Surface mine blasting hole layout method and system

Abstract

The invention provides a surface mine blasting hole layout method and system. The method is applied to the surface mine blasting hole layout system. The system comprises a probe vehicle provided with a GR, an SWL, a GPS receiver, an LS and an ES. The method comprises the following steps: detecting geologic structure information under a strip mine mountain through GR; sound waves are emitted to rocks under the strip mine mountain through the SWL, and reflected waves of the sound waves are captured; determining mechanical parameters of the rock based on the reflected waves; acquiring position information of the probe vehicle through a GPS receiver; scanning and surveying landforms around the surface mine through LS to obtain fluctuation and slope data of the landforms; monitoring meteorological information around the surface mine through the ES; determining geological target blasting parameters of the surface mine based on the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the fluctuation and gradient data of the landform and the meteorological information around the surface mine; the target blasting parameters are used for carrying out blasting hole layout on the surface mine.

Description

Technical Field

[0001] The present invention relates to the technical field of open-pit mining, and in particular to an open-pit mine blasting hole layout method and system. Background Art

[0002] Blasting is a critical initial step in open-pit mining, and the quality of blasthole layout directly impacts the success of the entire mining process. Traditionally, blasthole layout planning relies primarily on manual labor. Engineers rely on years of accumulated field experience, using simple tools such as tape measures, levels, and compasses, combined with visual observations of the mine's topography and geology, to determine parameters such as blasthole location, depth, and angle.

[0003] This traditional approach has several significant drawbacks: 1. Manual methods can only capture superficial information about the rock surface, making it difficult to accurately detect geological structures such as faults, karst caves, and weak interlayers deep underground. Furthermore, they cannot precisely quantify key mechanical properties of the rock, such as hardness, elastic modulus, and Poisson's ratio. This results in significant uncertainty in blasting results, resulting in excessively large ore fragments, skyrocketing secondary crushing costs, or excessive ore crushing, leading to resource waste. 2. Manual planning, lacking real-time monitoring and rapid response mechanisms, cannot adapt blasthole layouts to newly revealed geological conditions in a timely and accurate manner. This not only significantly reduces mining efficiency with increasing depth, but also creates additional safety hazards, such as increased risk of slope instability, due to poor blasting results. 3. Traditional manual layout planning often focuses on the current blasting task, viewing each blasthole placement in isolation. This fails to consider the overall mining situation and comprehensively balance complex factors such as explosives costs, mining efficiency, resource recovery, ecological and environmental protection, and safe production. This short-sighted behavior ultimately resulted in high mining costs, low resource utilization efficiency, and unnecessary damage to the mining ecological environment. Summary of the Invention

[0004] The present invention provides a method and system for arranging blastholes for open-pit mines, which are used to solve technical problems such as high open-pit mine blasting mining costs, low resource utilization efficiency, and unnecessary damage to the mine ecological environment when arranging blastholes for open-pit mines.

[0005] An embodiment of the present invention provides a method for blasthole layout in an open-pit mine, which is applied to an open-pit mine blasthole layout system. The system includes a probe vehicle equipped with a geological radar (GR), a sonic well logging instrument (SWL), a high-precision GPS receiver, a laser scanner (LS), and an environmental sensor (ES). The method includes: Detecting geological structure information under the open-pit mine by using the GR; transmitting sound waves to rocks beneath the open-pit mine through the SWL and capturing reflected waves of the sound waves; determining mechanical parameters of the rocks based on the reflected waves; Obtaining the location information of the probe car through the GPS receiver; Scanning and mapping the topography around the open-pit mine using the LS to obtain data on the undulation and slope of the topography; monitoring meteorological information around the open-pit mine through the ES; The target blasting parameters of the open-pit mine geology are determined based on the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine; the target blasting parameters are used to arrange blasting holes for the open-pit mine.

[0006] In some embodiments, the geological structure information includes rock layer information, cavity information, and fault information; and detecting the geological structure information under the open-pit mine by the GR includes: The GR uses ultra-wideband pulse technology to detect the rock layer information, the cavity information, and the fault information under the open-pit mine.

[0007] In some embodiments, the mechanical parameters of the rock include a hardness parameter and an elastic modulus parameter of the rock; and determining the mechanical parameters of the rock based on the reflected wave includes: determining the longitudinal wave velocity and the shear wave velocity of the rock based on the reflected wave; The hardness parameter and elastic modulus parameter of the rock are determined according to the longitudinal wave velocity and the shear wave velocity.

[0008] In some embodiments, the location information includes three-dimensional location coordinate information; and obtaining the location information of the probe vehicle through the GPS receiver includes: The three-dimensional position coordinate information of the probe vehicle is obtained by using the carrier phase difference technology through the GPS receiver.

[0009] In some embodiments, the meteorological information includes air temperature information, humidity information, and air pressure information; and monitoring the meteorological information around the open-pit mine by the ES includes: The temperature information, the humidity information, and the air pressure information around the open-pit mine are monitored by the ES.

[0010] In some embodiments, the transmission method determines target blasting parameters of the open-pit mine geology based on the geological structure information, the mechanical parameters of the rock, the position information of the exploration vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine, including: A preset graph neural network (GNN) is used to deeply fuse the geological structure information, the mechanical parameters of the rock, the location information of the exploration vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain geological image data of the open-pit mine; Obtaining historical mining data of the open-pit mine; The target blasting parameters are determined based on the geological image data and the historical mining data.

[0011] In some embodiments, the geological image data includes spatial features; the historical mining data includes temporal features; and determining the target blasting parameters based on the geological image data and the historical mining data includes: The preset spatiotemporal coupled deep reinforcement learning STC-DRL is used to learn the spatial features and the temporal features to obtain the target blasting parameters.

[0012] In some embodiments, the method further comprises: Preprocessing the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain preprocessed geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine; Determining the contribution parameters of the pre-processed geological structure information, rock mechanical parameters, rover position information, topographic undulations and slope data, and meteorological information around the open-pit mine based on a preset multi-head attention mechanism; Modifying the GNN based on the contribution parameter to obtain a modified GNN; According to the modified GNN, the pre-processed geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine are deeply integrated to obtain the geological image data of the open-pit mine.

[0013] In some embodiments, the adopting of a preset spatiotemporal coupled deep reinforcement learning (STC-DRL) to learn the spatial features and the temporal features to obtain the target blasting parameters includes: The STC-DRL is used to perform model training on the spatial features and the temporal features to obtain a blast hole layout strategy for the open-pit mine; the blast hole layout strategy includes the three-dimensional coordinates, hole diameter, hole depth, inclination angle, and explosive charge of each blast hole; The target blasting parameters are determined according to the blasting hole layout strategy.

