Green intelligent blasting rock breaking system and method

Abstract

The invention discloses a green intelligent blasting rock breaking system and method, and relates to the technical field of excavation rock breaking, the system comprises a data acquisition module, an intelligent decision module and an unmanned engineering machinery construction module; wherein the data acquisition module is used for monitoring a construction site and engineering mechanical equipment in real time by using a laser radar and a binocular camera and acquiring construction site data, and the construction site data comprises engineering geological data, azimuth angle data of the engineering mechanical equipment and image video data; the intelligent decision-making module is used for extracting different types of construction parameters in the construction site data and analyzing and evaluating the blasting effect in combination with historical data; and the unmanned engineering machinery construction module is used for executing the construction task and calling and controlling engineering machinery equipment to implement blasting construction operation. Through cooperative work of multiple modules, a brand-new green intelligent blasting rock breaking system is formed, the construction efficiency is remarkably improved, the influence on the environment is reduced, and the construction safety and the intelligent level are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of excavation and rock breaking, and in particular to a green and intelligent blasting and rock breaking system and method. Background Art

[0002] Excavation and rock breaking operations play an indispensable role in modern engineering construction, especially in the fields of tunnel construction, mining and urban underground engineering. With the acceleration of urbanization, the demand for underground engineering is increasing, which makes the efficiency and safety of rock breaking operations particularly important. However, existing excavation and rock breaking technologies face multiple challenges, which not only affect the efficiency of construction, but also pose potential threats to safety and the environment.

[0003] Traditional rock breaking methods mainly rely on manual operation. This dependence makes the construction efficiency limited by the experience and skills of workers. Under complex geological conditions, such as hard rock or abundant groundwater, manual operation often cannot achieve efficient and accurate construction, resulting in extended construction period and increased costs. In addition, manual operation has limited response capabilities in dealing with emergencies, and often cannot take effective measures in time, increasing the risk of accidents. Under harsh geological conditions, construction accidents frequently occur, workers' safety is seriously threatened, and the safety of the construction site is greatly reduced.

[0004] At the same time, traditional blasting and mechanical excavation operations are often accompanied by problems such as high energy consumption, noise and dust pollution. This not only affects the working environment of the construction site, but also has a significant impact on the surrounding residents and the ecological environment. The use of high energy consumption has increased construction costs, and in the context of the current increasing environmental awareness, traditional excavation methods are difficult to meet the requirements of modern engineering for environmentally friendly construction. Noise and dust pollution problems not only affect the health of construction workers, but may also cause ecological damage to the surrounding environment, which has put many projects under pressure from public opinion.

[0005] In addition, the existing excavation equipment has limited collaborative working capabilities. The lack of effective linkage mechanisms between traditional mechanical equipment often leads to waste of resources and delayed construction progress. During the construction process, the lack of coordination between equipment makes the coordination between various links not tight enough, which affects the construction efficiency and prolongs the construction period. This situation is particularly prominent in large-scale engineering projects, which often require more time and human resources for coordination and scheduling.

[0006] Therefore, there is an urgent need for a green and intelligent blasting and rock breaking system and method to solve the current technical problems. Summary of the invention

[0007] Based on this, it is necessary to provide a green and intelligent blasting and rock breaking system and method to address the above technical problems.

[0008] In a first aspect, the present invention provides a green intelligent blasting and rock breaking system, the system comprising: a data acquisition module, an intelligent decision-making module and an unmanned engineering machinery construction module;

[0009] The data acquisition module is used to use laser radar and binocular cameras to monitor the construction site and engineering machinery and equipment in real time and obtain construction site data. The construction site data includes engineering geological data, azimuth angle data of engineering machinery and equipment, and image and video data.

[0010] Intelligent decision-making module, which extracts different types of construction parameters from construction site data, combines historical data, analyzes and evaluates blasting effects, and optimizes and adjusts construction parameters and construction plans;

[0011] The unmanned engineering machinery construction module is used to execute construction tasks, call and control engineering machinery equipment to implement blasting construction operations, and conduct unmanned, green and intelligent blasting construction.

[0012] Furthermore, the data acquisition module includes: a drone monitoring unit, a machinery monitoring unit and a central control room;

[0013] Among them, the drone monitoring unit is used to use the laser radar and binocular camera carried by the drone to obtain the three-dimensional site data of the construction site, the azimuth angle data of the construction machinery and equipment, and the image and video data in the construction site;

[0014] The mechanical monitoring unit is used to obtain the three-dimensional rock mass data and image and video data of the rock mass during the blasting operation by using the laser radar and binocular camera carried by the engineering machinery;

[0015] The central control room is used to control drones and engineering machinery equipment to perform data monitoring tasks, and receive monitoring data uploaded in real time by the drone monitoring unit and the machinery monitoring unit.

[0016] Furthermore, the intelligent decision-making module includes a parameter setting unit, a digital simulation unit, a model evaluation unit, a drilling control unit, a blasting control unit, a crushing control unit and a feedback optimization unit;

[0017] The parameter setting unit is used to set the initial construction parameters of the engineering machinery and equipment;

[0018] The digital simulation unit is used to construct a three-dimensional digital model based on the collected engineering geological data and azimuth angle data, simulate the geological characteristics and mechanical operations of the construction site through the three-dimensional digital model, synchronize the actual construction process in real time, and use the simulation results to guide the construction operation;

[0019] The model evaluation unit is used to establish a neural network model, identify the geometric features of cracks in image and video data, evaluate crack indicators, and evaluate blasting effects based on the crack indicators.

