A mine intelligent blasting management system and method

Abstract

The present invention provides a mine intelligent blasting management system and method. The system includes: a data acquisition module for collecting and preprocessing mine rock mass data and blasting data to establish a blasting database; a three-dimensional model construction module for constructing a digital three-dimensional rock mass model of the blasting area based on the mine rock mass data; an intelligent decision-making module for establishing an economic mathematical model based on the digital three-dimensional rock mass model of the blasting area and the collected blasting data of the entire blasting process, making intelligent decisions on mine blasting and obtaining decision results; and a blasting effect evaluation module for performing blasting based on the decision results and evaluating the blasting effects, thereby completing the management of intelligent mine blasting. Through intelligent management of mine blasting, blasting decisions are optimized and blasting costs are reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of mine blasting, and in particular relates to a mine intelligent blasting management system and method. Background Art

[0002] Mine blasting operations are highly dangerous, and their management is consequently complex. Existing technologies typically manage blasting operations using time cards or other methods, with paper and pen records kept of blasting operations. During a complete blasting operation, the complexity of the process can lead to personnel not strictly following blasting safety procedures, falsifying blasting data and time. This lacks effective oversight and management, creating potential safety hazards. Furthermore, traditional mine management cannot calculate blasting costs, hindering blasting decision-making. Summary of the Invention

[0003] In view of the shortcomings of the existing technology, the present invention provides a mine intelligent blasting management system and method, which reduces blasting costs through intelligent management and decision-making of mine blasting.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] A mine intelligent blasting management system, comprising: a data acquisition module, a three-dimensional model building module, an intelligent decision-making module and a blasting effect evaluation module;

[0006] The data acquisition module is used to collect and pre-process mine rock mass data and blasting data to establish a blasting database;

[0007] The three-dimensional model building module is used to build a digital three-dimensional rock model of the blasting area based on the mine rock data;

[0008] An intelligent decision-making module is used to establish an economic mathematical model based on the digital three-dimensional rock model of the blasting area and the blasting data collected throughout the blasting process, make intelligent decisions on mine blasting, and obtain decision results;

[0009] The blasting effect evaluation module is used to carry out blasting based on the decision result and evaluate the blasting effect, thereby completing the management of intelligent blasting in mines.

[0010] Preferably, the mine rock mass data includes mine geological parameters, rock mass characteristics, rock mass distribution and mine image data; wherein the mine image data is obtained by oblique photography;

[0011] The blasting data includes data on drilling, blasting, loading, transporting, breaking and grinding processes;

[0012] The data acquisition module also collects other relevant data; the other relevant data include climate conditions in the mine blasting area, personnel data and various equipment data;

[0013] The data acquisition module includes a data acquisition unit, a data preprocessing unit and a database construction unit;

[0014] The data acquisition unit is used to acquire the rock mass data, the blasting data and the other related data;

[0015] The data preprocessing unit is used to clean all collected data and fill in missing values ​​to obtain preprocessed data;

[0016] The database construction unit is used to store the pre-processed data accordingly.

[0017] Preferably, the three-dimensional model construction module includes an image data correction unit, a triangle matching unit, an initial model construction unit, a mapping unit and a final model construction unit;

[0018] The image data correction unit is configured to establish a coordinate matrix of the mine image data, obtain distortion data in the mine image data based on the coordinate matrix, and correct the distortion data to obtain the corrected mine image data;

[0019] The triangle matching unit is used to perform aerial triangulation connection point matching based on the corrected mine image data to obtain the orientation elements of the mine image data;

[0020] The initial model building unit is used to build an initial three-dimensional real scene model based on the orientation elements;

[0021] The mapping unit is configured to perform data mapping on the initial three-dimensional real scene model based on the mine rock mass data; and to establish three-dimensional mine coordinates and two-dimensional texture coordinates based on the mine image data, perform texture mapping, and obtain a three-dimensional real scene model;

[0022] The final model construction unit is used to interpolate the unknown points of the three-dimensional real scene model based on the bilinear interpolation method to obtain a digital three-dimensional rock model of the blasting area.

[0023] Preferably, the intelligent decision-making module includes a cost calculation unit, a mathematical model construction unit and a decision-making unit;

[0024] a cost calculation unit, configured to calculate mining costs based on the blasting data; wherein the mining costs include drilling costs, blasting costs, shoveling costs, transportation costs, crushing costs, and grinding costs;

[0025] A mathematical model building unit, configured to calculate the weight of each cost in the mining cost based on an entropy weight method, and to build the economic mathematical model using a linear regression equation;

[0026] The decision-making unit is used to obtain the comprehensive capital cost of mine blasting based on the economic mathematical model; based on the comprehensive capital cost, make cost-considered blasting intelligent decisions to obtain the decision results.

