A method and system for optimizing the control of the mine blasting process based on big data

Abstract

A method and system for optimizing the control of the mine blasting process based on big data, which relates to the field of mine blasting technology; constructing a three-dimensional twin model of the mine and obtaining a number of blasting points and their effective blasting areas, obtaining the borehole lithology information and the effective blasting area of each blasting point, constructing a charge evaluation model in combination with historical blasting parameters, using the charge evaluation model to obtain the expected charge amount in combination with the first blasting parameter, obtaining the gravel throwing duration of each effective blasting area, constructing a primer evaluation model in combination with the borehole lithology information and historical blasting parameters, using the primer evaluation model to obtain the expected primer time difference in combination with the second blasting parameter, generating recommended blasting parameters according to the expected charge amount and the expected primer time difference and giving feedback; capable of outputting the corresponding expected charge amount according to the expected blasting area, and improving the charging efficiency of mine blasting operations and the overall blasting effect.

Description

Technical Field

[0001] The present invention relates to the technical field of mine blasting, and specifically to an optimization method and system for mine blasting process control based on big data. Background Art

[0002] Controlling and optimizing the mine blasting process is an integrated solution that combines advanced technologies and intelligent algorithms, aiming to improve the efficiency and blasting effect of mine blasting. By using technologies such as big data and machine learning, it is possible to control and optimize each link in the mine blasting process separately, and generate an optimal blasting plan to guide on-site operations, achieving the maximization of the blasting effect;

[0003] The prior art lacks an effective control method for the initiation time difference between adjacent drill holes, and is unable to provide corresponding control methods based on the possible rock fragmentation throwing conditions of each drill hole. If the initiation time difference is too short, the blasting effect of the next drill hole will inevitably be weakened. If the initiation time difference is too long, the overall blasting efficiency will be significantly reduced;

[0004] For mine blasting operations, the charge amount will directly affect the blasting effect. The prior art lacks an effective control method for the charge amount and is unable to provide the corresponding charge amount according to the expected blasting area. In view of the deficiencies of the prior art, the present invention provides an optimization method and system for mine blasting process control based on big data. Summary of the Invention

[0005] The purpose of the present invention is to provide an optimization method and system for mine blasting process control based on big data.

[0006] The purpose of the present invention can be achieved through the following technical solutions: An optimization system for mine blasting process control based on big data, including the following modules:

[0007] A data acquisition module, used to obtain the historical blasting data of the mine, and construct a corresponding three-dimensional twin model in combination with the geological lithology information, and obtain a number of blasting points and their corresponding effective blasting areas in the three-dimensional twin model;

[0008] A first evaluation module, used to obtain the drill hole lithology information and effective blasting area of each blasting point, and construct a corresponding charge evaluation model in combination with its historical blasting parameters, obtain the expected blasting area of the mine blasting operation, and use the charge evaluation model to obtain the corresponding expected charge amount in combination with the first blasting parameter;

[0009] A second evaluation module, used to obtain the rock fragmentation throwing duration in each effective blasting area, and construct a corresponding initiation evaluation model in combination with its drill hole lithology information and historical blasting parameters, and use the initiation evaluation model to obtain the corresponding expected initiation time difference in combination with the second blasting parameter;

[0010] A scheme generation module, configured to generate corresponding recommended blasting parameters according to the expected charge amount and the expected initiation time difference and give feedback.

[0011] Further, the process of obtaining the historical blasting data of the mine and constructing a corresponding three-dimensional twin model in combination with the geological lithology information includes:

[0012] The historical blasting data includes drilling parameters, charging parameters, initiation parameters, blasting range, and blasting degree, and the geological lithology information includes rock type, geological age, rock composition, rock structure, weathering degree, and hardness;

[0013] Using three-dimensional geological modeling technology to construct a corresponding three-dimensional geological model according to the geological lithology information of a single historical blasting data, using simulation software to simulate the mine blasting process in the three-dimensional geological model, and continuously adjusting the simulation parameters in the three-dimensional geological model until the simulation result is consistent with the corresponding historical blasting data to obtain the three-dimensional twin model.

