A method, system, electronic device and medium for determining tunnel blasting parameters

Abstract

The present invention discloses a method, system, electronic device and medium for determining tunnel blasting parameters, and relates to the field of tunnel blasting. The method comprises obtaining the current face image and historical operation data of the tunnel to be blasted; determining the blasting geometric parameters of the blasthole according to the current face image and the previous cycle historical operation data, using the blasting geometric parameter intelligent selection model; controlling the rock drilling trolley to drill according to the blasting geometric parameters of the blasthole, and recording the drilling operation information; determining the surrounding rock information according to the drilling operation information and the previous cycle historical operation data, using the surrounding rock information perception model; further determining the charge amount and the blasting time interval according to the previous cycle historical operation information and the surrounding rock information through the charge amount and the detonation parameter prediction model, and then performing simulation, adjusting the determined charge amount and blasting time interval, and determining the optimal charge amount and the optimal blasting time interval. The present invention improves the accuracy of tunnel blasting parameter optimization and further improves the tunnel blasting efficiency.

Description

Technical Field

[0001] The present invention relates to the field of tunnel blasting, and in particular to a method, system, electronic equipment and medium for determining tunnel blasting parameters. Background Art

[0002] As China's transportation network continues to be built, transportation infrastructure continues to expand to complex geological areas such as mountainous areas. Transportation projects, especially tunnel projects, are gradually developing in the direction of length, size and depth. In order to shorten the length of the route, it is often necessary to dig tunnels in the mountains. At present, there are common methods such as drilling and blasting, TBM, and shield method in tunnel construction. However, drilling and blasting is still one of the mainstream methods due to its construction flexibility and economy. The key to drilling and blasting is blasting design. The selection of blasting parameters such as the main blasthole spacing, blasthole arrangement, depth, angle, and charge amount in blasting design is an important aspect.

[0003] Blasting technology, as an important means in tunnel construction, plays an indispensable role. Reasonable optimization of tunnel blasting scheme design can not only improve production efficiency, but also help improve the safe production environment of tunnels. For different geological environments, choosing reasonable blasting parameters is the key to safe and efficient production. Therefore, the research on blasting parameter optimization is a topic of great concern. Tunnel blasting design mainly adopts empirical method or engineering analogy method for design. The optimization of construction site mostly adopts feedback optimization of blasting effect. With the improvement of computer technology, the blasting scheme is optimized through numerical simulation. However, these methods often face problems such as small data, high cost and poor efficiency. With the rapid development of tunnel construction machinery and equipment and artificial intelligence technology, the optimization selection of blasting parameters has gradually been combined with intelligent algorithms and simulation in recent years, which has also become a trend in the optimization of blasting parameters in mechanized construction tunnels.

[0004] The blasting environment of mountain tunnels is complex. If the blasting parameters are not selected properly and the blasting quality deviates, it will lead to low utilization of tunnel blastholes, serious over-excavation and under-excavation of surrounding rock, and increase the difficulty of later support, which will not only affect the tunnel excavation efficiency and tunnel construction progress, but also increase production costs. In order to control the quality of tunnel blasting excavation, it is necessary to optimize and control the blasting parameters and blasting effects.

[0005] There are many ways to optimize blasting parameters. Traditional methods include engineering analogy and empirical formula optimization, as well as on-site blasting test optimization, which often require a lot of resources to obtain a lot of data, high cost, low efficiency, and unstable optimization results. Therefore, in order to improve production efficiency and reduce costs, we choose to combine artificial intelligence and simulation to optimize blasting parameters.

[0006] The invention patent with Chinese patent number CN20221034239.5 discloses a method for matching and optimizing blasting parameters according to drilling parameters. Drilling parameters and tunnel section parameters are input, and the expert system automatically matches blasting parameters, and optimizes blasting parameters according to overbreak and underbreak parameters after blasting. The expert system is trained through the following steps: obtaining a training set, which includes the collected tunnel section parameters, drilling parameters and overbreak and underbreak parameters; inputting the training set into the constructed expert system, and constructing a mapping relationship between tunnel section parameters, drilling parameters and overbreak and underbreak parameters and blasting methods, blasthole layout positions and charge amounts. The blasting scheme is automatically matched according to the tunnel section parameters, the drilling parameters of the drilling rig drill and the overbreak and underbreak parameters of the upper cycle. After the blasting scheme is implemented, the lower cycle blasting parameters can be optimized according to the overbreak and underbreak scanning results, ensuring the adaptability of the blasting scheme to the current surrounding rock and geological environment, reducing the overbreak and underbreak volume, reducing the processing cost, and improving the construction speed and quality.