[0027] Preferably, the blasting effect evaluation module includes a feature extraction unit, an effect evaluation unit and a decision optimization unit;

[0028] The feature extraction unit is used to extract features of rock fragments between two drill holes after blasting based on image features, obtain blasting fragment size distribution features, and calculate the single-hole blasting range based on the blasting fragment size distribution features;

[0029] The effect evaluation unit is used to evaluate the blasting effect based on the blasting fragmentation distribution characteristics and the single-hole blasting range;

[0030] The decision optimization unit is used to optimize intelligent decision-making using a neural network based on the blasting effect evaluation result, and to predict the blasting effect based on the optimized intelligent decision.

[0031] Preferably, the decision optimization unit includes an optimization weight acquisition subunit, a fitness calculation subunit, an optimization subunit and a decision optimization subunit;

[0032] The optimization weight acquisition subunit is used to optimize the loss function of the deep neural network using the SGD optimizer and obtain the SGD optimization weight;

[0033] The fitness calculation subunit is used to obtain individual fitness values ​​based on the SGD optimization weight and the encoding method of deep neural network evolution; wherein each individual represents an updated network weight; and the individual fitness value is the comprehensive capital cost of mine blasting;

[0034] The optimization subunit is used to optimize the individual fitness value using the SGD optimizer to obtain the best individual;

[0035] The decision optimization subunit is used to obtain the final optimized network weight encoding based on the best individual, thereby completing the optimization of the intelligent decision.

[0036] A mine intelligent blasting management method, applied to the above system, comprises the following steps:

[0037] Collect and pre-process mine rock mass data and blasting data to establish a blasting database; the mine rock mass data includes mine geological parameters, rock mass characteristics, rock mass distribution, and mine image data; wherein the mine image data is obtained using oblique photography;

[0038] Based on the mine rock mass data, construct a digital three-dimensional rock mass model of the blasting area;

[0039] Based on the digital three-dimensional rock model of the blasting area and the blasting data collected throughout the blasting process, an economic mathematical model is established to make intelligent decisions on mine blasting and obtain decision results;

[0040] Blasting is carried out based on the decision results, and the blasting effect is evaluated to complete the management of intelligent blasting in mines.

[0041] Preferably, the method for constructing a digital three-dimensional rock model of the blasting area is:

[0042] establishing a coordinate matrix of the mine image data, acquiring distortion data in the mine image data based on the coordinate matrix, and correcting the distortion data to obtain the corrected mine image data;

[0043] Performing aerial triangulation connection point matching based on the corrected mine image data to obtain orientation elements of the mine image data;

[0044] Based on the orientation elements, construct an initial three-dimensional real scene model;

[0045] The mine rock mass data is used to perform data mapping on the initial three-dimensional real scene model; and based on the mine image data, three-dimensional coordinates and two-dimensional texture coordinates of the mine are established, and texture mapping is performed to obtain a three-dimensional real scene model;

[0046] Based on the bilinear interpolation method, the unknown points of the three-dimensional real scene model are interpolated to obtain a digital three-dimensional rock model of the blasting area.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The three-dimensional model construction module of the present invention is used to construct a digital three-dimensional rock model of the blasting area based on the mine rock data; the present invention supports the implementation of intelligent blasting by three-dimensional twinning of the blasting area, and can significantly optimize the blasting effect by relying on the three-dimensional rock model. (2) The intelligent decision-making module of the present invention is used to establish an economic mathematical model based on the digital three-dimensional rock model of the blasting area and the blasting data of the entire blasting process collected, make intelligent decisions on mine blasting, and obtain decision results; the economic mathematical model of the present invention can make intelligent overall arrangements for blasting work based on the acquired multi-source data, reasonably allocate resources, reduce mining costs, and improve efficiency. (3) The blasting effect evaluation module of the present invention is used to perform blasting based on the decision results, evaluate the blasting effect, and complete the management of intelligent mine blasting. The present invention continuously optimizes the blasting plan through the evaluation of the blasting effect, and promotes the development of intelligent mine blasting. In summary, the present invention uses a combination of multiple models to perform intelligent data calculation on mines, thereby obtaining the best, reasonable and effective blasting plan, achieving safe supervision of blasting operations, and reducing accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0049] Figure 1 The figure is a structural diagram of a mine intelligent blasting management system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0050] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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 ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0051] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0052] Example 1

