A method and system for delay selection between mine blasting holes

Abstract

The present invention relates to the technical field of delay selection between mine blasting holes, specifically a method and system for delay selection between mine blasting holes, the method comprising the following steps: obtaining geological multidimensional data, processing the geological multidimensional data to obtain geological multidimensional target data; establishing an improved optimization algorithm and setting optimization conditions; establishing a parameter update model based on the improved optimization algorithm and the optimization conditions; obtaining an optimal delay parameter through the parameter update model, and obtaining an optimal delay prediction model based on the optimal delay parameter; training and predicting the optimal delay prediction model based on the geological multidimensional target data to select the optimal blasting delay time between mine blasting holes. The present invention optimizes the geological multidimensional data, establishes a parameter optimization model to screen and adjust the optimal parameter combination of the model to obtain an optimal prediction model, enhances the reliability of the delay prediction result, and reduces the prediction error caused by improper human experience judgment.

Description

Technical Field

[0001] The invention relates to the technical field of time delay selection between mine blasting holes, and in particular to a time delay selection method and system between mine blasting holes. Background Art

[0002] In mining operations, explosive blasting is an indispensable part, and the precise control of the inter-hole delay strategy plays a decisive role in the blasting effect. It can not only significantly reduce the vibration intensity generated during the blasting process and protect the safety of surrounding structures, but also significantly improve the efficiency and uniformity of rock crushing. Under specific optimization conditions, it can also enhance the main frequency effect of blasting to ensure that the blasting operation reaches the preset ideal state. However, the existing technology usually relies on the operator's experience and judgment in designing the blasting delay time, lacks scientificity and systematicity, and often leads to unsatisfactory blasting results, which not only affects the mining efficiency, but also increases safety hazards and cost burdens.

[0003] In order to determine the optimal blasting delay time, the blasting effects under different delay conditions are now mostly simulated through model experiments and numerical simulation methods. However, the above methods are limited by computing resources and can only simulate a limited number of blastholes, which makes it difficult to fully reflect the complex situation of actual multi-row blasting. On the other hand, even if there are sufficient computing resources, high-precision and large-scale numerical simulations take a lot of time, making it difficult to respond to on-site needs in a timely manner and formulate flexible blasting plans.

[0004] Existing technologies are mainly based on subjective judgment based on experience and rely on numerical simulations performed by computers. In the process of obtaining the most optimal blasting delay time, there are limitations to varying degrees, including but not limited to excessive subjectivity, limited simulation accuracy, and excessive time consumption. Therefore, a more scientific, efficient, and flexible delay selection method is urgently needed to optimize the blasting design scheme and improve the overall efficiency of mining. Summary of the invention

[0005] In view of the defects of existing methods and the shortcomings of practical applications, in order to reduce the errors caused by human subjective experience, reduce the intensity of blasting vibration and improve the quality of blasting, the present invention optimizes the geological multidimensional data, substitutes the data into the algorithm to optimize the hyperparameters and trains through a neural network to obtain regression prediction results, and finally determines the optimal blasting delay time in combination with the error analysis evaluation results. In the first aspect, the present invention provides a delay selection method between mine blasting holes, the method comprising the following steps: obtaining geological multidimensional data, processing the geological multidimensional data to obtain geological multidimensional target data; establishing an improved optimization algorithm and setting optimization conditions; establishing a parameter update model based on the improved optimization algorithm and the optimization conditions; obtaining the optimal delay parameters through the parameter update model, and obtaining the optimal delay prediction model based on the optimal delay parameters; training and predicting the optimal delay prediction model based on the geological multidimensional target data to select the optimal blasting delay time between mine blasting holes. The present invention processes and analyzes geological multidimensional data and combines it with an improved optimization algorithm to find the optimal blasting delay parameters for current geological conditions to obtain the best blasting delay time, which can optimize the distribution of blasting energy, reduce invalid explosions, and improve the overall blasting effect.

[0006] Optionally, the processing of the geological multidimensional data to obtain geological multidimensional target data includes: performing a first data processing on the geological multidimensional data and obtaining a first data processing result, wherein the first data processing includes an algorithm for removing outliers; performing a second data processing on the first data processing result and obtaining geological multidimensional target data, wherein the second data processing includes an algorithm for interpolating missing values. The present invention removes outliers and interpolates missing values ​​from geological multidimensional data, which helps to obtain high-quality geological multidimensional target data and improve the accuracy and reliability of data analysis results.