[0014] The embodiment of the present invention further provides a blasthole layout system for open-pit mines, the system comprising a probe vehicle equipped with a geological radar GR, a sonic well logging instrument SWL, a high-precision GPS receiver, a laser scanner LS, and an environmental sensor ES, and a processing device; the probe vehicle is connected to the processing device; wherein, The GR is used to detect geological structure information under the open-pit mine; The SWL is configured to transmit sound waves to rocks beneath the open-pit mine and capture reflected waves of the sound waves; determine mechanical parameters of the rocks based on the reflected waves; The GPS receiver is used to obtain the location information of the probe car; The LS is used to scan and map the topography around the open-pit mine to obtain data on the undulation and slope of the topography; The ES is used to monitor meteorological information around the open-pit mine; The processing device is used to determine the target blasting parameters of the open-pit mine geology based on the geological structure information, the mechanical parameters of the rock, the position information of the detection vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine; the target blasting parameters are used to arrange blasting holes in the open-pit mine.

[0015] An embodiment of the present invention provides an open-pit mine blasting hole layout device, which includes: a processor and a memory for storing a computer program that can be run on the processor, wherein the processor is used to execute the steps of any of the above methods when running the computer program.

[0016] An embodiment of the present invention provides a storage medium having a computer program stored thereon; when the computer program is executed by a processor, the steps of any one of the above methods are implemented.

[0017] An embodiment of the present invention provides a method and system for blast hole layout in an open-pit mine, and the method is applied to the blast hole layout system in an open-pit mine; the system includes a probe vehicle equipped with a GR, a SWL, a GPS receiver, a LS, and an ES; the method includes detecting geological structure information under the open-pit mine through the GR; emitting sound waves to the rocks under the open-pit mine through the SWL and capturing the reflected waves of the sound waves; determining the mechanical parameters of the rocks based on the reflected waves; obtaining the position information of the probe vehicle through the GPS receiver; scanning and mapping the terrain and landforms around the open-pit mine through the LS to obtain the undulation and slope data of the terrain and landforms; monitoring the meteorological information around the open-pit mine through the ES; determining the target blasting parameters of the open-pit mine geology based on the geological structure information, the mechanical parameters of the rocks, the position information of the probe vehicle, the undulation and slope data of the terrain and landforms, and the meteorological information around the open-pit mine; the target blasting parameters are used to layout blast holes for the open-pit mine. By adopting the technical solution of the present invention, the geological structure information under the open-pit mine is detected by GR; the SWL transmits sound waves to the rocks under the open-pit mine and captures the reflected waves of the sound waves; the mechanical parameters of the rocks are determined based on the reflected waves; the GPS receiver obtains the position information of the detection vehicle; the LS scans and maps the topography around the open-pit mine to obtain the undulation and slope data of the topography; the ES monitors the meteorological information around the open-pit mine; based on the geological structure information, the mechanical parameters of the rocks, the position information of the detection vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine, the target blasting parameters for blasting hole layout of the open-pit mine are determined, that is, through the intelligent adaptive open-pit mine blasting hole layout, the mining cost is reduced, the resource utilization efficiency is improved, and the mine ecological environment is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic diagram of a flow chart of a method for blast hole layout in an open-pit mine provided by an embodiment of the present invention; Figure 2 This is a flow chart of the blasting parameter optimization process in the blast hole layout method for open pit mine blasting; Figure 3 This is a flow chart of the algorithm in the open-pit mine blasting hole layout method according to an embodiment of the present application; Figure 4 This is a schematic diagram of an application scenario of the open-pit mine blasting hole layout method according to an embodiment of the present application; Figure 5 A schematic structural diagram of a blasthole layout system for open-pit mine blasting provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0020] The various specific technical features in the various embodiments described in the specific implementation methods can be combined in various ways without contradiction. For example, different implementation methods can be formed by combining different specific technical features. In order to avoid unnecessary repetition, the various possible combinations of the specific technical features in the present invention will not be described separately.

[0021] It should also be noted here that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions of the present invention are shown in the drawings, while other details that are not closely related to the present invention are omitted.

[0022] In addition, it should be noted that the terms "include", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the following description, the terms "first\second\..." involved are merely used to distinguish different objects and do not indicate that there is any similarity or connection between the objects. It should be understood that the directions described by the directional nouns such as "above", "below", "inside" and "outside" are all directions in normal use.

[0023] The present invention provides a method for arranging blastholes in an open-pit mine. Figure 1 As shown, Figure 1 A flow chart of a method for blasthole layout in an open-pit mine provided in an embodiment of the present invention is applied to an open-pit mine blasthole layout system. The system includes a probe vehicle equipped with a geological radar GR, a sonic logging instrument SWL, a high-precision GPS receiver, a laser scanner LS, and an environmental sensor ES. The method includes: Step S101: Detecting geological structure information under the open-pit mine through the GR.

[0024] Step S102: transmitting sound waves to the rocks under the open-pit mine through the SWL and capturing reflected waves of the sound waves; and determining mechanical parameters of the rocks based on the reflected waves.

[0025] Step S103: obtaining the location information of the probe car through the GPS receiver; Step S104: Scan and map the topography around the open-pit mine using the LS to obtain data on the undulation and slope of the topography; Step S105: monitoring meteorological information around the open-pit mine through the ES; Step S106: Determine target blasting parameters for the open-pit mine geology based on the geological structure information, the mechanical parameters of the rock, the location information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine; the target blasting parameters are used to arrange blastholes for the open-pit mine.

[0026] It should be noted that the open-pit mine blasting hole layout system can be determined according to actual conditions and is not limited here. As an example, the open-pit mine blasting hole layout system can be an intelligent adaptive open-pit mine blasting hole layout optimization system.

[0027] The system includes a probe vehicle equipped with a geological radar GR, a sonic logging instrument SWL, a high-precision GPS receiver, a laser scanner LS and an environmental sensor ES; wherein, the probe vehicle can be understood as a multifunctional probe vehicle, which uses ultra-wideband pulse technology to penetrate deep into the rock, clearly image the complex underground geological structure, and accurately locate key information such as faults, caves, and rock stratification.

[0028] In step S101, the specific detection process for detecting geological structure information beneath the open-pit mine using the GR can be determined based on actual conditions and is not limited herein. As an example, the geological structure information may include rock layer information, cavity information, and fault information; and detecting the geological structure information beneath the open-pit mine using the GR may include detecting the rock layer information, cavity information, and fault information beneath the open-pit mine using ultra-wideband pulse technology using the GR.

[0029] In step S102, the specific process for determining the mechanical parameters of the rock based on the reflected wave can be determined based on actual conditions and is not limited herein. As an example, the mechanical parameters of the rock include a hardness parameter and an elastic modulus parameter of the rock; determining the mechanical parameters of the rock based on the reflected wave may include: determining the longitudinal wave velocity and the shear wave velocity of the rock based on the reflected wave; and determining the hardness parameter and the elastic modulus parameter of the rock based on the longitudinal wave velocity and the shear wave velocity.

[0030] In step S103, the specific process of obtaining the probe vehicle's location information via the GPS receiver can be determined based on actual circumstances and is not limited herein. As an example, the location information includes three-dimensional position coordinate information; and obtaining the probe vehicle's location information via the GPS receiver may include obtaining the three-dimensional position coordinate information of the probe vehicle via the GPS receiver using a carrier phase difference technique.

