Sample optimization method and device for micro-seismic data set and electronic equipment
By performing sample optimization on the microseismic dataset, generating a new target sample and balancing the dataset, the problem of fewer impact ground pressure samples in the microseismic dataset is solved, and the accuracy and efficiency of impact ground pressure type prediction is improved.
Patent Information
- Application Number
- CN202411890595.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-20
- Publication Date
- 2025-04-22
AI Technical Summary
In the intelligent prediction of impact ground pressure type, the samples of impact ground pressure and large energy events in micro-seismic data account for only a small part, resulting in extremely poor prediction probability during model training, and it is impossible to accurately predict impact ground pressure.
A sample optimization method for microseismic data set is proposed. By determining the reference sample characteristics of the mining area, monitoring microseismic events, dividing minority and majority-class samples, and by generating new target samples, balancing the data set, improving prediction accuracy.
By optimizing the microseismic dataset, the model can learn the characteristics of each category more accurately, improving the accuracy and efficiency of impact pressure type prediction.
Smart Images
Figure CN119862417A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of rock burst prediction and prevention, and in particular to a sample optimization method, device and electronic equipment for a microseismic data set. Background Art
[0002] Intelligent prediction of rock burst types is a necessary means to improve the efficiency and accuracy of rock burst type prediction. It requires monitoring data of real rock burst and high-energy events on site as samples for model training and prediction.
[0003] However, when conducting intelligent prediction of rock burst types, the samples of rock burst and high-energy events only account for a small part of the microseismic dataset used. Directly using the original microseismic dataset to train the model will result in an extremely poor prediction probability of the model for rock burst events, making it impossible to accurately predict rock burst. Summary of the invention
[0004] The purpose of this application is to solve one of the technical problems in the related art at least to some extent.
[0005] To this end, a first objective of the present application is to propose a sample optimization method for a microseismic dataset, so as to optimize an unbalanced microseismic dataset into a balanced microseismic dataset.
[0006] The second objective of the present application is to provide a sample optimization device for microseismic data sets.
[0007] The third objective of the present application is to provide an electronic device.
[0008] A fourth objective of the present application is to provide a computer-readable storage medium.
[0009] To achieve the above-mentioned purpose, the first aspect of the present application proposes a sample optimization method for a microseismic data set, including: determining reference sample features of microseismic data based on reference mine information and reference microseismic events in a mining area; monitoring microseismic events in the mining area to determine a microseismic data set, and determining sample features corresponding to the microseismic data set based on the mine information of the mining area, the microseismic data set and the reference features, wherein the sample features include at least the number of samples and the strong spatiotemporal features of the samples; determining minority class samples and majority class samples in the microseismic data set based on the number of samples and a set number threshold; determining the total number of samples to be synthesized, and determining the proportion of majority class samples in the neighboring samples based on the neighboring samples of the minority class samples; standardizing the proportion of majority class samples to obtain a standardized proportion of majority class samples; determining the number of target new samples to be generated based on the total number of samples and the standardized proportion of majority class samples, and generating target new samples based on the minority class samples and the target number of new samples to optimize the microseismic data set.
[0010] To achieve the above-mentioned purpose, the second embodiment of the present application proposes a sample optimization device for a microseismic data set, including: a first determination module, used to determine the reference sample characteristics of the microseismic data based on the reference mine information and reference microseismic events of the mining area; a second determination module, used to monitor the microseismic events in the mining area, determine the microseismic data set, and determine the sample characteristics corresponding to the microseismic data set based on the mine information of the mining area, the microseismic data set and the reference characteristics, wherein the sample characteristics at least include the number of samples and the strong spatiotemporal characteristics of the samples; a third determination module, used to determine the sample characteristics of the microseismic data set based on the number of samples and a set number threshold value, determine the minority class samples and the majority class samples in the microseismic data set; a fourth determination module, used to determine the total number of samples to be synthesized, and based on the neighboring samples of the minority class samples, determine the proportion of the majority class samples in the neighboring samples; a processing module, used to standardize the proportion of the majority class samples to obtain the standardized proportion of the majority class samples; a generation module, used to determine the target number of new samples to be generated according to the total number of samples and the standardized proportion of the majority class samples, and generate the target new samples based on the minority class samples and the target number of new samples, so as to optimize the microseismic data set.