[0020] A drilling control unit is used to extract drilling parameters from the construction parameters, drive the drilling machine to perform drilling operations, and monitor the actual drilling effect of the drilling machine during the drilling operation in real time;

[0021] A blasting control unit, used to set blasting parameters of a gas blasting machine according to the drilling depth in the drilling parameters, and drive the gas blasting machine to perform blasting operations;

[0022] The crushing control unit is used to control the blasting machine to perform secondary crushing operations according to the adjusted construction parameters, and coordinate the excavator and the loading and transporting vehicle to perform the transportation task;

[0023] The feedback optimization unit is used to record the construction parameters adjusted and controlled during the actual construction process, and to adjust and improve the construction plan in the future construction process through data accumulation and analysis.

[0024] Furthermore, the digital simulation unit includes an engineering geological model, an engineering machinery model and a simulation engine;

[0025] The engineering geological model is used to establish a three-dimensional model of the ore body where the construction site is located based on the three-dimensional site data and the three-dimensional rock mass data as an engineering geological model;

[0026] The engineering machinery model is used to identify the azimuth angle of the engineering machinery equipment based on the azimuth angle data of the engineering machinery equipment and establish the engineering machinery model of the engineering machinery equipment;

[0027] The simulation engine is used to keep pace with the actual construction process, identify differences in actual construction operations through simulation of the 3D digital model, and dynamically adjust construction parameters.

[0028] Furthermore, in synchronization with the actual construction process, the 3D digital model is used to simulate and emulate the actual construction process, identify the differences in actual construction operations, and dynamically adjust construction parameters including:

[0029] Integrate the engineering geological model and the engineering machinery model to construct a three-dimensional digital model of the construction site. This model is based on real-time updated three-dimensional site data, three-dimensional rock mass data, and the azimuth angle data of engineering machinery and equipment, and is updated synchronously with the actual construction process.

[0030] The simulation cycle of the three-dimensional digital model is set according to the construction stage. Through the deep neural network algorithm, the physical behavior at the future moment is predicted based on the current construction parameters, and the blasting effect during the simulation blasting process is output;

[0031] Compare the simulation results at the future moment with the feedback data collected during the actual construction process. By identifying the differences between the simulation prediction results and the actual construction operations, calculate the comparison difference value, which includes the deviation of crack extension, the difference of rock mass fragmentation size and the deviation of equipment operation trajectory.

[0032] Based on the results of the difference analysis, the optimization algorithm is used to adjust the construction parameters to reduce the error between the simulation and the actual operation. Through multiple iterative optimizations, the optimal construction plan is gradually approached.

[0033] Furthermore, the model evaluation unit includes a data preprocessing subunit, a feature learning subunit, and an evaluation prediction subunit;

[0034] The data preprocessing subunit is used to identify cracks in the image and video data and to mark the structural information of the cracks in the image and video data;

[0035] The feature learning subunit is used to gradually extract the geometric features of the cracks from the annotated structural information through multi-layer convolution and deconvolution operations of the neural network;

[0036] The evaluation and prediction subunit is used to identify the geometric characteristics of cracks through deep learning and model training, predict the development trend of the blasting process as a crack indicator, and optimize the actual construction parameters based on the results of the crack indicators; crack indicators include the number of cracks, fractal dimension, blasting porosity, phase intersection and fragmentation.

[0037] Furthermore, through the multi-layer convolution and deconvolution operations of the neural network, the geometric features of the cracks are gradually extracted from the annotated structural information, including:

[0038] Extracting video frames containing cracks from the image video data as crack images, and performing normalization and denoising processing on the crack images through image preprocessing;

[0039] The size and step length of the convolution kernel are set according to the scale of the crack geometric features. The feature map generated by each convolution layer is defined as the information of different levels of the crack. Through filtering of multiple convolution layers, the local edge, morphology and texture information of the crack in the crack image are gradually extracted as local features.

[0040] Through the deconvolution operation of the deconvolution layer, the local features compressed in the convolution process are spatially restored to restore the three-dimensional spatial structure of the crack. In the process of three-dimensional spatial structure restoration, the output results are characterized as the geometric characteristics of the crack in combination with the annotated geometric information, where the geometric characteristics include crack length, average crack width and crack degree.

[0041] Furthermore, through the deconvolution operation of the deconvolution layer, the local features compressed in the convolution process are spatially restored to restore the three-dimensional spatial structure of the crack. In the process of restoring the three-dimensional spatial structure, the output results are characterized as geometric features of the crack in combination with the annotated geometric information, including:

[0042] Read edge information of the crack image, extract the contours of each edge in the crack image, calculate the perimeter of each contour, and add the perimeters of all contours to obtain the overall perimeter of all cracks in the crack image, and then convert it into the actual length of the crack according to the actual size as the crack length;

[0043] A linear structure element is set, and the cracks in the crack image are filled by dilation operation to obtain the center line of the crack image. The dilated crack image is summed along the column direction to obtain the number of white pixels in each row, and the crack width information is calculated. The actual width of the crack is generated by conversion of the actual size as the average width of the crack.

[0044] Taking pixels as units, the total crack area and the total cross-sectional area in the crack image are counted, and the ratio of the total crack area to the total cross-sectional area is calculated as the crack degree of each crack image.

[0045] Furthermore, the calculation formula of the crack length is:

[0046]

[0047] Where P represents the crack length in the crack image; Represents the Euclidean distance between two adjacent pixels; Represents the line between the first pixel and the last pixel in the contour; (x i ,y i ) represents the i-th pixel in the crack image; n represents the number of pixels at the edge of the crack contour;

[0048] The calculation formula for the crack width is:

[0049]

[0050] In the formula, ω a represents the crack width of the ath row; m represents the number of columns of the crack image; dilateEdge(α, b) represents the pixel value of the crack image at position (a, b) after dilation;

[0051] The calculation formula of crack degree is:

[0052]

[0053] In the formula, r i represents the total crack area in the crack image; Si represents the total area of ​​the crack image; S 1i Indicates the degree of cracking of the image.