[0031] In step S104, the specific acquisition process of scanning and mapping the topography surrounding the open-pit mine using the LS to obtain the undulation and slope data of the topography can be determined based on actual conditions and is not limited here. As an example, scanning and mapping the topography surrounding the open-pit mine using the LS to obtain the undulation and slope data of the topography can include performing a detailed scanning and mapping of the topography surrounding the open-pit mine using the LS to obtain the undulation and slope data of the topography.

[0032] In step S105, the specific monitoring process of monitoring the meteorological information around the open-pit mine via the ES can be determined based on actual conditions and is not limited herein. As an example, the meteorological information includes air temperature information, humidity information, and air pressure information; and monitoring the meteorological information around the open-pit mine via the ES may include monitoring the temperature information, humidity information, and air pressure information around the open-pit mine via the ES.

[0033] In step S106, the specific determination process of determining the target blasting parameters of the open-pit mine geology based on the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine can be determined according to actual conditions and is not limited here. As an example, the determination of the target blasting parameters of the open-pit mine geology based on the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine can include using a preset graph neural network GNN to deeply fuse the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain geological image data of the open-pit mine; obtaining historical mining data of the open-pit mine; and determining the target blasting parameters based on the geological image data and the historical mining data.

[0034] An embodiment of the present invention provides a method for arranging blastholes for blasting in an open-pit mine, wherein GR detects geological structure information under the open-pit mine; SWL emits sound waves to rocks under the open-pit mine and captures reflected waves of the sound waves; mechanical parameters of the rocks are determined based on the reflected waves; a GPS receiver obtains position information of the probe vehicle; LS scans and maps the terrain and landforms around the open-pit mine to obtain undulation and slope data of the terrain and landforms; ES monitors meteorological information around the open-pit mine; and target blasting parameters for arranging blastholes for blasting in the open-pit mine are determined based on the geological structure information, the mechanical parameters of the rocks, the position information of the probe vehicle, the undulation and slope data of the terrain and landforms, and the meteorological information around the open-pit mine. That is, through intelligent and adaptive open-pit mine blasthole layout, mining costs are reduced, resource utilization efficiency is improved, and the mine ecological environment is improved.

[0035] In some embodiments, the geological structure information includes rock layer information, cavity information, and fault information; and detecting the geological structure information under the open-pit mine by the GR includes: The GR uses ultra-wideband pulse technology to detect the rock layer information, the cavity information, and the fault information under the open-pit mine.

[0036] In this embodiment, the geological structure information includes rock stratification information, cavity information, and fault information; wherein, the geological structure information may also include initial geology; the geological structure information may be understood as key information; the rock stratification information may be referred to as rock stratification; the cavity information may include cave information, which may be referred to as cave; and the fault information may be referred to as fault.

[0037] The use of ultra-wideband pulse technology by the GR to detect the rock stratification information, the cavity information, and the fault information under the open-pit mine can be understood as the GR in the multi-functional detection vehicle using ultra-wideband pulse technology to detect the rock stratification information, the cavity information, and the fault information under the open-pit mine.

[0038] In actual applications, the multi-functional exploration vehicle uses ultra-wideband pulse technology to penetrate deep into the rock, clearly image the complex underground geological structure, and accurately locate key information such as faults, caves, and rock stratification.

[0039] In some embodiments, the mechanical parameters of the rock include a hardness parameter and an elastic modulus parameter of the rock; and determining the mechanical parameters of the rock based on the reflected wave includes: determining the longitudinal wave velocity and the shear wave velocity of the rock based on the reflected wave; The hardness parameter and elastic modulus parameter of the rock are determined according to the longitudinal wave velocity and the shear wave velocity.

[0040] In this embodiment, the specific process for determining the longitudinal and shear wave velocities of the rock based on the reflected waves can be determined based on actual conditions and is not limited herein. As an example, determining the longitudinal and shear wave velocities of the rock based on the reflected waves can be understood as an acoustic logging instrument transmitting acoustic waves to the rock and capturing reflected waves, thereby accurately measuring the longitudinal and shear wave velocities of the rock based on the propagation characteristics of acoustic waves in different media.

[0041] The specific determination process of determining the hardness parameter and elastic modulus parameter of the rock based on the longitudinal wave velocity and the shear wave velocity can be determined based on actual conditions and is not limited here. As an example, determining the hardness parameter and elastic modulus parameter of the rock based on the longitudinal wave velocity and the shear wave velocity can be calculating the hardness parameter and elastic modulus parameter of the rock based on the longitudinal wave velocity and the shear wave velocity.

[0042] In practical applications, as an example, sonic logging instruments emit sound waves to rocks and capture reflected waves. Based on the propagation characteristics of sound waves in different media, they accurately measure the longitudinal and transverse wave velocities of rocks, and then deduce mechanical parameters such as the hardness and elastic modulus of the rocks, providing a scientific basis for the formulation of blasting plans.

[0043] In some embodiments, the location information includes three-dimensional location coordinate information; and obtaining the location information of the probe vehicle through the GPS receiver includes: The three-dimensional position coordinate information of the probe vehicle is obtained by using the carrier phase difference technology through the GPS receiver.

[0044] In this embodiment, the specific process for determining the three-dimensional position coordinate information of the probe vehicle using the carrier phase differential technology by the GPS receiver can be determined based on actual conditions and is not limited herein. As an example, the process for obtaining the three-dimensional position coordinate information of the probe vehicle using the carrier phase differential technology by the GPS receiver can be understood as the process of a high-precision GPS receiver using the carrier phase differential technology to lock the three-dimensional precise position coordinates of the probe vehicle in real time.

[0045] In practical applications, high-precision GPS receivers use carrier phase differential technology to lock the three-dimensional precise position coordinates of the probe vehicle in real time.

[0046] In some embodiments, the meteorological information includes air temperature information, humidity information, and air pressure information; and monitoring the meteorological information around the open-pit mine by the ES includes: The temperature information, the humidity information, and the air pressure information around the open-pit mine are monitored by the ES.

[0047] In this embodiment, the meteorological information includes air temperature information, humidity information and air pressure information; wherein, the air temperature information can be simply referred to as air temperature; the humidity information can be simply referred to as humidity; and the air pressure information can be simply referred to as air pressure.

[0048] In actual applications, environmental sensors constantly monitor meteorological conditions such as air temperature, humidity, and air pressure, providing auxiliary reference for accurate prediction of blasting effects.

[0049] In some embodiments, the transmission method determines target blasting parameters of the open-pit mine geology based on the geological structure information, the mechanical parameters of the rock, the position information of the exploration vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine, including: A preset graph neural network (GNN) is used to deeply fuse the geological structure information, the mechanical parameters of the rock, the location information of the exploration vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain geological image data of the open-pit mine; Obtaining historical mining data of the open-pit mine; The target blasting parameters are determined based on the geological image data and the historical mining data.

[0050] In this embodiment, the preset graph neural network GNN can be determined according to actual conditions and is not limited here. As an example, the preset graph neural network GNN can include a modified graph neural network GNN.