[0014] Further, the process of obtaining several blasting points and their corresponding effective blasting areas in the three-dimensional twin model includes:

[0015] Obtaining the drilling layout map under a single historical blasting data in the three-dimensional twin model, and randomly selecting the distribution positions of several drill holes in the drilling layout map as the blasting points;

[0016] Obtaining the rock fragmentation distribution under the single historical blasting data in the three-dimensional twin model, including the large block rate and the small block rate, and taking a single blasting point as the center, and successively making circles outward with preset fixed distances R, 2R, 3R,..., nR as the radii;

[0017] Obtaining the local large block rate of the local area corresponding to each preset fixed distance, comparing the local large block rates between the adjacent local areas of the single blasting point, and when the local large block rate of the local area corresponding to nR is greater than the local large block rate of the local area corresponding to (n - 1)R, taking the circular area with a radius of (n - 1)R centered on the single blasting point as the effective blasting area of the single blasting point.

[0018] Further, the process of obtaining the drilling lithology information and the effective blasting area of each blasting point and constructing a corresponding charge evaluation model in combination with its historical blasting parameters includes:

[0019] Obtaining the drilling lithology information of the single blasting point according to the core dug during drilling of the single blasting point, including the density, porosity, joint fracture rate, compressive strength, and tensile strength of the core;

[0020] Take the area of the effective blasting area corresponding to the single blasting point as the effective blasting area, and take the hole depth, hole diameter, charge amount, charge structure, charge density and length, and initiation method of the single blasting point in the historical blasting data as the historical blasting parameters;

[0021] Generate a first evaluation set according to the borehole lithology information, historical blasting parameters and their corresponding effective blasting areas of different blasting points, and divide the first evaluation set into a first training set and a first test set;

[0022] Construct a first convolutional neural network, take the different borehole lithology information and historical blasting parameters in the first training set as the input data of the first convolutional neural network, and take the corresponding effective blasting area in the first training set as the output data of the first convolutional neural network;

[0023] Train the first convolutional neural network to obtain an initial first convolutional neural network, use the first test set to verify the model of the initial first convolutional neural network, and output the initial first convolutional neural network with a first test error threshold less than or equal to the preset value as the charge evaluation model.

[0024] Furthermore, the process of obtaining the expected blasting area of the mine blasting operation and using the charge evaluation model to obtain the corresponding expected charge amount in combination with the first blasting parameters includes:

[0025] In subsequent mine blasting operations, take the effective blasting area that a single drill hole is expected to achieve as the expected blasting area, obtain the borehole lithology information of the single drill hole, and set the first blasting parameters, including the preset hole depth, hole diameter, charge structure, charge density and length, and the adjustable charge amount;

[0026] Input the borehole lithology information and the first blasting parameters of the single drill hole into the charge evaluation model, use the charge evaluation model to output the corresponding effective blasting area, continuously increase the charge amount in the first blasting parameters, and when the effective blasting area output by the charge evaluation model is equal to the expected blasting area, take the charge amount at this time as the expected charge amount corresponding to the expected blasting area.

[0027] Furthermore, the process of obtaining the gravel throwing duration in each effective blasting area and constructing the corresponding initiation evaluation model in combination with its borehole lithology information and historical blasting parameters includes:

[0028] In the 3D twin model, take the moment when the single blasting point starts blasting as the blasting start moment, monitor the simulated large block rate of its effective blasting area in real time during the blasting process of the single blasting point, and obtain the historical large block rate of its effective blasting area in the historical blasting data;

[0029] Take the ratio between the simulated large block rate and the historical large block rate as the gravel throwing rate of its effective blasting area, set a threshold for the gravel throwing rate, take the moment when the gravel throwing rate is equal to the gravel throwing rate threshold as the throwing completion moment, and take the time interval between the blasting start moment and the throwing completion moment as the gravel throwing duration;

