Water pressure blasting simulation method based on big data

Abstract

The invention discloses a water pressure blasting simulation method based on big data, which comprises the following steps: acquiring blasting area data and preprocessing to obtain preprocessed data; constructing a three-dimensional model of a blasting area based on the preprocessed data, and performing a tunnel water pressure blasting experiment on the three-dimensional model of the blasting area to obtain experimental data; establishing an initial tunnel water pressure blasting effect prediction model, and training the initial tunnel water pressure blasting effect prediction model through the experimental data to obtain a tunnel water pressure blasting effect prediction model; and performing blasting simulation through the tunnel water pressure blasting effect prediction model to obtain predicted blasting data. According to the method, blasting simulation is carried out by constructing the tunnel water pressure blasting effect prediction model, blasting can be predicted in advance, and a theoretical basis is provided for blasting testing.

Description

Technical Field

[0001] The present invention belongs to the technical field of data simulation blasting, and in particular relates to a water pressure blasting simulation method based on big data. Background Art

[0002] The application of hydraulic blasting technology in tunnel excavation construction can improve work efficiency, relieve construction period pressure, save costs, and protect the environment. In traditional tunnel hydraulic blasting practice, the selection and setting of blasting parameters generally rely on empirical formulas or results obtained through small-scale tests. This method has obvious limitations, because empirical formulas often cannot fully consider all factors that affect the blasting effect, and small-scale tests are difficult to accurately replicate the complex geological conditions and environmental variables during actual blasting. Therefore, traditional methods are not only difficult to achieve precise control of blasting effects, but also have low resource utilization, limited optimization of blasting design, and may also lead to increased safety risks.

[0003] With the development of big data technology, in order to improve the accuracy and efficiency of tunnel hydraulic blasting, the development of a tunnel hydraulic blasting simulation method based on big data has become an important demand in the industry. Big data technology can process and analyze massive amounts of data, including geological data, physical and mechanical property data, and environmental condition data. Using this data, a more accurate blasting model can be constructed, and previously unknown blasting laws and influencing factors can be discovered through data analysis, providing a scientific basis for the optimization of blasting parameters. However, there is still a lack of a systematic simulation method. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a water pressure blasting simulation method based on big data to solve the problems existing in the above-mentioned prior art.

[0005] To achieve the above object, the present invention provides a hydraulic blasting simulation method based on big data, comprising the following steps:

[0006] Collect blasting area data and pre-process them to obtain pre-processed data;

[0007] Constructing a three-dimensional model of the blasting area based on the preprocessed data, and conducting a tunnel water pressure blasting experiment on the three-dimensional model of the blasting area to obtain experimental data;

[0008] Establishing an initial tunnel water pressure blasting effect prediction model, training the initial tunnel water pressure blasting effect prediction model through the experimental data, and obtaining a tunnel water pressure blasting effect prediction model;

[0009] The tunnel hydraulic blasting effect prediction model is used to perform blasting simulation to obtain predicted blasting data.

[0010] Preferably, the blasting area data includes geological structure data, rock physical and mechanical properties and environmental conditions;

[0011] The geological structure data include stratum structure, rock type and degree of fracture development;

[0012] The rock physical and mechanical properties include rock density, elastic modulus, Poisson's ratio, compressive strength and tensile strength;

[0013] The environmental conditions include the temperature of the blasting area.

[0014] Preferably, the preprocessing method includes performing missing value processing and data standardization on the blasting area data; and obtaining key features of the standardized blasting area data by principal component analysis.

[0015] Preferably, the method of obtaining key features by principal component analysis includes:

[0016] The principal components of the standardized blasting area data were extracted, and the variance and covariance of each principal component were calculated;

[0017] The appropriate number of principal components is selected according to the variance of the principal components.

[0018] The eigenvector corresponding to the selected principal component is calculated, and the standardized data is projected onto the eigenvector to obtain the score of each sample on the selected principal component, and the blasting area data is reduced to the principal component.

[0019] Preferably, the initial tunnel hydraulic blasting effect prediction model is a network model based on a support vector machine.

[0020] Preferably, the method for obtaining the tunnel hydraulic blasting effect prediction model includes training and verifying the trained model by a bootstrap method, and the training and verification methods include:

[0021] Randomly extract a sample from the original dataset, copy it and put it back into the dataset, and then repeat this process until a new dataset of the same size as the original dataset is generated;

[0022] Further dividing the new data set into multiple subsets for cross-validation;

[0023] Each time, select a subset as the validation set, and merge the remaining subsets into the training set. Train the model and evaluate its performance on the validation set. Repeat the evaluation process until each subset has been used as a validation set.