[0007] Optionally, the algorithm for removing outliers satisfies the following relationship:

[0008] ,

[0009] in, Indicates the normal value range of the data. represents the average value of any blasting characteristic parameter, represents standard deviation;

[0010] The missing value interpolation algorithm satisfies the following relationship:

[0011] ,

[0012] in, Indicates the pending The The data of blasting parameters, Indicates The The data of blasting parameters, Indicates The The data of blasting parameters, The invention significantly improves the quality of geological multidimensional data by eliminating outliers and interpolating missing values, making it more accurate, complete and reliable, which is helpful for subsequent data analysis and model building, and provides a more scientific decision-making basis for geological engineering such as mine blasting.

[0013] Optionally, the establishing of the improved optimization algorithm includes: introducing a whale optimization algorithm, improving the whale optimization algorithm, and obtaining an improved optimization algorithm; the improved optimization algorithm satisfies the following relationship;

[0014] ,

[0015] in, represents the next position of the individual after iteration, represents the position of the individual in the current iteration, represents a constant, Indicates the position of another randomly selected whale. The optimization algorithm of the present invention has a powerful global search capability. By improving the above algorithm, its search efficiency in complex geological data can be further enhanced, and the solution space can be explored more comprehensively, so as to find a better blasting delay parameter.

[0016] Optionally, the setting of the optimization condition includes: setting a first optimization condition, the first optimization condition satisfies ; Set a second optimization condition, the second optimization condition satisfies The present invention sets different conditions to control the search process more finely, so that the algorithm adopts a more radical search strategy, thereby obtaining a wider solution space.

[0017] Optionally, the establishing of the parameter update model based on the improved optimization algorithm and the optimization condition includes: establishing a first parameter update model based on the improved optimization algorithm and the first optimization condition; establishing a second parameter update model based on the improved optimization algorithm and the second optimization condition; and obtaining a parameter update model by combining the first parameter update model and the second parameter update model. The present invention establishes models for different conditions, and can adjust parameters more accurately to ensure the optimal search effect at different stages.

[0018] Optionally, the parameters update the first model to satisfy the following relationship:

[0019]

[0020] in, represents the next position of the individual after iteration, represents the optimal position of the whale in the current iteration, Represents a random number between [0, 1], represents the position of another whale chosen at random;

[0021] The parameters update the second model to satisfy the following relationship:

[0022]

[0023] in, represents the next position of the individual after iteration, represents the optimal position of the whale in the current iteration, Indicates a coefficient that gradually decreases as the number of iterations increases. The present invention combines the two models to achieve high efficiency and accuracy in the search process, thereby improving the overall optimization effect, making the method of the present invention flexible and adaptable when dealing with complex problems.

[0024] Optionally, obtaining the optimal delay parameters through the parameter updating model includes: optimizing the model parameters of the delay prediction model using the parameter updating model, and obtaining the optimal delay parameters, wherein the model parameters include the number of hidden units, the maximum training cycle, and the initial learning rate. The present invention can systematically adjust the key parameters of the delay prediction model through the parameter updating model, thereby improving the accuracy and reliability of the prediction results.

[0025] Optionally, the training and prediction of the optimal time delay prediction model based on the geological multi-dimensional target data to select the optimal blasting delay time between mine blasting holes includes: training and predicting the optimal time delay prediction model based on the geological multi-dimensional target data, and the prediction evaluation indicators include mean absolute error, root mean square error, mean absolute percentage error and determination coefficient; combining the geological multi-dimensional target data, the optimal time delay prediction model and the prediction evaluation indicators to select the optimal blasting delay time between mine blasting holes. The present invention trains and predicts the optimal time delay prediction model based on geological multi-dimensional target data to select the optimal blasting delay time between mine blasting holes, which improves the prediction accuracy, scientificity and objectivity.