[0030] Generate a second evaluation set based on the borehole lithology information, historical blasting parameters, and their corresponding gravel throwing durations of different effective blasting areas, and divide the second evaluation set into a second training set and a second test set;

[0031] Construct a second convolutional neural network, take the different borehole lithology information and historical blasting parameters in the second training set as the input data of the second convolutional neural network, and take the corresponding gravel throwing duration in the second training set as the output data of the second convolutional neural network;

[0032] Train the second convolutional neural network to obtain an initial second convolutional neural network, use the second test set to verify the model of the initial second convolutional neural network, and output the initial second convolutional neural network with an error less than or equal to the preset second test error threshold as the detonation evaluation model.

[0033] Further, the process of using the detonation evaluation model to combine with the second blasting parameters to obtain the corresponding expected detonation time difference includes:

[0034] In subsequent mine blasting operations, obtain the borehole lithology information of a single borehole and its expected charge amount under the expected blasting area, and set the second blasting parameters, including the preset hole depth, hole diameter, charge structure, charge density and length, and the corresponding expected charge amount;

[0035] Input the borehole lithology information and the second blasting parameters of the single borehole into the detonation evaluation model, use the detonation evaluation model to output the corresponding gravel throwing duration, and take the obtained gravel throwing duration as the expected detonation time difference between the single borehole and the next adjacent borehole in the detonation sequence.

[0036] Further, the process of generating the corresponding recommended blasting parameters based on the expected charge amount and the expected detonation time difference and giving feedback includes:

[0037] In subsequent mine blasting operations, reverse infer the hole spacing and row spacing of adjacent boreholes according to the expected blasting area of each borehole, use the charge evaluation model to obtain the expected charge amount of each borehole, use the detonation evaluation model to obtain the expected detonation time difference between two adjacent boreholes in the detonation sequence, generate the recommended blasting parameters for the mine blasting operation and feedback them to the relevant personnel;

[0038] The recommended blasting parameters include preset hole depth, hole diameter, charging structure, charging density and length, initiation method, initiation sequence, as well as the expected charge amount for each drill hole and the hole spacing, row spacing, and expected initiation time difference between adjacent drill holes.

[0039] An optimization method for controlling the mine blasting process based on big data, comprising the following steps:

[0040] Step S1: Obtain the historical blasting data of the mine, and construct a corresponding three-dimensional twin model in combination with geological lithology information, and obtain a number of blasting points and their corresponding effective blasting areas in the three-dimensional twin model;

[0041] Step S2: Obtain the borehole lithology information and effective blasting area of each blasting point, and construct a corresponding charge evaluation model in combination with its historical blasting parameters, obtain the expected blasting area of the mine blasting operation, and use the charge evaluation model to obtain the corresponding expected charge amount in combination with the first blasting parameter;

[0042] Step S3: Obtain the gravel throwing duration in each effective blasting area, and construct a corresponding initiation evaluation model in combination with its borehole lithology information and historical blasting parameters, and use the initiation evaluation model to obtain the corresponding expected initiation time difference in combination with the second blasting parameter;

[0043] Step S4: Generate corresponding recommended blasting parameters according to the expected charge amount and the expected initiation time difference and give feedback.

[0044] Compared with the prior art, the beneficial effects of the present invention are:

[0045] By constructing a corresponding three-dimensional twin model according to the historical blasting data, the present invention can obtain different blasting points and effective blasting areas therein, construct a charge evaluation model according to the corresponding relationship between the borehole lithology information, historical blasting parameters and effective blasting area of different blasting points, and can directly output the corresponding expected charge amount according to the expected effective blasting area;

[0046] By obtaining the gravel throwing duration of each effective blasting area in the three-dimensional twin model, constructing an initiation evaluation model according to the corresponding relationship between the borehole lithology information, historical blasting parameters and gravel throwing duration of different effective blasting areas, the corresponding gravel throwing duration can be obtained in combination with the expected charge amount and used as the expected initiation time difference between adjacent drill holes. Using the above two models to obtain the expected charge amount of each drill hole and the expected initiation time difference of different drill holes is beneficial to greatly improve the charging efficiency of the mine blasting operation and the overall blasting effect. Description of the Drawings