[0054] In a second aspect, the present invention further provides a green and intelligent blasting and rock breaking method, the method comprising the following steps:

[0055] S 1. Through the data acquisition module, real-time collection of engineering geological data, azimuth angle data of engineering machinery and equipment, and image and video data in the construction site;

[0056] S2. Set the initial construction parameters of the engineering machinery and equipment, perform the blasting and rock breaking construction tasks, and collect real-time feedback data of the engineering machinery and equipment during the blasting process;

[0057] S3. Based on the collected engineering geological data, a three-dimensional digital model is constructed, which integrates the real-time updated engineering geological data and azimuth angle data to keep in sync with the actual construction process, and adjusts the construction parameters of engineering machinery and equipment through simulation of the three-dimensional digital model;

[0058] S4. By constructing a neural network model, extracting geometric features from image and video data, and setting crack indicators, the blasting effect of blasting construction performed by engineering machinery and equipment is evaluated;

[0059] S5. According to the optimized construction parameters, the blasting machine is controlled to perform secondary crushing, and the excavator and the loading and transporting vehicle are coordinated to perform the transportation task;

[0060] S6. Record the construction parameters adjusted and controlled during the actual construction process, and adjust and improve the construction plan in the future construction process through data accumulation and analysis.

[0061] The beneficial effects of the present invention are:

[0062] 1. Through the collaborative work of the data acquisition module, the intelligent decision-making module and the unmanned engineering machinery collaborative construction module, a new green intelligent blasting and rock breaking system is formed, which can significantly improve construction efficiency, reduce the impact on the environment, and enhance the safety and intelligence level of construction.

[0063] 2. By combining advanced digital simulation technology and deep neural network algorithms, and through the continuous accumulation and analysis of various data including drilling parameters, blasting parameters, engineering machinery data, rock mass characteristic data, etc. during the construction process, the intelligent decision-making module continuously adjusts and improves the construction plan, and gradually realizes the precise optimization of construction parameters, thereby achieving the goal of overall green excavation. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0065] Figure 1 is a principle block diagram of a green intelligent blasting and rock breaking system according to an embodiment of the present invention;

[0066] Figure 2 It is a flow chart of a green and intelligent rock breaking method by blasting according to an embodiment of the present invention.

[0067] Figure numbers: 1. Data acquisition module; 2. Intelligent decision-making module; 3. Unmanned engineering machinery construction module. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0069] See also Figure 1 , provides a green and intelligent blasting and rock breaking system, which includes: a data acquisition module 1, an intelligent decision-making module 2 and an unmanned engineering machinery construction module 3.

[0070] The data acquisition module 1 is used to monitor the construction site and engineering machinery and equipment in real time using a laser radar and a binocular camera to acquire construction site data. The construction site data includes engineering geological data, azimuth angle data of engineering machinery and equipment, and image and video data.

[0071] In the description of the present invention, the data acquisition module 1 includes: a drone monitoring unit (not shown in the figure), a machinery monitoring unit (not shown in the figure) and a central control room (not shown in the figure).

[0072] The drone monitoring unit is used to obtain the three-dimensional site data of the construction site, the azimuth angle data of the construction machinery and equipment, and the image and video data within the construction site by using the laser radar and binocular camera carried by the drone.

[0073] Specifically, the drones are equipped with laser radars and binocular cameras to measure the distance and perform three-dimensional reconstruction of the ore bodies, steps, boulders, pits, and surrounding vegetation to be excavated, obtain three-dimensional site data, and transmit it to the central control room. At the same time, the drones should have the function of autonomous flight path planning, and adjust the flight route in real time based on the three-dimensional terrain data of the construction site and the progress of the construction to ensure coverage of the entire construction area. The task allocation of the drones should be highly matched with the operation progress of the construction machinery. The system will prioritize the dispatch of drones to monitor specific areas based on the operation of the construction machinery and equipment.

[0074] The mechanical monitoring unit is used to obtain three-dimensional rock data and image video data of the rock during blasting operations using the laser radar and binocular camera carried by engineering machinery.

[0075] Specifically, engineering machinery and equipment include drilling machines, gas blasting machines, blasting machines, etc. Through the laser radar and binocular camera equipped with the engineering machinery and equipment, more detailed data on rock types, crack distribution, size and orientation of pits or boulders on the rock surface, etc. can be obtained and transmitted to the central control room in real time as three-dimensional rock data.

[0076] The central control room is used to control drones and engineering machinery equipment to perform data monitoring tasks, and receive monitoring data uploaded in real time by the drone monitoring unit and the machinery monitoring unit.

[0077] Intelligent decision-making module 2 extracts different types of construction parameters from the construction site data, combines historical data, analyzes and evaluates the blasting effect, and optimizes and adjusts the construction parameters and construction plans.

[0078] In the description of the present invention, the intelligent decision-making module 2 includes a parameter setting unit (not shown in the figure), a digital simulation unit (not shown in the figure), a model evaluation unit (not shown in the figure), a drilling control unit (not shown in the figure), a blasting control unit (not shown in the figure), a crushing control unit (not shown in the figure) and a feedback optimization unit (not shown in the figure).

[0079] The parameter setting unit is used to set the initial construction parameters of the engineering machinery equipment.