[0051] The use of a preset graph neural network GNN to deeply fuse the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain the geological image data of the open-pit mine can be determined according to actual conditions and is not limited here. As an example, the use of a preset graph neural network GNN to deeply fuse the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain the geological image data of the open-pit mine can be to use the modified graph neural network GNN's cross-modal feature alignment method to deeply fuse the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain the geological image data of the open-pit mine. This process can effectively extract the spatial features of geological data and dynamically weight the contribution of different sensor data through a multi-head attention mechanism. The nodes in the network represent rock units, and the edges represent the relationships between rock units. The modified graph neural network update formula can be referred to: ; Among them: h v (l) Represents the feature vector of node v in layer l. N(v) represents the set of neighbor nodes of node v. c uv is the normalization coefficient between node v and node u. (l) and b (l) is the weight matrix and bias term of the lth layer. This can more accurately extract the complex relationships and geological structures of rock layers, significantly improving the ability to perceive complex geological conditions.

[0052] The specific process for obtaining the historical mining data of the open-pit mine can be determined based on actual circumstances and is not limited here. As an example, the historical mining data of the open-pit mine can be obtained from a server. In actual applications, a cluster of multiple high-performance servers equipped with large-capacity solid-state drives are used to store massive amounts of historical geological data, previous blasting plans and their effect feedback data, etc., providing rich data support for the training of deep learning models.

[0053] The specific process for determining the target blasting parameters based on the geological image data and the historical mining data can be determined based on actual circumstances and is not limited herein. As an example, the geological image data includes spatial features; the historical mining data includes temporal features; and determining the target blasting parameters based on the geological image data and the historical mining data may include: using a preset spatiotemporal coupled deep reinforcement learning (STC-DRL) method to learn the spatial features and the temporal features to obtain the target blasting parameters.

[0054] In some embodiments, the geological image data includes spatial features; the historical mining data includes temporal features; and determining the target blasting parameters based on the geological image data and the historical mining data includes: The preset spatiotemporal coupled deep reinforcement learning STC-DRL is used to learn the spatial features and the temporal features to obtain the target blasting parameters.

[0055] In this embodiment, the specific learning process of using the preset spatiotemporal coupling deep reinforcement learning STC-DRL to learn the spatial features and the temporal features to obtain the target blasting parameters can be determined according to actual conditions and is not limited here. As an example, the specific learning process of using the preset spatiotemporal coupling deep reinforcement learning STC-DRL to learn the spatial features and the temporal features to obtain the target blasting parameters can include: using the STC-DRL to perform model training on the spatial features and the temporal features to obtain the blasting hole layout strategy of the open-pit mine; the blasting hole layout strategy includes the three-dimensional coordinates, aperture, hole depth, inclination angle, and explosive loading of each blasting hole; and determining the target blasting parameters based on the blasting hole layout strategy.

[0056] In some embodiments, the method further comprises: Preprocessing the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain preprocessed geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine; Determining the contribution parameters of the pre-processed geological structure information, rock mechanical parameters, rover position information, topographic undulations and slope data, and meteorological information around the open-pit mine based on a preset multi-head attention mechanism; Modifying the GNN based on the contribution parameter to obtain a modified GNN; According to the modified GNN, the pre-processed geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine are deeply integrated to obtain the geological image data of the open-pit mine.

[0057] In this embodiment, the preprocessing can be determined according to actual conditions and is not limited here. As an example, the preprocessing may include noise removal, outlier correction, deep cleaning, calibration, and fusion processing.

[0058] The specific process for determining the contribution parameters of the preprocessed geological structure information, rock mechanical parameters, rover position information, topographical undulations and slope data, and meteorological information surrounding the open-pit mine based on the preset multi-head attention mechanism can be determined based on actual conditions and is not limited here. As an example, the contribution of different sensor data can be dynamically weighted using the multi-head attention mechanism.

[0059] The specific correction process of the GNN obtained by correcting the GNN based on the contribution parameter can be determined according to the actual situation and is not limited here. As an example, the GNN obtained by correcting the GNN based on the contribution parameter can be corrected by using a preset algorithm based on the contribution parameter to obtain a corrected GNN; wherein, the preset algorithm can be determined according to the actual situation and is not limited here. As an example, the preset algorithm can refer to ; Among them: h v (l) Represents the feature vector of node v in layer l. N(v) represents the set of neighbor nodes of node v. c uv is the normalization coefficient between node v and node u. (l) and b (l) is the weight matrix and bias term of the lth layer.

[0060] In practical applications, relevant data is collected through the aforementioned advanced methods, and then deeply integrated using a modified graph neural network (GNN) cross-modal feature alignment method. This method effectively extracts the spatial features of geological data and dynamically weights the contributions of different sensor data through a multi-head attention mechanism. Nodes in the network represent rock units, and edges represent relationships between rock units. The modified GNN update formula is: ; Among them: h v (l) Represents the feature vector of node v in layer l. N(v) represents the set of neighbor nodes of node v. c uv is the normalization coefficient between node v and node u. (l)and b (l) is the weight matrix and bias term of the lth layer. This can more accurately extract the complex relationships and geological structures of rock layers, significantly improving the ability to perceive complex geological conditions.

[0061] In some embodiments, the adopting of a preset spatiotemporal coupled deep reinforcement learning (STC-DRL) to learn the spatial features and the temporal features to obtain the target blasting parameters includes: The STC-DRL is used to perform model training on the spatial features and the temporal features to obtain a blast hole layout strategy for the open-pit mine; the blast hole layout strategy includes the three-dimensional coordinates, hole diameter, hole depth, inclination angle, and explosive charge of each blast hole; The target blasting parameters are determined according to the blasting hole layout strategy.

[0062] It should be noted that this system is built on the cutting-edge artificial intelligence architecture of deep neural networks. Its core algorithm cleverly combines the unique advantages of convolutional neural networks (CNNs) and recurrent neural networks (RNNs). CNNs are used to automatically extract features from geological and topographic image data, accurately identifying rock structural patterns and terrain variations. RNNs excel at processing time series data and, combined with the historical mining process, predict geological trends. Taking into account multi-dimensional constraints such as explosive performance, cost control, mining schedule requirements, and ecological and environmental protection, the model is trained using a reinforcement learning algorithm, enabling it to autonomously learn optimal blasthole layout strategies. The model outputs an optimized blasthole layout plan that includes detailed parameters such as each blasthole's precise three-dimensional coordinates (X, Y, and Z axis position with millimeter-level accuracy), hole diameter (dynamically adjusted based on rock hardness and explosive type), hole depth (precisely calculated based on geological stratification and geological conditions), inclination angle (adapted to terrain slope and rock direction), and explosive charge (minimized while ensuring blasting effectiveness). To better optimize blasthole layout, the optimization correlation engine of spatiotemporal coupled deep reinforcement learning (STC-DRL) is used. By combining the spatial characteristics of geological image data with time series data from historical mining processes, changes in geological conditions can be fully captured. This allows for more accurate prediction of changes in geological conditions and dynamic optimization of blasthole layout. ;in, is the importance sampling ratio; is the advantage function, representing the relative advantage of choosing action at in state st. By maximizing the reward function R (which takes into account factors such as green ore size, explosives cost, and safety), the reinforcement learning module enables the system to automatically optimize the blasting layout based on real-time data and historical feedback. .