[0026] In the second aspect, the present invention also provides a time delay selection system between mine blasting holes, which can efficiently execute a time delay selection method between mine blasting holes provided by the present invention, the system includes an input device, a processor, an output device and a memory, wherein the input device, the processor, the output device and the memory are interconnected, the memory includes a computer-readable storage medium as described in the first aspect of the present invention, the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions. The system provided by the present invention has a compact structure, strong applicability, and greatly improves the operating efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 A flow chart of a method for selecting a time delay between mine blasting holes according to the present invention;

[0028] Figure 2 It is a flow chart of a parameter updating model in the method for selecting the delay between mine blasting holes of the present invention;

[0029] Figure 3 It is a structural schematic diagram of the delay selection system between mine blasting holes of the present invention. DETAILED DESCRIPTION

[0030] The specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described herein are only for illustration and are not intended to limit the present invention. In the following description, a large number of specific details are set forth in order to provide a thorough understanding of the present invention. However, it is obvious to those of ordinary skill in the art that these specific details do not need to be adopted to implement the present invention. In other examples, in order to avoid confusing the present invention, known circuits, software or methods are not specifically described.

[0031] Throughout the specification, references to "one embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment or example is included in at least one embodiment of the present invention. Therefore, the phrases "in one embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily all refer to the same embodiment or example. In addition, particular features, structures, or characteristics may be combined in one or more embodiments or examples in any suitable combination and / or subcombination. In addition, it should be understood by those of ordinary skill in the art that the figures provided herein are for illustrative purposes and that the figures are not necessarily drawn to scale.

[0032] See also Figure 1In order to improve the safety and effectiveness of blasting operations and reduce the impact of subjective human judgment on the results, the present invention optimizes geological multidimensional data to ensure the accuracy and comprehensiveness of the data; then the data is input into the improved algorithm to adjust the model parameters, so that the model can more accurately predict the complex relationship between geological conditions and blasting delay time, thereby improving the overall quality of blasting operations and providing a strong guarantee for mine safety and efficient mining. The present invention provides a method for selecting a delay between mine blasting holes, and the above-mentioned method for selecting a delay between mine blasting holes includes the following steps:

[0033] S1. Acquire geological multidimensional data and process the geological multidimensional data to obtain geological multidimensional target data. The implementation steps and specific contents are as follows:

[0034] In order to comprehensively and accurately collect multi-dimensional data related to geological and blasting operations. In the embodiment, the geological multi-dimensional data mainly covers geological characteristics, such as the specific type of rock, hardness level and degree of joint development; the specific arrangement details of the blastholes, including but not limited to the hole size, hole depth, distance between holes, spacing between rows, angle setting of blastholes, selection of filling materials and filling length; explosive parameters, such as explosive type, charge amount per hole, charge density, etc., as well as the core parameters of blasting design such as historical inter-hole delay data and the blasting mode adopted.

[0035] In an optional embodiment, rock drilling equipment can be used to drill underground rock samples, including but not limited to mechanical hand-pushed drills, hydraulic drills, and rotary drills. Then, a rock analyzer, such as an X-ray diffractometer, an electron microscope, etc., is used to conduct a detailed analysis of the rock samples to determine the rock type, mineral composition, and chemical properties; a total station can be used to measure the surface topography, slope, altitude, etc., and satellite positioning technology can also be used to achieve rapid and accurate measurement of the blasthole position; a multi-sensor system carried by an unmanned aerial vehicle or a manned aircraft, including but not limited to a laser radar, a high-definition camera, etc., can be used to obtain a large range of geological and topographic data and blasthole layout from the air. In addition, a ground-based three-dimensional laser scanner can be used to obtain a fine three-dimensional model of the surface, including information such as rock surface morphology and joint distribution.

[0036] In the embodiment, relevant monitoring equipment is arranged to obtain multi-dimensional geological data. Different geological conditions and blasting requirements require the selection or combination of different data acquisition equipment, thereby ensuring and improving the efficiency and accuracy of data acquisition, and providing strong support for geological exploration and blasting design plans.

[0037] Then, the first data processing is performed on the geological multidimensional data to obtain the first data processing result. In the embodiment, the first data processing mainly includes removing abnormal values, and its specific implementation content is as follows:

[0038] In order to ensure the robustness and accuracy of subsequent data analysis, outliers in geological multidimensional data are caused by measurement errors, entry errors, or extreme natural events.