[0080] Specifically, the parameter setting unit provides instruction parameters to the drilling machine based on construction experience and technical theory, including drilling depth, drilling diameter, hole spacing, row spacing, etc., and the drilling machine performs automatic drilling construction; the gas blasting machine automatically sets parameters such as the gas blasting pressure and gas impact amount according to the drilling depth to perform blasting operations.

[0081] The digital simulation unit is used to construct a three-dimensional digital model based on the collected engineering geological data and azimuth angle data. The three-dimensional digital model is used to simulate the geological characteristics and mechanical operations of the construction site, and the actual construction process is synchronized in real time, and the simulation results are used to guide the construction operations.

[0082] In the description of the present invention, the digital simulation unit includes an engineering geological model (not shown in the figure), an engineering machinery model (not shown in the figure) and a simulation engine (not shown in the figure).

[0083] Among them, the engineering geological model is used to establish a three-dimensional model of the ore body where the construction site is located based on the three-dimensional site data and the three-dimensional rock mass data, as an engineering geological model.

[0084] The engineering machinery model is used to identify the azimuth angle of the engineering machinery equipment based on the azimuth angle data of the engineering machinery equipment and establish the engineering machinery model of the engineering machinery equipment.

[0085] The simulation engine is used to keep pace with the actual construction process, identify differences in actual construction operations through simulation of the 3D digital model, and dynamically adjust construction parameters.

[0086] In the description of the present invention, keeping in sync with the actual construction process, identifying the differences in actual construction operations through simulation of the three-dimensional digital model, and dynamically adjusting the construction parameters include:

[0087] Step S101: Integrate the engineering geological model and the engineering machinery model to form a three-dimensional digital model of the construction site, and integrate the real-time updated three-dimensional site data, three-dimensional rock mass data and the azimuth angle data of the engineering machinery equipment to keep pace with the actual construction process. Through this three-dimensional digital model, the system can monitor the geological and mechanical operation conditions in the construction area in real time, ensuring a high degree of consistency between the virtual model and the actual construction scene.

[0088] Step S102, setting the simulation cycle of the three-dimensional digital model according to the construction stage, using the deep neural network algorithm, based on the current construction parameters, predicting the physical behavior at the future moment, and outputting the blasting effect during the simulation blasting process.

[0089] Specifically, the system sets the simulation cycle of the three-dimensional digital model according to the construction stage. Through the deep neural network algorithm, based on the current construction parameters (such as drilling depth, blasting pressure, mechanical position, etc.), the system can predict the physical behavior at future times and simulate the crack expansion path, rock mass fragmentation degree, equipment operation status and other results during the blasting process. These simulation results provide a basis for predicting subsequent construction decisions.

[0090] Step S103, compare the simulation results at the future moment with the feedback data collected during the actual construction process, and calculate the comparison difference value by identifying the difference between the simulation prediction results and the actual construction operation. The comparison difference value includes the deviation of crack extension, the difference in rock fragmentation size, and the deviation of equipment operation trajectory.

[0091] Step S104: Based on the difference analysis results, the construction parameters are adjusted using an optimization algorithm to reduce the error between the simulation and the actual operation, and the optimal construction plan is gradually approached through multiple iterations of optimization.

[0092] Specifically, the three-dimensional digital model built by the digital simulation unit is used to simulate the geological characteristics and mechanical operations of the construction site. Through the simulation engine, the real-time updated geological data and mechanical dynamics are integrated, and the virtual three-dimensional digital model can not only keep pace with the actual construction process, but also predict the physical behavior at future moments based on the current data, such as crack expansion after blasting and rock mass fragmentation. In the prediction process, the deep neural network simulates the blasting effect at future moments based on historical data and current construction parameters (such as blasting pressure, drilling depth, etc.), including physical behaviors such as the expansion path of cracks and rock mass fragmentation. The simulation results are used to guide the next construction operation. In order to ensure the accuracy of the model prediction, the simulation results will be compared with the real-time feedback data after the actual blasting. The real-time feedback data includes the actual number of cracks after blasting, the fragmentation size, and other actual operation data corresponding to the simulation results. By comparing the simulation prediction with the actual results, the intelligent decision-making module can identify the differences, analyze the causes, and dynamically adjust the construction parameters.

[0093] Finally, the system can adjust the construction parameters using optimization algorithms based on the difference analysis results. Specifically, the system automatically adjusts key parameters such as drilling depth, blasting pressure, and hole spacing to reduce the error between simulation and actual operation. Through multiple iterations of optimization, the system gradually approaches the optimal construction plan to ensure the accuracy and efficiency of subsequent construction operations.

[0094] The model evaluation unit is used to establish a neural network model, identify the geometric features of cracks in image and video data, evaluate crack indicators, and evaluate blasting effects based on the crack indicators.

[0095] In the description of the present invention, the model evaluation unit includes a data preprocessing subunit (not shown in the figure), a feature learning subunit (not shown in the figure) and an evaluation prediction subunit (not shown in the figure).

[0096] The data preprocessing subunit is used to identify cracks in the image and video data and to mark the structural information of the cracks in the image and video data.

[0097] Specifically, the data preprocessing subunit annotates the collected image or video data. The annotated content includes basic structural information such as the crack boundary, morphology, opening, trace length, and inclination. These geometric features are the basis for evaluating indicators such as the number of cracks, fractal dimension, bursting fracture degree, phase intersection, and fragmentation. The annotated crack boundary and morphological information can help the network learn the spatial distribution and expansion trend of the cracks; characteristics such as opening and inclination can reflect the mechanical properties of the cracks. Through this annotated information, the neural network can accurately learn the geometric characteristics of the cracks;

[0098] The feature learning subunit is used to gradually extract the geometric features of the cracks from the annotated structural information through multi-layer convolution and deconvolution operations of the neural network.