[0047] Figure 1 It is the schematic diagram of the present invention. Detailed Embodiment

[0048] As shown Figure 1 in the figure, an optimization system for controlling the mine blasting process based on big data includes the following modules:

[0049] The data acquisition module is used to obtain the historical blasting data of the mine, construct a corresponding three-dimensional twin model in combination with the geological lithology information, and obtain several blasting points and their corresponding effective blasting areas in the three-dimensional twin model;

[0050] The first evaluation module is used to obtain the borehole lithology information and effective blasting area of each blasting point, construct a corresponding charge evaluation model in combination with its historical blasting parameters, obtain the expected blasting area of the mine blasting operation, and obtain the corresponding expected charge amount by combining the charge evaluation model with the first blasting parameter;

[0051] The second evaluation module is used to obtain the gravel throwing duration in each effective blasting area, construct a corresponding initiation evaluation model in combination with its borehole lithology information and historical blasting parameters, and obtain the corresponding expected initiation time difference by combining the initiation evaluation model with the second blasting parameter;

[0052] The scheme generation module is used to generate corresponding recommended blasting parameters according to the expected charge amount and expected initiation time difference and give feedback.

[0053] It should be further noted that in the specific implementation process, the process of obtaining the historical blasting data of the mine and constructing a corresponding three-dimensional twin model in combination with the geological lithology information includes:

[0054] The historical blasting data includes drilling parameters, charging parameters, initiation parameters, blasting range, and blasting degree. The geological lithology information refers to the rock characteristics and properties corresponding to each historical blasting data, including rock type, geological age, rock composition, rock structure, weathering degree, and hardness;

[0055] The drilling parameters include hole depth, hole diameter, hole spacing, and row spacing. The charging parameters include charge amount, charging structure, charge density, and length. The initiation parameters include initiation method, initiation sequence, and initiation time difference. The blasting range refers to the broken range and area of the rock after blasting. The blasting degree refers to the rock fragment size distribution after blasting, including the large block rate and small block rate;

[0056] Using three-dimensional geological modeling technology, construct a three-dimensional geological model of the corresponding mine according to the geological lithology information of a single historical blasting data, including descriptions of the morphology, structure, and properties of the geological body. Use simulation software to simulate the mine blasting process in this three-dimensional geological model, including the explosion process of explosives, the rock fragmentation process, and the throwing and accumulation process after blasting;

[0057] In the three-dimensional geological model, continuously adjust its simulation parameters until the simulation results match the corresponding historical blasting data. At this time, the three-dimensional geological model is used as the three-dimensional twin model under this historical blasting data, and the three-dimensional twin models of different historical blasting data are obtained respectively.

[0058] It should be further noted that in the specific implementation process, the process of obtaining several blasting points and their corresponding effective blasting areas in the three-dimensional twin model includes:

[0059] Obtain the drilling layout map under a single historical blasting data in the three-dimensional twin model. The drilling layout map refers to the distribution positions of each drilling deployed during mine blasting operations. Randomly select the distribution positions of several drillings in the drilling layout map as blasting points, and obtain their corresponding effective blasting areas with each blasting point as the center;

[0060] Obtain the rock fragment size distribution under this historical blasting data in the three-dimensional twin model. The rock fragment size distribution refers to the distribution of rock blocks of different sizes formed during the process of rock fragmentation and accumulation, including the large block rate and the small block rate. The crushed stones formed by a single drilling after blasting during mine blasting operations will be thrown and accumulated around;

[0061] Among them, the larger-volume crushed stones mostly fall nearby, and the smaller-volume crushed stones mostly fall farther away. Therefore, the effective blasting area can be determined by analyzing the change of the large block rate around a single drilling. Taking any blasting point as an example, with this blasting point as the center of the circle, draw circles outward in turn with preset fixed distances R, 2R, 3R,..., nR as the radii, where n is a natural number greater than 1;

