Process parameter optimization method for electric pulse rock breaking

Abstract

The invention discloses a technological parameter optimization method for electric pulse rock breaking. The method comprises the steps that electric tree characteristic parameters under different technological parameters are obtained; establishing a process parameter prediction model by taking the process parameters as input and the electrical tree characteristic parameters as output; based on a process parameter prediction model, an improved NSGA-2 multi-objective optimization algorithm is utilized to optimize process parameters, and the process comprises the following steps: providing a selection, intersection and variation process fusing a tabu search thought, namely, in a filial generation population generation process, a process of generating a filial generation population is provided; a solution with better fitness than a parent and a suboptimal solution within the number limit of flouting criteria serve as offspring solutions and are added into a taboo table, repeated acquisition is avoided, the local search ability of the algorithm is enhanced, a crowding degree value calculation method is improved based on the super-lattice thought to complete offspring population selection and construction, and the global search ability of the algorithm is improved. According to the method, the technological parameters are optimized through the improved NSGA-2 multi-objective optimization algorithm, the optimal technological parameter combination can be obtained, and the rock breaking efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric pulse rock breaking, and in particular to a method for optimizing process parameters for electric pulse rock breaking. Background Art

[0002] Electric pulse rock breaking is influenced by multiple factors, including rock heterogeneity, discharge parameters, and electrode spacing. Selecting an appropriate combination of process parameters can improve the efficiency of electric pulse rock breaking by enhancing the electrical breakdown process. Existing research on the effects of electric pulse rock breaking on electrical tree branches typically focuses on qualitative analysis, but lacks a mathematical model to quantitatively examine the relationship between electric pulse rock breaking process parameters and these characteristics. This lacks guidance on how to select a combination of process parameters to improve rock breaking efficiency. Summary of the Invention

[0003] The purpose of the present invention is to provide a process parameter optimization method for electric pulse rock breaking, which optimizes the process parameters by using an improved NSGA-2 multi-objective optimization algorithm to obtain the optimal process parameter combination to improve the rock breaking efficiency.

[0004] The first aspect of the present invention provides a method for optimizing process parameters for electric pulse rock breaking, the method comprising: Obtaining characteristic parameters of electric tree branches under different process parameters, which are process parameters for electric pulse rock breaking, i.e., process parameters for generating electric pulses for rock breaking; A process parameter prediction model is established with process parameters as input and electrical tree characteristic parameters as output; Based on the process parameter prediction model, the improved NSGA-2 multi-objective optimization algorithm is used to optimize the process parameters, including: Initialize the population and taboo table, where each individual in the population corresponds to a set of process parameters; The process parameter prediction model is used to obtain the electrical tree characteristic parameters corresponding to each individual, and the population is quickly non-dominated sorted based on the electrical tree characteristic parameters to obtain the non-dominated level corresponding to each individual; Based on the electrical tree characteristic parameters corresponding to each individual, the crowding distance of each individual is determined as the neighborhood crowding value of the individual; and based on the individual density of each individual in the high-dimensional space of the electrical tree characteristic parameters, the global crowding value of each individual is determined; Based on the non-dominated rank, neighborhood crowding value and global crowding value corresponding to each individual, an elite preservation strategy is used to select individuals from the population to form the parent population; Individuals are selected from the parent population for crossover operation to generate offspring, and the offspring's fitness is determined based on the offspring's non-dominated rank, neighborhood crowding value, and global crowding value: when the offspring's fitness is higher than the average fitness of the parent, the offspring is retained; when the offspring's fitness is lower than the average fitness of the parent and the offspring's occurrence count does not exceed the taboo count, the offspring is retained and added to the taboo table; when the offspring's fitness is lower than the average fitness of the parent and the offspring's occurrence count exceeds the taboo count, the offspring's parent is treated as an offspring and retained; wherein, the offspring's occurrence count is determined based on the taboo table; Perform mutation operation on the retained offspring to obtain the offspring population; The parent population and the child population are merged into a new population. The elite preservation strategy is continued to be used to select individuals from the new population to form the parent population and perform crossover and mutation. This is repeated until the termination condition is met, and the optimal process parameter combination is finally obtained. The obtained optimal process parameter combination is used to generate rock-breaking electric pulses.

[0005] In the above solution, the process parameters include at least one of initial voltage, circuit inductance, circuit capacitance, electrode spacing and discharge times.

[0006] In the above solution, the characteristic parameters of the electrical tree branches include at least two of the maximum width of the electrical tree branches, the average depth of the electrical tree branches, and the maximum depth of the electrical tree branches.

[0007] In the above scheme, the characteristic parameters of electrical treeing under different process parameters are obtained based on orthogonal experimental design.