[0039] First, the text data of geological multidimensional data is converted into numerical data for subsequent statistical analysis or machine learning modeling, which mainly involves text encoding technology, such as unique hot encoding, label encoding, bag of words model, word frequency-inverse document frequency conversion, etc., based on which the delay selection method of the present invention can be applied to different scenarios. When the text information is represented digitally, in this embodiment, the explosive type and its digital encoding relationship are as follows: [emulsion explosive, colloid explosive, ammonium ladder explosive, slurry explosive] = [1, 2, 3, 4], which refers to emulsion explosive (1), colloid explosive (2), ammonium ladder explosive (3), slurry explosive (4). In addition to directly mapping to a digital sequence, this corresponding relationship can also be expressed in a variety of ways to increase readability, understandability or meet the optimized application scenario.

[0040] The first data processing is performed on the geological multidimensional data, that is, the outliers in the geological multidimensional data need to be eliminated, and the first data processing result of the geological multidimensional data is obtained based on the outlier elimination algorithm, and the following relationship needs to be satisfied:

[0041] ,

[0042] in, Indicates the normal value range of the data. represents the average value of any blasting characteristic parameter, represents standard deviation;

[0043] The normal value range of data refers to the normal fluctuation range of geological multidimensional data or its data characteristics determined by applying statistical methods. Using the normal value range of data as a reference standard for the data is conducive to improving the quality and accuracy of geological multidimensional data.

[0044] In this embodiment, different blasting characteristic parameters of the multidimensional data set have unique variability, which can be quantified by calculating the standard deviation of the characteristic parameters. The standard deviation is an important indicator for measuring the degree of data dispersion, and can reflect the degree of deviation between any column of data values ​​and their average value based on each column of data.

[0045] Next, the second data processing is performed on the first data processing result to obtain geological multi-dimensional target data. In the embodiment, the second data processing mainly includes interpolating missing values, and its specific implementation content is as follows:

[0046] In the embodiment, the second data processing is performed on the result of the first data processing, that is, the geological multidimensional data after the missing values ​​are identified is selectively interpolated to obtain the geological multidimensional target data.

[0047] In an optional embodiment, the missing values ​​can be filled with the average of other non-missing values ​​in the variable; similar to mean interpolation, the median can be used as the filling value: for categorical variables, the mode, that is, the value with the most occurrences, can be used to fill the missing values; the missing values ​​can also be predicted based on the known values ​​calculated based on the non-missing features according to the most similar neighbors; in addition, multiple complete data sets can be created, each of which has different interpolations for missing values, and then the above data sets are statistically analyzed, and the results are combined to obtain the final data processing results. In practical applications, filling missing values ​​can solve the problem of data integrity, but it may also introduce certain deviations or uncertainties. Therefore, when selecting an interpolation method, the characteristics of the data and the analysis requirements should be considered to ensure the reliability of the data processing results.

[0048] The missing value interpolation algorithm in the embodiment needs to satisfy the following relationship:

[0049] ,

[0050] in, Indicates the pending The The data of blasting parameters, Indicates The The data of blasting parameters, Indicates The The data of blasting parameters, Characterizes the blasting parameters.

[0051] In a multidimensional dataset, the blasting parameter value optimized in a certain dimension or optimized position. The dimension refers to a certain dimension or optimized position of the geological multidimensional dataset, and i refers to the characteristics of the above blasting parameters. When processing blasting parameter data, special attention should be paid to the accuracy, completeness and consistency of the relevant data to avoid deviations or errors in the analysis results caused by data problems, so as to ensure that the data meets the needs of analysis or modeling.

[0052] Furthermore, the method for acquiring geological multidimensional target data in this embodiment is only an optional condition of the present invention. In one or some other embodiments, the method for acquiring multidimensional target data can be flexibly adjusted according to the actual situation of geological data and multidimensional data, optimized according to the characteristics of geological data, and unnecessary calculation and data conversion steps can be reduced, thereby improving the overall data processing efficiency and enhancing the accuracy and reliability of multidimensional target data.

[0053] S2. Establish an improved optimization algorithm and set optimization conditions. The specific steps and implementation contents are as follows:

[0054] In an optional embodiment, the whale optimization algorithm is first introduced, and the whale optimization algorithm is improved to obtain an improved optimization algorithm, the specific contents of which are as follows:

[0055] In the embodiment, a whale optimization algorithm is introduced and improved. The whale optimization algorithm is a meta-heuristic optimization algorithm inspired by the hunting behavior of humpback whales, which simulates the hunting strategy used by humpback whales when looking for prey.