[0062] Obtain the local large block rates of the local areas corresponding to each preset fixed distance. Among them, the local area corresponding to R is circular, and the local area corresponding to nR is an annular ring. The local large block rates of different local areas decrease as the preset fixed distance increases. When the local large block rate does not decrease but increases, it means that the local area contains large crushed stones formed by other drillings;

[0063] Continuously compare the local large block rates between adjacent local areas of this blasting point. When the local large block rate of the local area corresponding to nR is greater than the local large block rate of the local area corresponding to (n - 1)R, take the circular area with a radius of (n - 1)R centered on this blasting point as the effective blasting area of this blasting point, and obtain the effective blasting areas of each blasting point under different historical blasting data respectively.

[0064] It should be further noted that in the specific implementation process, the process of obtaining the drilling lithology information and the effective blasting area of each blasting point, and constructing the corresponding charge evaluation model in combination with its historical blasting parameters includes:

[0065] Taking any blasting point as an example, the borehole lithology information of the blasting point is obtained according to the core excavated during drilling at the blasting point. The borehole lithology information includes the density, porosity, joint fracture rate, compressive strength, and tensile strength of the core;

[0066] Taking the area of the effective blasting area corresponding to the blasting point as its effective blasting area, and taking the hole depth, hole diameter, charge amount, charge structure, charge density and length, and initiation method of the blasting point in the historical blasting data as its historical blasting parameters;

[0067] Generating a first evaluation set according to the borehole lithology information, historical blasting parameters, and corresponding effective blasting areas of different blasting points, and dividing the first evaluation set into a first training set and a first test set;

[0068] Constructing a first convolutional neural network, taking different borehole lithology information and historical blasting parameters in the first training set as the input data of the first convolutional neural network, and taking the corresponding effective blasting area in the first training set as the output data of the first convolutional neural network;

[0069] Training the first convolutional neural network to obtain an initial first convolutional neural network, and using the first test set to verify the model of the initial first convolutional neural network, and outputting the initial first convolutional neural network with a first test error threshold less than or equal to the preset value as the charge evaluation model.

[0070] It should be further noted that in the specific implementation process, the process of obtaining the expected blasting area of the mine blasting operation and using the charge evaluation model to obtain the corresponding expected charge amount in combination with the first blasting parameters includes:

[0071] In subsequent mine blasting operations, taking the effective blasting area that relevant personnel expect a single borehole to achieve as its expected blasting area, obtaining the borehole lithology information according to the core of the single borehole, and setting the first blasting parameters. The first blasting parameters include the preset hole depth, hole diameter, charge structure, charge density and length, and the adjustable charge amount;

[0072] Inputting the obtained borehole lithology information and the first blasting parameters into the charge evaluation model, using the charge evaluation model to output the corresponding effective blasting area, continuously increasing the charge amount in the first blasting parameters, and when the effective blasting area output by the charge evaluation model is equal to the expected blasting area, taking the charge amount at this time as the expected charge amount corresponding to the expected blasting area.

[0073] It should be further noted that in the specific implementation process, the process of obtaining the gravel throwing duration in each effective blasting area and constructing the corresponding initiation evaluation model in combination with its borehole lithology information and historical blasting parameters includes:

[0074] In the three-dimensional twin model, the corresponding moment when a single blasting point starts blasting is taken as the blasting start moment. The large block rate of the effective blasting area of this single blasting point during the blasting process is monitored in real time and marked as the simulated large block rate. The final large block rate of the effective blasting area in the historical blasting data is obtained and marked as the historical large block rate;

[0075] The ratio between the simulated large block rate and the historical large block rate is taken as the gravel throwing rate of the effective blasting area. A gravel throwing rate threshold is set, and the corresponding moment when the gravel throwing rate is equal to the gravel throwing rate threshold is taken as the throwing completion moment. The time interval between the blasting start moment and the throwing completion moment is taken as the corresponding gravel throwing duration;