[0099] Specifically, feature learning is to gradually extract the geometric features of cracks from the labeled data through multi-layer convolution and deconvolution operations of the neural network. Specifically, the convolution layer extracts the local features of the cracks, such as edges and openings, and the deconvolution layer is used to restore the spatial structure of the cracks to ensure the integrity of the geometric features. The goal of feature learning is to evaluate crack indicators such as the number of cracks, fractal dimension, burst porosity, phase intersection, and fragmentation by identifying these geometric features.

[0100] In the description of the present invention, the geometric features of the cracks are gradually extracted from the annotated structural information through multi-layer convolution and deconvolution operations of the neural network, including:

[0101] Step S201: extracting video frames containing cracks from image video data as crack images, and performing normalization and denoising on the crack images through image preprocessing.

[0102] Step S202: setting the size and step length of the convolution kernel according to the scale of the crack geometric features, defining the feature map generated by each convolution layer as information of different levels of the crack, and gradually extracting the local edge, morphology and texture information of the crack in the crack image as local features through filtering of multiple convolution layers.

[0103] Step S203: Through the deconvolution operation of the deconvolution layer, the local features compressed in the convolution process are spatially restored to restore the three-dimensional spatial structure of the crack. In the process of restoring the three-dimensional spatial structure, the output results are characterized as geometric features of the crack in combination with the annotated geometric information, where the geometric features include crack length, average crack width and crack degree.

[0104] In the description of the present invention, the local features compressed in the convolution process are spatially restored through the deconvolution operation of the deconvolution layer to restore the three-dimensional spatial structure of the crack. In the process of restoring the three-dimensional spatial structure, the output results are characterized as geometric features of the crack in combination with the annotated geometric information, including:

[0105] Read edge information of the crack image, extract the contours of each edge in the crack image, calculate the perimeter of each contour, and add the perimeters of all contours to obtain the overall perimeter of all cracks in the crack image, and then convert it into the actual length of the crack according to the actual size as the crack length;

[0106] A linear structure element is set, and the cracks in the crack image are filled by dilation operation to obtain the center line of the crack image. The dilated crack image is summed along the column direction to obtain the number of white pixels in each row, and the crack width information is calculated. The actual width of the crack is generated by conversion of the actual size as the average width of the crack.

[0107] Taking pixels as units, the total crack area and the total cross-sectional area in the crack image are counted, and the ratio of the total crack area to the total cross-sectional area is calculated as the crack degree of each crack image.

[0108] Specifically, the calculation formula for the crack length is:

[0109]

[0110] Where P represents the crack length in the crack image; Represents the Euclidean distance between two adjacent pixels; Represents the line between the first pixel and the last pixel in the contour; (x i ,y i ) represents the i-th pixel in the crack image; n represents the number of pixels at the edge of the crack contour;

[0111] The calculation formula for the crack width is:

[0112]

[0113] In the formula, ω a represents the crack width of the ath row; m represents the number of columns of the crack image; dilateEdge(α, b) represents the pixel value of the crack image at position (a, b) after dilation;

[0114] The calculation formula of crack degree is:

[0115]

[0116] In the formula, r i represents the total crack area in the crack image; S i represents the total area of ​​the crack image; S 1i Indicates the degree of cracking of the image.

[0117] The evaluation and prediction subunit is used to identify the geometric characteristics of cracks through deep learning and model training, predict the development trend of the blasting process, use it as a crack indicator, and optimize the actual construction parameters based on the results of the crack indicators. Crack indicators include the number of cracks, fractal dimension, blasting porosity, phase intersection, and fragmentation.

[0118] Specifically, the model evaluation unit uses the Unet recognition method, including the backbone feature extraction part, the enhanced feature extraction part and the prediction part, to identify and evaluate the cracks formed during the blasting rock breaking process. The backbone feature extraction part of the Unet model is used to extract the preliminary features of the cracks in the image or video, including the crack opening, trace length, inclination, etc.

[0119] The enhanced feature extraction part further conducts deep learning and optimization on these preliminary features. Through the learning of a large amount of training data, it can autonomously identify the geometric parameters of cracks and predict the development trend of these cracks during the blasting process, including the prediction of crack extension path, crack growth rate and crack intersection degree. In the prediction part, the neural network can not only identify the basic geometric characteristics of cracks, but also propose a series of crack indicators based on these characteristics, including the number of cracks, fractal dimension (used to characterize the complexity and irregularity of cracks), blasting porosity (volume proportion of cracks in the rock mass after blasting), phase intersection (reflecting the degree of mutual intersection between cracks) and fragmentation (used to measure the size distribution of rock fragments formed after blasting). The calculation method of these crack indicators depends on the learning mechanism of the Unet model.

[0120] It should be noted that geometric features are the basis for further calculation and derivation of crack indicators. The following is a specific example of crack indicator identification:

[0121] 1. Number of cracks: By identifying the crack boundaries in the image, the system uses the two parameters of opening and trace length to distinguish independent cracks. The system identifies multiple continuous crack edges, determines the starting and ending points of the cracks, and calculates the total number of cracks. For example, when there are multiple independent values ​​of opening and trace length in a certain area, the system identifies multiple independent cracks.

[0122] 2. Fractal dimension: Fractal dimension is used to describe the complexity and irregularity of cracks. It is calculated based on the length of the crack and the complexity of the crack shape. For example, if a crack has a long trace and a more curved and irregular shape, the fractal dimension of the crack is higher; on the contrary, regular and straight cracks have a lower fractal dimension.

