An intelligent optimization method for tunnel smooth blasting design parameters based on deep learning and neural networks

Abstract

The present invention is a method for optimizing tunnel smooth blasting design parameters based on deep learning and neural networks. This method, which belongs to the field of tunnel blasting construction technology, addresses the current optimization design problem of tunnel blasting design parameters. The method comprises the following steps: loading processed image data into an SP-Yolov8 model and a Yolov8-seg model, outputting the main structural surface inclination category and main structural surface area respectively, extracting the main structural surface information to obtain the main structural surface position, inputting the blasting design parameters, integrity coefficient, uniaxial compressive strength, main structural surface inclination category, and main structural surface position into a DBN model, outputting the maximum linear overexcavation prediction value, comparing the maximum linear overexcavation prediction value with the limit value, and outputting the final blasting design parameters. The method of the present invention can quickly and accurately obtain the final blasting design parameters, and has the advantages of accurate feature information extraction, high robustness, and fast convergence speed.

Description

Technical Field

[0001] The present invention relates to the technical field of tunnel blasting construction, and in particular to a method for optimizing design parameters of tunnel smooth blasting based on deep learning and neural networks. Background Art

[0002] Geological work runs through the entire process of tunnel investigation, design and construction. Due to the complexity and variability of geological conditions, the surrounding rock actually revealed during construction often differs significantly from the geological survey data, resulting in irrational construction parameters, economic losses, delays in construction schedules, safety accidents and other problems. The drilling and blasting method is widely used in tunnel construction due to its low cost and strong geological applicability. However, due to the "instantaneous" nature of the work, it is often difficult to accurately control in engineering applications, and problems such as over-excavation, under-excavation and block size inappropriateness often occur. These two problems will directly affect the safety, quality and efficiency of tunnel excavation.

[0003] Tunnel overexcavation has long been a major challenge facing drilling and blasting in my country. This not only increases construction costs but also easily causes localized stress concentrations in the surrounding rock, thereby impacting overall rock stability. To predict and control overexcavation caused by drilling and blasting, Ibarra et al. predicted overexcavation based on the relationship between surrounding rock classification systems and unit explosive consumption. They found that increasing unit explosive consumption reduced overexcavation, while increasing the Q value led to increased underexcavation. Singh et al. conducted a series of experiments to investigate the factors influencing overexcavation damage and found that all influencing parameters can be categorized into three categories: rock characteristics, explosives, and blasting mode. Furthermore, existing research has shown that overexcavation is influenced by numerous factors, including plugging depth, plugging material, the last charge versus the total charge, special charges, special charges per delayed charge, number of drill rows, rock strength, rock quality index, rock weathering, and groundwater. Minimizing the impact of blasting damage on the surrounding rock during blasting excavation, reducing overexcavation, and ensuring surrounding rock stability have become fundamental issues in tunnel blasting construction.

[0004] In recent years, information technology, particularly artificial intelligence, has experienced explosive growth. Cross-disciplinary research has been conducted across various fields and disciplines, and has been applied to civil and mining engineering. Xiao Qinghua, for example, has introduced deep learning and image recognition techniques to automatically extract and segment tunnel overbreak and underbreak images, enabling intelligent statistics of overbreak and underbreak volume and intelligent classification of blasting excavation quality. Wu Renjie et al., Nie Jun et al., and Ku-latilake et al. have developed prediction models for blasting fragment size using principal component analysis, gene expression programming algorithms, and neural networks, respectively.

[0005] Currently, computer vision inspection based on deep learning methods has been widely applied in many areas, including tunnel lining defect detection and in-tunnel vehicle identification. Theoretical research and engineering experience have demonstrated the feasibility of applying this technology to tunnel construction. However, tunnel blasting design parameters are a multi-parameter, high-dimensional, nonlinear problem. To ultimately determine the optimal smooth blasting design parameters for the corresponding surrounding rock conditions and achieve high-quality tunnel blasting, optimizing these parameters using on-site tunnel face image data remains a challenge in current underground tunnel construction. Summary of the Invention

[0006] In response to the defects in the above-mentioned prior art, the purpose of the present invention is to propose a tunnel smooth blasting design parameter optimization method based on deep learning and neural networks, and to improve the deep learning Yolov8 model to obtain the main structural surface inclination classification and identification and main structural surface information extraction in the tunnel face image. Then, the blasting design parameter optimization dataset is input into the DBN deep neural network for training to obtain the optimal final blasting design parameters to solve the problems existing in the background technology.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] In one aspect, the present invention discloses a method for optimizing design parameters of tunnel smooth blasting based on deep learning and neural networks, comprising the following steps:

[0009] S1: Select blasting design parameters, integrity coefficient and uniaxial compressive strength;

[0010] S2: Acquire tunnel face image data and preprocess;

[0011] S3: Load the pre-processed tunnel face image data into the SP-Yolov8 model and the Yolov8-seg model, and output the main structural surface inclination category and main structural surface area accordingly;

[0012] S4: extracting the main structural surface information from the main structural surface area to obtain the main structural surface position;

[0013] S5: Input the blasting design parameters, integrity coefficient, uniaxial compressive strength, main structural surface inclination type, and main structural surface position into the optimized DBN model, and output the maximum linear overexcavation prediction value;

[0014] S6: If the output maximum linear over-excavation prediction value is higher than the limit, modify the blasting design parameters and return to step S5 to re-predict until the output maximum linear over-excavation prediction value is less than or equal to the limit, then output the final blasting design parameters.

[0015] As a further preferred solution of the above technical solution: in step S2, the tunnel face image data is subjected to sliding window cutting processing, enhancement processing and normalization processing in sequence, wherein the formula for normalization processing is:

[0016]

[0017] In the formula: y is the value of the data after normalization; x is the original data before normalization, x max is the maximum value of the data before normalization; x min is the minimum value of the data before normalization.

[0018] A further preferred solution is: in step S3, the SP-Yolov8 model is obtained by optimizing the Backbone network structure of the Yolov8 model, and the SP-Yolov8 model is used to classify and identify the main structural surface inclination of the tunnel face. The specific steps are as follows:

[0019] S31: Add an Identity layer in parallel to the three asymmetric convolution blocks to form the SP-DWConv2D convolution layer;

[0020] S32: Bring the SP-DWConv2D convolutional layer into the Bottleneck structure in the Backbone network structure, and replace it upwards in sequence to form a new SP-C2f module network structure;

[0021] S33: Replace the C2f module in the Yolov8 model with the SP-C2f module network structure to obtain the SP-Yolov8 model.

[0022] A further preferred solution is: in step S32, the SP-DWConv2D convolution layer includes a Split layer at the input front end, 3 depth convolution kernels, 1 Identity layer and a concat layer at the back end for feature fusion, wherein the sizes of the 3 depth convolution kernels are 3×3 convolution kernel, 1×9 strip kernel and 9×1 strip kernel respectively.

[0023] A further preferred solution is: in step S3, when classifying and identifying the inclination angle of the main structural surface, a linear frame is used to mark and classify the main structural surface in the tunnel face image, and the results of the marking and classification are saved in a JSON file. The results of the marking and classification are mobilized together with the tunnel face image through the SP-Yolov8 model to identify the main structural surface inclination angle category.

[0024] A further preferred solution is: in step S3, based on the Head structure in the Yolov8 model, by adding the aspect ratio constraint condition and the distance constraint condition of the main structural surface to the Soft-NMS filtering algorithm, the Yolov8-seg model is obtained after optimization, and the false main structural surfaces are eliminated with the Yolov8-seg model to obtain the main structural surface area. The specific steps include:

[0025] First, based on the aspect ratio constraint, the aspect ratio of the upper left corner coordinates and the lower right corner coordinates of the detection box is calculated. Then, the aspect ratio constraint is introduced and the aspect ratio of each detection box is calculated again. If the aspect ratio of the upper left corner coordinates and the lower right corner coordinates of the detection box is greater than or equal to the aspect ratio constraint, the detection box with such a large range or useless is eliminated to obtain a new detection box set;

[0026] Then, based on the distance constraint, the probability scores of the new detection boxes are sorted from high to low, and the center coordinates x are calculated based on the upper left corner coordinates and the lower right corner coordinates of the detection box. o1 、x o2 , with x o1 As the midpoint, the distance constraint R is introduced. When x o2 with x o1 When the distance is less than or equal to R, the detection box with a lower probability score is eliminated.