[0008] In the above solution, a process parameter prediction model is established based on Gaussian process regression, and each electrical tree characteristic parameter corresponds to a process parameter prediction model.

[0009] In the above scheme, the global crowding value of each individual is determined based on the individual density of each individual in the high-dimensional space of electrical tree characteristic parameters, including: Constructing a high-dimensional space of characteristic parameters of electrical tree branches, where the dimension of the space is the number of characteristic parameters of electrical tree branches; Divide the space into multiple hypercube grids; Determine the hypercube grid where each individual is located and its adjacent hypercube grids; The total number of individuals in the hypercube grid where the individual is located and its adjacent hypercube grids is counted as the global crowding value of the individual.

[0010] In the above scheme, based on the non-dominated rank, neighborhood crowding value, and global crowding value corresponding to each individual, an elite preservation strategy is used to select individuals from the population to form the parent population, including: First, all individuals corresponding to the entire non-dominated level are placed into the parent population in descending order of non-dominated level, until all individuals corresponding to a certain non-dominated level cannot be placed into the parent population; For this non-dominated level, the total crowding value is calculated based on the individual neighborhood crowding value and the global crowding value; wherein, the larger the neighborhood crowding value, the larger the total crowding value, and the smaller the global crowding value, the larger the total crowding value; Then, in descending order of total crowding value, the individuals corresponding to the non-dominated level are placed into the parent population in sequence until the number of parent population is sufficient.

[0011] In the above scheme, the offspring's fitness is determined based on its non-dominated level, neighborhood crowding value, and global crowding value, including: Place the offspring generated by the crossover operation into the parent population and determine the offspring's non-dominated rank, neighborhood crowding value, and global crowding value; The fitness is determined based on the weighted summation method; the lower the non-dominated level, the greater the fitness; the larger the neighborhood crowding value, the greater the individual fitness; the smaller the global crowding value, the greater the individual fitness; and the weight of the non-dominated level is greater than the weights of the neighborhood crowding value and the global crowding value.

[0012] According to a second aspect of the present invention, an electronic device is provided, comprising: a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the programs or instructions are executed by the processor, the steps of the process parameter optimization method for electric pulse rock breaking described in any one of the first aspects are implemented.

[0013] According to a third aspect of the present invention, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the process parameter optimization method for electric pulse rock breaking described in any one of the first aspects are implemented.

[0014] In general, the above technical solutions conceived by the present invention can achieve the following beneficial effects compared with the prior art: The present invention first establishes a process parameter prediction model based on Gaussian process regression. This process parameter prediction model can accurately predict the corresponding electrical tree characteristic parameters based on the process parameters. Then, based on this process parameter prediction model, the process parameters are optimized using an improved NSGA-2 multi-objective optimization algorithm to obtain the optimal process parameter combination. The improved NSGA-2 multi-objective optimization algorithm proposes a selection, crossover, and mutation process that integrates the concept of taboo search. Specifically, when generating the offspring population, solutions with better fitness than the parent and suboptimal solutions within the limit of the number of contempt criteria are selected as offspring solutions and added to the taboo table to avoid repeated acquisition and enhance the algorithm's local search capability. Furthermore, based on the super-grid concept, an improved method for calculating the congestion value is used to complete the offspring population selection and construction, improving the algorithm's global search capability.

[0015] The present invention optimizes process parameters by using an improved NSGA-2 multi-objective optimization algorithm, thereby obtaining an optimal combination of process parameters and improving rock breaking efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Schematic diagram of the original NSGA-2 algorithm flow; Figure 2 A schematic diagram of a flow chart of an improved NSGA-2 multi-objective optimization algorithm provided in an embodiment of the present application; FIG3 is a diagram of hyperparameter optimization of a process parameter prediction model provided in an embodiment of the present application; FIG3(a) is a diagram of the maximum width hyperparameter optimization, FIG3(b) is a diagram of the average depth hyperparameter optimization, and FIG3(c) is a diagram of the maximum depth hyperparameter optimization; FIG4 is a fitting test diagram of a process parameter prediction model prediction effect provided in an embodiment of the present application; FIG4(a) is a maximum width fitting test diagram, FIG4(b) is an average depth fitting test diagram, and FIG4(c) is a maximum depth fitting test diagram; Figure 5 A Pareto front diagram for multi-objective optimization of electrical tree characteristic parameters provided in an embodiment of the present application; FIG6 is a three-dimensional cloud diagram showing the crushing effect before and after process parameter optimization according to an embodiment of the present application; FIG6(a) is a three-dimensional cloud diagram of the crushed rock sample before optimization, and FIG6(b) is a three-dimensional cloud diagram of the crushed rock sample after optimization; Figure 7 A schematic diagram of the hardware structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Based on the embodiments provided in this application, all other embodiments obtained by those of ordinary skill in the art without making creative work are within the scope of protection of the present invention.