[0123] 3. Explosion porosity: Based on the opening and trace length, the system calculates the total volume percentage of the cracks (i.e., the crack degree). For example, if the crack opening in a certain area is large and the trace length is long, the explosion porosity in this area will be higher, reflecting that the volume percentage of the rock mass cracks after blasting is large and the rock mass is more fragmented.

[0124] 4. Intersection degree: By identifying the strike and dip of multiple fractures, the system calculates the number of intersections and angles between fractures. For example, when the trace lengths of two fractures tend to intersect and the dip angles are close, the system identifies them as fracture intersections. The higher the intersection degree, the more fractures intersect each other and the smaller the rock mass fragments.

[0125] 5. Fragmentation: Combined with the crack trace length, intersection and opening, the system predicts the size distribution of rock fragments formed after blasting. If the crack opening is large, the trace length is long and there are many intersections, the system predicts that the rock mass will be broken into smaller fragments; conversely, if the cracks are sparse and there are few intersections, the fragmentation is large and the fragment size is large.

[0126] Through the recognition and calculation of these specific geometric features, the intelligent decision-making module can automatically generate and evaluate the above-mentioned crack indicators, and adjust the construction parameters of the gas blasting machine (such as gas blasting pressure, drilling depth, hole spacing, etc.) based on the evaluation results to optimize the blasting effect and achieve the optimization of construction parameters.

[0127] The drilling control unit is used to extract the drilling parameters in the construction parameters, drive the drilling machine to perform the drilling operation, and monitor the actual drilling effect of the drilling machine during the drilling operation in real time.

[0128] The blasting control unit is used to set the blasting parameters of the gas blasting machine according to the drilling depth in the drilling parameters, and drive the gas blasting machine to perform the blasting operation.

[0129] The crushing control unit is used to control the blasting machine to perform secondary crushing operations according to the adjusted construction parameters, and coordinate the excavator and the loading and transporting vehicle to perform transportation tasks.

[0130] The feedback optimization unit is used to record the construction parameters adjusted and controlled during the actual construction process, and to adjust and improve the construction plan in the future construction process through data accumulation and analysis.

[0131] The unmanned engineering machinery construction module 3 is used to execute construction tasks, call and control engineering machinery equipment to implement blasting construction operations, and conduct unmanned green and intelligent blasting construction.

[0132] Specifically, the unmanned engineering machinery construction module 3 is mainly composed of drilling machines, gas blasting machines, blasting machines, excavators and loading and transport vehicles, etc., which are used to perform green and intelligent blasting construction. Combining advanced digital simulation technology and deep neural network algorithms, through the continuous accumulation and analysis of various data including drilling parameters, blasting parameters, engineering machinery data, rock mass characteristic data, etc. during the construction process, the intelligent decision-making module continuously adjusts and improves the construction plan, gradually realizes the precise optimization of construction parameters, and thus achieves the goal of overall green excavation.

[0133] See also Figure 2 , and also provides a green and intelligent blasting and rock breaking method, the method comprising the following steps:

[0134] S 1. Through the data acquisition module, the engineering geological data, the azimuth angle data of the engineering machinery and equipment, and the image and video data in the construction site are collected in real time.

[0135] S2. Set the initial construction parameters of the engineering machinery and equipment, perform the blasting and rock breaking construction tasks, and collect real-time feedback data of the engineering machinery and equipment during the blasting process.

[0136] S3. Based on the collected engineering geological data, a three-dimensional digital model is constructed, integrating the real-time updated engineering geological data and azimuth angle data, keeping in sync with the actual construction process, and adjusting the construction parameters of the engineering machinery and equipment through simulation of the three-dimensional digital model.

[0137] In the description of the present invention, based on the collected engineering geological data, a three-dimensional digital model is constructed, the engineering geological data updated in real time is integrated with the azimuth angle data, and the actual construction process is synchronized. Through the simulation of the three-dimensional digital model, the construction parameters of the engineering machinery equipment are adjusted, which includes the following steps:

[0138] S3 1. Based on the three-dimensional site data and three-dimensional rock mass data in the engineering geological data, an engineering geological model of the ore body where the construction site is located is established, and based on the azimuth angle data of the engineering machinery and equipment, the azimuth angle of the engineering machinery and equipment is identified, and an engineering machinery model of the engineering machinery and equipment is established.

[0139] S32. Integrate the engineering geological model and the engineering machinery model to form a three-dimensional digital model of the construction site, and integrate the real-time updated three-dimensional site data, three-dimensional rock mass data and azimuth angle data of the engineering machinery equipment to keep in sync with the actual construction process.

[0140] S33. Set the simulation cycle of the three-dimensional digital model, and use the deep neural network to predict the physical behavior at a future moment based on the construction parameters at the current moment, and simulate the blasting effect at the future moment.

[0141] S34. Compare the simulation results at the future moment with the actual feedback data, identify the difference between the actual construction operation and the simulation results, and dynamically adjust the construction parameters according to the comparison difference value.

[0142] S4. By constructing a neural network model, extracting geometric features from image and video data, and setting crack indicators, the blasting effect of engineering machinery and equipment performing blasting construction is evaluated.

[0143] In the description of the present invention, by constructing a neural network model, extracting geometric features in image video data, and setting crack indicators, evaluating the blasting effect of blasting construction performed by engineering machinery and equipment includes the following steps:

[0144] S4 1. Acquire image and video data, identify cracks in the image and video data, and annotate structural information of the cracks in the image and video data.