[0011] To achieve the above-mentioned purpose, the third aspect embodiment of the present application proposes an electronic device, comprising: a processor; and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory, so that the processor can execute the sample optimization method of the microseismic data set described in the first aspect embodiment above.
[0012] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer instructions are used to enable the computer to execute the sample optimization method for the microseismic data set described in the first aspect of the present application.
[0013] The sample optimization method, device, electronic device and storage medium of the microseismic data set provided in the present application determine the sample characteristics of the microseismic data set corresponding to the mining area, and determine the minority class samples and the majority class samples in the microseismic data set. Furthermore, the total number of samples to be synthesized is determined based on the minority class samples and the majority class samples, and the proportion of majority class samples in the neighboring samples of the minority class samples is determined, and the proportion of majority class samples is standardized. The number of target new samples to be generated can be determined based on the total number of samples and the proportion of standardized majority class samples, so as to generate target new samples based on the number of minority class samples and target new samples in the microseismic data set. By generating target new samples and adding the generated target new samples to the microseismic data set, an unbalanced microseismic data set can be optimized into a balanced microseismic data set, which helps to learn the characteristics of each category when using the optimized data set to predict the type of rock burst, so as to improve the accuracy of the prediction, thereby improving the prediction performance of the model and improving the prediction efficiency of the model.
[0014] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:
[0016] Figure 1 A schematic flow chart of a sample optimization method for a microseismic data set provided in an embodiment of the present application;
[0017] Figure 2 A schematic diagram of optimization of an unbalanced microseismic data set provided in an embodiment of the present application;
[0018] Figure 3 A schematic flow chart of another sample optimization method for a microseismic dataset provided in an embodiment of the present application;
[0019] Figure 4 A projection result diagram of a microseismic event in space provided in an embodiment of the present application;
[0020] Figure 5 A graph showing the relationship between the energy of a microseismic event and the time interval provided in an embodiment of the present application;
[0021] Figure 6A schematic flow chart of another sample optimization method for a microseismic data set provided in an embodiment of the present application;
[0022] Figure 7 A schematic diagram of a process for optimizing a microseismic data set provided in an embodiment of the present application;
[0023] Figure 8 A schematic diagram of the structure of a sample optimization device for a microseismic dataset provided in an embodiment of the present application. DETAILED DESCRIPTION
[0024] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.
[0025] The sample optimization method and device for a microseismic dataset according to an embodiment of the present application will be described below with reference to the accompanying drawings.
[0026] Figure 1 is a flow chart showing a sample optimization method for a microseismic data set according to an exemplary embodiment. Figure 1 As shown, the sample optimization method of the microseismic dataset in the embodiment of the present application includes but is not limited to the following steps:
[0027] S101, determining reference sample characteristics of microseismic data based on reference mine information and reference microseismic events in a mining area.
[0028] It should be noted that the sample optimization method for the microseismic data set provided in the embodiment of the present application is performed by an electronic device, which may be a terminal device. Optionally, the terminal device may be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device may be a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook or a personal digital assistant (PDA), etc., and the non-mobile electronic device may be a personal computer (PC), a television, etc. The embodiment of the present application does not specifically limit this.
[0029] In some implementations, historical mining data of the mining area, such as quantitative mine geological information and mine mining information, can be obtained as reference mine information of the mining area, and historical microseismic events can be selected from historical microseismic data as reference microseismic events.
[0030] Furthermore, by extracting features from the reference mine information and the reference microseismic events, reference sample features corresponding to different microseismic data can be determined. Alternatively, a reference microseismic event can be used as a sample, and the reference sample features can be determined by extracting features from the reference microseismic event.
[0031] Optionally, the reference sample characteristics include, but are not limited to: waveform characteristics, spectrum characteristics, spatiotemporal intensity characteristics of the microseismic event, the number of samples corresponding to the microseismic event, etc.
[0032] S102, monitor microseismic events in the mining area, determine a microseismic data set, and determine sample features corresponding to the microseismic data set based on mine information, the microseismic data set and reference features in the mining area, where the sample features at least include sample quantity and sample spatiotemporal strength features.
[0033] In some implementations, microseismic events in the mining area can be detected based on geophones. By setting up multiple geophones in the mining area, microseismic waves in the mining area are received, and the waveform, amplitude, frequency, energy, source location, seismic time and other information of the microseismic waves are determined as a microseismic data set. Optionally, microseismic events can also be added as samples to the microseismic data set. Optionally, microseismic events include but are not limited to: rock burst hazard events, large energy events, small energy events, etc.