[0027] A further preferred solution is: in step S4, the step of extracting the main structural surface information includes:

[0028] S41: Input the height Y of the tunnel face image and the height h of the tunnel face into the SP-YOLOv8 model, and use the coordinates of the lower right corner endpoint of the tunnel face image as the origin;

[0029] S42: Segment the tunnel face image using the YOLOv8-seg model, expand the main structural surface to form a mask area, and the coordinates of the lower right corner of the mask area box are (x1, y1);

[0030] S43: extract the main structural surface area and perform image processing using binarization;

[0031] S44: The skeleton of the binarized image is extracted using the Zhang-Suen algorithm. The region to be extracted is iterated multiple times, and a trace is formed by indenting from the edge to the center. The coordinates of the right endpoint of the trace are (x2, y2).

[0032] S45: The pixel height y=y1+y2 of the right endpoint of the main structural surface is obtained by using the image pixel height. The actual height H=y×(h / Y) of the right endpoint of the main structural surface is obtained according to the ratio of the actual tunnel face height, thereby obtaining the position of the main structural surface.

[0033] A further preferred solution is: in step S5, the DBN model consists of a two-layer RBM and a single-layer BP neural network, and the PSO algorithm is used to assist the DBN model in finding the number of nodes in the first hidden layer, the number of nodes in the second hidden layer, the learning rate and the number of training times in the hyperparameters.

[0034] On the other hand, the present invention discloses an electronic device, including a processor and a memory, wherein the memory stores computer instructions, and the processor is used to execute the computer instructions stored in the memory to implement the steps of any of the above-mentioned methods for optimizing design parameters of tunnel smooth blasting based on deep learning and neural networks.

[0035] On the other hand, the present invention also discloses a computer-readable storage medium storing computer instructions, which are used to enable a computer to execute the steps of any of the above-mentioned methods for optimizing design parameters of tunnel smooth blasting based on deep learning and neural networks.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. The present invention studies the intelligent optimization of tunnel smooth blasting design parameters and constructs a blasting design parameter optimization method based on the combination of deep learning and deep neural network. First, the deep learning network based on the YOLOv8 model is optimized, and the optimized SP-YOLOv8 model is used to obtain the main structural surface inclination classification, and the optimized YOLOv8-seg model is used to obtain the main structural surface area. The main structural surface position is obtained by combining the YOLOv8 model with the YOLOv8-seg model. In this way, a tunnel smooth blasting design parameter optimization data set is constructed to determine the input and output parameters of the DBN neural network model. After multiple training of the DBN model, the optimal final blasting design parameters are obtained.

[0038] 2. The SP-Yolov8 model of the present invention can effectively reduce the image preprocessing steps in traditional methods, and can still complete the main structural surface information extraction when there are many image interference factors. It has the advantages of accurate feature information extraction and fast convergence speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required in the embodiments.

[0040] Figure 1 This is a flow chart of a method for optimizing design parameters of tunnel smooth blasting based on deep learning and neural networks according to the present invention;

[0041] Figure 2 are the input parameters of the tunnel smooth blasting design parameter optimization model of the present invention;

[0042] Figure 3 The present invention is a process of performing sliding window cutting processing on tunnel face image data;

[0043] Figure 4 This is a network structure diagram of the SP-DWConv2D convolutional layer of the present invention;

[0044] Figure 5 A diagram showing the process of replacing the C2f module of the YOLOv8 model with the SP-C2f module network to obtain the SP-YOLOv8 model.

[0045] Figure 6 The network structure of the DBN model of the present invention;

[0046] Figure 7 These are the training and validation fitting curves for the BP model, CNN model, and DBN model in the present invention. DETAILED DESCRIPTION

[0047] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments, not all embodiments. For those skilled in the art, other embodiments can be derived without inventive effort.

[0048] Reference Figure 1-7 The present invention provides a method for optimizing design parameters of tunnel smooth blasting based on deep learning and neural network, comprising the following steps:

[0049] S1: Select blasting design parameters, integrity coefficient and uniaxial compressive strength;

[0050] S2: Acquire tunnel face image data and preprocess;

[0051] S3: Load the pre-processed tunnel face image data into the SP-Yolov8 model and the Yolov8-seg model, and output the main structural surface inclination category and main structural surface area accordingly;

[0052] S4: extracting the main structural surface information from the main structural surface area to obtain the main structural surface position;

[0053] S5: Input the blasting design parameters, integrity coefficient, uniaxial compressive strength, main structural surface inclination type, and main structural surface position into the optimized DBN model, and output the maximum linear overexcavation prediction value;

[0054] S6: If the output maximum linear over-excavation prediction value is higher than the limit, modify the blasting design parameters and return to step S5 to re-predict until the output maximum linear over-excavation prediction value is less than or equal to the limit, then output the final blasting design parameters.