[0063] In practical applications, an open-pit mine blasting hole layout method can be specifically an intelligent adaptive open-pit mine blasting hole layout optimization method, including ① data acquisition and fusion algorithm, ② data transmission network, ③ data processing and analysis center, ④ intelligent decision engine and ⑤ visual interactive terminal.

[0064] ① Data Acquisition and Fusion Algorithm: The multi-functional rover utilizes ultra-wideband pulse technology to penetrate deep into the rock, clearly imaging the complex underground geological structure and accurately locating key information such as faults, caves, and rock layers. The sonic logging instrument transmits sound waves into the rock and captures the reflected waves. Based on the propagation characteristics of sound waves in different media, it accurately measures the rock's longitudinal and shear wave velocities. This allows the calculation of mechanical parameters such as the rock's hardness and elastic modulus, providing a scientific basis for blasting plan formulation. A high-precision GPS receiver uses carrier phase differential technology to lock the rover's precise three-dimensional position in real time, and collaborates with a laser scanner to conduct detailed scanning and mapping of the surrounding terrain. Meanwhile, environmental sensors constantly monitor meteorological conditions such as air temperature, humidity, and air pressure, providing auxiliary reference for accurate prediction of blasting effects. Each sensor is embedded with an intelligent microprocessor that performs preliminary analysis of the collected raw data, removing noise interference and correcting outliers, ensuring the purity and reliability of the data transmitted to the backend. These advanced methods collect relevant data, which is then deeply fused using a modified graph neural network (GNN) cross-modal feature alignment method. This method can effectively extract the spatial features of geological data and dynamically weight the contribution of different sensor data through a multi-head attention mechanism. The nodes in the network represent rock units, and the edges represent the relationships between rock units. The modified graph neural network update formula is: in: h v (l) Represents the feature vector of node v in layer l.

[0065] N(v) represents the set of neighbor nodes of node v.

[0066] c uv is the normalization coefficient between node v and node u.

[0067] w (l) and b (l) is the weight matrix and bias term of the lth layer.

[0068] This allows for more accurate extraction of the complex relationships and geological structures of rock strata, significantly improving the ability to perceive complex geological conditions.

[0069] ②Data transmission network: Using 5G high-speed communication network, massive, real-time data can be quickly transmitted to the data processing and analysis center to ensure the timeliness of information and inject strong impetus into subsequent decision-making analysis based on real-time data.

[0070] ③ Data Processing and Analysis Center: Consists of multiple high-performance server clusters and equipped with large-capacity solid-state drives, it is used to store massive historical geological data of the mine, previous blasting plans and their effect feedback data, etc., providing rich data support for the training of deep learning models; the multi-core processor performs deep cleaning, calibration, and fusion processing on the incoming data, and uses a feature extraction algorithm based on data mining to convert multi-source data from different sensors into a unified feature vector, preparing for the operation of the subsequent intelligent decision-making engine.

[0071] ④ Intelligent Decision-Making Engine: Built on the cutting-edge artificial intelligence architecture of deep neural networks, its core algorithm cleverly combines the unique advantages of convolutional neural networks (CNNs) and recurrent neural networks (RNNs). CNNs are used to automatically extract features from geological and topographic image data, accurately identifying rock structural patterns and terrain variations. RNNs excel at processing time series data and, combined with historical mining processes, predict geological trends. Taking into account multi-dimensional constraints such as explosive performance, cost control, mining schedule requirements, and ecological and environmental protection, the model is trained using a reinforcement learning algorithm, enabling it to autonomously learn optimal blasthole layout strategies. The model outputs an optimized blasthole layout plan that includes detailed parameters such as the precise three-dimensional coordinates of each blasthole (precise X, Y, and Z axis position with millimeter-level accuracy), hole diameter (dynamically adjusted based on rock hardness and explosive type), hole depth (precisely calculated based on geological stratification and geological conditions), inclination angle (adapted to terrain slope and rock direction), and explosive charge (minimized while ensuring blasting effectiveness). To better optimize blasthole layout, the optimization correlation engine of spatiotemporal coupled deep reinforcement learning (STC-DRL) is used. By combining the spatial characteristics of geological image data with time series data from historical mining processes, changes in geological conditions can be fully captured. This allows for more accurate prediction of changes in geological conditions and dynamic optimization of blasthole layout.

[0072] in, is the importance sampling ratio; is the advantage function, which represents the relative advantage of selecting action at in state st.

[0073] By maximizing the reward function R (taking into account factors such as green ore size, explosives cost, and safety), the reinforcement learning module enables the system to automatically optimize the blasting layout based on real-time data and historical feedback.

[0074] .

[0075] This content can be combined Figure 2 To understand, Figure 2 This is a flow chart of the blasting parameter optimization process in the blasthole layout method for open-pit mine blasting.

[0076] ⑤ Visual interactive terminal: It uses an industrial-grade touch screen and runs professional visualization software. On the one hand, it presents the blasting hole layout plan generated by the intelligent decision-making engine in three-dimensional dynamic graphics, simulating realistic blasting effects, which is convenient for engineers to review, evaluate and adjust. On the other hand, it can be directly connected to the drilling operation machinery control system to transmit the optimized blasting hole layout parameters in real time, realizing precise guidance of automated drilling operations and ensuring that the actual construction is in perfect accordance with the design plan.

[0077] The algorithm is composed as follows Figure 3 As shown, Figure 3 This is a flow chart of the algorithm in the open-pit mine blasting hole layout method according to an embodiment of the present application.

[0078] As an example, you can combine Figure 4 To understand, Figure 4 This is a schematic diagram of the application scenario of the method for blasting hole layout in open-pit mines according to the embodiment of the present application; During the mining preparation stage: a multifunctional detection vehicle is dispatched to the starting position of the operating area to be mined, all sensors are activated, and the vehicle drives steadily according to the preset scientific route to collect initial geological, topographical and environmental data in all directions. After the collected data is preliminarily processed by the sensor's built-in microprocessor, it is transmitted to the data processing and analysis center in real time and at high speed via the 5G network. The server cluster deeply cleans, calibrates, and fuses the data, properly stores it on the local hard disk, and inputs it into the intelligent decision-making engine. The intelligent decision-making engine relies on deep learning models and massive historical data to quickly generate the first round of blasting hole layout plans, which are pushed to the visual interactive terminal. The technical personnel conduct prudent reviews and fine-tuning based on the actual mining needs and safety regulations. After confirmation, the plans are transmitted to the drilling operation machinery control system.

[0079] Continuous optimization stage of the mining process: After each blasting operation and ore removal, the probe car collects data from the newly exposed working surface and surrounding areas, repeating the above data transmission and processing process. The intelligent decision-making engine dynamically optimizes the layout of subsequent blasting holes based on new data and feedback on the mining progress. After further review by technical personnel, the drilling operation instructions are updated to continuously push the mining process forward steadily. The system is regularly updated in an all-round way. Based on indicators such as actual blasting effects, mining efficiency improvements, and cost control effectiveness, big data analysis technology is used to deeply explore potential problems, and targeted adjustments and optimizations are made to the deep learning model to continuously improve system performance until the area is mined, and then the above process is repeated in the next working area.