[0018] Obviously, the drawings described below are merely examples or embodiments of the present application. Those skilled in the art can, without inventive effort, apply the present application to other similar scenarios based on these drawings. Furthermore, it is also understood that, although the effort involved in such a development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, changes in design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as an insufficiency of the content disclosed in this application.

[0019] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it refer to independent or alternative embodiments that are mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments unless there is a conflict.

[0020] Unless otherwise defined, technical or scientific terms used herein shall have the ordinary meaning as understood by persons of ordinary skill in the art to which this application belongs. The terms "a," "an," "an," "the," and similar expressions used herein do not denote quantitative limitations and may refer to either the singular or the plural. The terms "comprise," "include," "have," and any variations thereof, used herein, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or modules (units) is not limited to the listed steps or units but may also include steps or units not listed, or may include other steps or units inherent to the process, method, product, or apparatus. The terms "connected," "connected," "coupled," and similar expressions used herein are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. As used herein, "plurality" means two or more. "And / or" describes an association between associated objects, indicating that three possible relationships exist. For example, "A and / or B" may mean: A exists alone; A and B exist simultaneously; or B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific order for the objects.

[0021] This application provides a process parameter prediction model and process parameter optimization method for a joint simulation platform for electric pulse rock breaking. A process parameter prediction model is established, combining predictions of the electrical breakdown process with predictions of the discharge rock breaking process. Based on this model, a multi-objective optimization algorithm is proposed to optimize the process parameter combination to enhance rock breaking effectiveness. First, an orthogonal experiment is used to obtain a process parameter optimization dataset for five process parameters that influence the generation of breakdown electrical trees: initial voltage, circuit inductance, circuit capacitance, electrode spacing, and number of discharges. A significance analysis is then performed to determine the specific effects of each process parameter on the characteristic parameters of the breakdown electrical trees. Subsequently, a process parameter prediction model is established based on Gaussian process regression, achieving the best performance across all evaluation metrics compared to fitting models such as BP neural networks and support vector machines. Furthermore, to address the performance degradation of the traditional NSGA-2 multi-objective optimization algorithm for three-dimensional objective optimization, an improved NSGA-2 multi-objective optimization algorithm is proposed, combining the tabu search principle with an improved congestion value algorithm. This algorithm demonstrates superior qualitative and quantitative performance on three-dimensional objective test functions compared to the original algorithm. Finally, the improved NSGA-2 algorithm was used to optimize the process parameters, and the effectiveness of the optimization algorithm in enhancing the rock breaking effect was verified through experiments.

[0022] This application proposes a process parameter prediction model based on Gaussian process regression and a multi-objective optimization algorithm based on an improved NSGA-2, and performs multi-objective optimization of process parameters. This application first obtains a data set through orthogonal experiments and verifies the correlation between the data set and the electrical breakdown phenomenon through significance analysis, providing a data source for subsequent modeling and optimization.

[0023] First, the initial voltage, circuit inductance, circuit capacitance, electrode spacing, and number of discharges are taken as process parameters to be optimized. In order to obtain the characteristic data of the electrical dendrites under different process parameters, representative experimental groups are selected from all experimental groups based on orthogonality to conduct experiments. This can achieve results equivalent to those of a comprehensive experiment with a small number of orthogonal experiments. The independent variables in the simulation data set obtained through the orthogonal experiment are the process parameters of the discharge process, and the dependent variables are the characteristic parameters of the electrical dendrites. The existing research on electric pulse rock breaking prediction models lacks the demonstration that process parameters are the main factors affecting the breakdown of electrical dendrites. At the same time, there is a certain degree of randomness in the process of electrical dendrite growth, which also interferes with the analysis of the influence of process parameters on the characteristics of electrical dendrites. Therefore, before establishing the process parameter prediction model and multi-objective optimization, a data set significance analysis is performed to prove that process parameters have a significant impact on the breakdown of electrical dendrites.

[0024] Based on the concept of orthogonal experimentation, this application designed 81 sets of simulation experiments to obtain data sets corresponding to process parameters such as initial voltage, circuit inductance, circuit capacitance, electrode spacing, and number of discharges, and the maximum width, average depth, and maximum depth of electrical dendrites, as shown in Tables 1 and 2. By analyzing the data distribution characteristics, the correlation between process parameters and rock breaking performance was determined, as shown in Tables 3, 4, and 5.

