate of a mineral is greater than 0, the mineral will dissolve. The actual amount of each mineral dissolved is calculated and recorded in the actual dissolved mineral reduction data and saved in HDF5 format.

[0165] For each grid cell, simulate the mineral metasomatism process based on the instantaneous reaction rate data and the actual dissolved mineral reduction data. For example, if a certain mineral dissolves, the chemical components it releases may cause precipitation or metasomatism of other minerals. Consider the effect of solid phase diffusion on the mineral metasomatism process. For example, solid phase diffusion can affect the rate and range of mineral metasomatism. Record the mineral content changes caused by metasomatism reactions in each grid cell into the metasomatism reaction mineral change data and save it in HDF5 format.

[0166] Extract the mineral composition data of each grid cell in the last time step from the initial chemical composition model. For example, read the content of each mineral in each grid cell in the last time step.

[0167] For each grid cell, the mineral composition is updated according to the mineral composition of the previous time step, the actual precipitation mineral increment data, the actual dissolved mineral reduction data, and the metasomatic reaction mineral change data. For example, the actual precipitation mineral increment is added to the mineral composition of the previous time step, the actual dissolved mineral reduction is subtracted from the mineral composition of the previous time step, and the metasomatic reaction mineral change data is applied to the mineral composition. The updated mineral composition data is recorded into the current time step mineral composition data and saved in HDF5 format.

[0168] Compare the mineral composition of the previous time step with the mineral composition data of the current time step, and record the changes in the mineral content in each grid cell. For example, calculate the increase or decrease in the content of each mineral in each grid cell. Record these changes into the mineral phase distribution change data and save them in HDF5 format.

[0169] Preferably, step S3 comprises the following steps:

[0170] Step S31: acquiring historical exploration data and historical mining data; extracting geological features from the dynamic geological model, historical exploration data and historical mining data, and preparing machine learning training data to obtain a machine learning training data set;

[0171] Step S32: training a reserve prediction model according to the machine learning training data set to obtain a reserve prediction model;

[0172] Step S33: inputting the physical field parameters output by the dynamic geological model into the reserve prediction model, performing probabilistic resource prediction, and performing uncertainty quantification to obtain probabilistic resource prediction results;

[0173] Step S34: construct a probability resource map according to the probability resource prediction results to obtain a probability resource map.

[0174] In an embodiment of the present invention, existing drilling data, geochemical measurement data, geophysical data and historical mining data of the target mining area are collected. Physical field parameters corresponding to each drilling position, such as temperature, pressure, permeability, and different mineral contents, are extracted from the dynamic geological model. These parameters are associated with information such as ore grade and lithology in the drilling data. The geochemical measurement data is spatially interpolated to generate a continuous geochemical anomaly field, and the geochemical anomaly values ​​corresponding to each drilling position are extracted as features. Similarly, the geophysical data is processed to extract the geophysical anomaly values ​​corresponding to the drilling position. The average grade, mining volume and other information of each mining block are extracted from the historical mining data and associated with the spatial position. All extracted feature data are integrated with the ore grade data of the borehole to form a machine learning training data set, which is saved in CSV format, where each row represents a borehole sample and each column represents a feature or target variable (ore grade).

[0175] The machine learning training data set is divided into a training set and a test set with a ratio of 8:2. The training set is used to train a random forest regression model for predicting ore grade. The parameters of the random forest model are set, such as the number of trees is 100 and the maximum number of features for each node is the square root of the total number of features. The trained model is used to predict the test set and evaluate the performance of the model, such as calculating the root mean square error (RMSE) and the coefficient of determination (R-squared). The trained random forest regression model is saved as a pkl file, which is the reserve prediction model.

[0176] Extract the physical field parameters of the entire area from the dynamic geological model, such as temperature, pressure, permeability, and different mineral contents, etc., consistent with the way of extracting features in step S31. Take these parameters as input, use the trained reserve prediction model (random forest regression model) for prediction, and obtain the ore grade prediction value of each grid cell. In order to quantify the uncertainty of the prediction, the Bootstrap method is used. The training set is sampled 100 times with replacement, and a new random forest model is trained each time. Use these 100 models to predict the physical field parameters of the dynamic geological model to obtain 100 grade prediction values ​​for each grid cell. Calculate the mean and standard deviation of the grade prediction value of each grid cell, respectively as the probabilistic resource prediction value and uncertainty index. Save the probabilistic resource prediction value and uncertainty index in HDF5 format to form the probabilistic resource prediction result.