[0007] The invention patent with Chinese patent number CN20221099927.8 discloses a method for dynamic design of blasting parameters based on drilling parameters. The method includes dividing the surrounding rock grade area of ​​the tunnel section for this blasting, adjusting the blasting parameter design of the section for this cycle, drilling, obtaining the drilling data of the section for this cycle, and then comparing it with the drilling data of the previous cycle, dynamically adjusting the surrounding rock grade zoning, charging and networking design parameters of the tunnel section for this cycle, and then charging, networking and detonating. The beneficial effect of this patent is that by dynamically adjusting the blasting parameters through drilling parameters, reasonable blasting parameters are selected according to the actual situation on site, so that the designed blasting parameters are more in line with the actual situation on site, ensuring the rationality, scientificity and accuracy of the selected blasting parameters, and more accurately controlling the accuracy of the excavation contour line, reducing over-excavation and under-excavation, and reducing the subsequent tunnel section processing costs.

[0008] However, the above method only optimizes the blasting parameters once through drilling data, etc., without testing and simulating the obtained blasting parameters to deduce the blasting effect, thereby optimizing the blasting parameters twice; the determination and classification of the surrounding rock grade is only based on drilling information, without considering the operation information and blasting effect information of the previous cycle; the blasting parameters are not trained and predicted based on the mechanical parameters of the surrounding rock, etc. Therefore, the above method only optimizes once, without considering the influence of other information, which will lead to inaccurate optimization of blasting parameters and unsatisfactory blasting effect. Summary of the invention

[0009] The object of the present invention is to provide a method, system, electronic equipment and medium for determining tunnel blasting parameters, so as to improve the accuracy of tunnel blasting parameter optimization.

[0010] To achieve the above object, the present invention provides the following solutions:

[0011] A method for determining tunnel blasting parameters, comprising:

[0012] Acquire the current face image and historical operation data of the tunnel to be blasted; the historical operation data includes tunnel geometry information, face mechanics information, face scanning image information, drilling rig operation information and blasting effect information;

[0013] According to the current face image and the previous cycle historical operation data, the blasting geometric parameters of the blasthole are determined by using the blasting geometric parameter intelligent selection model; the blasting geometric parameters include blasthole spacing, blasthole arrangement and blasthole angle; the blasting geometric parameter intelligent selection model is obtained by training the first neural network model using the first training data set; the first training data set includes the training face image, the training historical operation data and the corresponding blasting geometric parameters; the first neural network model is a deep learning neural network model;

[0014] Controlling the drilling rig to drill according to the blasting geometric parameters of the blasthole and recording drilling operation information; the drilling operation information includes mechanical impact pressure, rotation pressure, jacking pressure, water pressure, water flow, drilling speed, drilling position and drilling depth;

[0015] According to the drilling operation information and the previous cycle historical operation data, the surrounding rock information is determined by using the surrounding rock information perception model; the surrounding rock information includes surrounding rock mechanical parameters, surrounding rock types and rock mass structural surface information; wherein the surrounding rock information perception model is obtained by training the second neural network model using the second training data set; the second training data set includes training drilling operation information, training historical operation data and corresponding surrounding rock information; the second neural network model is a hybrid model of LSTM-CNN-deep neural network model;

[0016] According to the surrounding rock information, the historical operation data of the previous cycle and the blasting geometric parameters, the charge amount and the blasting time interval are determined by using the charge amount and detonation parameter prediction model; wherein the charge amount and detonation parameter prediction model is obtained by training a third neural network model using a third training data set; the third training data set includes surrounding rock information for training, historical operation data for training, blasting geometric parameters for training and corresponding charge amounts and blasting time intervals; the third neural network model is a deep learning neural network model;

[0017] According to the charge amount and the blasting time interval, a blasting simulation is performed using a blasting effect deduction model to determine simulated blasting effect information; the blasting effect information includes over-break, under-break, rock mass fragmentation degree and blasthole utilization rate; wherein the blasting effect deduction model is established using LS-DYNA software, with charge amount and blasting time interval as input and blasting effect information as output;

[0018] The charge amount and the blasting time interval are adjusted according to the simulated blasting effect information to determine the optimal charge amount and the optimal blasting time interval.