[0025] Table 1 Independent variable factor level table

[0026] Table 2 Average and standard deviation of electrical tree characteristic parameters

[0027] Table 3. Significance analysis of the maximum width of electrical trees

[0028] Table 4. Significance analysis of average depth of electrical treeing

[0029] Table 5. Analysis of the significance of the maximum depth of electrical treeing

[0030] During the modeling process, a process parameter prediction model was established based on Gaussian process regression. Comparisons with other fitting models confirmed that the Gaussian process regression model had better fitting accuracy. 81 sets of orthogonal experimental data were used as the training data for the Gaussian process regression model. Model training and optimization were performed based on a regression learner, with training performed within a selected range for each hyperparameter. Training time was not restricted. To avoid excessive training time, the number of iterations was set to 30, taking into account both the optimal selection of hyperparameters and the training duration. Figure 3 shows the hyperparameter optimization diagram for the process parameter prediction model. Figure 3(a) shows the maximum width hyperparameter optimization diagram, Figure 3(b) shows the average depth hyperparameter optimization diagram, and Figure 3(c) shows the maximum depth hyperparameter optimization diagram.

[0031] After determining the hyperparameters, a process parameter prediction model based on Gaussian process regression was established to verify the effectiveness of hyperparameter optimization. Three independent Gaussian process regression models were developed for the three characteristic parameters of electrical dendrites (maximum width, average depth, and maximum depth). Each model used the same input variables—five process parameters (initial voltage, circuit inductance, circuit capacitance, electrode spacing, and number of discharges)—and the output variables each corresponded to one characteristic parameter of the electrical dendrites.

[0032] The predicted response-true response distribution diagram and model fitting analysis diagram are established for the three sets of output characteristic parameters to verify the effect of the established process parameter prediction model. In order to better visualize the prediction effect of the optimization model and intuitively understand the model fitting effect, the characteristic parameters of the predicted breakdown electric tree are used as three prediction targets to draw the prediction effect fitting diagram of the process parameter prediction model. The true value of each target characteristic parameter is used as the horizontal coordinate, and the predicted value of the process parameter prediction model is used as the vertical coordinate, and the range and proportion of the two coordinate axes are equal, that is, the distribution is The points on this diagonal line are the reference points for perfect predictions without errors. Figure 4 shows a fit test diagram for the prediction effect of a process parameter prediction model provided in an embodiment of this application; Figure 4(a) shows the maximum width fit test diagram, Figure 4(b) shows the average depth fit test diagram, and Figure 4(c) shows the maximum depth fit test diagram. The fit of the prediction model to each target characteristic parameter is analyzed by observing the distribution of the predicted response values ​​of the optimization model along the diagonal line.

[0033] While training the Gaussian process regression model, considering that there is no ready-made qualitative conclusion on the effect of process parameter combination on electrical tree branches, in addition to the multiple linear regression used in significance analysis, considering the possible linear and nonlinear situations between data sets, decision tree, SVM and BP neural network are added as training models to compare the training effects. 2 The test is used as an evaluation indicator to compare the quality of the training results. The training results are shown in Tables 6, 7 and 8.

[0034] Table 6 Evaluation table of the prediction effect of the maximum width of electrical trees under different models

[0035] Table 7 Evaluation of the prediction effect of average depth of electrical tree under different models

[0036] Table 8 Evaluation of the prediction effect of the maximum depth of electrical trees under different models

[0037] From the evaluation indicators of different model training effects, it can be seen that for the three training targets of maximum width, average depth and maximum depth of electrical tree branches, the Gaussian process regression has the best indicators compared with other models and is suitable for the needs of training models. At the same time, due to the nonlinearity and small sample characteristics of the data, the R 2 The fitting effect is worse than that of Gaussian process regression, and BP neural network is limited by the small number of samples and has the worst fitting ability for data sets. Therefore, this application uses the Gaussian process regression model.

[0038] In optimizing the parameters of the electric pulse rock breaking process, the maximum width, average depth, and maximum depth of the electric tree branches are all key positive indicators of the electric pulse rock breaking effect. The optimization goal is to maximize all three simultaneously through process parameter adjustments. Due to the synergies and conflicts between these objectives, it is necessary to improve the NSGA-2 multi-objective optimization algorithm to generate a Pareto optimal solution set. A balanced parameter combination is selected based on specific project requirements to ultimately achieve a comprehensive improvement in rock breaking efficiency.

[0039] Figure 1 This is the original NSGA-2 algorithm flow chart, such as Figure 1 As shown in the figure, the basic idea is: first, randomly generate an initial population, each individual in the population is a solution, then perform non-dominated sorting on the initial population, and obtain the next generation population through genetic operations. Introduce the elite strategy, merge the initial population and the next generation population and perform fast non-dominated sorting, then assign crowding values ​​to individuals in the same non-dominated layer, and select the best solution based on the dominance relationship and crowding value between individuals in the population. Individuals form a new initial population, and then the next generation population is obtained through genetic operations. This step is repeated until the termination condition is met.