[0056] The whale optimization algorithm mainly consists of three main stages: surrounding prey, bubble net attack and searching for prey. Surrounding prey means that the whale group moves towards the best whale, that is, the current optimal solution; bubble net attack means that the whales perform spiral movements around the prey to simulate the formation of a bubble net. As the whale group gradually narrows the encirclement to more accurately locate the prey, the whales will randomly search for new locations to increase the diversity of exploration.

[0057] In order to further improve the local search capability, in each iteration, a fine search is performed on the small neighborhood around the current optimal solution to find possible better solutions. A restart mechanism is also set up. When the algorithm falls into a local optimum, the algorithm is restarted and the positions of some or all whales are reinitialized to jump out of the local optimum.

[0058] The above improved optimization algorithm satisfies the following relationship:

[0059]

[0060] in, represents the next position of the individual after iteration, represents the position of the individual in the current iteration, represents a constant, represents the position of another whale chosen at random.

[0061] By introducing The parameter algorithm can conduct a wider search on a global scale, thereby increasing the possibility of discovering new solution areas. By adjusting the constant value, the algorithm can balance between exploring new solution areas and developing known excellent solutions, which helps to avoid premature convergence while maintaining the convergence speed of the algorithm. In addition, the position can be randomly selected and updated in combination with the positions of other whales, which helps to maintain the diversity of the population and avoid falling into the local optimal solution.

[0062] Then, you need to set the optimization conditions, the specific contents are as follows:

[0063] In the bubble network attack spiral update stage of the optimization algorithm, the optimization condition is introduced to apply different update strategies according to different conditions, so that the algorithm can adapt to different optimization problems or optimization processes and different stages. Set the first optimization condition. In the embodiment, the first optimization condition satisfies ; Set the second optimization condition. In the embodiment, the second optimization condition satisfies .

[0064] In the spiral update process of the optimization algorithm, if the first optimization condition is met, the first model is updated with parameters; conversely, if the second optimization condition is met, the second model is updated with parameters. Different update strategies are adopted for different conditions, so that the algorithm can search in the solution space more efficiently. For example, a wider search range is used in the exploration phase, while the search range is narrowed and fine-tuning is focused on in the development phase. In the embodiment, the optimization conditions are set to make the design of the algorithm more flexible, and then the condition threshold and update strategy can be adjusted according to actual problems and needs to obtain a more effective optimization algorithm.

[0065] Furthermore, the improved method of the optimization algorithm in the present embodiment is only an optional condition of the present invention. In one or some other embodiments, the improved method of the optimization algorithm can be replaced according to the optimization requirements and model parameter characteristics. By adjusting the improved method of the optimization algorithm according to the optimization problem and model characteristics, the behavior of the algorithm can be adjusted more accurately to obtain better performance.

[0066] S3. Based on the above improved optimization algorithm optimization conditions, a parameter update model is established. The specific implementation steps and related contents are as follows:

[0067] Based on the above improved optimization algorithm and the first optimization condition, a first parameter updating model is established. When the first optimization condition is met, that is, , the above parameters update the first model to satisfy the following relationship:

[0068]

[0069] in, represents the next position of the individual after iteration, represents the optimal position of the whale in the current iteration, Represents a random number between [0, 1], represents the position of another whale chosen at random;

[0070] Based on the above improved optimization algorithm and the second optimization condition, the second model is established with parameters updated. When the second optimization condition is met, that is, , the above parameters update the second model to satisfy the following relationship:

[0071]

[0072] in, represents the next position of the individual after iteration, represents the optimal position of the whale in the current iteration, Represents a coefficient that gradually decreases as the number of iterations increases.

[0073] The above, and A random number between [0-1].

[0074] The coefficients that gradually decrease as the number of iterations increases need to satisfy the following relationship:

[0075]

[0076] in, represents the coefficient that gradually decreases as the number of iterations increases. represents the number of iterations, Indicates the maximum number of iterations.

[0077] In the spiral update phase of the optimization algorithm, a random number r in the interval [0,1] is introduced so that the update strategy can be selected in a conditional probability manner. If r is less than 0.5, that is Then update according to the first model:

[0078]

[0079] in, represents the next position of the individual after iteration, represents the optimal position of the whale in the current iteration, Represents a random number between [0, 1], Represents the position of another randomly selected whale; to execute the update strategy, it means that the algorithm tends to adopt a more exploratory update method at this time, allowing the algorithm to conduct a wide search.