[0024] Record the model accuracy of each cross-validation, summarize all cross-validation accuracies, and conduct a comprehensive evaluation.

[0025] Preferably, the tunnel hydraulic blasting simulation method further comprises optimizing the tunnel hydraulic blasting effect prediction model through actual blasting data. The method for optimizing the tunnel hydraulic blasting effect prediction model comprises:

[0026] When blasting is performed in the blasting area, actual blasting data is collected, and parameters of the tunnel water pressure blasting effect prediction model are adjusted based on the actual blasting data and the predicted blasting data to obtain an optimized tunnel water pressure blasting effect prediction model;

[0027] Based on the optimized tunnel hydraulic blasting effect prediction model, the predicted blasting area is simulated, the blasting result is predicted, and the above steps are iterated for continuous optimization.

[0028] Preferably, the continuously optimized updating method further comprises:

[0029] The performance indicators of the tunnel water pressure blasting effect prediction model before and after the update are compared to confirm whether the update has achieved the expected effect; if the update causes new problems, the rollback mechanism is triggered to restore the tunnel water pressure blasting effect prediction model before the update.

[0030] Compared with the prior art, the present invention has the following advantages and technical effects:

[0031] The present invention discloses a water pressure blasting simulation method based on big data, comprising the following steps: collecting blasting area data and preprocessing it to obtain preprocessed data; constructing a three-dimensional model of the blasting area based on the preprocessed data, and conducting a tunnel water pressure blasting experiment on the three-dimensional model of the blasting area to obtain experimental data; establishing an initial tunnel water pressure blasting effect prediction model, and training the initial tunnel water pressure blasting effect prediction model through the experimental data to obtain a tunnel water pressure blasting effect prediction model; and conducting a blasting simulation through the tunnel water pressure blasting effect prediction model to obtain predicted blasting data. The present invention can predict blasting in advance by constructing a tunnel water pressure blasting effect prediction model to perform blasting simulation, provide a theoretical basis for blasting testing, significantly improve the success rate and safety of blasting operations, and effectively reduce unnecessary resource consumption to achieve optimal resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0033] Figure 1 This is a flow chart of a water pressure blasting simulation method based on big data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0034] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0035] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0036] Embodiment 1

[0037] like Figure 1 As shown, in this embodiment, a water pressure blasting simulation method based on big data is provided, which is characterized by comprising the following steps:

[0038] Collect blasting area data and pre-process them to obtain pre-processed data;

[0039] Constructing a three-dimensional model of the blasting area based on the preprocessed data, and conducting a tunnel water pressure blasting experiment on the three-dimensional model of the blasting area to obtain experimental data;

[0040] Establishing an initial tunnel water pressure blasting effect prediction model, training the initial tunnel water pressure blasting effect prediction model through the experimental data, and obtaining a tunnel water pressure blasting effect prediction model;

[0041] The tunnel hydraulic blasting effect prediction model is used to perform blasting simulation to obtain predicted blasting data.

[0042] Furthermore, the blasting area data includes geological structure data, rock physical and mechanical properties and environmental conditions;

[0043] The geological structure data include stratum structure, rock type and degree of fracture development;

[0044] The rock physical and mechanical properties include rock density, elastic modulus, Poisson's ratio, compressive strength and tensile strength;

[0045] The environmental conditions include the temperature of the blasting area.

[0046] The specific methods to obtain these data are:

[0047] Acquisition of geological structure data: Intelligent lithology identification: With the development and use of intelligent drilling rigs, the lithology data of blasting boreholes can be accurately obtained. The intelligently identified blasthole data is stored and managed by establishing a blasthole database. Using the blasthole lithology data as a sample, the inverse square method of distance is used to interpolate the solid units within the blasting area to generate a three-dimensional solid model of the blasted rock mass.

[0048] The acquisition of rock physical and mechanical property data includes: Indoor tests: including rock density, elastic modulus, Poisson's ratio, and compressive and tensile strength tests. Field tests: such as rock pressure tests and shear tests, can measure the mechanical properties of rock in situ.

[0049] The acquisition of environmental condition data includes: Temperature monitoring: By arranging temperature sensors, the temperature changes in the blasting area can be monitored in real time.