[0177] Read the probabilistic resource prediction results, including the predicted mean and standard deviation of ore grade for each grid cell. Use a 3D visualization library, such as Mayavi, to visualize the predicted mean ore grade in 3D space, for example, using different colors to represent different grade levels. Overlay the visualization of the probabilistic resource prediction results on the geological model, for example, representing high-grade areas in red and low-grade areas in blue. Visualize the predicted standard deviation of ore grade for each grid cell as an uncertainty map, for example, using color depth to represent the size of uncertainty, with higher uncertainty and darker colors. Integrate the 3D visualization results of the predicted mean ore grade and the uncertainty map into a probabilistic resource map and save it as an image or HTML format.

[0178] Preferably, step S4 comprises the following steps:

[0179] Step S41: obtaining a mineral mining plan; applying mining plan boundary conditions to the dynamic geological model according to the mineral mining plan to obtain a mining boundary condition model;

[0180] Step S42: performing a mining disturbance multi-physics field coupling simulation according to the mining boundary condition model and the dynamic geological model to obtain a mining disturbance response field;

[0181] Step S43: performing geological model deformation update on the dynamic geological model according to the mining disturbance response field to obtain a deformation updated geological model;

[0182] Step S44: performing a probability resource map disturbance adjustment on the probability resource map according to the mining disturbance response field to obtain a disturbance adjusted resource map;

[0183] Step S45: Calculate the reserves affected by the disturbance according to the deformation updated geological model, the disturbance adjusted resource map, and the mineral mining plan to obtain the reserves affected by the disturbance;

[0184] Step S46: dynamically updating the deformation updated geological model according to the disturbance adjusted resource map to obtain an updated dynamic geological model;

[0185] Step S47: collecting actual production data and monitoring data of the target geological area, and integrating the measured data to obtain a measured data set;

[0186] Step S48: performing prediction result comparison and error analysis on the reserves affected by the disturbance, the updated dynamic geological model and the measured data set to obtain a model prediction error report;

[0187] Step S49: adjusting the model parameters of the updated dynamic geological model according to the model prediction error report to obtain an adjusted dynamic geological model.

[0188] In an embodiment of the present invention, a mineral mining plan for a period of time in the future is obtained, including information such as the daily mining area, mining method (such as open-site method, backfill method), and mining volume. The mining plan is converted into boundary conditions required for numerical simulation. For example, for open-site mining, the boundary of the mining area is set as a free surface, and the corresponding grid cells are removed. For backfill mining, it is necessary to define the mechanical parameters of the backfill material, and add the backfill material to the model after mining. These boundary condition information are applied to the dynamic geological model to generate a mining boundary condition model, which is saved in VTK format, including the geological model and corresponding boundary condition information for each time step.

[0189] Use COMSOL Multiphysics software to import the mining boundary condition model and dynamic geological model. Set the parameters of multi-physics field coupling simulation, such as time step, solver type, etc. Activate the solid mechanics module, fluid flow module, and heat transfer module to simulate the stress changes, groundwater flow, and temperature changes during the mining process. Run numerical simulations to calculate the changes in each physical field in the model at each time step, such as stress field, displacement field, seepage field, temperature field, etc. The simulation results are output in VTK format, including the physical field data of each grid unit at each time step, forming the mining disturbance response field.

[0190] Read the displacement field data of each time step in the mining disturbance response field. Apply the displacement field data to the dynamic geological model and update the coordinates of each grid node to achieve the deformation update of the geological model. For example, add the initial coordinates of each node to its corresponding displacement vector to obtain the new coordinates of the node. Regenerate the mesh of the geological model based on the updated node coordinates to obtain the deformation updated geological model and save it in VTK format.

[0191] Read the stress field and displacement field data of each time step in the mining disturbance response field. Based on the stress field and displacement field data, evaluate the impact of mining disturbance on the ore body. For example, the ore body in the high stress or large displacement area may be broken or displaced, which will affect its grade and mineability. According to the evaluation results, adjust the ore grade prediction value and uncertainty index of the corresponding area in the probability resource map. For example, reduce the grade prediction value of the broken area and increase the uncertainty index. Save the adjusted probability resource map in image or HTML format to form a disturbance-adjusted resource map.