[0055] like Figure 2 As shown in Figure 1, when considering the input parameters of the tunnel smooth blasting design parameter optimization model, there are mainly two aspects: surrounding rock quality parameters and blasting design parameters. The surrounding rock quality parameters include the integrity coefficient X1 and uniaxial compressive strength X2 (MPa) in step S1, as well as the main structural surface inclination X3 and main structural surface position X4 in step S2, which have a greater impact on the maximum linear over-excavation in tunnel blasting construction. The blasting design parameters include the peripheral eye charge X5 (kg), peripheral eye hole spacing X6 (cm), outer ring auxiliary eye hole spacing X7 (cm), outer ring auxiliary eye charge X8 (kg), smooth blasting layer thickness X9 (cm), peripheral eye charge at the structural surface X1 (kg), and the surrounding eye charge at the structural surface X2 (cm). 10 (kg), the amount of charge of the outer ring auxiliary eye at the structural surface X 11 (kg), number of blastholes X 12 、Total explosive quantity X 13 (kg) and cycle footage X 14 (cm), and the above 14 parameters are used as input parameters of the tunnel blasting design parameter optimization model.

[0056] In step S2, tunnel face image data is obtained from the construction site and preprocessed. Considering that the tunnel face surrounding rock images collected according to the acquisition standard are large in size, with complex backgrounds and a small proportion of pixels on the main structural surface, if the model is directly trained on the tunnel blasting design parameter optimization model, there are problems such as excessive time consumption and difficulty in determining the model structure and parameters. Therefore, it is necessary to perform sliding window cutting processing on each tunnel face image. The sliding window cutting processing process is as follows: Figure 3 shown.

[0057] After performing sliding window cutting on the tunnel face image, image data enhancement technology is used to increase the quantity and quality of the tunnel face surrounding rock image database. To improve the convergence speed and calculation accuracy of the tunnel blasting design parameter optimization model, the enhanced data is normalized to unify the dimension and variation range. The "mapminmax" normalization formula in MATLAB is used for data normalization. The processed data range is [0, 1]. The normalization formula is as follows:

[0058]

[0059] In the formula: y is the value of the data after normalization; x is the original data before normalization, x max is the maximum value of the data before normalization; x minis the minimum value of the data before normalization.

[0060] After normalization, a data set suitable for the subsequent prediction of maximum linear over-excavation neural network is obtained, namely, the tunnel blasting maximum linear over-excavation data set.

[0061] In step S3, the SP-Yolov8 model is obtained by optimizing the Backbone network structure of the Yolov8 model, and the SP-Yolov8 model is used to classify and identify the inclination angle of the main structural surface of the tunnel face.

[0062] In the YOLOv8 model, the backbone network structure is mainly used to extract image feature information. When the backbone network structure is used to train the main structural surface inclination classification and recognition dataset, the image feature information is lost.

[0063] The core of the Backbone network is the C2f module, which is used to convert the output of the convolutional layer into the input of the fully connected layer. It is the key module that causes missed detection and false detection after YOLOv8 model training. In the C2f module, it is the Bottleneck structure that can improve the effect. Therefore, this application improves the performance of the Bottleneck structure in the tunnel face main structure surface inclination classification and recognition dataset by adjusting the number and size of the convolution kernels in the Bottleneck structure.

[0064] When optimizing the Bottleneck structure, an Identity (identity mapping) layer is added in parallel to the asymmetric convolution block, and a structural plane-DWConv2D (SP-DWConv2D) convolution layer is designed to optimize the classification and recognition dataset of the main structural surface inclination of the tunnel face. The principle of the SP-DWConv2D convolution layer is as follows: Figure 4 shown.

[0065] More specifically, the SP-DWConv2D convolution layer includes a Split layer at the input front end, three depth convolution kernels, an Identity layer, and a concat layer for feature fusion at the back end. The sizes of the three depth convolution kernels are 3×3 convolution kernels, 1×9 strip kernels, and 9×1 strip kernels, respectively. The SP-DWConv2D convolution layer is brought into the Bottleneck structure and replaced upward in sequence to form a new structural plane-C2f (SP-C2f) module network structure. The C2f module of YOLOv8 is replaced with the SP-C2f module network to obtain the optimized YOLOv8 model, as shown in Figure 2. Figure 5 As shown in the figure, this is the SP-YOLOv8 model, which is the tunnel blasting design parameter optimization model mentioned above.