[0080] The embodiment of the present invention also provides an open-pit mine blasting hole layout system 500, such as Figure 5 As shown, Figure 5 A schematic structural diagram of a blasthole layout system for open-pit mine blasting is provided in an embodiment of the present invention; the structure and functions of the system are exemplarily described below in conjunction with various embodiments.

[0081] The system 500 includes a probe vehicle 501 equipped with a geological radar GR5011, a sonic logging instrument SWL5012, a high-precision GPS receiver 5013, a laser scanner LS5014 and an environmental sensor ES5015, and a processing device 502; the probe vehicle 501 is connected to the processing device 502; wherein, The GR5011 is used to detect geological structure information under the open-pit mine; The SWL5012 is used to transmit sound waves to the rocks under the open-pit mine and capture the reflected waves of the sound waves; and determine the mechanical parameters of the rocks based on the reflected waves; The GPS receiver 5013 is used to obtain the location information of the probe car; The LS5014 is used to scan and map the terrain around the open-pit mine to obtain the undulation and slope data of the terrain; The ES5015 is used to monitor meteorological information around the open-pit mine; The processing device 502 is used to determine the target blasting parameters of the open-pit mine geology based on the geological structure information, the mechanical parameters of the rock, the position information of the exploration vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine; the target blasting parameters are used to arrange blasting holes for the open-pit mine.

[0082] It should be noted that the open-pit mine blasting hole layout system can be determined according to actual conditions and is not limited here. As an example, the open-pit mine blasting hole layout system can be an intelligent adaptive open-pit mine blasting hole layout optimization system.

[0083] The system includes a probe vehicle equipped with a geological radar GR, a sonic logging instrument SWL, a high-precision GPS receiver, a laser scanner LS and an environmental sensor ES; wherein, the probe vehicle can be understood as a multifunctional probe vehicle, which uses ultra-wideband pulse technology to penetrate deep into the rock, clearly image the complex underground geological structure, and accurately locate key information such as faults, caves, and rock stratification.

[0084] The specific detection process for detecting geological structure information beneath the open-pit mine using the GR can be determined based on actual circumstances and is not limited herein. As an example, the geological structure information may include rock layer information, cavity information, and fault information; and detecting the geological structure information beneath the open-pit mine using the GR may include detecting the rock layer information, cavity information, and fault information beneath the open-pit mine using ultra-wideband pulse technology using the GR.

[0085] The specific process for determining the mechanical parameters of the rock based on the reflected wave can be determined based on actual conditions and is not limited herein. As an example, the mechanical parameters of the rock include a hardness parameter and an elastic modulus parameter of the rock; determining the mechanical parameters of the rock based on the reflected wave may include: determining the longitudinal wave velocity and the shear wave velocity of the rock based on the reflected wave; and determining the hardness parameter and the elastic modulus parameter of the rock based on the longitudinal wave velocity and the shear wave velocity.

[0086] The specific process for obtaining the location information of the probe vehicle via the GPS receiver can be determined based on actual circumstances and is not limited herein. As an example, the location information includes three-dimensional position coordinate information; and obtaining the location information of the probe vehicle via the GPS receiver may include obtaining the three-dimensional position coordinate information of the probe vehicle via the GPS receiver using a carrier phase differential technique.

[0087] The specific acquisition process of scanning and mapping the topography and landforms around the open-pit mine using the LS to obtain the undulation and slope data of the topography and landforms can be determined based on actual conditions and is not limited here. As an example, scanning and mapping the topography and landforms around the open-pit mine using the LS to obtain the undulation and slope data of the topography and landforms can include performing a detailed scanning and mapping of the topography and landforms around the open-pit mine using the LS to obtain the undulation and slope data of the topography and landforms.

[0088] The specific monitoring process for monitoring the meteorological information around the open-pit mine via the ES can be determined based on actual circumstances and is not limited herein. As an example, the meteorological information includes air temperature information, humidity information, and air pressure information; and monitoring the meteorological information around the open-pit mine via the ES may include monitoring the temperature information, humidity information, and air pressure information around the open-pit mine via the ES.

[0089] The specific determination process of determining the target blasting parameters of the open-pit mine geology based on the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine can be determined according to actual conditions and is not limited here. As an example, the determination of the target blasting parameters of the open-pit mine geology based on the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine can include using a preset graph neural network GNN to deeply fuse the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain geological image data of the open-pit mine; obtaining historical mining data of the open-pit mine; and determining the target blasting parameters based on the geological image data and the historical mining data.

[0090] In some embodiments, the geological structure information includes rock stratification information, cavity information, and fault information; the GR5011 is also used to use ultra-wideband pulse technology to detect the rock stratification information, cavity information, and fault information under the open-pit mine.

[0091] In this embodiment, the geological structure information includes rock stratification information, cavity information, and fault information; wherein, the geological structure information may also include initial geology; the geological structure information may be understood as key information; the rock stratification information may be referred to as rock stratification; the cavity information may include cave information, which may be referred to as cave; and the fault information may be referred to as fault.

[0092] The use of ultra-wideband pulse technology by the GR to detect the rock stratification information, the cavity information, and the fault information under the open-pit mine can be understood as the GR in the multi-functional detection vehicle using ultra-wideband pulse technology to detect the rock stratification information, the cavity information, and the fault information under the open-pit mine.

[0093] In actual applications, the multi-functional exploration vehicle uses ultra-wideband pulse technology to penetrate deep into the rock, clearly image the complex underground geological structure, and accurately locate key information such as faults, caves, and rock stratification.

[0094] In some embodiments, the mechanical parameters of the rock include hardness parameters and elastic modulus parameters of the rock; the SWL5012 is also used to determine the longitudinal wave velocity and transverse wave velocity of the rock based on the reflected wave; and determine the hardness parameters and elastic modulus parameters of the rock according to the longitudinal wave velocity and the transverse wave velocity.

[0095] In this embodiment, the specific process for determining the longitudinal and shear wave velocities of the rock based on the reflected waves can be determined based on actual conditions and is not limited herein. As an example, determining the longitudinal and shear wave velocities of the rock based on the reflected waves can be understood as an acoustic logging instrument transmitting acoustic waves to the rock and capturing reflected waves, thereby accurately measuring the longitudinal and shear wave velocities of the rock based on the propagation characteristics of acoustic waves in different media.

[0096] The specific determination process of determining the hardness parameter and elastic modulus parameter of the rock based on the longitudinal wave velocity and the shear wave velocity can be determined based on actual conditions and is not limited here. As an example, determining the hardness parameter and elastic modulus parameter of the rock based on the longitudinal wave velocity and the shear wave velocity can be calculating the hardness parameter and elastic modulus parameter of the rock based on the longitudinal wave velocity and the shear wave velocity.