[0034] In some implementations, by real-time monitoring of the mining area, the mine information of the mining area can be determined. By extracting features from the mine information and the microseismic data set, and matching the extracted features with the reference features, a matched feature can be obtained, and the matched feature can be used as a sample feature corresponding to the microseismic event in the microseismic data set.
[0035] Optionally, the sample characteristics include, but are not limited to: waveform characteristics, spectrum characteristics, spatiotemporal intensity characteristics, sample quantity, etc. of the microseismic event.
[0036] S103, determining minority class samples and majority class samples in the microseismic dataset based on the number of samples and a set number threshold.
[0037] In some implementations, the samples in the microseismic dataset may be divided into minority class samples and majority class samples according to the number of samples in the microseismic dataset. Alternatively, the minority class samples and majority class samples in the microseismic dataset may be determined based on the sample type and a set number threshold.
[0038] Optionally, the sample type is the type of microseismic event, including: rock burst hazard event, high energy event, and low energy event. By determining the number of samples corresponding to different sample types and setting a number threshold, the samples in the microseismic dataset can be divided into minority class samples and majority class samples.
[0039] Optionally, if the number of samples is greater than the number threshold, the sample is determined to be a majority class sample, otherwise it is a minority class sample. For example, if the number threshold is 100, the number of samples of rock burst danger is 20, and the number of samples of small energy events is 200, then it can be determined that small energy events are majority class samples and rock burst is a minority class sample.
[0040] S104, determining the total number of samples to be synthesized, and determining the proportion of majority class samples in the neighboring samples based on the neighboring samples of the minority class samples.
[0041] S105, standardizing the proportion of majority class samples to obtain a standardized proportion of majority class samples.
[0042] In some implementations, in order to balance the samples in the microseismic dataset, target new samples can be generated and added to the microseismic dataset to optimize the microseismic dataset. Optionally, the total number of samples to be synthesized can be determined based on the number of minority class samples and the number of majority class samples in the microseismic dataset.
[0043] Optionally, by determining the nearest neighbor samples of any minority class sample, determining the proportion of majority class samples from the nearest neighbor samples, and normalizing the proportion of majority class samples in the nearest neighbor samples, the number of target new samples to be generated is determined based on the total number of samples and the standardized proportion of majority class samples. Optionally, by determining the samples within a set range around the minority class samples, and using the samples as the nearest neighbor samples of the minority class samples. Optionally, by determining the data space of the microseismic data set, and determining any minority class sample in the data space, the nearest neighbor samples can be determined by setting a range around the minority class sample.
[0044] Furthermore, the number of neighbor samples and majority class samples included in the neighbor samples may be determined, so as to determine the proportion of majority class samples based on the number of neighbor samples and majority class samples.
[0045] Furthermore, the proportion of majority class samples and the sum of the proportion of majority class samples of the number of minority class samples can be determined, so as to determine the standardized proportion of majority class samples based on the sum of the proportion of majority class samples and the proportion of majority class samples of the number of minority class samples.
[0046] S106, determining the target number of new samples to be generated according to the total number of samples and the proportion of standardized majority class samples, and generating target new samples based on the minority class samples and the target number of new samples to optimize the microseismic dataset.
[0047] It is understandable that when there are minority class samples and majority class samples in the microseismic dataset, the microseismic dataset becomes unbalanced. In the subsequent prediction process of rock burst occurrence areas and types, the unbalanced dataset may cause the model prediction to be biased towards the majority class, thereby affecting the generalization ability of the model and the prediction performance for the minority class. Therefore, the microseismic dataset can be optimized by regenerating and adding samples in the microseismic dataset to achieve a balanced state between minority class samples and majority class samples.
[0048] In some implementations, the number of target new samples to be generated can be determined based on the standardized proportion of majority class samples and the total number of samples, and then the target number of target new samples can be generated based on the minority class samples and the number of target new samples in the microseismic dataset, and the target new minority class samples can be added to the microseismic dataset to achieve optimization of the microseismic dataset.
[0049] For example, Figure 2 Schematic diagram of the optimization of the unbalanced microseismic dataset shown. Figure 2 In , the minority class samples are dangerous events, and the majority class samples are non-dangerous events. Figure 2 It can be seen that before the microseismic data set is optimized, there are only 5 dangerous events and the distribution is not concentrated enough, which is significantly different from non-dangerous events. After the microseismic data set model is optimized using the method provided in the embodiment of the present application, the number of dangerous events increases to 13 and the distribution is concentrated. The microseismic data set is optimized, which is helpful for the subsequent use of the optimized microseismic data set to predict the area and type of rock burst.