[0145] S42. Using the convolution layer of the neural network, extract the local features of the cracks in the image and video data. Using the deconvolution layer, restore the spatial structure of the cracks to obtain complete geometric features, and the geometric features include the boundary, shape, opening, trace length and inclination of the cracks.

[0146] S43. Through deep learning and model training, the geometric characteristics of cracks are identified, the development trend of the blasting process is predicted, and the crack indicators are used as crack indicators. Based on the results of the crack indicators, the actual construction parameters are optimized. The crack indicators include the number of cracks, fractal dimension, blasting porosity, phase intersection and fragmentation.

[0147] S5. According to the optimized construction parameters, the blasting machine is controlled to perform secondary crushing, and the excavator and the loading and transporting vehicle are coordinated to perform the transportation task.

[0148] S6. Record the construction parameters adjusted and controlled during the actual construction process, and adjust and improve the construction plan in the future construction process through data accumulation and analysis.

[0149] In summary, with the help of the above technical solution of the present invention, through the collaborative work of the data acquisition module, the intelligent decision-making module and the unmanned engineering machinery collaborative construction module, a new green intelligent blasting and rock breaking system is formed, which can significantly improve the construction efficiency, reduce the impact on the environment, and enhance the safety and intelligence level of construction. By combining advanced digital simulation technology and deep neural network algorithms, through the continuous accumulation and analysis of various data including drilling parameters, blasting parameters, engineering machinery data, rock mass characteristic data, etc. during the construction process, the intelligent decision-making module continuously adjusts and improves the construction plan, gradually realizes the precise optimization of construction parameters, and thus achieves the goal of overall green excavation.

[0150] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

Claims

1. A green and intelligent blasting and rock breaking system, characterized in that: The system includes: data acquisition module, intelligent decision-making module and unmanned engineering machinery construction module; The data acquisition module is used to monitor the construction site and engineering machinery and equipment in real time using a laser radar and a binocular camera to acquire construction site data, wherein the construction site data includes engineering geological data, azimuth angle data of engineering machinery and equipment, and image and video data; The intelligent decision-making module extracts different types of construction parameters from the construction site data, combines historical data, analyzes and evaluates the blasting effect, and optimizes and adjusts the construction parameters and construction plans; The unmanned engineering machinery construction module is used to execute construction tasks, call and control engineering machinery equipment to implement blasting construction operations, and conduct unmanned green and intelligent blasting construction.

2. A green and intelligent blasting and rock breaking system according to claim 1, characterized in that: The data acquisition module includes: a drone monitoring unit, a machinery monitoring unit and a central control room; The drone monitoring unit is used to obtain three-dimensional site data of the construction site, azimuth angle data of the construction machinery and equipment, and image and video data in the construction site by using the laser radar and binocular camera carried by the drone; The mechanical monitoring unit is used to obtain three-dimensional rock mass data and image video data of the rock mass during the blasting operation by using the laser radar and binocular camera carried by the engineering machinery; The central control room is used to control the UAV and engineering machinery equipment to perform data monitoring tasks, and receive the monitoring data uploaded in real time by the UAV monitoring unit and the machinery monitoring unit.

3. A green and intelligent blasting and rock breaking system according to claim 2, characterized in that: The intelligent decision-making module includes a parameter setting unit, a digital simulation unit, a model evaluation unit, a drilling control unit, a blasting control unit, a crushing control unit and a feedback optimization unit; Wherein, the parameter setting unit is used to set the initial construction parameters of the engineering machinery equipment; The digital simulation unit is used to construct a three-dimensional digital model based on the collected engineering geological data and azimuth angle data, simulate the geological characteristics and mechanical operations of the construction site through the three-dimensional digital model, synchronize the actual construction process in real time, and use the simulation results to guide the construction operation; The model evaluation unit is used to establish a neural network model, identify crack geometric features in image and video data, evaluate crack indicators, and evaluate blasting effects based on the crack indicators. The drilling control unit is used to extract drilling parameters from the construction parameters, drive the drilling machine to perform the drilling operation, and monitor the actual drilling effect of the drilling machine during the drilling operation in real time; The blasting control unit is used to set the blasting parameters of the gas blasting machine according to the drilling depth in the drilling parameters, and drive the gas blasting machine to perform the blasting operation; The crushing control unit is used to control the blasting machine to perform secondary crushing operations according to the adjusted construction parameters, and coordinate the excavator and the loading and transporting vehicle to perform the transportation task; The feedback optimization unit is used to record the construction parameters adjusted and controlled during the actual construction process, and adjust and improve the construction plan in the future construction process through data accumulation and analysis.

4. A green and intelligent blasting and rock breaking system according to claim 3, characterized in that: The digital simulation unit includes an engineering geological model, an engineering machinery model and a simulation engine; The engineering geological model is used to establish a three-dimensional model of the ore body where the construction site is located based on the three-dimensional site data and the three-dimensional rock mass data, as an engineering geological model; The engineering machinery model is used to identify the azimuth angle of the engineering machinery equipment based on the azimuth angle data of the engineering machinery equipment and establish the engineering machinery model of the engineering machinery equipment; The simulation engine is used to keep in sync with the actual construction process, identify the differences in actual construction operations through simulation of the three-dimensional digital model, and dynamically adjust the construction parameters.