[0080] On the contrary, if r is not less than 0.5, that is Then update according to the second model:

[0081]

[0082] in, represents the next position of the individual after iteration, represents the optimal position of the whale in the current iteration, Represents a coefficient that gradually decreases as the number of iterations increases. To execute the update strategy, it reflects that the algorithm tends to adopt a more developmental or refined update strategy to promote the algorithm's local refined search.

[0083] The parameter updating first model and the parameter updating second model are combined to obtain a parameter updating model.

[0084] In this embodiment, an innovative parameter update strategy is proposed, which combines the first parameter update model and the second parameter update model. That is, according to the current iteration state such as the quality of the solution, the convergence speed, etc. or the preset conditions, the algorithm dynamically selects to apply one of the update models for parameter adjustment. The advantages of the two update strategies can be fully utilized to obtain a more comprehensive and effective parameter update model. In turn, the overall performance and stability of the parameter update model are improved.

[0085] In an optional embodiment, the optimization process of the parameter update model is as follows, see Figure 2 .

[0086] based on Figure 2 It can be seen that the evolutionary algorithm is used for solving in the embodiment, where the initial population size is N, the maximum number of iterations is tmax, and a series of parameters such as a, vectors A and C and two update models are defined, namely the first parameter update model and the second parameter update model of this embodiment. In addition, random numbers p and r are introduced, as well as existing optimization algorithm models and improved models, to match appropriate optimization update models based on different parameter conditions.

[0087] Figure 2 In it, N represents the size of the initial population, tmax represents the maximum number of iterations, a, vectors A and C represent the parameters and vectors required to define or calculate the algorithm, and t represents the number of iterations.

[0088] For each individual (or solution) in the population, a random number is generated through a random number generator If the present invention is used to establish an update model, r is required, and a random number generator is used to make ,The use of random numbers in the embodiment contributes to the diversity and ,exploration of the algorithm, where parameters and vectors can be adjusted according to ,specific problems to optimize performance.

[0089] If p<0.5 and , then update the position according to the original formula, perform random search, and output the result when the number of iterations is less than the maximum number of iterations. ), press the improved update model, that is, use the optimized and improved update model to update the position.

[0090] If p 0.5, and r is less than 0.5, that is, p 0.5 and , then update the position of the parameters by updating the first modulus; otherwise, if p 0.5, r is not less than 0.5, that is, p 0.5 and , update the second model according to the parameters to update the position.

[0091] Furthermore, the method for establishing the parameter updating model in the present embodiment is only an optional condition of the present invention. In one or some other embodiments, the method for establishing the parameter updating model can be optimized according to the parameter optimization requirements and the model establishment situation, so that the framework of the parameter updating model is more flexible and adaptable, which helps to deal with complex optimization problems in different fields and improve the universality and application value of the algorithm.

[0092] S4. Obtain the optimal delay parameter through the parameter update model, and obtain the optimal delay prediction model based on the optimal delay parameter. The specific steps and related contents are as follows:

[0093] The parameter update model is used to optimize the model parameters of the delay prediction model and obtain the optimal delay parameters. The model parameters in the embodiment mainly include the number of hidden units, the maximum training cycle, and the initial learning rate. The specific implementation content is as follows:

[0094] In this embodiment, the model parameters of the delay prediction model are adjusted and optimized based on the parameter adjustment framework. The delay prediction model is a BP neural network model, and the embodiment mainly optimizes three core parameters in the BP neural network, including the number of hidden units, the maximum training cycle, and the initial learning rate.

[0095] Through the parameter update model established above, the influence of different parameters on model performance was analyzed, and the optimal configuration of each was determined. Subsequently, the geological multi-dimensional target data was input into the optimized BP model for training to ensure that the model can accurately learn from the data and generate reliable prediction results.

[0096] Furthermore, in this embodiment, the specific content of the model parameters is only an optional condition of the present invention. In one or some other embodiments, the specific content of the model parameters can be replaced according to the parameter optimization requirements and the prediction model structure, which can more accurately control the learning process and performance of the model, and help improve the prediction accuracy and robustness of the model.