[0192] According to the mineral mining plan, determine the mining area and mining volume for each time step. Update the geological model according to the deformation and adjust the resource map according to the disturbance, and calculate the volume and average grade of the remaining ore for each time step. Multiply the volume of the remaining ore by the average grade to obtain the reserves of the remaining ore. Consider the mining losses, such as the dilution and loss rate of the ore, and correct the calculated reserves. Record the reserve data for each time step to obtain the reserves affected by the disturbance and save it in CSV format.

[0193] Apply the updated ore body information in the disturbance adjustment resource map, such as grade distribution and geometry, to the deformation update geological model. Update the ore grade values ​​of the corresponding areas in the geological model. If the ore body has significant displacement or deformation, the mesh of the geological model needs to be updated to accurately reflect the changes in the ore body. Save the updated geological model in VTK format to form an updated dynamic geological model.

[0194] Collect actual production data of the mine, such as daily mining volume, ore grade, waste rock volume, etc. Collect monitoring data of the mine, such as surface subsidence, groundwater level change, rock deformation, etc. Sort and integrate actual production data and monitoring data to form a measured data set and save it in CSV format.

[0195] Compare the reserves affected by the disturbance with the production volume and ore grade in the actual production data to calculate the reserve prediction error. Compare the physical field data in the updated dynamic geological model, such as stress field, displacement field, seepage field, etc., with the monitoring data in the measured data set to calculate the model prediction error. Analyze the source and size of the prediction error, such as which areas or physical fields have large prediction errors. Organize the error analysis results into a report to form a model prediction error report.

[0196] According to the model prediction error report, adjust the parameters of the updated dynamic geological model. For example, if it is found that the reserve prediction error in some areas is large, the geological model parameters of the area can be adjusted, such as lithology, grade, fault, etc. If it is found that the prediction error of some physical fields is large, the model parameters of the physical fields can be adjusted, such as permeability, Young's modulus, etc. The adjusted parameters are applied to the updated dynamic geological model to obtain the adjusted dynamic geological model and save it in VTK format.

[0197] Preferably, step S42 includes the following steps:

[0198] Step S421: configuring the ground stress change caused by mining according to the mining boundary condition model and the dynamic geological model to obtain a ground stress change model;

[0199] Step S422: extracting geological model parameters from the dynamic geological model to obtain rock mechanics parameters, hydrogeological parameters, thermal parameters, fault mechanics parameters and fault geometry information;

[0200] Step S423: solving the deformation field of the rock and soil mass caused by mining according to the geostress variation model and rock mechanics parameters to obtain the deformation field data of the rock and soil mass;

[0201] Step S424: using the rock and soil deformation field data as input, updating the hydrogeological parameters to obtain updated hydrogeological parameters; simulating the impact of mining on the groundwater flow field based on the updated hydrogeological parameters to obtain groundwater flow field data;

[0202] Step S425: performing coupled simulation of temperature field changes on the mining boundary condition model according to thermal parameters, rock and soil deformation field data, and groundwater flow field data to obtain temperature field data;

[0203] Step S426: performing fault activation and surface settlement analysis according to the rock and soil deformation field data and fault mechanics parameters to obtain fault activation and surface settlement analysis results;

[0204] Step S427: integrating the rock and soil deformation field data, groundwater flow field data, temperature field data, and fault activation and surface settlement analysis results into a mining disturbance response field to obtain a mining disturbance response field.

[0205] In an embodiment of the present invention, the mining area information of each time step is read from the mining boundary condition model. In the dynamic geological model, the units corresponding to the mining area are removed to form a free boundary. While removing the units, the initial stress state of these units is recorded. The initial stress state is applied to the remaining model units as a boundary condition to simulate the stress redistribution caused by mining. The stress change information of each time step is saved in the ground stress change model and stored in HDF5 format.

[0206] Extract rock mechanics parameters corresponding to different lithological units from the dynamic geological model, such as Young's modulus, Poisson's ratio, density, tensile strength, compressive strength, internal friction angle and cohesion. Extract hydrogeological parameters such as permeability and porosity corresponding to different lithological units. Extract thermal parameters such as thermal conductivity, specific heat capacity, thermal expansion coefficient, etc. corresponding to different lithological units. Extract fault mechanics parameters such as friction angle and cohesion of the fault plane, as well as geometric information such as the position, strike, and dip of the fault plane. Save the extracted parameter information as CSV files.