[0050] In the sample optimization method of the microseismic data set provided in the embodiment of the present application, the sample characteristics of the microseismic data set corresponding to the mining area are determined, and the minority class samples and the majority class samples in the microseismic data set are determined. Further, the total number of samples to be synthesized is determined based on the minority class samples and the majority class samples, and the proportion of the majority class samples in the neighboring samples of the minority class samples is determined, and the proportion of the majority class samples is standardized. The number of target new samples to be generated in the microseismic data set can be determined based on the total number of samples and the proportion of the standardized majority class samples, and the target new samples can be generated to achieve a balanced state of the minority class samples and the majority class samples. The unbalanced microseismic data set can be optimized into a balanced microseismic data set, which is helpful to learn the characteristics of each category when using the optimized data set to predict the rock burst occurrence area and type, so as to improve the accuracy of the prediction, thereby improving the prediction performance of the model and improving the prediction efficiency of the model.
[0051] Figure 3 is a flow chart showing a sample optimization method for a microseismic data set according to an exemplary embodiment. Figure 3As shown, the sample optimization method of the microseismic dataset in the embodiment of the present application includes but is not limited to the following steps:
[0052] S301, determining reference sample characteristics of microseismic data based on reference mine information and reference microseismic events in the mining area.
[0053] In the embodiment of the present application, the implementation method of step S301 can be implemented by any method in the embodiments of the present application, which is not limited here and will not be repeated.
[0054] S302, monitor the microseismic events in the mining area, determine the microseismic data set, and determine the sample characteristics corresponding to the microseismic data set based on the mine information, microseismic data set and reference characteristics of the mining area, where the sample characteristics at least include the sample quantity and the sample spatiotemporal intensity characteristics.
[0055] In the embodiment of the present application, the implementation method of step S302 can be implemented by any method in the embodiments of the present application, which is not limited here and will not be repeated.
[0056] S303: Determine the spatiotemporal strong features of the microseismic events in the microseismic data set based on the sample features.
[0057] In some implementations, the spatiotemporal strong features include time features, space features, and energy features, and the time features, space features, and energy features of the microseismic event can be determined based on the sample features to obtain the spatiotemporal strong features of the microseismic event. Optionally, the spatial features can be determined based on the coordinate information of the microseismic event. The coordinate information can be the three-dimensional coordinates of the microseismic event.
[0058] In some implementations, the coordinate information of the microseismic event is determined and the coordinate information is normalized to obtain the normalized coordinates. Further, the energy corresponding to the microseismic event is determined based on the sample characteristics, and the energy and the normalized coordinates are projected into space to obtain the spatial characteristics.
[0059] Optionally, by projecting the energy and normalized coordinates of the microseismic events onto space, the distribution law of the microseismic events on a spatial scale can be determined and used as a spatial feature. Figure 4 The projection result of the microseismic event in space is shown in the figure. Figure 4 The distribution pattern of microseismic events on a spatial scale can be determined.
[0060] Optionally, the time information of the microseismic event can be determined based on the sample characteristics to determine the correspondence between the time information and the energy as the time feature. The time information can be the time interval, that is, by determining the correspondence between the microseismic event and the time interval, the distribution law of the microseismic event on the time scale can be determined as the time feature. Figure 5 The relationship between the energy of microseismic events and the time interval is shown in Figure 2. Figure 5 The distribution pattern of microseismic events on the time scale can be determined.
[0061] Optionally, the energy information of the microseismic event can be determined based on the sample characteristics to determine the energy distribution law of the microseismic event as the energy feature. The energy information can be the source energy of the microseismic event, that is, the energy distribution law of the microseismic event can be determined as the energy feature by determining the corresponding relationship between the energy size of the microseismic event and the time interval. Figure 5 The relationship between the energy of microseismic events and the time interval is shown in Figure 2. Figure 5 The energy distribution law of microseismic events can be determined.
[0062] S304, based on the strong spatiotemporal features, determining the first strong spatiotemporal features of the minority class samples and the second strong spatiotemporal features of the majority class samples.