[0019] Optionally, adjusting the charge amount and the blasting time interval according to the simulated blasting effect information to determine an optimal charge amount and an optimal blasting time interval specifically includes:

[0020] Determine whether the simulated blasting effect information meets the blasting standard and obtain the judgment result;

[0021] If the judgment result is yes, blasting the tunnel to be blasted according to the charge amount and the blasting time interval;

[0022] If the judgment result is no, the charge amount and the blasting time interval are updated, and the step of "performing blasting simulation according to the charge amount and the blasting time interval using a blasting effect deduction model to determine simulated blasting effect information" is returned until the simulated blasting effect information meets the blasting standard.

[0023] Optionally, blasting the tunnel to be blasted according to the charge amount and the blasting time interval, and then further comprising:

[0024] Record blasting effect information;

[0025] The blasting effect information, blasting geometric parameters of the blasthole, drilling operation information, surrounding rock information, charge amount and blasting time interval are updated to the historical operation data.

[0026] A tunnel blasting parameter determination system, comprising:

[0027] A data acquisition module is used to acquire the current face image and historical operation data of the tunnel to be blasted; the historical operation data includes tunnel geometry information, face mechanics information, face scanning image information, drilling rig operation information and blasting effect information;

[0028] A blasthole parameter determination module is used to determine the blasting geometric parameters of the blasthole according to the current face image and the previous cycle historical operation data, using the blasting geometric parameter intelligent selection model; the blasting geometric parameters include blasthole spacing, blasthole arrangement and blasthole angle; the blasting geometric parameter intelligent selection model is obtained by training the first neural network model using the first training data set; the first training data set includes the training face image, the training historical operation data and the corresponding blasting geometric parameters; the first neural network model is a deep learning neural network model;

[0029] A drilling module is used to control the drilling rig to drill according to the blasting geometric parameters of the blasthole and record drilling operation information; the drilling operation information includes mechanical impact pressure, rotation pressure, jacking pressure, water pressure, water flow, drilling speed, drilling position and drilling depth;

[0030] A surrounding rock information determination module is used to determine the surrounding rock information according to the drilling operation information and the previous cycle historical operation data using a surrounding rock information perception model; the surrounding rock information includes surrounding rock mechanical parameters, surrounding rock types and rock mass structural surface information; wherein the surrounding rock information perception model is obtained by training the second neural network model using a second training data set; the second training data set includes training drilling operation information, training historical operation data and corresponding surrounding rock information; the second neural network model is a hybrid model of LSTM-CNN-deep neural network model;

[0031] A blasting parameter determination module, for determining the charge amount and blasting time interval according to the surrounding rock information, the previous cycle historical operation data and the blasting geometric parameters, using the charge amount and detonation parameter prediction model; wherein the charge amount and detonation parameter prediction model is obtained by training a third neural network model using a third training data set; the third training data set includes surrounding rock information for training, historical operation data for training, blasting geometric parameters for training and corresponding charge amounts and blasting time intervals; the third neural network model is a deep learning neural network model;

[0032] A blasting simulation module is used to perform blasting simulation according to the charge amount and the blasting time interval using a blasting effect deduction model to determine simulated blasting effect information; the blasting effect information includes over-break, under-break, rock mass fragmentation degree and blasthole utilization rate; wherein the blasting effect deduction model is established using LS-DYNA software with charge amount and blasting time interval as input and blasting effect information as output;

[0033] The blasting parameter adjustment module is used to adjust the charge amount and the blasting time interval according to the simulated blasting effect information to determine the optimal charge amount and the optimal blasting time interval. An electronic device includes: a memory and a processor, the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the above-mentioned tunnel blasting parameter determination method.

[0034] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for determining tunnel blasting parameters is implemented.