[0053] like Figure 1 As shown, a mine intelligent blasting management system includes: a data acquisition module, a three-dimensional model building module, an intelligent decision-making module and a blasting effect evaluation module;

[0054] Data acquisition module, used to collect and pre-process mining rock mass data and blasting data, and establish a blasting database;

[0055] In a further embodiment, the mine rock mass data includes mine geological parameters, rock mass characteristics, rock mass distribution, and mine image data; wherein the mine image data is obtained by oblique photography;

[0056] Blasting data includes data on drilling, blasting, loading, transporting, breaking, and grinding processes;

[0057] The data acquisition module also collects other relevant data; other relevant data include climate conditions in the mine blasting area, personnel data, and various equipment data;

[0058] The data acquisition module includes a data acquisition unit, a data preprocessing unit and a database construction unit;

[0059] Data acquisition unit, used to collect rock mass data, blasting data and other related data;

[0060] The data preprocessing unit is used to clean all collected data and fill in missing values ​​to obtain preprocessed data;

[0061] The database construction unit is used to store the pre-processed data accordingly.

[0062] 3D model building module, used to build a digital 3D rock model of the blasting area based on mine rock data;

[0063] A further embodiment is that the three-dimensional model construction module includes an image data correction unit, a triangle matching unit, an initial model construction unit, a mapping unit, and a final model construction unit;

[0064] An image data correction unit is used to establish a coordinate matrix of the mine image data, obtain distortion data in the mine image data based on the coordinate matrix, and correct the distortion data to obtain corrected mine image data;

[0065] In this embodiment, the process of establishing the coordinate matrix of the mine image data includes:

[0066] Establish the pixel coordinate system (u, v) of the image data, and the origin of the pixel coordinate system uv is O o , the horizontal coordinate u and the vertical coordinate V are the row and column of the image respectively. In the visual processing library OpenCV, u corresponds to x and v corresponds to y.

[0067] The coordinates of O1 in the uv coordinate system, assuming d x and d yRepresents the physical size of each pixel on the horizontal axis x and vertical axis y, in millimeters per pixel. The relationship between the coordinates of O1 in the image coordinate system (x, y) and in the pixel coordinate system (u, v) is:

[0068]

[0069] Assuming the unit in the physical coordinate system is millimeters, then d x The unit is mm / pixel, d x and d y The actual size of the pixels on the photosensitive chip in the device capturing the image connects the pixel coordinate system to the real-size coordinate system. u0 and v0 represent the horizontal and vertical pixel differences between the coordinates of the image's center pixel and the coordinates of the image's dot pixel. After deriving this formula, linear algebra is used to express the coordinate equation of the mine image data in matrix form:

[0070]

[0071] The triangulation matching unit is used to match aerial triangulation tie points based on the corrected mine image data, obtaining orientation elements from the mine image data. To create a 3D real-world model of the mine, mine image data is collected from multiple angles, and aerial triangulation is performed using the resulting multiple, highly overlapping, continuous oblique image models. Based on the number of oblique cameras and the distribution of the camera areas, a large triangulation area is divided into multiple smaller areas, generating a multi-angle, multi-point connected image network structure. This improves mine image coverage and reduces data errors. Triangulation tie point matching is performed based on the image network structure.

[0072] The initial model building unit is used to build an initial three-dimensional real scene model based on the orientation elements.

[0073] The mapping unit is used to perform data mapping on the initial three-dimensional real scene model based on the mine rock mass data; and to establish the mine three-dimensional coordinates and two-dimensional texture coordinates based on the mine image data, perform texture mapping, and obtain the three-dimensional real scene model.

[0074] The final model construction unit is used to interpolate unknown points in the 3D real-world model using bilinear interpolation to obtain a digital 3D rock model of the blasting area. This interpolation process can first be performed using the Bessel formula to perform linear interpolation, and then the Bessel formula is processed using the central difference format to obtain the second-order difference formula.

[0075] The intelligent decision-making module is used to establish an economic mathematical model based on the digital three-dimensional rock model of the blasting area and the blasting data collected from the entire blasting process, make intelligent decisions on mine blasting, and obtain decision results.

[0076] A further implementation method is that the intelligent decision-making module includes a cost calculation unit, a mathematical model construction unit and a decision-making unit.