[0063] In some implementations, after determining the strong spatiotemporal features, the first strong spatiotemporal features of minority class samples and the second strong spatiotemporal features of majority class samples can be determined based on samples corresponding to different microseismic events in the microseismic dataset.
[0064] S305: Based on the first strong spatiotemporal feature and the second strong spatiotemporal feature, assess the danger of rock burst during the mining process.
[0065] It is understandable that low-energy microseismic events that occur regularly and within a certain range in the mine during the mining process can be regarded as a stable state, that is, the majority class sample. High-energy dangerous events such as rock burst can obviously be regarded as a special sample that is far away from the data distribution center in time and space scale and breaks the equilibrium state in the mine, that is, the minority class sample. Therefore, the danger of the mine during the mining process can be determined by comparing the strong spatiotemporal features of the minority class samples with the strong spatiotemporal features of the majority class samples.
[0066] In some implementations, when the majority class samples have a certain continuous regularity over time and space, and the minority class samples are far away from the area with these regularities, the area where the minority class samples are located has the risk of rock burst. Optionally, by comparing the first spatiotemporal strong feature and the second spatiotemporal strong feature, it is determined whether the minority class samples are far away from the majority class samples. If so, it is determined that the area where the minority class samples are located has the risk of rock burst.
[0067] S306 , determining minority class samples and majority class samples in the microseismic dataset based on the number of samples and a set number threshold.
[0068] In the embodiment of the present application, the implementation method of step S306 can be implemented by any method in the embodiments of the present application, which is not limited here and will not be repeated.
[0069] S307, determining the total number of samples to be synthesized, and based on the neighboring samples of the minority class samples, determining the proportion of the majority class samples in the neighboring samples.
[0070] In the embodiment of the present application, the implementation method of step S307 can be implemented by any method in the embodiments of the present application, which is not limited here and will not be repeated.
[0071] S308, normalizing the proportion of majority class samples to obtain a standardized proportion of majority class samples.
[0072] In the embodiment of the present application, the implementation method of step S308 can be implemented by any method in the various embodiments of the present application, which is not limited here and will not be repeated.
[0073] S309, determining the target number of new samples to be generated according to the total number of samples and the proportion of standardized majority class samples, and generating target new samples based on the minority class samples and the target number of new samples to optimize the microseismic dataset.
[0074] In the embodiment of the present application, the implementation method of step S309 can be implemented by any method in the embodiments of the present application, which is not limited here and will not be repeated.
[0075] In the sample optimization method of the microseismic data set provided in the embodiment of the present application, by determining the sample characteristics of the microseismic data set corresponding to the mining area, and determining the temporal and spatial strong characteristics of the minority class samples and the majority class samples based on the sample characteristics, the danger of rock burst in the mining process can be determined by comparing the temporal and spatial strong characteristics between the two. By determining the danger of rock burst, real-time monitoring and early warning of the mining process can be achieved, thereby enhancing the early warning capability. Performing rock burst hazard assessment can also optimize the mining strategy, thereby improving mining efficiency and economic benefits.
[0076] Figure 6 is a flow chart showing a sample optimization method for a microseismic data set according to an exemplary embodiment. Figure 6 As shown, the sample optimization method of the microseismic dataset in the embodiment of the present application includes but is not limited to the following steps:
[0077] S601, determining reference sample characteristics of microseismic data based on reference mine information and reference microseismic events in the mining area.
[0078] In the embodiment of the present application, the implementation method of step S601 can be implemented by any method in the embodiments of the present application, which is not limited here and will not be repeated.
[0079] S602, monitor the microseismic events in the mining area, determine the microseismic data set, and determine the sample characteristics corresponding to the microseismic data set based on the mine information, microseismic data set and reference characteristics of the mining area, where the sample characteristics at least include the number of samples and the temporal and spatial intensity characteristics of the samples.
[0080] In the embodiment of the present application, the implementation method of step S602 can be implemented by any method in the embodiments of the present application, which is not limited here and will not be repeated.
[0081] S603: Determine minority class samples and majority class samples in the microseismic dataset based on the number of samples and a set number threshold.
[0082] In the embodiment of the present application, the implementation method of step S603 can be implemented by any method in the embodiments of the present application, which is not limited here and will not be repeated.
[0083] S604, determining a first sample quantity of minority class samples and a second sample quantity of majority class samples.
[0084] In some implementations, the number of samples in the microseismic dataset that are minority class samples is obtained and used as the first sample number, and the number of samples in the microseismic dataset that are majority class samples is obtained and used as the second sample number.