[0040] In order to improve the local search capability of the algorithm and improve the optimization capability of the algorithm under three-dimensional objective functions, this application proposes an improved NSGA-2 algorithm, which is improved mainly in the following two aspects: (1) A selection, crossover, and mutation process that integrates the idea of ​​taboo search is proposed. In the process of generating offspring populations, solutions with better fitness than the parent and suboptimal solutions within the limit of the number of times the criterion is despised are taken as offspring solutions and added to the taboo table to avoid repeated acquisition, thereby enhancing the local search capability of the algorithm.

[0041] (2) Based on the super-lattice idea, the cyclic crowding sorting strategy is used to improve the crowding value calculation method to complete the offspring population selection and construction, thereby improving the global search capability of the algorithm.

[0042] The specific operations are as follows: (1) Integrating the selection, crossover, and mutation of taboo search ideas This application introduces taboo search to increase the local search capability of the algorithm, while retaining the traditional NSGA-2 process of generating offspring based on selection, crossover, and mutation. When generating offspring, if the offspring's fitness is worse than the parent or the fitness is not as good as the parent but does not exceed the number of times allowed by the description criteria, the offspring will be retained, otherwise the parent will be used as the offspring solution. This strategy can enhance the local development capability of individuals in the parent generation, while retaining the parent's advantage information and taking into account the diversity of the population. The specific strategy is as follows: Step 1: After obtaining the parent population using the elite preservation strategy, use the selection operator to select individuals as parents for the crossover process. Since the tabu search has high requirements for the initial solution, the probability of selecting higher-level individuals will be increased in the early selection process.

[0043] Step 2: Determine the contempt criterion and set an appropriate number of taboos. When the fitness of the solution obtained is higher than the average fitness of the parent generation, regardless of whether the solution exists in the taboo table, the solution is selected as a child. When the fitness of the solution obtained is lower than the average fitness of the parent generation but the current number of solutions does not exceed the taboo number, the solution is selected as a child solution and added to the taboo table. If the fitness of the solution obtained is lower than the average fitness of the parent generation and exceeds the taboo number, the parent solution is directly used as the child solution obtained this time. This prevents the duplication of high-quality solutions while ensuring the quality of the overall child solution set as much as possible.

[0044] Step 3: Perform mutation operation on the offspring solution set obtained in the previous step.

[0045] Step 4: Calculate the number of offspring populations according to the set requirements. When the number of offspring solutions meets the requirements, stop the selection, crossover, and mutation cycle to obtain offspring.

[0046] (2) Improvement of the calculation method of congestion value To address the problem of traditional NSGA-2 algorithms being less effective for three-dimensional optimization than for two-dimensional optimization, this application, based on the concept of hypergrids, divides the space into multiple hypercube grids to describe the density of target individuals to be optimized in three-dimensional space. Since the hypergrid concept can only describe the density of individuals within a space and cannot measure the distribution of individuals around them, an improved method for calculating the crowding value is used to select the offspring population. The basic strategy is as follows: Step 1: Divide the supergrid, standardize and round the objective function of each individual in each layer of the population, divide the objective function into intervals, and save the supergrid where the individual is located.

[0047] Step 2: Calculate the crowding distance of each individual as the individual neighborhood crowding value.

[0048] Step 3: Calculate the number of solutions in the individual and its adjacent superlattices as the individual's global crowding value.

[0049] Step 4: Perform fast non-dominated sorting. In addition to sorting by non-dominated rank in the original NSGA-2 algorithm, when the non-dominated ranks are equal, if the individuals The neighborhood crowding value is greater than that of individual The corresponding value and its global crowding value is less than the individual When the corresponding value is Better than individual .

[0050] The improved NSGA-2 algorithm process is as follows Figure 2 As shown. Therefore, the five process parameters in the orthogonal experimental data set, such as initial voltage, circuit inductance, circuit capacitance, electrode spacing, and discharge times, are used as the five decision variables of the input, and the maximum width, average depth, and maximum depth of the electric tree are used as the target features for multi-objective optimization. The model to be optimized is the Gaussian process regression model, and the optimization algorithm is the improved NSGA-2 algorithm. Figure 5 As shown in the figure, based on the improved algorithm, the process parameters were optimized with multiple objectives, and the Pareto front with the maximum width, average depth and maximum depth of the electric tree branches as the optimization objectives was generated, and the recommended process parameter combination for electric pulse rock breaking was obtained.