[0076] A second evaluation set is generated according to the borehole lithology information, historical blasting parameters and their corresponding gravel throwing durations of different effective blasting areas, and the second evaluation set is divided into a second training set and a second test set;

[0077] A second convolutional neural network is constructed. The different borehole lithology information and historical blasting parameters in the second training set are used as the input data of the second convolutional neural network, and the corresponding gravel throwing duration in the second training set is used as the output data of the second convolutional neural network;

[0078] The second convolutional neural network is trained to obtain an initial second convolutional neural network. The initial second convolutional neural network is verified using the second test set, and the initial second convolutional neural network with an output less than or equal to the preset second test error threshold is used as the detonation evaluation model.

[0079] It should be further noted that in the specific implementation process, the process of using the detonation evaluation model to obtain the corresponding expected detonation time difference in combination with the second blasting parameters includes:

[0080] In subsequent mine blasting operations, the borehole lithology information of a single borehole and its expected charge amount under the expected blasting area are obtained, and second blasting parameters are set. The second blasting parameters include the preset hole depth, hole diameter, charge structure, charge density and length, as well as the corresponding expected charge amount;

[0081] The obtained borehole lithology information and second blasting parameters are input into the detonation evaluation model. The detonation evaluation model is used to output the corresponding gravel throwing duration, and the obtained gravel throwing duration is used as the expected detonation time difference between this single borehole and the next adjacent borehole in the detonation sequence.

[0082] It should be further noted that in the specific implementation process, the process of generating corresponding recommended blasting parameters based on the expected charge amount and the expected detonation time difference and giving feedback includes:

[0083] In subsequent mine blasting operations, based on the expected blasting area of each drill hole, the hole spacing and row spacing between adjacent drill holes are inversely calculated. The hole spacing and row spacing between adjacent drill holes must meet two conditions: the expected blasting area of all drill holes can cover the entire operation area of the mine blasting operation and the total number of drill holes is the least;

[0084] Use the charge evaluation model to obtain the expected charge amount of each drill hole, and use the initiation evaluation model to obtain the expected initiation time difference between two adjacent drill holes in the initiation sequence. Based on this, generate the recommended blasting parameters for this mine blasting operation and feedback them to relevant personnel;

[0085] The recommended blasting parameters include the preset hole depth, hole diameter, charge structure, charge density and length, initiation method, initiation sequence, as well as the expected charge amount of each drill hole and the hole spacing, row spacing, and expected initiation time difference between adjacent drill holes.

[0086] An embodiment of the present invention further includes an optimization method for mine blasting process control based on big data, including the following steps:

[0087] Step S1: Obtain the historical blasting data of the mine, and construct a corresponding three-dimensional twin model in combination with geological lithology information. Obtain several blasting points and their corresponding effective blasting areas in the three-dimensional twin model;

[0088] Step S2: Obtain the drill hole lithology information and effective blasting area of each blasting point, and construct a corresponding charge evaluation model in combination with its historical blasting parameters. Obtain the expected blasting area of the mine blasting operation, and use the charge evaluation model to obtain the corresponding expected charge amount in combination with the first blasting parameter;

[0089] Step S3: Obtain the gravel throwing duration in each effective blasting area, and construct a corresponding initiation evaluation model in combination with its drill hole lithology information and historical blasting parameters. Use the initiation evaluation model to obtain the corresponding expected initiation time difference in combination with the second blasting parameter;

[0090] Step S4: Generate corresponding recommended blasting parameters based on the expected charge amount and expected initiation time difference and give feedback.