[0035] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0036] The tunnel blasting parameter determination method, system, electronic device and medium of the present invention obtain the current face image and historical operation data of the tunnel to be blasted; determine the blasting geometric parameters of the blasthole according to the current face image and the previous cycle historical operation data, using the blasting geometric parameter intelligent selection model; control the rock drilling trolley to drill according to the blasting geometric parameters of the blasthole, and record the drilling operation information; according to the drilling operation information and the previous cycle historical operation data, use the surrounding rock information perception model and the charge amount and detonation parameter prediction model to further determine the charge amount and blasting time interval, and then perform simulation to adjust the determined charge amount and blasting time interval to determine the optimal charge amount and the optimal blasting time interval. The present invention improves the accuracy of tunnel blasting parameter optimization and further improves tunnel blasting efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0038] Figure 1 A flow chart of a method for determining tunnel blasting parameters provided by the present invention;

[0039] Figure 2 A flow chart of the tunnel blasting parameter determination method of the present invention in practical application;

[0040] Figure 3 is the structural diagram of the first neural network model;

[0041] Figure 4 is the structural diagram of the second neural network model;

[0042] Figure 5 It is the structural diagram of the third neural network model;

[0043] Figure 6 This is a diagram of the neural network DNN layer structure. DETAILED DESCRIPTION

[0044] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] The object of the present invention is to provide a method, system, electronic equipment and medium for determining tunnel blasting parameters, so as to improve the accuracy of tunnel blasting parameter optimization.

[0046] In the existing technology, the tunnel drilling and blasting method is flexible, low-cost, and adaptable to complex geological conditions. Therefore, the drilling and blasting method is widely used in tunnel engineering construction. The blasting parameter optimization scheme mainly includes obtaining existing data information-determining surrounding rock parameters-determining and optimizing blasting parameters-evaluating blasting effects-further optimizing blasting parameters, etc. The current blasting parameter optimization method does not consider comprehensive information, the model optimization effect is imperfect, and it is not possible to fully consider and analyze the information obtained, resulting in defects in the determination of the full amount of surrounding rock information, insufficient prediction of blasting parameter information, and inability to further simulate and simulate according to the optimized blasting parameters, predict and evaluate the blasting effect, and further adjust and optimize the blasting parameters. It is easy to cause the surrounding rock information to be not detailed, the blasting parameter prediction to be inaccurate, and the effect evaluation to be incomplete, which affects the final optimization and determination of the blasting parameters, resulting in poor blasting effects in actual projects, affecting project progress and construction safety.

[0047] In view of these existing problems, the present invention proposes a method for determining tunnel blasting parameters. Based on the existing historical data and the existing tunnel blasting status information, an artificial intelligence method is used to consider and train the determination of surrounding rock information and the selection of blasting parameters; the blasting parameters are optimized and predicted through the surrounding rock mechanics information; the blasting parameters optimized in the first stage are simulated and the blasting effect is evaluated, and the blasting parameters are optimized for a second time, so as to determine the appropriate blasting parameters, ensure that the blasting effect matches the expected effect as much as possible, reduce the waste of resources in the blasting process, and improve the construction quality and safety.

[0048] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0049] Embodiment 1

[0050] The tunnel blasting parameter determination method mainly includes the intelligent selection model of blasting geometric parameters, the charge and detonation parameter prediction model, and the LS-DYNA effect deduction 2D process. After each cycle, the blasting effect is collected and processed and then input into the historical operation database. Two prediction models (the intelligent selection model of blasting geometric parameters and the charge and detonation parameter prediction model) are updated every 100m of excavation.

[0051] like Figure 1 and Figure 2 As shown, the method for determining tunnel blasting parameters provided by the present invention comprises:

[0052] Step 101: obtaining the current face image and historical operation data of the tunnel to be blasted; the historical operation data includes tunnel geometry information, face mechanics information, face scanning image information, drilling rig operation information and blasting effect information.

[0053] In practical applications, the establishment of a historical operation database is mainly to obtain tunnel geometry information, select blasting parameters in historical cycles, and collect and evaluate historical blasting effect data, including tunnel geometry information, face mechanics information, face scanning image information, drilling rig operation information, and blasting effect information.

[0054] Step 102: According to the current face image and the previous cycle historical operation data, the blasting geometry parameter intelligent selection model is used to determine the initial blasting geometry parameters of the blasthole; the blasting geometry parameters include blasthole spacing, blasthole arrangement and blasthole angle. The blasting geometry parameter intelligent selection model is obtained by training the first neural network model using the first training data set; the first training data set includes the training face image, the training historical operation data and the corresponding blasting geometry parameters. The first neural network model is a deep learning neural network model (DNN model), and the model structure diagram is shown in the figure Figure 3 As shown in Figure 1, the model takes tunnel geometry information (width, height, burial depth), tunnel footage, and rock mechanics information as input. After model learning and training, it outputs blasting geometry parameters such as blasthole spacing, blasthole arrangement, and blasthole angle. The structure of the DNN layer is as follows: Figure 6 As shown, it includes input layer, hidden layer and output layer.