[0207] The stress changes calculated in the geostress change model are used as boundary conditions and applied to the dynamic geological model. The model is solved using finite element analysis software, such as Abaqus, to calculate the deformation of the rock and soil mass caused by mining. During the calculation process, the rock mechanics parameters extracted in step S422 are used to assign corresponding material properties to different lithology units. The displacement of each grid node at each time step is calculated, and the results are saved in VTK format to form the rock and soil mass deformation field data.

[0208] Based on the deformation field data of the rock mass, the volume strain of each unit is calculated. The volume strain is associated with the change in permeability using empirical formulas or experimental data. For example, the cubic law can be used to describe the change in permeability of fractured rock with volume strain. The updated permeability value is assigned to the dynamic geological model to obtain the updated hydrogeological parameters. The groundwater flow field is simulated using the updated hydrogeological parameters, such as permeability and porosity, and mining boundary conditions (e.g., the location and flow rate of drainage holes). The groundwater flow equations, such as Darcy's law, are solved using finite element or finite difference methods, and the groundwater head and velocity of each grid cell at each time step are calculated. The results are saved in VTK format to form the groundwater flow field data.

[0209] Use COMSOL Multiphysics software to import the mining boundary condition model. Assign the thermal parameters extracted in step S422 to different lithology units in the model. Use the rock deformation field data calculated in step S423 and the groundwater flow field data calculated in step S424 as input to perform coupled simulation of temperature field changes. Consider the heat generated during the mining process, the thermal convection caused by groundwater flow, and the heat conduction caused by rock deformation. Solve the heat conduction equation, calculate the temperature of each grid unit at each time step, and save the results in VTK format to form temperature field data.

[0210] Extract the displacement information of the grid nodes near the fault from the deformation field data of the rock mass. Calculate the shear stress and normal stress of each node on the fault plane based on the fault mechanics parameters, such as friction angle and cohesion. Use the Coulomb friction criterion to determine whether the fault has slipped. If the shear stress is greater than the friction coefficient multiplied by the normal stress plus the cohesion, the fault is considered to have slipped. Calculate the fault slip and save the results as a CSV file. Extract the vertical displacement of the surface grid nodes from the deformation field data of the rock mass and calculate the surface settlement. Integrate the surface settlement and fault slip data to generate fault activation and surface settlement analysis results, and save them in CSV format.

[0211] The rock and soil deformation field data calculated in step S423, the groundwater flow field data calculated in step S424, the temperature field data calculated in step S425, and the fault activation and surface settlement analysis results calculated in step S426 are integrated together to form a mining disturbance response field. All data are saved in HDF5 format for easy use in subsequent steps.

[0212] Preferably, step S421 is specifically:

[0213] According to the mining boundary condition model, the excavation unit of the dynamic geological model is identified, and the free surface is set to obtain the excavation unit identification result and the initial free surface model;

[0214] A step-by-step excavation simulation is performed according to the excavation unit identification results, and stress release is performed according to the initial free surface model to obtain a step-by-step excavation stress release model;

[0215] The constitutive model of the filling body is constructed according to the mining boundary condition model and the dynamic geological model to obtain the constitutive model parameters of the filling body; the contact between the filling body and the surrounding rock is simulated according to the constitutive model parameters of the filling body to obtain the contact model between the filling body and the surrounding rock;

[0216] Dynamically adjust the boundary conditions of complex mining processes according to the mining boundary condition model and the dynamic geological model to obtain a dynamic mining boundary condition sequence;

[0217] According to the mining boundary condition model and the dynamic geological model, the equivalent simulation of blasting disturbance is carried out to obtain the blasting equivalent load model;

[0218] The initial free surface model, gradual excavation stress release model, filling body and surrounding rock contact model, dynamic mining boundary condition sequence and blasting equivalent load model are integrated to generate the ground stress change model caused by mining.

[0219] In an embodiment of the present invention, the mining boundary condition model is read to obtain the spatial extent of the mining area at each time step. The geometric shape of the mining area is compared with the grid of the dynamic geological model to identify the grid cells located in the mining area. These cells are marked as excavation cells, and the surfaces of the excavation cells are set as free surfaces, that is, no constraints are imposed on these surfaces to simulate the cavities formed after mining. The excavation unit identification results and the initial free surface model are saved as a VTK format file, which contains the ID of the excavation unit and the geometric information of the free surface.

[0220] The excavation units are removed step by step according to the excavation sequence set in the mining plan. In each excavation step, a part of the excavation units is removed and the stress distribution of the model is recalculated. After removing the excavation units, zero stress boundary conditions are applied on the boundaries corresponding to the initial free surface model to simulate stress release. The stress changes of each excavation step are recorded and the results are saved in the step-by-step excavation stress release model and stored in HDF5 format.