[0097] S5. Based on the geological multi-dimensional target data, the optimal delay prediction model is trained and predicted to select the optimal blasting delay time between mine blasting holes. The implementation steps and related contents are as follows:

[0098] The optimal delay prediction model is trained and predicted based on geological multi-dimensional target data. The prediction evaluation indicators include mean absolute error, root mean square error, mean absolute percentage error and determination coefficient. The relevant contents are as follows:

[0099] In this embodiment, a multi-dimensional error evaluation system is also constructed to comprehensively analyze the performance of the prediction model. The system mainly includes the absolute error MAE, which is mainly used to measure the absolute average of the difference between the predicted value and the actual value; the root mean square error RMSE, which emphasizes the sensitivity to large errors and is used to evaluate the overall accuracy of the model; the average relative error MAPE, which intuitively displays the relative size of the prediction deviation in percentage form, is convenient for comparison across different dimensional data sets; in addition, the R² score is introduced as a key indicator to measure the goodness of fit between the model prediction value and the actual value. The closer its value is to 1, the better the model prediction performance.

[0100] The formula for absolute error MAE is as follows:

[0101]

[0102] in, represents the true value, Represents the predicted value.

[0103] The formula for root mean square error RMSE is as follows:

[0104]

[0105] in, represents the true value, represents the predicted value;

[0106] The formula for mean relative error MAPE is as follows:

[0107]

[0108] in, represents the true value, Represents the predicted value.

[0109] The formula is as follows:

[0110]

[0111] in, represents the true value, represents the predicted value, Represents the mean of the true values.

[0112] Furthermore, the error analysis method of the present invention is only an optional condition of the present invention. In one or some other embodiments, the error analysis method can be adjusted according to the optimization objectives and prediction results of the prediction model. Different prediction models and optimization objectives have different sensitivities and focuses on errors. Adjusting the error analysis method according to the specific situation can more accurately evaluate the performance of the model in different aspects, thereby more effectively guiding the optimization and improvement of the model.

[0113] Finally, the optimal blasting delay time between mine blasting holes is selected by combining geological multi-dimensional target data, the optimal delay prediction model and prediction evaluation indicators.

[0114] In the process of optimizing mine blasting operations, in order to solve the difficult problem of delay time selection between blasting holes, the present invention proposes a delay selection method between mine blasting holes. This method deeply integrates the detailed analysis of geological multi-dimensional target data, the construction of the optimal delay prediction model and a rigorous prediction evaluation index system, and determines the optimal blasting delay time through scientific calculation rather than relying on human subjective experience.

[0115] The present invention first converts complex text information into digital data that can be processed by the algorithm. The above process ensures the accuracy and processability of the information. Subsequently, the digital historical mine blasting data is used to deeply train the neural network model with algorithm-optimized hyperparameters. This neural network model, as a core prediction tool, has a strong learning ability and can capture the complex relationship between various influencing factors and delay time in blasting operations, and generate accurate regression prediction results based on this.

[0116] In the embodiment, in order to verify the accuracy and reliability of the prediction results, this paper introduces diversified error analysis indicators including absolute error MAE, root mean square error RMSE, average relative error MAPE and R² score. The above indicators can not only comprehensively evaluate the difference between the predicted value and the actual value, but also deeply analyze the goodness of fit and prediction accuracy of the model, providing a solid scientific basis for the final determination of the optimal blasting delay time.

[0117] Based on this, it not only effectively reduces the errors caused by subjective lack of experience or deviation, but also significantly reduces the vibration intensity during the blasting process, thereby protecting the safety of the surrounding environment and facilities. At the same time, more accurate blasting delay time selection also further improves the overall quality of blasting operations, optimizes blasting effects, and significantly improves the efficiency and safety of mining.

[0118] In summary, the present invention proposes an intelligent selection method for delay between mine blasting holes that integrates geological data analysis, intelligent prediction model construction and comprehensive error evaluation, which provides technical support for mine blasting operations and realizes the intelligent selection of the optimal blasting delay time between mine blasting holes.