[0055] In practical applications, historical operation data and current tunnel face photos (current tunnel face images) are input into the blasting geometry parameter intelligent selection model to output the blasting geometry parameters of the blastholes, including blasthole spacing, blasthole arrangement, blasthole angle and other parameters.

[0056] The intelligent selection model of blasting geometric parameters is obtained through the following steps, including: obtaining data (a first training data set), including tunnel section parameters (tunnel geometric structure information, tunnel face scanning image information, tunnel face mechanical information), drilling parameters (rock drilling rig operation information) and other training historical operation data, training tunnel face images and corresponding blasting geometric parameters; inputting tunnel section parameters, drilling parameters and previous cycle historical operation parameters into the intelligent selection model of blasting geometric parameters, and constructing a mapping relationship between tunnel section parameters, drilling parameters and historical operation parameters and blasthole spacing, blasthole arrangement and blasthole angle; outputting the blasting geometric parameters of the blastholes based on the model, and obtaining a preliminary blasting plan.

[0057] The blasting scheme provided by the intelligent selection model of blasting geometric parameters is a recommended scheme. After the scheme is confirmed by the construction personnel, the drilling rig will carry out drilling operations according to this scheme.

[0058] Step 103: Control the drilling rig to drill according to the blasting geometric parameters of the blasthole and record the drilling operation information; the drilling operation information includes mechanical impact pressure, rotation pressure, jacking pressure, water pressure, water flow, drilling speed, drilling position and drilling depth.

[0059] Step 104: Based on the drilling operation information and the previous cycle historical operation data, the surrounding rock information perception model is used to determine the surrounding rock information; the surrounding rock information includes surrounding rock mechanical parameters, surrounding rock types and rock mass structural surface information. The surrounding rock information perception model is obtained by training the second neural network model using the second training data set; the second training data set includes the training drilling operation information, the training historical operation data and the corresponding surrounding rock information; the second neural network model is a hybrid model of the LSTM-CNN-deep neural network model, and the model structure diagram is as follows Figure 4 As shown in the figure, the model takes drilling operation information (mechanical impact pressure, rotation pressure, jacking pressure, water pressure, water flow, drilling speed, etc.) and face image information as model input, and outputs surrounding rock mechanical parameters and rock structure surface information after feature extraction by LSTM-CNN model.

[0060] In practical applications, the drilling operation information is input into the surrounding rock intelligent perception model to identify the surrounding rock mechanical parameters, surrounding rock types and rock structure surface information.

[0061] The surrounding rock information perception model is obtained through the following steps, including: acquiring data (a second training data set), inputting drilling operation information and historical operation data into the surrounding rock information perception model, constructing a mapping relationship between the drilling operation information and historical operation data and surrounding rock mechanical parameters, surrounding rock types and rock structure surface information; and outputting surrounding rock information based on the model.

[0062] Step 105: Determine the charge amount and blasting time interval using the charge amount and detonation parameter prediction model based on the surrounding rock information, the previous cycle historical operation data, and the blasting geometric parameters. The charge amount and detonation parameter prediction model is obtained by training the third neural network model using the third training data set; the third training data set includes surrounding rock information for training, historical operation data for training, blasting geometric parameters for training, and corresponding charge amount and blasting time interval; the third neural network model is a deep learning neural network model (DNN model), and the model structure is as follows: Figure 5 As shown in the figure, the model uses the outputs (blasting geometric parameters and surrounding rock information) of the two models (DNN model and LSTM-CNN model) as the input of this model. After model learning and training, it outputs the charge information and detonation time interval information of each blasthole including the slot hole, peripheral hole, auxiliary hole, and bottom plate hole.

[0063] In practical applications, the surrounding rock mechanical parameters, surrounding rock types, rock mass structural surface information, historical operation data of the previous cycle, and the blasting geometry data determined in this cycle are input into the charge and detonation parameter prediction model to obtain blasting parameters such as charge and blasting time interval.