[0051] The actual discharge rock breaking experiment verified that the optimized process parameter combination enhanced the rock breaking effect, as shown in Figure 6. The comparison of the broken rock samples before and after optimization is shown in Figure 6 (a) and Figure 6 (b), respectively.

[0052] In summary, this application proposes a process parameter prediction model based on Gaussian process regression and a multi-objective optimization algorithm based on improved NSGA-2, and performs multi-objective optimization of process parameters.

[0053] Specifically, the process of optimizing process parameters using the improved NSGA-2 multi-objective optimization algorithm is as follows: (1) Initialize the population and taboo table. Each individual in the population corresponds to a set of process parameters.

[0054] (2) The process parameter prediction model is used to obtain the electrical tree characteristic parameters corresponding to each individual, and the population is quickly non-dominated sorted based on the electrical tree characteristic parameters to obtain the non-dominated level corresponding to each individual. The lower the non-dominated level (Pareto level), the better the individual i The better the corresponding electric tree characteristic parameters are, the better the corresponding rock breaking effect will be.

[0055] (3) Determine the crowding distance of each individual based on the electrical tree characteristic parameters corresponding to each individual, as the neighborhood crowding value of the individual. i Individuals, their neighborhood crowdedness It can be expressed as:

[0056] in, To optimize the target number, this embodiment sets it to 3. For the i The individual's m The denominator is used for normalization to avoid the influence of the target dimension. Usually, two adjacent individuals (such as the first and ), forming a local sliding window. In the three-dimensional objective function (high-voltage electric pulse rock breaking), is the maximum width of the electrical tree, is the average depth of electrical tree, is the maximum depth of electrical tree branches. and For the m The maximum and minimum target values. The larger it is, the sparser the neighborhood solutions are and the better the diversity is.

[0057] At the same time, based on the individual density of each individual in the high-dimensional space of electrical tree characteristic parameters, the global crowding value of each individual is determined as follows: Constructing a high-dimensional space of characteristic parameters of electrical tree branches, where the dimension of the space is the number of characteristic parameters of electrical tree branches; Divide the space into multiple hypercube grids; Determine the hypercube grid where each individual is located and its adjacent hypercube grids; The total number of individuals in the hypercube grid where the individual is located and its adjacent hypercube grids is counted as the global crowding value of the individual.

[0058] In this embodiment, for individual The global congestion of is defined as the sum of the number of solutions in its superlattice and its adjacent superlattices, and the formula is as follows:

[0059] in, For individuals The hypercube grid where is a set of adjacent superlattices, such as individuals The 6 super grids in front, behind, left, right, top, and bottom of the hypercube grid or the 28 super grids adjacent to it, It is a statistic of the number of individuals in the super grid.

[0060] (4) Based on the non-dominated rank, neighborhood crowding value, and global crowding value corresponding to each individual, an elite preservation strategy is used to select individuals from the population to form the parent population, as follows: First, all individuals corresponding to the entire non-dominated level are placed into the parent population in descending order of non-dominated level, until all individuals corresponding to a certain non-dominated level cannot be placed into the parent population; For this non-dominant level, if the individual The neighborhood crowding value is greater than that of individual The corresponding value and its global crowding value is less than the individual When the corresponding value is Better than individual In order to facilitate the selection of individuals to form the parent population, the total crowding value is calculated based on the individual's neighborhood crowding value and the global crowding value; the larger the neighborhood crowding value, the larger the total crowding value, and the smaller the global crowding value, the larger the total crowding value; Then, in descending order of total crowding value, the individuals corresponding to the non-dominated level are placed into the parent population in sequence until the number of parent population is sufficient.

[0061] (5) Select individuals from the parent population to perform crossover operations to generate offspring, and determine the offspring's fitness based on the offspring's non-dominated level, neighborhood crowding value, and global crowding value: when the offspring's fitness is higher than the average fitness of the parent, the offspring is retained; when the offspring's fitness is lower than the average fitness of the parent and the offspring's occurrence times do not exceed the taboo times, the offspring is retained and added to the taboo table; when the offspring's fitness is lower than the average fitness of the parent and the offspring's occurrence times exceed the taboo times, the offspring's parent is treated as an offspring and retained; wherein, the offspring's occurrence times are determined based on the taboo table.

[0062] Among them, the calculation of individual i The fitness is as follows: Place the offspring generated by the crossover operation into the parent population and determine the offspring's non-dominated rank, neighborhood crowding value, and global crowding value; The fitness is determined based on the weighted summation method; the lower the non-dominated level, the greater the fitness; the larger the neighborhood crowding value, the greater the individual fitness; the smaller the global crowding value, the greater the individual fitness; and the weight of the non-dominated level is greater than the weights of the neighborhood crowding value and the global crowding value.