[0119] See also Figure 3 In an optional embodiment, the present invention further provides a delay selection system between mine blasting holes, the delay selection system between mine blasting holes comprising a processor, an input device, an output device and a memory, the processor, the input device, the output device and the memory being interconnected, wherein the memory is used to store a computer program, the computer program comprising program instructions, the processor being configured to call the program instructions, and executing the specific steps of the delay selection method between mine blasting holes and related embodiments provided by the present invention. The delay selection system between mine blasting holes of the present invention has a complete structure and is objectively stable.

[0120] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A method for delay selection between mine blasting holes, characterized in that: The steps include: Acquiring geological multidimensional data, and processing the geological multidimensional data to obtain geological multidimensional target data; Establish and improve optimization algorithms and set optimization conditions; Establishing a parameter updating model based on the improved optimization algorithm and the optimization conditions; Obtaining an optimal delay parameter through the parameter updating model, and obtaining an optimal delay prediction model according to the optimal delay parameter; The optimal delay prediction model is trained and predicted based on the geological multi-dimensional target data to select the optimal blasting delay time between mine blasting holes; The establishment of the improved optimization algorithm comprises: Introducing a whale optimization algorithm, improving the whale optimization algorithm, and obtaining an improved optimization algorithm; The improved optimization algorithm satisfies the following relationship: ， in, represents the next position of the individual after iteration, represents the position of the individual in the current iteration, represents a constant, represents the position of another whale chosen at random; The setting optimization conditions include: A first optimization condition is set, and the first optimization condition satisfies ,in A random factor representing a conditional judgment; A second optimization condition is set, and the second optimization condition satisfies in A random factor representing a conditional judgment; The establishing of a parameter updating model based on the improved optimization algorithm and the optimization conditions comprises: Establishing parameters to update a first model based on the improved optimization algorithm and the first optimization condition; Establishing parameters based on the improved optimization algorithm and the second optimization condition to update the second model; Combining the first parameter updating model and the second parameter updating model to obtain a parameter updating model; The parameters update the first model to satisfy the following relationship: ， in, represents the next position of the individual after iteration, represents the optimal position of the whale in the current iteration, Represents a random number between [0, 1], represents the position of another whale chosen at random; The parameters update the second model to satisfy the following relationship: ， in, represents the next position of the individual after iteration, represents the optimal position of the whale in the current iteration, Represents a coefficient that gradually decreases as the number of iterations increases.

2. The method for selecting the delay between mine blasting holes according to claim 1, characterized in that: The processing of the geological multi-dimensional data to obtain geological multi-dimensional target data includes: Performing a first data processing on the geological multidimensional data and obtaining a first data processing result, wherein the first data processing includes an algorithm for eliminating outliers; The second data processing is performed on the first data processing result to obtain geological multi-dimensional target data, and the second data processing includes an algorithm for interpolating missing values.

3. The method for selecting the delay between mine blasting holes according to claim 2, characterized in that: The algorithm for removing outliers satisfies the following relationship: ， in, Indicates the normal value range of the data. represents the average value of any blasting characteristic parameter, represents standard deviation; The missing value interpolation algorithm satisfies the following relationship: ， in, Indicates the pending The The data of blasting parameters, Indicates The The data of blasting parameters, Indicates The The data of blasting parameters, Characterizes the blasting parameters.

4. The method for selecting the time delay between mine blasting holes according to claim 1, characterized in that: The obtaining of the optimal delay parameter by updating the model through the parameters includes: The parameter updating model is used to optimize the model parameters of the delay prediction model and obtain the optimal delay parameters, wherein the model parameters include the number of hidden units, the maximum training cycle, and the initial learning rate.

5. The method for selecting the delay between mine blasting holes according to claim 1, characterized in that: The training and prediction of the optimal delay prediction model based on the geological multi-dimensional target data to select the optimal blasting delay time between mine blasting holes includes: Based on the geological multi-dimensional target data, the optimal delay prediction model is trained and predicted, and the prediction evaluation indicators include mean absolute error, root mean square error, mean absolute percentage error and determination coefficient; The optimal blasting delay time between mine blasting holes is selected by combining the geological multi-dimensional target data, the optimal delay prediction model and the prediction evaluation index.

6. A time-delay selection system between mine blasting holes, characterized in that: The system includes a processor, an input device, an output device and a memory, wherein the processor, the input device, the output device and the memory are interconnected, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions to execute the delay selection method between mine blasting holes as described in any one of claims 1 to 5.

Images