[0097] In practical applications, as an example, sonic logging instruments emit sound waves to rocks and capture reflected waves. Based on the propagation characteristics of sound waves in different media, they accurately measure the longitudinal and transverse wave velocities of rocks, and then deduce mechanical parameters such as the hardness and elastic modulus of the rocks, providing a scientific basis for the formulation of blasting plans.

[0098] In some embodiments, the location information includes three-dimensional location coordinate information; the GPS receiver 5013 is further configured to acquire the three-dimensional location coordinate information of the probe vehicle using carrier phase differential technology.

[0099] In this embodiment, the specific process for determining the three-dimensional position coordinate information of the probe vehicle using the carrier phase differential technology by the GPS receiver can be determined based on actual conditions and is not limited herein. As an example, the process for obtaining the three-dimensional position coordinate information of the probe vehicle using the carrier phase differential technology by the GPS receiver can be understood as the process of a high-precision GPS receiver using the carrier phase differential technology to lock the three-dimensional precise position coordinates of the probe vehicle in real time.

[0100] In practical applications, high-precision GPS receivers use carrier phase differential technology to lock the three-dimensional precise position coordinates of the probe vehicle in real time.

[0101] In some embodiments, the meteorological information includes air temperature information, humidity information, and air pressure information; the ES5015 is further used to monitor the temperature information, humidity information, and air pressure information around the open-pit mine.

[0102] In this embodiment, the meteorological information includes air temperature information, humidity information and air pressure information; wherein, the air temperature information can be simply referred to as air temperature; the humidity information can be simply referred to as humidity; and the air pressure information can be simply referred to as air pressure.

[0103] In actual applications, environmental sensors constantly monitor meteorological conditions such as air temperature, humidity, and air pressure, providing auxiliary reference for accurate prediction of blasting effects.

[0104] In some embodiments, the processing device 502 is also used to use a preset graph neural network GNN to deeply fuse the geological structure information, the mechanical parameters of the rock, the location information of the probe vehicle, the undulation and slope data of the terrain, and the meteorological information around the open-pit mine to obtain geological image data of the open-pit mine; obtain historical mining data of the open-pit mine; and determine the target blasting parameters based on the geological image data and the historical mining data.

[0105] In this embodiment, the preset graph neural network GNN can be determined according to actual conditions and is not limited here. As an example, the preset graph neural network GNN can include a modified graph neural network GNN.

[0106] The use of a preset graph neural network GNN to deeply fuse the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain the geological image data of the open-pit mine can be determined according to actual conditions and is not limited here. As an example, the use of a preset graph neural network GNN to deeply fuse the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain the geological image data of the open-pit mine can be to use the modified graph neural network GNN's cross-modal feature alignment method to deeply fuse the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain the geological image data of the open-pit mine. This process can effectively extract the spatial features of geological data and dynamically weight the contribution of different sensor data through a multi-head attention mechanism. The nodes in the network represent rock units, and the edges represent the relationships between rock units. The modified graph neural network update formula can be referred to: ; Among them: h v (l) Represents the feature vector of node v in layer l. N(v) represents the set of neighbor nodes of node v. c uv is the normalization coefficient between node v and node u. (l) and b (l) is the weight matrix and bias term of the lth layer. This can more accurately extract the complex relationships and geological structures of rock layers, significantly improving the ability to perceive complex geological conditions.

[0107] The specific process for obtaining the historical mining data of the open-pit mine can be determined based on actual circumstances and is not limited here. As an example, the historical mining data of the open-pit mine can be obtained from a server. In actual applications, a cluster of multiple high-performance servers equipped with large-capacity solid-state drives are used to store massive amounts of historical geological data, previous blasting plans and their effect feedback data, etc., providing rich data support for the training of deep learning models.

[0108] The specific process for determining the target blasting parameters based on the geological image data and the historical mining data can be determined based on actual circumstances and is not limited herein. As an example, the geological image data includes spatial features; the historical mining data includes temporal features; and determining the target blasting parameters based on the geological image data and the historical mining data may include: using a preset spatiotemporal coupled deep reinforcement learning (STC-DRL) method to learn the spatial features and the temporal features to obtain the target blasting parameters.

[0109] In some embodiments, the geological image data includes spatial features; the historical mining data includes temporal features; the processing device 502 is also used to use a preset spatiotemporal coupling deep reinforcement learning STC-DRL to learn the spatial features and the temporal features to obtain the target blasting parameters.

[0110] In this embodiment, the specific learning process of using the preset spatiotemporal coupling deep reinforcement learning STC-DRL to learn the spatial features and the temporal features to obtain the target blasting parameters can be determined according to actual conditions and is not limited here. As an example, the specific learning process of using the preset spatiotemporal coupling deep reinforcement learning STC-DRL to learn the spatial features and the temporal features to obtain the target blasting parameters can include: using the STC-DRL to perform model training on the spatial features and the temporal features to obtain the blasting hole layout strategy of the open-pit mine; the blasting hole layout strategy includes the three-dimensional coordinates, aperture, hole depth, inclination angle, and explosive loading of each blasting hole; and determining the target blasting parameters based on the blasting hole layout strategy.

[0111] In some embodiments, the processing device 502 is also used to preprocess the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain the preprocessed geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine; determine the contribution parameters of the preprocessed geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine based on a preset multi-head attention mechanism; correct the GNN based on the contribution parameters to obtain a corrected GNN; and deeply fuse the preprocessed geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine according to the corrected GNN to obtain geological image data of the open-pit mine.

[0112] In this embodiment, the preprocessing can be determined according to actual conditions and is not limited here. As an example, the preprocessing may include noise removal, outlier correction, deep cleaning, calibration, and fusion processing.

[0113] The specific process for determining the contribution parameters of the preprocessed geological structure information, rock mechanical parameters, rover position information, topographical undulations and slope data, and meteorological information surrounding the open-pit mine based on the preset multi-head attention mechanism can be determined based on actual conditions and is not limited here. As an example, the contribution of different sensor data can be dynamically weighted using the multi-head attention mechanism.

[0114] The specific correction process of the GNN obtained by correcting the GNN based on the contribution parameter can be determined according to the actual situation and is not limited here. As an example, the GNN obtained by correcting the GNN based on the contribution parameter can be corrected by using a preset algorithm based on the contribution parameter to obtain a corrected GNN; wherein, the preset algorithm can be determined according to the actual situation and is not limited here. As an example, the preset algorithm can refer to ; Among them: h v (l) Represents the feature vector of node v in layer l. N(v) represents the set of neighbor nodes of node v. c uv is the normalization coefficient between node v and node u. (l) and b (l) is the weight matrix and bias term of the lth layer.

[0115] In practical applications, relevant data is collected through the aforementioned advanced methods, and then deeply integrated using a modified graph neural network (GNN) cross-modal feature alignment method. This method effectively extracts the spatial features of geological data and dynamically weights the contributions of different sensor data through a multi-head attention mechanism. Nodes in the network represent rock units, and edges represent relationships between rock units. The modified GNN update formula is: ; Among them: h v (l) Represents the feature vector of node v in layer l. N(v) represents the set of neighbor nodes of node v. c uv is the normalization coefficient between node v and node u. (l) and b (l) is the weight matrix and bias term of the lth layer. This can more accurately extract the complex relationships and geological structures of rock layers, significantly improving the ability to perceive complex geological conditions.