[0063] Specifically, the formula is as follows:

[0064] Where, is the weight coefficient, satisfying ,generally Dominant (in this embodiment ), ensuring that non-dominated ranks are prioritized to balance local and global diversity.

[0065] (6) Perform mutation operations on the retained offspring to obtain the offspring population.

[0066] (7) The parent population and the child population are merged to form a new population. The elite preservation strategy is continued to be used to select individuals from the new population to form the parent population and perform crossover and mutation. This is repeated until the termination condition is met, and the optimal process parameter combination is finally obtained.

[0067] In addition, combined Figure 2 The process parameter optimization method for electric pulse rock breaking described in the embodiments of the present application can be implemented by a computer device. Figure 7 Schematic diagram of the hardware structure of the computer device of the embodiment of the present application. Figure 7 As shown, the device may include a processor 201 and a memory 202 storing computer program instructions.

[0068] Specifically, the processor 201 may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0069] Memory 202 may include a large-capacity memory for data or instructions. By way of example, and not limitation, memory 202 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 202 may include removable or non-removable (or fixed) media. Where appropriate, memory 202 may be internal or external to the data processing device. In certain embodiments, memory 202 is non-volatile memory. In certain embodiments, memory 202 includes read-only memory (ROM) and random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM) or a flash memory (FLASH), or a combination of two or more of these. Under appropriate circumstances, the RAM can be a static random access memory (SRAM) or a dynamic random access memory (DRAM), where the DRAM can be a fast page mode dynamic random access memory (FPMDRAM), an extended data out dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0070] The memory 202 may be used to store or cache various data files that need to be processed and / or used for communication, as well as possible computer program instructions executed by the processor 201 .

[0071] The processor 201 reads and executes computer program instructions stored in the memory 202 to implement any one of the process parameter optimization methods for electric pulse rock breaking in the above embodiments.

[0072] In some embodiments, the computer device may further include a communication interface 203 and a bus 200. Figure 7 As shown, the processor 201 , the memory 202 , and the communication interface 203 are connected via a bus 200 and communicate with each other.

[0073] The communication interface 203 is used to implement communication between the various modules, devices, units, and / or devices in the embodiments of the present application. The communication interface 203 can also implement data communication with other components such as: external devices, image / data acquisition equipment, databases, external storage, and image / data processing workstations.

[0074] Bus 200 includes hardware, software, or both, and couples components of a computer device to each other. Bus 200 includes, but is not limited to, at least one of the following: a data bus, an address bus, a control bus, an expansion bus, and a local bus. By way of example, and not limitation, bus 200 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Bus 200 may include one or more buses, where appropriate. Although embodiments herein describe and illustrate a particular bus, this application contemplates any suitable bus or interconnect.

[0075] The computer device can execute the process parameter optimization method for electric pulse rock breaking in the embodiment of the present application, thereby realizing the combination of Figure 2 A process parameter optimization method for electric pulse rock breaking is described.

[0076] In addition, in conjunction with the process parameter optimization method for electric pulse rock breaking in the above-mentioned embodiments, embodiments of the present application may provide a computer-readable storage medium for implementation. The computer-readable storage medium stores computer program instructions; when executed by a processor, the computer program instructions implement any of the process parameter optimization methods for electric pulse rock breaking in the above-mentioned embodiments.

[0077] It should be noted that the various technical features of the above-described embodiments can be combined in any manner. To simplify the description, not all possible combinations of the various technical features in the above-described embodiments are described. However, as long as there are no contradictions in the combination of these technical features, they should be considered to be within the scope of this specification. In addition, according to the needs of implementation, the various steps / components described in this application can be split into more steps / components, and two or more steps / components or partial operations of steps / components can be combined into new steps / components to achieve the purpose of the present invention.

[0078] Those skilled in the art will readily understand that the above-described embodiments merely represent several implementation methods of the present application, and their descriptions are relatively specific and detailed, but should not be construed as limiting the scope of the invention. It should be noted that a person of ordinary skill in the art may make several variations and improvements without departing from the concept of the present application, and these variations and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be based on the appended claims.