[0077] The cost calculation unit is used to calculate the mining cost based on the blasting data; wherein the mining cost includes drilling cost, blasting cost, shoveling cost, transportation cost, crushing cost and grinding cost.

[0078] The mathematical model construction unit is used to calculate the weight of each cost in the mining cost based on the entropy weight method, and use the linear regression equation to construct an economic mathematical model; specifically, the application process of the entropy weight method is: obtaining the numerical value of each cost, performing data standardization, and obtaining standardized data.

[0079] Based on the standardized data, calculate the information entropy of each cost;

[0080] Based on the information entropy of each cost, the weight of each cost is determined.

[0081] The decision-making unit is used to obtain the comprehensive capital cost of mine blasting based on the economic mathematical model; based on the comprehensive capital cost, it makes intelligent blasting decisions that take cost into consideration and obtains decision results.

[0082] The blasting effect evaluation module is used to carry out blasting based on decision-making results, evaluate the blasting effect, and complete the management of intelligent blasting in mines.

[0083] A further embodiment is that the blasting effect evaluation module includes a feature extraction unit, an effect evaluation unit, and a decision optimization unit;

[0084] A feature extraction unit is used to extract features of rock fragments between two drill holes after blasting based on image features, obtain blasting fragment size distribution features, and calculate the single-hole blasting range based on the blasting fragment size distribution features;

[0085] In this example, an image of the rock mass after blasting is collected, grayscaled, and denoised. The color gradient of the image is calculated. The interface between the coal and rock blocks is where the color gradient changes significantly. The interface is used to segment the different blocks. The segmented image is binarized to obtain the optimal pixel threshold. Based on the optimal pixel threshold, the rock blocks are identified and the connected regions of the image are obtained. The connected regions are defined as a rock block. The parameters of the connected regions are statistically analyzed (area, center coordinates, eccentricity, circularity, diameter, etc.) to obtain the rock block size x.

[0086] The fractal theory is used to analyze the fragmentation distribution characteristics of rock blocks. The fractal dimension D is proportional to the linear characteristic size x of the fragment and the number N(x) of fragments larger than this size:

[0087] N(x)∝x -D.

[0088] Differentiate the above relationship and combine it with the particle size distribution function to obtain the ratio of the volume of rock blocks with a size smaller than x to the total volume;

[0089] Based on the ratio of the volume of rock blocks smaller than x to the total volume, a fractal model of the size distribution of the blasted rock mass is obtained;

[0090] The natural logarithm of the fractal model is taken to calculate the fractal dimension of the rock fragments, and then the size distribution characteristics of the blasted rock mass are obtained.

[0091] Effect evaluation unit, used to evaluate blasting effects based on blasting fragmentation distribution characteristics and single-hole blasting range;

[0092] The decision optimization unit is used to optimize intelligent decision-making based on the blasting effect evaluation results using a neural network, and to predict the blasting effect based on the optimized intelligent decision.

[0093] A further embodiment is that the decision optimization unit includes an optimization weight acquisition subunit, a fitness calculation subunit, an optimization subunit and a decision optimization subunit;

[0094] The optimization weight acquisition subunit is used to optimize the loss function of the deep neural network using the SGD optimizer and obtain the SGD optimized weights;

[0095] The fitness calculation subunit is used to obtain individual fitness values ​​based on the SGD optimization weight and the encoding method of deep neural network evolution; each individual represents the updated network weight; the individual fitness value is the comprehensive capital cost of mine blasting;

[0096] The optimization subunit is used to optimize the individual fitness value using the SGD optimizer to obtain the best individual;

[0097] The decision optimization subunit is used to obtain the final optimized network weight encoding based on the best individual and complete the optimization of intelligent decision-making.

[0098] Example 2

[0099] A mine intelligent blasting management method, applied to a system, comprises the following steps:

[0100] Collect and pre-process mine rock mass data and blasting data to establish a blasting database; mine rock mass data includes mine geological parameters, rock mass characteristics, rock mass distribution, and mine image data; among them, mine image data is obtained using oblique photography;

[0101] Based on the mine rock mass data, a digital 3D rock mass model of the blasting area is constructed;

[0102] Based on the digital 3D rock mass model of the blasting area and the blasting data collected throughout the entire blasting process, an economic mathematical model is established to make intelligent decisions on mine blasting and obtain decision results;

[0103] Blasting is carried out based on the decision-making results, and the blasting effect is evaluated to complete the management of intelligent blasting in mines.