[0091] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An optimization system for controlling the process of mine blasting based on big data, characterized in that, It includes the following modules: A data acquisition module, which is used to obtain the historical blasting data of the mine, construct a corresponding three-dimensional twin model in combination with the geological lithology information, and obtain several blasting points and their corresponding effective blasting areas in the three-dimensional twin model; A first evaluation module, which is used to obtain the borehole lithology information and effective blasting area of each blasting point, construct a corresponding charge evaluation model in combination with its historical blasting parameters, obtain the expected blasting area of the mine blasting operation, and use the charge evaluation model to obtain the corresponding expected charge amount in combination with the first blasting parameter; A second evaluation module, which is used to obtain the gravel throwing duration in each effective blasting area, construct a corresponding initiation evaluation model in combination with its borehole lithology information and historical blasting parameters, and use the initiation evaluation model to obtain the corresponding expected initiation time difference in combination with the second blasting parameter; A scheme generation module, which is used to generate corresponding recommended blasting parameters according to the expected charge amount and expected initiation time difference and give feedback; Among them, the process of constructing the corresponding initiation evaluation model includes: Taking the moment when a single blasting point starts blasting in the three-dimensional twin model as the blasting start moment, real-time monitoring the simulated large block rate of the effective blasting area of the single blasting point during the blasting process, and obtaining the historical large block rate of its effective blasting area in the historical blasting data; Taking the ratio between the simulated large block rate and the historical large block rate as the gravel throwing rate of its effective blasting area, setting a gravel throwing rate threshold, taking the moment when the gravel throwing rate is equal to the gravel throwing rate threshold as the throwing completion moment, and taking the time interval between the blasting start moment and the throwing completion moment as the gravel throwing duration; Generating a second evaluation set according to the borehole lithology information, historical blasting parameters and their corresponding gravel throwing durations of different effective blasting areas, and dividing the second evaluation set into a second training set and a second test set; Constructing a second convolutional neural network, taking the different borehole lithology information and historical blasting parameters in the second training set as the input data of the second convolutional neural network, and taking the corresponding gravel throwing duration in the second training set as the output data of the second convolutional neural network; Training the second convolutional neural network to obtain an initial second convolutional neural network, using the second test set to verify the model of the initial second convolutional neural network, and outputting the initial second convolutional neural network with an error less than or equal to the preset second test error threshold as the initiation evaluation model.

2. The optimization system for controlling the mine blasting process based on big data according to claim 1, characterized in that, The process of constructing the corresponding three-dimensional twin model includes: The historical blasting data includes drilling parameters, charging parameters, initiation parameters, blasting range, and blasting degree, and the geological lithology information includes rock type, geological age, rock composition, rock structure, weathering degree, and hardness; Using three-dimensional geological modeling technology to construct a corresponding three-dimensional geological model according to the geological lithology information of a single historical blasting data, using simulation software to simulate the mine blasting process in the three-dimensional geological model, and continuously adjusting the simulation parameters in the three-dimensional geological model until the simulation result is consistent with the corresponding historical blasting data to obtain the three-dimensional twin model.

3. The optimization system for controlling the mine blasting process based on big data according to claim 2, wherein, The process of obtaining several blasting points and their corresponding effective blasting areas includes: Obtain the drilling layout map under a single historical blasting data in the 3D twin model, and randomly select the distribution positions of several drill holes in the drilling layout map as blasting points; Obtain the rock fragmentation distribution under the single historical blasting data in the 3D twin model, including the large block ratio and small block ratio. Taking a single blasting point as the center, draw circles outward in turn with preset fixed distances R, 2R, 3R, ……, nR as the radii; Obtain the local large block ratio of the local area corresponding to each preset fixed distance, compare the local large block ratios between adjacent local areas of the single blasting point. When the local large block ratio of the local area corresponding to nR is greater than that of the local area corresponding to (n - 1)R, take the circular area with a radius of (n - 1)R centered on the single blasting point as the effective blasting area of the single blasting point.