[0066] The SP-YOLOv8 model is then used to classify and identify the inclination angles of the tunnel's main structural surface. Before classification and identification, the main structural surface is marked in the tunnel face image using a linear frame and classified according to the angle.

[0067] Specifically, for the input parameter of the main structural surface inclination X3, engineering geology divides rock masses into five categories according to the structural surface inclination, namely, nearly horizontal rock layers (0°~5°), gently inclined rock layers (5°~30°), inclined rock layers (30°~70°), steep rock layers (70°~85°), and vertical rock layers (85°~90°). This application draws on the rock mass classification of engineering geology and divides the main structural surface of the tunnel into five categories, numbering them 1, 2, 3, 4, and 5 respectively. The main structural surface dataset is generated according to the VOC dataset format, and the main structural surface dataset is input into the SP-YOLOv8 model to output the main structural surface inclination type of the tunnel face.

[0068] For the input parameter of the main structural surface position X4, a labeling box is made according to the segmentation and extraction of the main structural surface. Specifically, the main structural surface is marked in the tunnel face image in the form of a linear frame. The labeling box should cover the main structural surface and require that the labeling box contain the complete main structural surface and the surrounding small area rock background. When there are multiple main structural surfaces in a tunnel face image, they are annotated one by one with a single linear labeling box and saved in the form of a JSON file. Then, the labeling box and the image are called together in deep learning, and the height of the main structural surface is calculated and recorded. The tunnel face area is classified according to the calculated height value, and the entire upper step tunnel face is divided into three areas according to height. The area within 100 cm from the arch is the arch area, the middle and lower part of the step is 200 cm from the bottom of the step, and the rest is the arch spandrel area. The recorded main structural surface heights are classified according to the above design requirements and recorded as 11, 12, and 13 respectively.

[0069] Since the Head structure in the original YOLOv8 model performs poorly in the recognition of the main structural surface area, false detection and re-detection may occur. Therefore, the Head structure of the YOLOv8 model is optimized. The Soft-NMS filtering algorithm with aspect ratio constraints and distance constraints is added to the Head structure. After optimization, the YOLOv8-seg model is obtained. The aspect ratio constraint is used to eliminate some detection frames that are too large or useless. After the aspect ratio constraint is eliminated, a new set of detection frames will be output. On this basis, the distance constraint is introduced to further eliminate false main structural surfaces from the newly output detection frames to reduce the false detection rate and re-detection rate of the model. The specific method is as follows:

[0070] Based on the aspect ratio constraint algorithm, the aspect ratio of the upper left corner coordinates and the lower right corner coordinates of the detection box is calculated. Then, the aspect ratio constraint is introduced and the aspect ratio of each detection box is calculated again. If the aspect ratio of the upper left corner coordinates and the lower right corner coordinates of the detection box is greater than or equal to the aspect ratio constraint, the detection box with such a large range or useless is eliminated to obtain a new detection box set;

[0071] Then, based on the constraint algorithm of the distance constraint, the probability scores of the new detection boxes are sorted from high to low, and the center coordinates x are calculated based on the upper left corner coordinates and the lower right corner coordinates of the detection box. o1 、x o2 , with x o1 As the midpoint, the distance constraint R is introduced. When x o2 with x o1 When the distance is less than or equal to R, the detection box with a lower probability score is eliminated.

[0072] Furthermore, the optimized YOLOv8-seg model is used to further segment and extract the main structural surface. After obtaining the main structural surface area, the image is segmented by the YOLOv8-seg model to form a mask image. The mask image is binarized, and then the Zhang-Suen skeleton extraction algorithm is used to obtain the endpoint coordinates of the main structural surface, thereby extracting the main structural surface information and obtaining the main structural surface position. Specifically:

[0073] The SP-YOLOv8 model first inputs the tunnel face image height Y and the tunnel face height h, with the coordinates of the lower right corner of the tunnel face image as the origin. The tunnel face image is then segmented using the YOLOv8-seg model, expanding the main structural surface to form a mask region with the lower right corner coordinates (x1, y1). The main structural surface region is extracted and binarized. Skeleton extraction is performed on the binarized image using the Zhang-Suen algorithm. Multiple iterations are performed on the region to be extracted, indenting from the edge to the center to form a trace. The right endpoint of the trace is now (x2, y2). The pixel height of the right endpoint of the main structural surface is calculated from the image pixel height, y = y1 + y2. Based on the ratio of the actual tunnel face height, the actual height of the right endpoint of the main structural surface is calculated as H = y × (h / Y), thus obtaining the main structural surface position. This method reduces the image preprocessing steps required in traditional methods and can still complete the main structural surface information extraction even in the presence of numerous image interference factors.