[0104] A further embodiment is to construct a digital three-dimensional rock model of the blasting area by:

[0105] Establishing a coordinate matrix of the mine image data, obtaining distortion data in the mine image data based on the coordinate matrix, and correcting the distortion data to obtain corrected mine image data;

[0106] Perform aerial triangulation connection point matching based on the corrected mine image data to obtain the orientation elements of the mine image data;

[0107] Based on the orientation elements, construct the initial three-dimensional real scene model;

[0108] The mine rock mass data is used to map the initial 3D real scene model; based on the mine image data, the mine 3D coordinates and 2D texture coordinates are established, and texture mapping is performed to obtain the 3D real scene model;

[0109] Based on the bilinear interpolation method, the unknown points of the 3D real scene model are interpolated to obtain a digital 3D rock model of the blasting area.

[0110] The embodiments described above are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by persons skilled in the art should fall within the scope of protection defined by the claims of the present invention.

Claims

1. A mine intelligent blasting management system, characterized in that: include: Data acquisition module, 3D model building module, intelligent decision-making module and blasting effect evaluation module; The data acquisition module is used to collect and pre-process mine rock mass data and blasting data to establish a blasting database; The three-dimensional model building module is used to build a digital three-dimensional rock model of the blasting area based on the mine rock data; The intelligent decision-making module is used to establish an economic mathematical model based on the digital three-dimensional rock model of the blasting area and the blasting data collected for the entire blasting process, make intelligent decisions on mine blasting, and obtain decision results; The blasting effect evaluation module is used to perform blasting based on the decision result and evaluate the blasting effect to complete the management of intelligent blasting in the mine; The mine rock mass data includes mine geological parameters, rock mass characteristics, rock mass distribution and mine image data; wherein the mine image data is obtained by oblique photography; The blasting data includes data on drilling, blasting, loading, transporting, breaking and grinding processes; The data acquisition module also collects other relevant data; the other relevant data include climate conditions in the mine blasting area, personnel data and various equipment data; The data acquisition module includes a data acquisition unit, a data preprocessing unit and a database construction unit; The data acquisition unit is used to acquire the rock mass data, the blasting data and the other related data; The data preprocessing unit is used to clean all collected data and fill in missing values ​​to obtain preprocessed data; The database construction unit is used to store the pre-processed data accordingly; The three-dimensional model construction module includes an image data correction unit, a triangle matching unit, an initial model construction unit, a mapping unit and a final model construction unit; The image data correction unit is configured to establish a coordinate matrix of the mine image data, obtain distortion data in the mine image data based on the coordinate matrix, and correct the distortion data to obtain the corrected mine image data. The process of establishing the coordinate matrix of the mine image data includes: Establish the pixel coordinate system (u, v) of the image data, and the origin of the pixel coordinate system uv is O o , the horizontal coordinate u and the vertical coordinate V are the row and column of the image respectively. In the visual processing library OpenCV, u corresponds to x and v corresponds to y; The coordinates of O1 in the uv coordinate system, assuming d x and d y Represents the physical size of each pixel on the horizontal axis x and vertical axis y, in millimeters per pixel. The relationship between the coordinates of O1 in the image coordinate system (x, y) and in the pixel coordinate system (u, v) is: Assuming the unit in the physical coordinate system is millimeters, then d x The unit is mm / pixel, d x and d y It represents the actual size of the pixel on the photosensitive chip in the device that collects the image, and it connects the pixel coordinate system with the real-size coordinate system. u0 and v0 represent the horizontal and vertical pixel differences between the coordinates of the center pixel of the image and the coordinates of the pixel of the image point. After deriving this formula, we use the knowledge of linear algebra to express the coordinate equation of the mine image data in matrix form: The triangle matching unit is used to perform aerial triangulation connection point matching based on the corrected mine image data to obtain the orientation elements of the mine image data; The initial model building unit is used to build an initial three-dimensional real scene model based on the orientation elements; The mapping unit is configured to perform data mapping on the initial three-dimensional real scene model based on the mine rock mass data; and to establish three-dimensional mine coordinates and two-dimensional texture coordinates based on the mine image data, perform texture mapping, and obtain a three-dimensional real scene model; The final model construction unit is used to interpolate the unknown points of the three-dimensional real scene model based on the bilinear interpolation method to obtain a digital three-dimensional rock model of the blasting area; The intelligent decision-making module includes a cost calculation unit, a mathematical model construction unit and a decision-making unit; a cost calculation unit, configured to calculate mining costs based on the blasting data; wherein the mining costs include drilling costs, blasting costs, shoveling costs, transportation costs, crushing costs, and grinding costs; A mathematical model building unit, configured to calculate the weight of each cost in the mining cost based on an entropy weight method, and to build the economic mathematical model using a linear regression equation; A decision-making unit, configured to obtain a comprehensive capital cost of mine blasting based on the economic mathematical model; and to make a cost-considered intelligent blasting decision based on the comprehensive capital cost to obtain the decision result; The blasting effect evaluation module includes a feature extraction unit, an effect evaluation unit and a decision optimization unit; The feature extraction unit is used to extract features of rock fragments between two drill holes after blasting based on image features, obtain blasting fragmentation distribution features, and calculate the single-hole blasting range based on the blasting fragmentation distribution features; collect images of rock fragmentation after blasting, perform grayscale and denoising processing, calculate the image color change gradient, and the interface of the coal-rock block is where the color gradient changes greatly, and use the interface to segment different blocks; binarize the segmented image to obtain the optimal pixel threshold; identify the rock block based on the optimal pixel threshold to obtain the image connected area, where the connected area is a rock block, and perform statistics on the connected area parameters, such as area, center coordinates, eccentricity, circularity, and diameter, to obtain the rock block size x; The fractal theory is used to analyze the fragmentation distribution characteristics of rock blocks. The fractal dimension D is proportional to the linear characteristic size x of the fragment and the number N(x) of fragments larger than this size: N(x)∝x -D ; Differentiate the above relationship and combine it with the particle size distribution function to obtain the ratio of the volume of rock blocks with a size smaller than x to the total volume; Based on the ratio of the volume of rock blocks smaller than x to the total volume, a fractal model of the size distribution of the blasted rock mass is obtained; Taking the natural logarithm of the fractal model, the fractal dimension of the rock fragments is calculated, and then the size distribution characteristics of the blasted rock mass are obtained; The effect evaluation unit is used to evaluate the blasting effect based on the blasting fragmentation distribution characteristics and the single-hole blasting range; The decision optimization unit is used to optimize intelligent decision-making using a neural network based on the blasting effect evaluation result, and predict blasting effect based on the optimized intelligent decision; The decision optimization unit includes an optimization weight acquisition subunit, a fitness calculation subunit, an optimization subunit and a decision optimization subunit; The optimization weight acquisition subunit is used to optimize the loss function of the deep neural network using the SGD optimizer and obtain the SGD optimization weight; The fitness calculation subunit is used to obtain individual fitness values ​​based on the SGD optimization weight and the encoding method of deep neural network evolution; wherein each individual represents an updated network weight; and the individual fitness value is the comprehensive capital cost of mine blasting; The optimization subunit is used to optimize the individual fitness value using the SGD optimizer to obtain the best individual; The decision optimization subunit is used to obtain the final optimized network weight encoding based on the best individual, thereby completing the optimization of the intelligent decision.