[0064] The charge amount and detonation parameter prediction model is obtained through the following steps, including: acquiring data (the third training data set), inputting surrounding rock information, historical operation data and blasting geometry information determined in this cycle into the charge amount and detonation parameter prediction model, constructing a mapping relationship between surrounding rock information, historical operation data and blasting geometry information determined in this cycle and blasting parameters such as charge amount and blasting time interval; outputting blasting parameters based on the model, namely charge amount and blasting time interval.

[0065] Step 106: According to the charge amount and the blasting time interval, a blasting simulation is performed using a blasting effect deduction model to determine simulated blasting effect information; the blasting effect information includes over-break, under-break, rock mass fragmentation degree and blasthole utilization rate. The blasting effect deduction model is established using LS-DYNA software, with charge amount and blasting time interval as input and blasting effect information as output.

[0066] Step 107: adjusting the charge amount and the blasting time interval according to the simulated blasting effect information to determine the optimal charge amount and the optimal blasting time interval.

[0067] As an optional implementation, step 107 specifically includes:

[0068] Determine whether the simulated blasting effect information meets the blasting standard and obtain a judgment result.

[0069] If the judgment result is yes, the tunnel to be blasted is blasted according to the charge amount and the blasting time interval.

[0070] If the judgment result is no, the charge amount and the blasting time interval are updated, and the step of "performing blasting simulation according to the charge amount and the blasting time interval using a blasting effect deduction model to determine simulated blasting effect information" is returned until the simulated blasting effect information meets the blasting standard.

[0071] In practical applications, a 2D blasting effect deduction model is established for the obtained blasting parameters using the LS-DYNA program software. Through digital modeling, the blasting effect is simulated and calculated to determine whether the blasting effect meets the standard (the standard is that there is no over-excavation or under-excavation, and the rock fragmentation degree and blasthole utilization rate both reach the corresponding set thresholds). If it meets the standard, charging and blasting are carried out. After the blasting, the blasting effect information is collected, and the blasting effects such as over-excavation or under-excavation, rock fragmentation degree, and blasthole utilization rate are analyzed and evaluated, and the obtained blasting effect information is transferred to the historical data information database; if it does not meet the standard, the charging amount and blasting time interval are adjusted and optimized, and the blasting effect is simulated and judged again through LS-DYNA, and multiple cycles are performed until the blasting effect meets the standard, and the final blasting effect information is transferred to the historical data information database.

[0072] As an optional implementation manner, the tunnel to be blasted is blasted according to the charge amount and the blasting time interval, and then further comprising:

[0073] Record blasting effect information; update the blasting effect information, blasting geometry parameters of the blasthole, drilling operation information, surrounding rock information, charge amount and blasting time interval to historical operation data.

[0074] The present invention ensures the optimization efficiency by identifying the rock mechanical parameters, rock types and rock structure surface information based on the operation information of the drilling rig, and optimizing the blasting parameters in two stages. In the first stage, the drilling position is optimized, and the drilling operation is performed after optimization to obtain further drilling data; in the second stage, the rock parameters are predicted by the parameters of the completed hole, and then the charge amount and time interval are determined, which are input into the simulation model to optimize the charge amount and detonation parameters. Blasting optimization is performed in LS-DYNA, and a loss function f that takes into account the blasting effect (over-excavation, under-excavation, degree of fragmentation, etc.) is established to achieve the optimization of the blasting results.

[0075] Through blasting calculations based on artificial intelligence and simulation of construction information, the blasting parameters are continuously iteratively optimized in the tunnel blasting construction cycle. On the basis of obtaining the blasting parameters of the current stage using historical data, combined with the simulation effect, the parameters of the current stage are used again to re-optimize the blasting design parameters of the current stage. At the same time, the blasting data is dynamically updated to achieve continuous optimization of parameter optimization efficiency and blasting effect control.

[0076] Embodiment 2

[0077] In order to execute the method corresponding to the above embodiment 1 to achieve corresponding functions and technical effects, a tunnel blasting parameter determination system is provided below, including:

[0078] The data acquisition module is used to obtain the current face image and historical operation data of the tunnel to be blasted; the historical operation data includes tunnel geometry information, face mechanics information, face scanning image information, drilling rig operation information and blasting effect information.