[0050] Furthermore, the preprocessing method includes performing missing value processing and data standardization on the blasting area data; and obtaining key features of the standardized blasting area data through principal component analysis.

[0051] Furthermore, methods for obtaining key features through principal component analysis include:

[0052] First, the data from the blasting area needs to be standardized. This is to eliminate the effects of different dimensions and orders of magnitude so that each feature is on the same scale, thereby ensuring the fairness and accuracy of subsequent analysis. The standardized data will have zero mean and unit variance, which is an ideal starting point for principal component analysis.

[0053] Next, principal component extraction is performed on these standardized data. This step involves calculating the covariance matrix of the data and extracting the eigenvalues ​​and eigenvectors from it. The eigenvalues ​​represent the variance of the corresponding principal components, while the eigenvectors define the directions of these principal components. In this way, the contribution of each principal component to the total variation can be quantified, thereby determining which principal components are the most critical.

[0054] Choosing the right number of principal components is a key step. Usually, the number of principal components to be retained is determined based on the size of the eigenvalues ​​or the cumulative explained variance ratio. For example, if the first few principal components have explained most of the variance, then only these principal components can be retained to achieve data dimensionality reduction.

[0055] Once the number of principal components is determined, the eigenvectors corresponding to these principal components can be calculated. Then, the standardized data is projected onto these eigenvectors, so that the coordinates, or scores, of each sample in the new feature space can be obtained. These scores reflect the relative position of the samples on the selected principal components, which helps to understand the relationships and differences between samples.

[0056] Finally, through this projection, the original high-dimensional blow-up area data is reduced to the selected principal components. This not only reduces the complexity of the data, but also helps to discover the underlying patterns and structures in the data. The reduced-dimensional data can be used for further analysis, such as clustering, classification, or visualization, to support decision-making.

[0057] Furthermore, the initial tunnel hydraulic blasting effect prediction model is a network model based on support vector machine.

[0058] Further, first, a random sample is taken from the original dataset, copied and put back into the dataset, and then this process is repeated until a new dataset of the same size as the original dataset is generated. This method is called bootstrapping, and it can be used to evaluate the performance and stability of the model. In this way, multiple different datasets are generated, each of which contains all the information in the original dataset, but in different orders and combinations.

[0059] Next, the new dataset is further divided into multiple subsets for cross-validation. Cross-validation is a statistical method used to evaluate the performance of machine learning models. In this method, we divide the dataset into k subsets, select one subset each time as the validation set, and combine the remaining k-1 subsets into training sets. We then use the training set to train the model and evaluate its performance on the validation set. This process is repeated k times until each subset has been used as a validation set once.

[0060] The accuracy of the model is recorded during each cross-validation process. Accuracy is an important indicator for measuring model performance, which indicates the proportion of correct predictions made by the model. By comparing the accuracy on different subsets, we can understand whether the model performs consistently on different data subsets and whether there is an overfitting or underfitting problem.

[0061] Finally, all cross-validation accuracies are summarized and comprehensively evaluated. Comprehensive evaluation can help us understand the performance of the model more comprehensively and determine whether further adjustments to the model parameters or structure are needed. If the accuracies of all cross-validation are high and stable, then it can be considered that the model has good generalization ability and can accurately predict unknown data.

[0062] Furthermore, the tunnel hydraulic blasting simulation method further includes optimizing the tunnel hydraulic blasting effect prediction model through actual blasting data. The method for optimizing the tunnel hydraulic blasting effect prediction model includes:

[0063] When blasting is performed in the blasting area, actual blasting data is collected, and parameters of the tunnel water pressure blasting effect prediction model are adjusted based on the actual blasting data and the predicted blasting data to obtain an optimized tunnel water pressure blasting effect prediction model;

[0064] Based on the optimized tunnel hydraulic blasting effect prediction model, the predicted blasting area is simulated, the blasting result is predicted, and the above steps are iterated for continuous optimization.

[0065] Furthermore, the continuously optimized updating method further includes:

[0066] The performance indicators of the tunnel water pressure blasting effect prediction model before and after the update are compared to confirm whether the update has achieved the expected effect; if the update causes new problems, the rollback mechanism is triggered to restore the tunnel water pressure blasting effect prediction model before the update.

[0067] After updating the tunnel hydraulic blasting effect prediction model, it is necessary to conduct a detailed comparative analysis of its performance indicators. By comparing the model performance before and after the update, it can be determined whether the update has achieved the expected effect. If it is found that the updated model performs poorly in some aspects or even has new problems, then the rollback mechanism needs to be triggered.