Claims

1. A method for optimizing process parameters for electric pulse rock breaking, characterized in that: The method includes: Obtain characteristic parameters of electrical tree branches under different process parameters; A process parameter prediction model is established with process parameters as input and electrical tree characteristic parameters as output; Based on the process parameter prediction model, the improved NSGA-2 multi-objective optimization algorithm is used to optimize the process parameters, including: Initialize the population and taboo table, where each individual in the population corresponds to a set of process parameters; The process parameter prediction model is used to obtain the electrical tree characteristic parameters corresponding to each individual, and the population is quickly non-dominated sorted based on the electrical tree characteristic parameters to obtain the non-dominated level corresponding to each individual; Based on the electrical tree characteristic parameters corresponding to each individual, the crowding distance of each individual is determined as the neighborhood crowding value of the individual; and based on the individual density of each individual in the high-dimensional space of the electrical tree characteristic parameters, the global crowding value of each individual is determined; Based on the non-dominated rank, neighborhood crowding value and global crowding value corresponding to each individual, an elite preservation strategy is used to select individuals from the population to form the parent population; Individuals are selected from the parent population for crossover operation to generate offspring, and the offspring's fitness is determined based on the offspring's non-dominated rank, neighborhood crowding value, and global crowding value: when the offspring's fitness is higher than the average fitness of the parent, the offspring is retained; when the offspring's fitness is lower than the average fitness of the parent and the offspring's occurrence count does not exceed the taboo count, the offspring is retained and added to the taboo table; when the offspring's fitness is lower than the average fitness of the parent and the offspring's occurrence count exceeds the taboo count, the offspring's parent is treated as an offspring and retained; wherein, the offspring's occurrence count is determined based on the taboo table; Perform mutation operation on the retained offspring to obtain the offspring population; The parent population and the child population are merged into a new population. The elite preservation strategy is continued to be used to select individuals from the new population to form the parent population and perform crossover and mutation. This is repeated until the termination condition is met, and the optimal process parameter combination is finally obtained. The obtained optimal process parameter combination is used to generate rock-breaking electric pulses.

2. The process parameter optimization method for electric pulse rock breaking according to claim 1 is characterized in that: The process parameters include at least one of an initial voltage, an electric loop inductance, an electric loop capacitance, an electrode spacing, and a discharge number.

3. The process parameter optimization method for electric pulse rock breaking according to claim 1 is characterized in that: The characteristic parameters of the electrical tree include at least two of the maximum width of the electrical tree, the average depth of the electrical tree, and the maximum depth of the electrical tree.

4. The process parameter optimization method for electric pulse rock breaking according to claim 1, characterized in that: The characteristic parameters of electrical treeing under different process parameters were obtained based on orthogonal experimental design.

5. The process parameter optimization method for electric pulse rock breaking according to claim 1, characterized in that: A process parameter prediction model is established based on Gaussian process regression, and each electrical tree characteristic parameter corresponds to a process parameter prediction model.

6. The process parameter optimization method for electric pulse rock breaking according to claim 1, characterized in that: Based on the individual density of each individual in the high-dimensional space of electrical tree characteristic parameters, the global crowding value of each individual is determined, including: Constructing a high-dimensional space of characteristic parameters of electrical tree branches, where the dimension of the space is the number of characteristic parameters of electrical tree branches; Divide the space into multiple hypercube grids; Determine the hypercube grid where each individual is located and its adjacent hypercube grids; The total number of individuals in the hypercube grid where the individual is located and its adjacent hypercube grids is counted as the global crowding value of the individual.

7. The process parameter optimization method for electric pulse rock breaking according to claim 1, characterized in that: Based on the non-dominated rank, neighborhood crowding value, and global crowding value corresponding to each individual, an elite preservation strategy is used to select individuals from the population to form the parent population, including: First, all individuals corresponding to the entire non-dominated level are placed into the parent population in descending order of non-dominated level, until all individuals corresponding to a certain non-dominated level cannot be placed into the parent population; For this non-dominated level, the total crowding value is calculated based on the individual neighborhood crowding value and the global crowding value; wherein, the larger the neighborhood crowding value, the larger the total crowding value, and the smaller the global crowding value, the larger the total crowding value; Then, in descending order of total crowding value, the individuals corresponding to the non-dominated level are placed into the parent population in sequence until the number of parent population is sufficient.

8. The process parameter optimization method for electric pulse rock breaking according to claim 1, characterized in that: The fitness of the offspring is determined based on its non-dominated level, neighborhood crowding value, and global crowding value, including: Place the offspring generated by the crossover operation into the parent population and determine the offspring's non-dominated rank, neighborhood crowding value, and global crowding value; The fitness is determined based on the weighted summation method; the lower the non-dominated level, the greater the fitness; the larger the neighborhood crowding value, the greater the individual fitness; the smaller the global crowding value, the greater the individual fitness; and the weight of the non-dominated level is greater than the weights of the neighborhood crowding value and the global crowding value.

9. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the programs or instructions are executed by the processor, the steps of the process parameter optimization method for electric pulse rock breaking described in any one of claims 1 to 8 are implemented.

10. A readable storage medium, characterized in that: Programs or instructions are stored thereon, and when the programs or instructions are executed by the processor, the steps of the process parameter optimization method for electric pulse rock breaking described in any one of claims 1 to 8 are implemented.

Images