[0085] S605, determining the total number of samples based on the first sample number, the second sample number and the random number.
[0086] In some implementations, the total number of samples is determined by calculating the difference between the first sample number and the second sample number, and based on the difference and the random number. Optionally, the formula for calculating the total number of samples is as follows:
[0087] G=(m1-m s )×β(1)
[0088] Among them, G represents the total number of samples, m1 represents the number of second samples, and m s represents the first sample size, β represents a random number, and β∈[0,1].
[0089] S606: for any minority class sample, determine the neighboring samples of any minority class sample and the number of neighbors of the neighboring samples from the microseismic data set, and determine the number of majority class samples in the neighboring samples.
[0090] S607, performing a ratio operation on the number of nearest neighbors and the number of majority class samples to determine the proportion of majority class samples.
[0091] In some implementations, for any minority class sample, a sample within a set range around any minority class sample is determined, and the sample is used as a neighboring sample of the minority class sample, and the number of neighbors of the neighboring sample is determined. Alternatively, a data space of a microseismic data set is determined, and any minority class sample is determined in the data space, and a neighboring sample can be determined by setting a range around the minority class sample.
[0092] Further, the majority class samples among the neighbor samples are determined, and the number of majority class samples is counted to determine the proportion of majority class samples according to the number of neighbors and the number of majority class samples. The proportion of majority class samples can be calculated by performing a ratio operation on the number of neighbors and the number of majority class samples. Optionally, the formula for calculating the proportion of majority class samples is as follows:
[0093]
[0094] Among them, r i represents the proportion of majority class samples, K represents the number of neighbors, Δ i Represents the number of majority class samples among the neighboring samples.
[0095] S608, normalizing the proportion of majority class samples to obtain a standardized proportion of majority class samples.
[0096] S609: Determine the target number of new samples to be generated based on the standardized proportion of majority class samples and the total number of samples.
[0097] Optionally, a ratio operation may be performed on the proportion of majority class samples and the sum of the proportion of majority class samples of the number of minority samples, and the proportion of majority class samples may be standardized to obtain a standardized proportion of majority class samples. Optionally, the formula for standardizing the proportion of majority class samples is as follows:
[0098]
[0099] in, represents the proportion of standardized majority class samples, r i Represents the proportion of majority class samples.
[0100] Furthermore, the proportion of standardized majority class samples can be used as sample weights, and the target number of new samples to be generated can be calculated based on the total number of samples. Optionally, the formula for calculating the target number of new samples to be generated is as follows:
[0101]
[0102] Among them, g represents the number of target new samples, represents the sample weight, and G represents the total number of samples.
[0103] S610, determining any minority class sample among the neighboring samples corresponding to the minority class sample.
[0104] S611, generating a target number of new samples based on the minority class samples, any minority class samples and random numbers.
[0105] In some implementations, by selecting any minority class sample from the neighboring samples of the minority class sample, and based on the minority class samples in the microseismic dataset, a target number of new samples can be generated. Optionally, the formula for generating the target minority class samples is as follows:
[0106] s i =x i +(x zi -x i )×λ(5)
[0107] Among them, s i represents the target minority class sample, x i represents the i-th sample in the minority class sample, x zi represents any minority class sample, and λ represents a random number between 0 and 1.
[0108] In the sample optimization method of the microseismic data set provided in the embodiment of the present application, the sample characteristics of the microseismic data set corresponding to the mining area are determined, and the minority class samples and the majority class samples in the microseismic data set are determined. Further, the total number of samples to be synthesized is determined based on the minority class samples and the majority class samples, and the proportion of the majority class samples in the neighboring samples of the minority class samples is determined and standardized. The number of target new samples to be generated can be determined based on the total number of samples and the proportion of the standardized majority class samples, so as to generate target new samples based on the minority class samples and the number of target new samples in the microseismic data set. By generating target new samples and adding the generated target new samples to the microseismic data set, an unbalanced microseismic data set can be optimized into a balanced microseismic data set, which helps to learn the characteristics of each category when using the optimized data set to predict the area and type of rock burst, so as to improve the accuracy of the prediction, thereby improving the prediction performance of the model and improving the prediction efficiency of the model.