[0074] In step S5, the maximum linear over-excavation Y1 (cm) is selected as the output parameter of the tunnel smooth blasting design parameter optimization model, and the tunnel profile after blasting is measured using a cross-section scanner. The tunnel blasting maximum linear over-excavation data set obtained above is input into the DBN deep network model for maximum linear over-excavation prediction; the DBN deep neural network is composed of an unsupervised lower layer composed of several restricted Boltzmann machines (RBMs) and a supervised upper layer composed of a BP neural network. RBM is a Markov random field proposed by Professor Hinton and is an important component of the deep belief network (DBN). RBM is a neural network structure composed of two layers: a hidden layer (h) and an input layer (v). Its basic principle is to adjust the parameter settings through a large amount of training data so that the distribution probability represented by the RBM under the parameters is as consistent as possible with the training data.

[0075] In this invention, a variety of neural network models (such as BP model, CNN model, DBN model) are compared and selected, and the prediction result fitting curve is obtained through training and verification. The performance of the above three models is comprehensively evaluated, and the DBN model is finally determined. The comparison process of the above three models is referred to Figure 7 As shown, where: Figure 7 (a) Figure 7 (b) represents the predicted value fitting curves of the DBN model training results and verification results respectively; Figure 7 (c) Figure 7 Middle (d) represents the predicted value fitting curves of the training results and verification results of the BP model; (d) BP verification; Figure 7 Middle (e), Figure 7 (f) in the middle represents the predicted value fitting curves of the training results and verification results of the CNN model.

[0076] Depend on Figure 7 As can be seen, during the training and validation process, the scatter points of the three models mostly lie on both sides of the standard line. The DBN model's scatter points are closest to the standard line, followed by the BP model, and the CNN model's scatter points are farthest from the standard line. The fitted curves also show that the DBN model's fitted curve is closest to the standard line, but still exhibits underfitting. The BP and CNN model's fitted curves deviate significantly from the standard line, exhibiting severe underfitting. Combining these results, the DBN model exhibits the best fitting performance throughout the training and validation processes.

[0077] Typically, hyperparameters of the DBN model, such as the learning rate, number of hidden layer nodes, and number of training times, are set manually. This results in the maximum relative error not meeting the requirements of actual engineering use, so the hyperparameter settings need to be optimized.

[0078] In this invention, the particle swarm optimization algorithm (PSO) is used to optimize the DBN model, mainly to optimize the selection of hyperparameters to avoid the subjectivity of manual parameter determination. The PSO algorithm is used to find the number of nodes in the first hidden layer, the number of nodes in the second hidden layer, the learning rate and the number of training times in the DBN model hyperparameters. Figure 6 As shown in the figure, the DBN model in the present invention consists of a two-layer RBM and a single-layer BP neural network. The number of hidden layer nodes in each RBM layer is 20, the learning rate is 0.01, the number of RBM pre-training times is 500, and the number of reverse iterations is 3000. The maximum linear over-excavation dataset of tunnel blasting is input into the optimized DBN model, and the maximum linear over-excavation prediction value is output.

[0079] The maximum linear over-excavation prediction value obtained by the present invention is compared with the limit value. If the maximum linear over-excavation prediction value is higher than the limit value, the blasting design parameters are modified and the optimized DBN model is used again for prediction until the output maximum linear over-excavation prediction value is less than or equal to the limit value, and the final blasting design parameters are output.

[0080] It should be understood that the detailed description of the embodiments of the present invention provided in the accompanying drawings is only for further explanation of the present invention and is not intended to limit the scope of the invention as claimed, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, some non-essential improvements and adjustments made by ordinary technicians in this field based on the above content of the present invention fall within the scope of protection of the present invention.