[0221] Determine the constitutive model of the filling body according to the filling material type specified in the mining boundary condition model, such as cement mortar, tailings filling, etc. For example, for cement mortar, you can use the elastoplastic constitutive model and determine its elastic modulus, Poisson's ratio, yield strength and other parameters. For tailings filling, you can use the particle flow model and determine its particle size, friction coefficient, porosity and other parameters. Save the parameters of the filling body constitutive model as a CSV file.

[0222] In the dynamic geological model, the grid cells of the filling area are marked as filling cells. The constitutive model parameters of the filling body determined in step S421.3 are assigned to the filling cells. The contact relationship between the filling unit and the surrounding rock unit is set, for example, using a contact algorithm to simulate the interaction between the two. The contact algorithm needs to consider the mechanical properties of the filling body and the surrounding rock, as well as the friction coefficient and cohesion between the two. The contact model of the filling body and the surrounding rock is saved as a VTK format file, which contains the IDs of the filling unit and the surrounding rock unit, as well as the contact relationship information between the two.

[0223] For complex mining processes, such as layered mining and staged backfilling, the boundary conditions need to be adjusted dynamically according to the mining progress. For example, in the layered mining process, the excavation of each layer will change the boundary conditions of the model. In the staged backfilling process, the addition of backfill material will also change the boundary conditions of the model. According to the mining sequence and backfilling plan defined in the mining boundary condition model, the boundary conditions of the model are dynamically adjusted, such as the free face position, the range of the backfill body, etc. The boundary condition information of each time step is saved in the dynamic mining boundary condition sequence and stored in HDF5 format.

[0224] The equivalent load generated by blasting is calculated based on the blasting parameters defined in the mining boundary condition model, such as explosive type, charge, blast hole location, etc. The pressure waves and vibrations caused by blasting can be calculated using empirical formulas or numerical simulation methods. The calculated equivalent load is applied to the dynamic geological model to simulate the impact of blasting on the surrounding rock mass. The blasting equivalent load model is saved as an HDF5 format file, which contains information such as the location of each blasting point, load size, and action time.

[0225] The initial free surface model, the step-by-step excavation stress release model, the backfill and surrounding rock contact model, the dynamic mining boundary condition sequence, and the blasting equivalent load model are integrated together to generate a mining-induced ground stress change model. The model contains stress change information for each time step, including stress redistribution caused by excavation, backfilling, blasting, etc. The ground stress change model is saved in HDF5 format for easy use in subsequent steps.

[0226] Preferably, step S426 is specifically:

[0227] Constructing a fault geometry model based on fault geometry information and performing model mesh division to obtain a fault geometry model and a fault unit mesh;

[0228] Apply fault contact constraints according to the fault geometry model and fault mechanics parameters, and determine the fault constitutive model parameters to obtain the fault contact constraint model and fault constitutive model parameters;

[0229] According to the deformation field data of the rock and soil mass and the fault contact constraint model, the stress state of the fault plane is calculated for the fault unit grid to obtain the stress distribution data of the fault plane;

[0230] According to the fault contact constraint model, fault plane stress distribution data and fault constitutive model parameters, the fault stability is evaluated and the fault slip amount is calculated to obtain the fault stability evaluation results and fault slip amount data;

[0231] Extract surface settlement information from the rock and soil deformation field data to obtain surface settlement displacement data;

[0232] Conduct surface settlement analysis on the surface settlement displacement data and visualize the surface settlement to obtain the surface settlement analysis diagram and surface settlement curve;

[0233] The fault stability evaluation results, fault slip data, surface settlement displacement data, surface settlement analysis diagram and surface settlement curve are integrated to obtain the fault activation and surface settlement analysis results.

[0234] In an embodiment of the present invention, geometric information of the fault, including the position, strike, dip and inclination of the fault, is extracted from the dynamic geological model. Using this information, a geometric model of the fault is constructed in a three-dimensional modeling software (such as GOCAD), represented as a series of triangular or quadrilateral facets. The fault geometric model is meshed to generate a fault unit mesh. In order to ensure the calculation accuracy, the mesh is encrypted in the area near the fault, using a smaller unit size. The generated fault geometric model and fault unit mesh are saved as a VTK format file.