5. A green and intelligent blasting and rock breaking system according to claim 4, characterized in that: The above-mentioned synchronization with the actual construction process, identifying the differences in actual construction operations through simulation of the three-dimensional digital model, and dynamically adjusting the construction parameters include: Integrate the engineering geological model with the engineering machinery model to construct a three-dimensional digital model of the construction site, which is based on real-time updated three-dimensional site data, three-dimensional rock mass data and azimuth angle data of engineering machinery and equipment, and is updated synchronously with the actual construction process; The simulation cycle of the three-dimensional digital model is set according to the construction stage. Through the deep neural network algorithm, the physical behavior at the future moment is predicted based on the current construction parameters, and the blasting effect during the simulation blasting process is output; Compare the simulation results at the future moment with the feedback data collected during the actual construction process, and calculate the comparison difference value by identifying the difference between the simulation prediction results and the actual construction operation, wherein the comparison difference value includes the deviation of crack extension, the difference of rock mass fragmentation size and the deviation of equipment operation trajectory; Based on the results of the difference analysis, the optimization algorithm is used to adjust the construction parameters to reduce the error between the simulation and the actual operation. Through multiple iterative optimizations, the optimal construction plan is gradually approached.

6. A green and intelligent blasting and rock breaking system according to claim 3, characterized in that: The model evaluation unit includes a data preprocessing subunit, a feature learning subunit and an evaluation prediction subunit; The data preprocessing subunit is used to identify cracks in the image and video data, and to mark structural information of the cracks in the image and video data; The feature learning subunit is used to gradually extract the geometric features of the cracks from the annotated structural information through multi-layer convolution and deconvolution operations of the neural network; The evaluation and prediction subunit is used to identify the geometric characteristics of cracks through deep learning and model training, predict the development trend of the blasting process as a crack indicator, and optimize the actual construction parameters based on the results of the crack indicator; the crack indicators include the number of cracks, fractal dimension, blasting porosity, phase intersection and fragmentation.

7. A green and intelligent blasting and rock breaking system according to claim 6, characterized in that: The multi-layer convolution and deconvolution operations of the neural network are used to gradually extract the geometric features of the cracks from the annotated structural information, including: Extracting video frames containing cracks from the image video data as crack images, and performing normalization and denoising processing on the crack images through image preprocessing; The size and step length of the convolution kernel are set according to the scale of the crack geometric features, and the feature map generated by each convolution layer is defined as the information of different levels of the crack. Through filtering of multiple convolution layers, the local edge, morphology and texture information of the crack in the crack image are gradually extracted as local features; Through the deconvolution operation of the deconvolution layer, the local features compressed in the convolution process are spatially restored to restore the three-dimensional spatial structure of the crack. In the process of restoring the three-dimensional spatial structure, the output results are characterized as geometric features of the crack in combination with the annotated geometric information, wherein the geometric features include crack length, average crack width and crack degree.

8. The green and intelligent blasting and rock breaking system according to claim 6 is characterized in that: The deconvolution operation of the deconvolution layer is used to spatially restore the local features compressed in the convolution process, and restore the three-dimensional spatial structure of the crack. In the process of restoring the three-dimensional spatial structure, the output results are characterized as geometric features of the crack in combination with the annotated geometric information, including: Read edge information of the crack image, extract the contours of each edge in the crack image, calculate the perimeter of each contour, and add the perimeters of all contours to obtain the overall perimeter of all cracks in the crack image, and then convert it into the actual length of the crack according to the actual size as the crack length; A linear structure element is set, and the cracks in the crack image are filled by dilation operation to obtain the center line of the crack image. The dilated crack image is summed along the column direction to obtain the number of white pixels in each row, and the crack width information is calculated. The actual width of the crack is generated by conversion of the actual size as the average width of the crack. Taking pixels as units, the total crack area and the total cross-sectional area in the crack image are counted, and the ratio of the total crack area to the total cross-sectional area is calculated as the crack degree of each crack image.

9. The green and intelligent blasting and rock breaking system according to claim 6, characterized in that: The calculation formula of the crack length is: Where P represents the crack length in the crack image; Represents the Euclidean distance between two adjacent pixels; Represents the line between the first pixel and the last pixel in the contour; (x i ,y i ) represents the i-th pixel in the crack image; n represents the number of pixels at the edge of the crack contour; The calculation formula of the crack width is: In the formula, ω a represents the crack width of row a; m represents the number of columns of the crack image; dilateEdge(α, b) represents the pixel value of the crack image at position (a, b) after dilation; The calculation formula of the crack degree is: In the formula, r i represents the total crack area in the crack image; S i represents the total area of ​​the crack image; S 1i Indicates the degree of cracking of the image.

10. A green and intelligent blasting and rock breaking method, using the green and intelligent blasting and rock breaking system according to any one of claims 1 to 9, characterized in that: The method comprises the following steps: S1. Through the data acquisition module, real-time collection of engineering geological data, azimuth angle data of engineering machinery and equipment, and image and video data in the construction site; S2. Set the initial construction parameters of the engineering machinery and equipment, perform the blasting and rock breaking construction tasks, and collect real-time feedback data of the engineering machinery and equipment during the blasting process; S3. Based on the collected engineering geological data, a three-dimensional digital model is constructed, the engineering geological data and the azimuth angle data updated in real time are integrated, the model is synchronized with the actual construction process, and the construction parameters of the engineering machinery and equipment are adjusted through simulation of the three-dimensional digital model; S4. By constructing a neural network model, extracting geometric features from image and video data, and setting crack indicators, the blasting effect of blasting construction performed by engineering machinery and equipment is evaluated; S5. According to the optimized construction parameters, the blasting machine is controlled to perform secondary crushing, and the excavator and the loading and transporting vehicle are coordinated to perform the transportation task; S6. Record the construction parameters adjusted and controlled during the actual construction process, and adjust and improve the construction plan in the future construction process through data accumulation and analysis.

Images