[0079] A blasthole parameter determination module is used to determine the blasting geometric parameters of the blasthole according to the current face image and the previous cycle historical operation data, using the blasting geometric parameter intelligent selection model; the blasting geometric parameters include blasthole spacing, blasthole arrangement and blasthole angle; the blasting geometric parameter intelligent selection model is obtained by training the first neural network model using the first training data set; the first training data set includes a training face image, training historical operation data and corresponding blasting geometric parameters; the first neural network model is a deep learning neural network model.

[0080] The drilling module is used to control the drilling rig to drill according to the blasting geometric parameters of the blasthole and record the drilling operation information; the drilling operation information includes mechanical impact pressure, rotation pressure, jacking pressure, water pressure, water flow, drilling speed, drilling position and drilling depth.

[0081] The surrounding rock information determination module is used to determine the surrounding rock information based on the drilling operation information and the previous cycle historical operation data using the surrounding rock information perception model; the surrounding rock information includes surrounding rock mechanical parameters, surrounding rock types and rock structure surface information; wherein the surrounding rock information perception model is obtained by training the second neural network model using the second training data set; the second training data set includes training drilling operation information, training historical operation data and corresponding surrounding rock information; the second neural network model is a hybrid model of LSTM-CNN-deep neural network model.

[0082] The blasting parameter determination module is used to determine the charge amount and blasting time interval based on the surrounding rock information, the previous cycle historical operation data and the blasting geometric parameters, using the charge amount and detonation parameter prediction model; wherein the charge amount and detonation parameter prediction model is obtained by training a third neural network model using a third training data set; the third training data set includes training surrounding rock information, training historical operation data, training blasting geometric parameters and corresponding charge amount and blasting time interval; the third neural network model is a deep learning neural network model.

[0083] The blasting simulation module is used to perform blasting simulation according to the charge amount and the blasting time interval using a blasting effect deduction model to determine simulated blasting effect information; the blasting effect information includes over-excavation, under-excavation, rock mass fragmentation degree and blasthole utilization rate; wherein the blasting effect deduction model is established using LS-DYNA software, with charge amount and blasting time interval as input and blasting effect information as output.

[0084] The blasting parameter adjustment module is used to adjust the charge amount and the blasting time interval according to the simulated blasting effect information to determine the optimal charge amount and the optimal blasting time interval.

[0085] Embodiment 3

[0086] The present invention provides an electronic device, comprising: a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the tunnel blasting parameter determination method of embodiment 1.

[0087] Embodiment 4

[0088] The present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for determining tunnel blasting parameters of the first embodiment is implemented.

[0089] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0090] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for determining tunnel blasting parameters, characterized in that: include: Obtain the current face image and historical operation data of the tunnel to be blasted; The historical operation data includes tunnel geometry information, face mechanics information, face scanning image information, drilling rig operation information and blasting effect information; According to the current face image and the previous cycle historical operation data, the blasting geometric parameters of the blasthole are determined by using the blasting geometric parameter intelligent selection model; the blasting geometric parameters include blasthole spacing, blasthole arrangement and blasthole angle; the blasting geometric parameter intelligent selection model is obtained by training the first neural network model using the first training data set; the first training data set includes the training face image, the training historical operation data and the corresponding blasting geometric parameters; the first neural network model is a deep learning neural network model; Controlling the drilling rig to drill according to the blasting geometric parameters of the blasthole and recording drilling operation information; the drilling operation information includes mechanical impact pressure, rotation pressure, jacking pressure, water pressure, water flow, drilling speed, drilling position and drilling depth; According to the drilling operation information and the previous cycle historical operation data, the surrounding rock information is determined by using the surrounding rock information perception model; the surrounding rock information includes surrounding rock mechanical parameters, surrounding rock types and rock mass structural surface information; wherein the surrounding rock information perception model is obtained by training the second neural network model using the second training data set; the second training data set includes training drilling operation information, training historical operation data and corresponding surrounding rock information; the second neural network model is a hybrid model of LSTM-CNN-deep neural network model; According to the surrounding rock information, the historical operation data of the previous cycle and the blasting geometric parameters, the charge amount and the blasting time interval are determined by using the charge amount and detonation parameter prediction model; wherein the charge amount and detonation parameter prediction model is obtained by training a third neural network model using a third training data set; the third training data set includes surrounding rock information for training, historical operation data for training, blasting geometric parameters for training and corresponding charge amounts and blasting time intervals; the third neural network model is a deep learning neural network model; According to the charge amount and the blasting time interval, a blasting simulation is performed using a blasting effect deduction model to determine simulated blasting effect information; the blasting effect information includes over-break, under-break, rock mass fragmentation degree and blasthole utilization rate; wherein the blasting effect deduction model is established using LS-DYNA software, with charge amount and blasting time interval as input and blasting effect information as output; The charge amount and the blasting time interval are adjusted according to the simulated blasting effect information to determine the optimal charge amount and the optimal blasting time interval.