[0235] Contact constraints are imposed on the fault cell grid to simulate the interaction between the rock masses on both sides of the fault. The Coulomb friction model is used to describe the mechanical behavior of the fault and the fault constitutive model parameters are determined, including the friction angle and cohesion. These parameters can be obtained from laboratory tests or literature. The fault contact constraints and constitutive model parameters are saved as a CSV file, which contains the friction angle and cohesion of each fault cell.

[0236] Extract the displacement information of the fault unit grid nodes from the deformation field data of the rock and soil. Calculate the normal stress and shear stress of each node on the fault plane based on the displacement information and the fault contact constraint model. Save the calculated stress state data as an HDF5 format file, which contains the normal stress and shear stress of each fault unit.

[0237] The stability of the fault is evaluated based on the Coulomb friction criterion. The shear stress calculated from the stress distribution data on the fault plane is compared with the shear strength determined by the friction angle and cohesion. If the shear stress exceeds the shear strength, the fault is considered to have slipped. The slip criterion of the fault constitutive model can be used to calculate the slip amount. The fault stability evaluation results and the fault slip amount data are saved as a CSV file, which contains the stability state and slip amount of each fault unit.

[0238] Extract the vertical displacement information of the surface grid nodes from the deformation field data of the rock and soil mass, and save the displacement information as a CSV file to form the surface settlement displacement data.

[0239] Perform statistical analysis on the surface settlement displacement data, such as calculating the maximum settlement, average settlement, settlement range, etc. Use a drawing library, such as matplotlib, to visualize the surface settlement data as a surface settlement analysis diagram, such as using a color map to represent the spatial distribution of settlement. Draw a surface settlement curve to show the trend of settlement over time. Save the surface settlement analysis diagram and surface settlement curve as PNG and CSV format files.

[0240] Integrate the fault stability evaluation results, fault slip data, surface settlement displacement data, surface settlement analysis diagrams, and surface settlement curves to form fault activation and surface settlement analysis results. Save all data and charts in HDF5 format for easy use and analysis in subsequent steps.