[0068] The rollback mechanism is a safety measure that ensures that the system can be quickly restored to the state before the update when there is a problem with the update. In this case, it is necessary to immediately stop using the updated tunnel hydraulic blasting effect prediction model and quickly restore to the model before the update. This can prevent the problem caused by the update from having a greater impact on the entire system.

[0069] This embodiment uses a network model based on a support vector machine to predict unknown blasting effect indicators, such as footage per cycle, unit rock charge consumption, half-eye mark retention rate, dust concentration, etc., based on known engineering blasting parameters. This method can not only guide blasting design and improve the accuracy and safety of the design, but also predict and generate blasting effect samples of unknown working conditions by mining the inherent laws in the data under limited learning samples, thereby realizing comprehensive data analysis.

[0070] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A hydraulic blasting simulation method based on big data, characterized in that: The following steps are involved: Collect blasting area data and pre-process them to obtain pre-processed data; Constructing a three-dimensional model of the blasting area based on the preprocessed data, and conducting a tunnel water pressure blasting experiment on the three-dimensional model of the blasting area to obtain experimental data; Establishing an initial tunnel water pressure blasting effect prediction model, training the initial tunnel water pressure blasting effect prediction model through the experimental data, and obtaining a tunnel water pressure blasting effect prediction model; The tunnel hydraulic blasting effect prediction model is used to perform blasting simulation to obtain predicted blasting data.

2. The hydraulic blasting simulation method based on big data according to claim 1 is characterized in that: The blasting area data include geological structure data, rock physical and mechanical properties and environmental conditions; The geological structure data include stratum structure, rock type and degree of fracture development; The rock physical and mechanical properties include rock density, elastic modulus, Poisson's ratio, compressive strength and tensile strength; The environmental conditions include the temperature of the blasting area.

3. The hydraulic blasting simulation method based on big data according to claim 1 is characterized in that: The preprocessing method includes performing missing value processing and data standardization on the blasting area data; and obtaining key features of the standardized blasting area data through a principal component analysis method.

4. The hydraulic blasting simulation method based on big data according to claim 3 is characterized in that: The method for obtaining key features by principal component analysis includes: The principal components of the standardized blasting area data were extracted, and the variance and covariance of each principal component were calculated; selecting the number of principal components based on the variance of the principal components; The eigenvector corresponding to the selected principal component is calculated, and the standardized data is projected onto the eigenvector to obtain the score of each sample on the selected principal component, and the blasting area data is reduced to the principal component.

5. The hydraulic blasting simulation method based on big data according to claim 1 is characterized in that: The initial tunnel hydraulic blasting effect prediction model is a network model based on support vector machine.

6. The hydraulic blasting simulation method based on big data according to claim 1 is characterized in that: The method for obtaining the tunnel hydraulic blasting effect prediction model includes training and verifying the trained model through a bootstrap method, and the training and verification methods include: Randomly extract a sample from the original dataset, copy it and put it back into the dataset, and then repeat this process until a new dataset of the same size as the original dataset is generated; Further dividing the new data set into multiple subsets for cross-validation; Each time, select a subset as the validation set, and merge the remaining subsets into the training set. Train the model and evaluate its performance on the validation set. Repeat the evaluation process until each subset has been used as a validation set. Record the model accuracy of each cross-validation, summarize all cross-validation accuracies, and conduct a comprehensive evaluation.

7. The hydraulic blasting simulation method based on big data according to claim 1 is characterized in that: The tunnel hydraulic blasting simulation method further includes optimizing the tunnel hydraulic blasting effect prediction model through actual blasting data. The method for optimizing the tunnel hydraulic blasting effect prediction model includes: When blasting is performed in the blasting area, actual blasting data is collected, and parameters of the tunnel water pressure blasting effect prediction model are adjusted based on the actual blasting data and the predicted blasting data to obtain an optimized tunnel water pressure blasting effect prediction model; Based on the optimized tunnel hydraulic blasting effect prediction model, the predicted blasting area is simulated, the blasting result is predicted, and the above steps are iterated for continuous optimization.

8. The water pressure blasting simulation method based on big data according to claim 7 is characterized in that: The continuously optimized updating method further includes: The performance indicators of the tunnel water pressure blasting effect prediction model before and after the update are compared to confirm whether the update has achieved the expected effect; if the update causes new problems, the rollback mechanism is triggered to restore the tunnel water pressure blasting effect prediction model before the update.

Images