2. The method for determining tunnel blasting parameters according to claim 1, characterized in that: The charge amount and the blasting time interval are adjusted according to the simulated blasting effect information to determine the optimal charge amount and the optimal blasting time interval, specifically including: Determine whether the simulated blasting effect information meets the blasting standard and obtain the judgment result; If the judgment result is yes, blasting the tunnel to be blasted according to the charge amount and the blasting time interval; If the judgment result is no, the charge amount and the blasting time interval are updated, and the step of "performing blasting simulation according to the charge amount and the blasting time interval using the blasting effect deduction model to determine the simulated blasting effect information" is returned until the simulated blasting effect information meets the blasting standard.

3. The method for determining tunnel blasting parameters according to claim 2, characterized in that: Blasting the tunnel to be blasted according to the charge amount and the blasting time interval, and then further comprising: Record blasting effect information; The blasting effect information, blasting geometric parameters of the blasthole, drilling operation information, surrounding rock information, charge amount and blasting time interval are updated to the historical operation data.

4. A tunnel blasting parameter determination system, characterized in that: include: A data acquisition module is used to obtain the current face image and historical operation data of the tunnel to be blasted; The historical operation data includes tunnel geometry information, face mechanics information, face scanning image information, drilling rig operation information and blasting effect information; A blasthole parameter determination module is used to determine the blasting geometric parameters of the blasthole according to the current face image and the previous cycle historical operation data, using the blasting geometric parameter intelligent selection model; the blasting geometric parameters include blasthole spacing, blasthole arrangement and blasthole angle; the blasting geometric parameter intelligent selection model is obtained by training the first neural network model using the first training data set; the first training data set includes the training face image, the training historical operation data and the corresponding blasting geometric parameters; the first neural network model is a deep learning neural network model; A drilling module is used to control the drilling rig to drill according to the blasting geometric parameters of the blasthole and record drilling operation information; the drilling operation information includes mechanical impact pressure, rotation pressure, jacking pressure, water pressure, water flow, drilling speed, drilling position and drilling depth; A surrounding rock information determination module is used to determine the surrounding rock information according to the drilling operation information and the previous cycle historical operation data using a surrounding rock information perception model; the surrounding rock information includes surrounding rock mechanical parameters, surrounding rock types and rock mass structural surface information; wherein the surrounding rock information perception model is obtained by training the second neural network model using a second training data set; the second training data set includes training drilling operation information, training historical operation data and corresponding surrounding rock information; the second neural network model is a hybrid model of LSTM-CNN-deep neural network model; A blasting parameter determination module, for determining the charge amount and blasting time interval according to the surrounding rock information, the previous cycle historical operation data and the blasting geometric parameters, using the charge amount and detonation parameter prediction model; wherein the charge amount and detonation parameter prediction model is obtained by training a third neural network model using a third training data set; the third training data set includes surrounding rock information for training, historical operation data for training, blasting geometric parameters for training and corresponding charge amounts and blasting time intervals; the third neural network model is a deep learning neural network model; A blasting simulation module is used to perform blasting simulation according to the charge amount and the blasting time interval using a blasting effect deduction model to determine simulated blasting effect information; the blasting effect information includes over-break, under-break, rock mass fragmentation degree and blasthole utilization rate; wherein the blasting effect deduction model is established using LS-DYNA software with charge amount and blasting time interval as input and blasting effect information as output; The blasting parameter adjustment module is used to adjust the charge amount and the blasting time interval according to the simulated blasting effect information to determine the optimal charge amount and the optimal blasting time interval.

5. An electronic device, characterized in that: include: A memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the method for determining tunnel blasting parameters according to any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for determining tunnel blasting parameters according to any one of claims 1 to 3 is implemented.

Images