[0241] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0242] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for simulating geological mineral resource reserves, characterized in that: The following steps are involved: Step S1: collect multi-source heterogeneous data in the target geological area and perform comprehensive feature extraction to obtain a comprehensive data set; Step S2: constructing an initial static geological model according to the comprehensive data set to obtain an initial static geological model; constructing a multi-physics field coupling model according to the initial static geological model to obtain a multi-physics field coupling model grid; simulating the migration and enrichment of ore-forming fluids according to the multi-physics field coupling model grid to obtain a simulation result of ore-forming fluid migration; updating the geometric morphology of the geological body of the initial static geological model according to the simulation result of ore-forming fluid migration to obtain a dynamic geological model; Step S3: extracting geological features from the dynamic geological model, and training a reserve prediction model using machine learning to obtain a reserve prediction model; using the reserve prediction model to perform probabilistic resource prediction to obtain a probabilistic resource map; Step S4: obtaining a mineral mining plan; applying mining plan boundary conditions to the dynamic geological model according to the mineral mining plan to obtain a mining boundary condition model; performing mining disturbance multi-physical field coupling simulation according to the mining boundary condition model and the dynamic geological model to obtain a mining disturbance response field; calculating the reserves affected by the disturbance according to the mining disturbance response field, the dynamic geological model and the probabilistic resource map to obtain the reserves affected by the disturbance; performing a dynamic geological model update on the deformation update geological model to obtain an updated dynamic geological model; performing a prediction result error analysis according to the reserves affected by the disturbance and the updated dynamic geological model to obtain a model prediction error report; adjusting the model parameters of the updated dynamic geological model according to the model prediction error report to obtain an adjusted dynamic geological model to achieve the simulation task of geological mineral resource reserves, wherein step S4 is specifically as follows: Step S41: obtaining a mineral mining plan; applying mining plan boundary conditions to the dynamic geological model according to the mineral mining plan to obtain a mining boundary condition model; Step S42: Perform a mining disturbance multi-physical field coupling simulation based on the mining boundary condition model and the dynamic geological model to obtain a mining disturbance response field, wherein step S42 is specifically as follows: Step S421: configuring the ground stress change caused by mining according to the mining boundary condition model and the dynamic geological model to obtain a ground stress change model, wherein step S421 is specifically as follows: According to the mining boundary condition model, the excavation unit of the dynamic geological model is identified, and the free surface is set to obtain the excavation unit identification result and the initial free surface model; A step-by-step excavation simulation is performed according to the excavation unit identification results, and stress release is performed according to the initial free surface model to obtain a step-by-step excavation stress release model; The constitutive model of the filling body is constructed according to the mining boundary condition model and the dynamic geological model to obtain the constitutive model parameters of the filling body; the contact between the filling body and the surrounding rock is simulated according to the constitutive model parameters of the filling body to obtain the contact model between the filling body and the surrounding rock; According to the mining boundary condition model and the dynamic geological model, the boundary conditions of the complex mining process are dynamically adjusted to obtain a dynamic mining boundary condition sequence; According to the mining boundary condition model and the dynamic geological model, the equivalent simulation of blasting disturbance is carried out to obtain the blasting equivalent load model; The initial free surface model, the gradual excavation stress release model, the backfill and surrounding rock contact model, the dynamic mining boundary condition sequence and the blasting equivalent load model are integrated to generate the ground stress change model caused by mining, and the ground stress change model is obtained; Step S422: extracting geological model parameters from the dynamic geological model to obtain rock mechanics parameters, hydrogeological parameters, thermal parameters, fault mechanics parameters and fault geometry information; Step S423: solving the deformation field of the rock and soil mass caused by mining according to the geostress variation model and rock mechanics parameters to obtain the deformation field data of the rock and soil mass; Step S424: using the rock and soil deformation field data as input, updating the hydrogeological parameters to obtain updated hydrogeological parameters; The impact of mining on groundwater flow field is simulated based on the updated hydrogeological parameters to obtain groundwater flow field data; Step S425: performing coupled simulation of temperature field changes on the mining boundary condition model according to thermal parameters, rock and soil deformation field data, and groundwater flow field data to obtain temperature field data; Step S426: Perform fault activation and surface settlement analysis based on the rock and soil deformation field data and fault mechanics parameters to obtain fault activation and surface settlement analysis results, wherein step S426 is specifically as follows: Constructing a fault geometry model based on fault geometry information and performing model mesh division to obtain a fault geometry model and a fault unit mesh; Apply fault contact constraints according to the fault geometry model and fault mechanics parameters, and determine the fault constitutive model parameters to obtain the fault contact constraint model and fault constitutive model parameters; According to the deformation field data of the rock and soil mass and the fault contact constraint model, the stress state of the fault plane is calculated for the fault unit grid to obtain the stress distribution data of the fault plane; According to the fault contact constraint model, fault plane stress distribution data and fault constitutive model parameters, the fault stability is evaluated and the fault slip amount is calculated to obtain the fault stability evaluation results and fault slip amount data; Extract surface settlement information from the rock and soil deformation field data to obtain surface settlement displacement data; Conduct surface settlement analysis on the surface settlement displacement data and visualize the surface settlement to obtain the surface settlement analysis diagram and surface settlement curve; Integrate the fault stability evaluation results, fault slip data, surface settlement displacement data, surface settlement analysis diagrams and surface settlement curves to obtain the fault activation and surface settlement analysis results; Step S427: integrating the rock and soil deformation field data, groundwater flow field data, temperature field data, and fault activation and surface settlement analysis results into a mining disturbance response field to obtain a mining disturbance response field; Step S43: performing geological model deformation update on the dynamic geological model according to the mining disturbance response field to obtain a deformation updated geological model; Step S44: performing a probability resource map disturbance adjustment on the probability resource map according to the mining disturbance response field to obtain a disturbance-adjusted resource map; Step S45: Calculate the reserves affected by the disturbance according to the deformation updated geological model, the disturbance adjusted resource map, and the mineral mining plan to obtain the reserves affected by the disturbance; Step S46: dynamically updating the deformation updated geological model according to the disturbance adjusted resource map to obtain an updated dynamic geological model; Step S47: collecting actual production data and monitoring data of the target geological area, and integrating the measured data to obtain a measured data set; Step S48: performing prediction result comparison and error analysis on the reserves affected by the disturbance, the updated dynamic geological model and the measured data set to obtain a model prediction error report; Step S49: adjusting the model parameters of the updated dynamic geological model according to the model prediction error report to obtain an adjusted dynamic geological model.