[0116] In some embodiments, the processing device 502 is also used to use the STC-DRL to perform model training on the spatial features and the temporal features to obtain the blasting hole layout strategy of the open-pit mine; the blasting hole layout strategy includes the three-dimensional coordinates, hole diameter, hole depth, inclination angle, and explosive loading of each blasting hole; and the target blasting parameters are determined according to the blasting hole layout strategy.

[0117] It should be noted that this system is built on the cutting-edge artificial intelligence architecture of deep neural networks. Its core algorithm cleverly combines the unique advantages of convolutional neural networks (CNNs) and recurrent neural networks (RNNs). CNNs are used to automatically extract features from geological and topographic image data, accurately identifying rock structural patterns and terrain variations. RNNs excel at processing time series data and, combined with the historical mining process, predict geological trends. Taking into account multi-dimensional constraints such as explosive performance, cost control, mining schedule requirements, and ecological and environmental protection, the model is trained using a reinforcement learning algorithm, enabling it to autonomously learn optimal blasthole layout strategies. The model outputs an optimized blasthole layout plan that includes detailed parameters such as each blasthole's precise three-dimensional coordinates (X, Y, and Z axis position with millimeter-level accuracy), hole diameter (dynamically adjusted based on rock hardness and explosive type), hole depth (precisely calculated based on geological stratification and geological conditions), inclination angle (adapted to terrain slope and rock direction), and explosive charge (minimized while ensuring blasting effectiveness). To better optimize blasthole layout, the optimization correlation engine of spatiotemporal coupled deep reinforcement learning (STC-DRL) is used. By combining the spatial characteristics of geological image data with time series data from historical mining processes, changes in geological conditions can be fully captured. This allows for more accurate prediction of changes in geological conditions and dynamic optimization of blasthole layout. ;in, is the importance sampling ratio; is the advantage function, representing the relative advantage of choosing action at in state st. By maximizing the reward function R (which takes into account factors such as green ore size, explosives cost, and safety), the reinforcement learning module enables the system to automatically optimize the blasting layout based on real-time data and historical feedback. .

[0118] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for blast hole layout in an open pit mine, characterized in that: The invention is applied to a blasthole layout system for open-pit mines; the system includes a detection vehicle equipped with a geological radar GR, a sonic well logging instrument SWL, a high-precision GPS receiver, a laser scanner LS, and an environmental sensor ES; the method includes: Detecting geological structure information under the open-pit mine by using the GR; transmitting sound waves to rocks beneath the open-pit mine through the SWL and capturing reflected waves of the sound waves; determining mechanical parameters of the rocks based on the reflected waves; Obtaining the location information of the probe car through the GPS receiver; Scanning and mapping the topography around the open-pit mine using the LS to obtain data on the undulation and slope of the topography; monitoring meteorological information around the open-pit mine through the ES; The target blasting parameters of the open-pit mine geology are determined based on the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine; the target blasting parameters are used to arrange blasting holes for the open-pit mine.

2. The method according to claim 1, characterized in that The geological structure information includes rock layer information, cavity information, and fault information; the geological structure information under the open-pit mine detected by the GR includes: The GR uses ultra-wideband pulse technology to detect the rock layer information, the cavity information, and the fault information under the open-pit mine.

3. The method according to claim 1, characterized in that The mechanical parameters of the rock include a hardness parameter and an elastic modulus parameter of the rock; and determining the mechanical parameters of the rock based on the reflected wave includes: determining the longitudinal wave velocity and the shear wave velocity of the rock based on the reflected wave; The hardness parameter and elastic modulus parameter of the rock are determined according to the longitudinal wave velocity and the shear wave velocity.

4. The method according to claim 1, wherein The location information includes three-dimensional location coordinate information; the obtaining of the location information of the probe vehicle by the GPS receiver includes: The three-dimensional position coordinate information of the probe vehicle is obtained by using the carrier phase difference technology through the GPS receiver.

5. The method according to claim 1, characterized in that The meteorological information includes air temperature information, humidity information and air pressure information; the meteorological information around the open-pit mine monitored by the ES includes: The temperature information, the humidity information, and the air pressure information around the open-pit mine are monitored by the ES.

6. The method according to claim 1, wherein The transmission method determines target blasting parameters of the open-pit mine geology based on the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine, including: A preset graph neural network (GNN) is used to deeply fuse the geological structure information, the mechanical parameters of the rock, the location information of the exploration vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain geological image data of the open-pit mine; Obtaining historical mining data of the open-pit mine; The target blasting parameters are determined based on the geological image data and the historical mining data.

7. The method according to claim 6, characterized in that The geological image data includes spatial features; the historical mining data includes temporal features; and determining the target blasting parameters based on the geological image data and the historical mining data includes: The preset spatiotemporal coupled deep reinforcement learning STC-DRL is used to learn the spatial features and the temporal features to obtain the target blasting parameters.

8. The method according to claim 7, characterized in that The method further comprises: Preprocessing the geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine to obtain preprocessed geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine; Determining the contribution parameters of the pre-processed geological structure information, rock mechanical parameters, rover position information, topographic undulations and slope data, and meteorological information around the open-pit mine based on a preset multi-head attention mechanism; Modifying the GNN based on the contribution parameter to obtain a modified GNN; According to the modified GNN, the pre-processed geological structure information, the mechanical parameters of the rock, the position information of the probe vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine are deeply integrated to obtain the geological image data of the open-pit mine.

9. The method according to claim 7, characterized in that The preset spatiotemporal coupled deep reinforcement learning (STC-DRL) is used to learn the spatial features and the temporal features to obtain the target blasting parameters, including: The STC-DRL is used to perform model training on the spatial features and the temporal features to obtain a blast hole layout strategy for the open-pit mine; the blast hole layout strategy includes the three-dimensional coordinates, hole diameter, hole depth, inclination angle, and explosive charge of each blast hole; The target blasting parameters are determined according to the blasting hole layout strategy.

10. A blasthole layout system for open-pit mines, characterized in that: The system includes a probe vehicle equipped with a geological radar GR, a sonic logging instrument SWL, a high-precision GPS receiver, a laser scanner LS and an environmental sensor ES, and a processing device; the probe vehicle is connected to the processing device; wherein, The GR is used to detect geological structure information under the open-pit mine; The SWL is configured to transmit sound waves to rocks beneath the open-pit mine and capture reflected waves of the sound waves; determine mechanical parameters of the rocks based on the reflected waves; The GPS receiver is used to obtain the location information of the probe car; The LS is used to scan and map the topography around the open-pit mine to obtain data on the undulation and slope of the topography; The ES is used to monitor meteorological information around the open-pit mine; The processing device is used to determine the target blasting parameters of the open-pit mine geology based on the geological structure information, the mechanical parameters of the rock, the position information of the detection vehicle, the undulation and slope data of the topography, and the meteorological information around the open-pit mine; the target blasting parameters are used to arrange blasting holes in the open-pit mine.

Images