2. A mine intelligent blasting management method, characterized in that: The system according to claim 1 comprises the following steps: Collect and pre-process mine rock mass data and blasting data to establish a blasting database; the mine rock mass data includes mine geological parameters, rock mass characteristics, rock mass distribution, and mine image data; wherein the mine image data is obtained using oblique photography; Based on the mine rock mass data, construct a digital three-dimensional rock mass model of the blasting area; Based on the digital three-dimensional rock model of the blasting area and the blasting data collected throughout the blasting process, an economic mathematical model is established to make intelligent decisions on mine blasting and obtain decision results; Blasting is carried out based on the decision results, and the blasting effect is evaluated to complete the management of intelligent blasting in mines.

3. The mine intelligent blasting management method according to claim 2, characterized in that: The method for constructing a digital three-dimensional rock model of the blasting area is: establishing a coordinate matrix of the mine image data, acquiring distortion data in the mine image data based on the coordinate matrix, and correcting the distortion data to obtain the corrected mine image data; Performing aerial triangulation connection point matching based on the corrected mine image data to obtain orientation elements of the mine image data; Based on the orientation elements, construct an initial three-dimensional real scene model; The mine rock mass data is used to perform data mapping on the initial three-dimensional real scene model; and based on the mine image data, three-dimensional coordinates and two-dimensional texture coordinates of the mine are established, and texture mapping is performed to obtain a three-dimensional real scene model; Based on the bilinear interpolation method, the unknown points of the three-dimensional real scene model are interpolated to obtain a digital three-dimensional rock model of the blasting area.

Images