2. The method for simulating geological mineral resource reserves according to claim 1, characterized in that: Step S1 includes the following steps: Step S11: collecting geospatial data of the target geological area to obtain original geospatial data; Step S12: unifying the spatial reference of the original geographic spatial data and performing geometric correction to obtain geometrically corrected data; Step S13: standardizing the attribute data of the geometrically corrected data and performing semantic coordination to obtain standardized attribute data; Step S14: performing spatial interpolation and resampling on the standardized attribute data to obtain spatial consistency data; Step S15: fusing multi-source data on spatial consistency data and extracting data features to obtain a feature fusion data set; Step S16: construct a comprehensive data set for the feature fusion data set to obtain a comprehensive data set.

3. The method for simulating geological mineral resource reserves according to claim 1, characterized in that: Step S2 includes the following steps: Step S21: constructing an initial static geological model according to the comprehensive data set to obtain an initial static geological model; Step S22: applying geological history constraints to the initial static geological model to obtain a geological evolution stage model; Step S23: constructing a multi-physics field coupling model according to the geological evolution stage model to obtain a multi-physics field coupling model grid; Step S24: performing ore-forming fluid migration and enrichment simulation according to the multi-physics field coupling model grid to obtain ore-forming fluid migration simulation results; Step S25: updating the model attributes of the initial static geological model and updating the geometric shape of the geological body according to the ore-forming fluid migration simulation results to obtain a dynamic geological model.

4. The method for simulating geological mineral resource reserves according to claim 3, characterized in that: Step S24 includes the following steps: Step S241: setting the initial fluid and rock chemical compositions according to the multi-physics field coupling model grid to obtain an initial chemical composition model; Step S242: constructing a hydrothermal reaction network of the initial chemical composition model to obtain a hydrothermal reaction network model; Step S243: constructing chemical reaction control equations according to the hydrothermal reaction network model, and configuring the chemical reaction control equations on the multi-physics field coupling model grid to obtain a group of chemical reaction control equations; Step S244: numerically solving the chemical component migration and reaction of the chemical reaction control equations according to the multi-physics field coupling model grid and the initial chemical composition model to obtain chemical component concentration distribution data; Step S245: performing mineral dissolution, mineral precipitation and mineral replacement simulation according to the chemical component concentration distribution data and the hydrothermal reaction network model to obtain mineral phase distribution change data; Step S246: tracking the enrichment of ore-forming elements according to the chemical component concentration distribution data and the mineral phase distribution change data to obtain the enrichment distribution data of ore-forming elements; Step S247: Visualize and analyze the chemical component concentration distribution data, mineral phase distribution change data, and ore-forming element enrichment distribution data to obtain ore-forming fluid migration simulation results.

5. The method for simulating geological mineral resource reserves according to claim 4, characterized in that: Step S245 is specifically as follows: According to the chemical component concentration distribution data, the instantaneous reaction rates of dissolution reaction, precipitation reaction and replacement reaction are calculated for the hydrothermal reaction network model to obtain the instantaneous reaction rate data; Determine the potential precipitation minerals and supersaturation of the hydrothermal reaction network model, and obtain the potential precipitation mineral list and supersaturation data; According to the instantaneous reaction rate data, the potential precipitation mineral list and supersaturation data, the competitive precipitation process is simulated to obtain the actual precipitation mineral increment data; The mineral dissolution process is simulated based on the instantaneous reaction rate data to obtain the actual dissolved mineral reduction data; According to the instantaneous reaction rate data and the actual dissolved mineral reduction data, the mineral replacement process and solid phase diffusion influence are simulated to obtain the mineral change data of the replacement reaction; Extract the mineral composition of the previous time step from the initial chemical composition model to obtain the mineral composition of the previous time step; update the mineral composition of the grid unit according to the mineral composition of the previous time step, the actual precipitation mineral increment data, the actual dissolved mineral reduction data and the metasomatic reaction mineral change data to obtain the mineral composition data of the current time step; The mineral phase distribution change is recorded for the mineral composition of the previous time step and the mineral composition data of the current time step to obtain the mineral phase distribution change data.

6. The method for simulating geological mineral resource reserves according to claim 1, characterized in that: Step S3 includes the following steps: Step S31: acquiring historical exploration data and historical mining data; extracting geological features from the dynamic geological model, historical exploration data and historical mining data, and preparing machine learning training data to obtain a machine learning training data set; Step S32: training a reserve prediction model according to the machine learning training data set to obtain a reserve prediction model; Step S33: inputting the physical field parameters output by the dynamic geological model into the reserve prediction model, performing probabilistic resource prediction, and performing uncertainty quantification to obtain probabilistic resource prediction results; Step S34: construct a probability resource map according to the probability resource prediction results to obtain a probability resource map.

Images