4. The optimization system for controlling the mine blasting process based on big data according to claim 3, characterized in that, The process of constructing the corresponding charge evaluation model includes: Obtain the drilling lithology information of the single blasting point according to the core dug during drilling of the single blasting point, including the density, porosity, joint fracture rate, compressive strength, and tensile strength of the core; Take the area of the effective blasting area corresponding to the single blasting point as the effective blasting area, and take the hole depth, hole diameter, charge amount, charge structure, charge density and length, and initiation method of the single blasting point in the historical blasting data as historical blasting parameters; Generate a first evaluation set according to the drilling lithology information, historical blasting parameters and their corresponding effective blasting areas of different blasting points, and divide the first evaluation set into a first training set and a first test set; Construct a first convolutional neural network, take the different drilling lithology information and historical blasting parameters in the first training set as the input data of the first convolutional neural network, and take the corresponding effective blasting area in the first training set as the output data of the first convolutional neural network; Train the first convolutional neural network to obtain an initial first convolutional neural network, use the first test set to verify the model of the initial first convolutional neural network, and output the initial first convolutional neural network with an error less than or equal to the preset first test error threshold as the charge evaluation model.

5. The optimization system for controlling the mine blasting process based on big data according to claim 4, wherein The process of obtaining the expected blasting area and its corresponding expected charge amount includes: In subsequent mine blasting operations, take the effective blasting area that a single drill hole is expected to achieve as the expected blasting area, obtain the drilling lithology information of the single drill hole, and set the first blasting parameters, including the preset hole depth, hole diameter, charge structure, charge density and length, and the adjustable charge amount; Input the drilling lithology information and the first blasting parameters of the single drill hole into the charge evaluation model, use the charge evaluation model to output the corresponding effective blasting area, continuously increase the charge amount in the first blasting parameters. When the effective blasting area output by the charge evaluation model is equal to the expected blasting area, take the charge amount at this time as the expected charge amount corresponding to the expected blasting area.

6. The optimization system for controlling the mine blasting process based on big data according to claim 5, wherein, The process of obtaining the expected initiation time difference includes: In subsequent mine blasting operations, obtain the drilling lithology information of a single drill hole and its expected charge amount under the expected blasting area, and set the second blasting parameters, including the preset hole depth, hole diameter, charge structure, charge density and length, and the corresponding expected charge amount; Input the drilling lithology information and the second blasting parameter of the single drill hole into the detonation evaluation model, and use the detonation evaluation model to output the corresponding gravel throwing duration. Use the obtained gravel throwing duration as the expected detonation time difference between the single drill hole and the next adjacent drill hole in the detonation sequence.

7. An optimization system for controlling the process of mine blasting based on big data according to claim 6, characterized in that, The process of generating recommended blasting parameters and providing feedback includes: In subsequent mine blasting operations, based on the expected blasting area of each drill hole, reverse calculate the hole spacing and row spacing of adjacent drill holes. Use the charge evaluation model to obtain the expected charge amount of each drill hole, and use the detonation evaluation model to obtain the expected detonation time difference between two adjacent drill holes in the detonation sequence. Generate the recommended blasting parameters for the mine blasting operation and feedback them to relevant personnel; The recommended blasting parameters include the preset hole depth, hole diameter, charge structure, charge density and length, detonation method, detonation sequence, as well as the expected charge amount of each drill hole and the hole spacing, row spacing, and expected detonation time difference of adjacent drill holes.

8. An optimization method for controlling the mine blasting process based on big data, which is implemented based on the mine blasting process control optimization system described in any one of claims 1-7, characterized in that, The method includes: Step S1: Obtain the historical blasting data of the mine, and construct a corresponding three-dimensional twin model in combination with the geological lithology information. Obtain several blasting points and their corresponding effective blasting areas in the three-dimensional twin model; Step S2: Obtain the drilling lithology information and the effective blasting area of each blasting point, and construct a corresponding charge evaluation model in combination with its historical blasting parameters. Obtain the expected blasting area of the mine blasting operation, and use the charge evaluation model to obtain the corresponding expected charge amount in combination with the first blasting parameter; Step S3: Obtain the gravel throwing duration within each effective blasting area, and construct a corresponding detonation evaluation model in combination with its drilling lithology information and historical blasting parameters. Use the detonation evaluation model to obtain the corresponding expected detonation time difference in combination with the second blasting parameter; Step S4: Generate corresponding recommended blasting parameters based on the expected charge amount and the expected detonation time difference and provide feedback.

Images