Claims

1. A tunnel smooth blasting design parameter optimization method based on deep learning and neural network, characterized in that: The following steps are involved: S1: Select blasting design parameters, integrity coefficient and uniaxial compressive strength; S2: Acquire tunnel face image data and preprocess; S3: Load the pre-processed tunnel face image data into the SP-Yolov8 model and the Yolov8-seg model, and output the main structural surface inclination category and main structural surface area accordingly; In step S3, the SP-Yolov8 model is obtained by optimizing the Backbone network structure of the Yolov8 model, and the SP-Yolov8 model is used to classify and identify the main structural surface inclination of the tunnel face. The specific optimization steps are as follows: S31: Add an Identity layer in parallel to the three asymmetric convolution blocks to form the SP-DWConv2D convolution layer; S32: Bring the SP-DWConv2D convolutional layer into the Bottleneck structure and replace it upwards in sequence to form the SP-C2f module network structure; S33: Replace the C2f module in the Yolov8 model with the SP-C2f module network structure to obtain the SP-Yolov8 model; Based on the Head structure in the Yolov8 model, the Soft-NMS filtering algorithm is optimized by adding aspect ratio constraints and distance constraints on the main structural surface. The Yolov8-seg model is then used to eliminate false main structural surfaces and obtain the main structural surface area. The specific steps include: First, based on the aspect ratio constraint, the aspect ratio of the upper left corner coordinates and the lower right corner coordinates of the detection box is calculated. Then, the aspect ratio constraint is introduced and the aspect ratio of each detection box is calculated again. If the aspect ratio of the upper left corner coordinates and the lower right corner coordinates of the detection box is greater than or equal to the aspect ratio constraint, the detection box with such a large range or useless is eliminated to obtain a new detection box set; Then, based on the distance constraint, the probability scores of the new detection boxes are sorted from high to low, and the center coordinates of the detection boxes are calculated based on the coordinates of the upper left corner and the lower right corner. ,by As the midpoint, the distance constraint R is introduced. and When the distance is less than or equal to R, the detection frame with a lower probability score is eliminated; S4: extracting the main structural surface information from the main structural surface area to obtain the main structural surface position; S5: Input the blasting design parameters, integrity coefficient, uniaxial compressive strength, main structural surface inclination type, and main structural surface position into the optimized DBN model, and output the maximum linear overexcavation prediction value; S6: If the output maximum linear over-excavation prediction value is higher than the limit, modify the blasting design parameters and return to step S5 to re-predict until the output maximum linear over-excavation prediction value is less than or equal to the limit, then output the final blasting design parameters.

2. A method for optimizing design parameters of tunnel smooth blasting based on deep learning and neural network according to claim 1, characterized in that: In step S32, the SP-DWConv2D convolution layer includes a Split layer at the front end of the input, three depth convolution kernels, an Identity layer, and a concat layer at the back end for feature fusion. The sizes of the three depth convolution kernels are The convolution kernel, The zona flexor and of the zona nucleus.

3. A tunnel smooth blasting design parameter optimization method based on deep learning and neural network according to claim 2, characterized in that: In step S3, when classifying and identifying the main structural surface inclination angle, a linear frame is used to mark and classify the main structural surface in the tunnel face image, and the results of the marking and classification are saved in a JSON file. The SP-Yolov8 model is used to combine the marking and classification results with the tunnel face image to identify the main structural surface inclination angle category.

4. A method for optimizing design parameters of tunnel smooth blasting based on deep learning and neural network according to claim 1, characterized in that: In step S4, the step of extracting the main structural surface information from the main structural surface area includes: S41: Input the height Y of the tunnel face image and the height h of the tunnel face into the SP-YOLOv8 model, and use the coordinates of the lower right corner endpoint of the tunnel face image as the origin; S42: Segment the tunnel face image using the YOLOv8-seg model, and expand the main structural surface to form a mask area. The coordinates of the lower right corner of the mask area box are ; S43: extract the main structural surface area and perform image processing using binarization; S44: Use Zhang-Suen algorithm to extract skeleton of the binarized image. Iterate the area to be extracted multiple times and indent from the edge to the center to form a trace. The coordinates of the right endpoint of the trace are ; S45: Obtain the pixel height of the right endpoint of the main structural surface through the image pixel height , according to the ratio of the actual tunnel face height, calculate the actual height of the right end point of the main structural surface , thus obtaining the position of the main structural surface.

5. An electronic device comprising a processor and a memory, characterized in that: Computer instructions are stored in the memory, and the processor is used to run the computer instructions stored in the memory to implement the steps of a tunnel smooth blasting design parameter optimization method based on deep learning and neural network as described in any one of claims 1 to 4.

6. A computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to enable a computer to execute the steps of a tunnel smooth blasting design parameter optimization method based on deep learning and neural network as described in any one of claims 1 to 4.

Images