[0109] Figure 7The flowchart for optimizing the microseismic data set is shown. By determining the reference mine information and reference microseismic events in the mining area, the reference sample features are determined according to the reference mine information and the reference microseismic events. The microseismic data set is determined by real-time monitoring of the microseismic events in the mining area, and the sample features corresponding to the microseismic data set are matched with the reference sample features according to the microseismic data set and the mine information. Further, the distribution law of the microseismic events on the time, space and energy scales is determined according to the sample features as the strong spatiotemporal features, and by comparing the first strong spatiotemporal features of the minority class samples with the second strong spatiotemporal features of the majority class samples, the danger of rock burst during the mining process can be determined. Furthermore, for the microseismic data set with an unbalanced number of samples, the optimization of the microseismic data set is achieved by regenerating and adding samples in the microseismic data set. The method of adding minority class samples can be referred to the above embodiment and will not be repeated here.
[0110] In order to implement the above embodiment, the present application also proposes a sample optimization device for a microseismic dataset.
[0111] Figure 8 A schematic diagram of the structure of a sample optimization device for a microseismic dataset provided in an embodiment of the present application.
[0112] like Figure 8 As shown, the sample optimization device 800 of the microseismic data set includes:
[0113] A first determination module 801 is used to determine reference sample features of microseismic data based on reference mine information and reference microseismic events in a mining area;
[0114] A second determination module 802 is used to monitor microseismic events in the mining area, determine a microseismic data set, and determine sample features corresponding to the microseismic data set based on the mine information of the mining area, the microseismic data set and the reference features, wherein the sample features at least include sample quantity and sample spatiotemporal intensity features;
[0115] A third determination module 803 is used to determine the minority class samples and the majority class samples in the microseismic dataset based on the sample quantity and a set quantity threshold;
[0116] The fourth determination module 804 is used to determine the total number of samples to be synthesized, and determine the proportion of the majority class samples in the neighboring samples based on the neighboring samples of the minority class samples;
[0117] A processing module 805 is used to standardize the proportion of the majority class samples to obtain a standardized proportion of the majority class samples;
[0118] The generation module 806 is used to determine the target number of new samples to be generated according to the total number of samples and the proportion of the standardized majority class samples, and generate target new samples based on the minority class samples and the target number of new samples to optimize the microseismic dataset.
[0119] In a possible implementation of an embodiment of the present application, the second determination module 802 is also used to: determine the strong spatiotemporal characteristics of the microseismic events in the microseismic data set based on the sample characteristics; determine the first strong spatiotemporal characteristics of the minority class samples and the second strong spatiotemporal characteristics of the majority class samples based on the strong spatiotemporal characteristics; and evaluate the danger of rock burst in the mining process based on the first strong spatiotemporal characteristics and the second strong spatiotemporal characteristics.
[0120] In a possible implementation of an embodiment of the present application, the second determination module 802 is further used to: determine the coordinate information of the microseismic event, and normalize the coordinate information to obtain normalized coordinates; determine the energy corresponding to the microseismic event based on the sample characteristics, and project it into space based on the energy and the normalized coordinates to obtain the spatial characteristics; determine the time information of the microseismic event based on the sample characteristics to determine the correspondence between the time information and the energy as the time characteristics; determine the energy information of the microseismic event based on the sample characteristics to determine the energy distribution law of the microseismic event as the energy characteristics.
[0121] In a possible implementation of an embodiment of the present application, the fourth determination module 804 is also used to: determine the first sample number of the minority class samples and the second sample number of the majority class samples; and determine the total sample number based on the first sample number, the second sample number and a random number.
[0122] In a possible implementation of an embodiment of the present application, the fourth determination module 804 is further used to: for any minority class sample, determine the neighboring samples of any minority class sample and the number of neighbors of the neighboring samples from the microseismic data set, and determine the number of majority class samples in the neighboring samples; perform a ratio operation on the number of neighbors and the number of majority class samples to determine the proportion of the majority class samples.
[0123] In a possible implementation of the embodiment of the present application, the processing module 805 is also used to: perform a ratio operation on the proportion of the majority class samples and the total proportion of the majority class samples of the minority sample number, and normalize the proportion of the majority class samples to obtain the standardized proportion of the majority class samples.
[0124] In a possible implementation of the embodiment of the present application, the generation module 806 is further used to determine the target number of new samples to be generated based on the proportion of the standardized majority class samples and the total number of samples.
[0125] In a possible implementation of an embodiment of the present application, the generation module 806 is further used to: determine any minority class sample among the neighboring samples corresponding to the minority class sample; and generate the target new sample quantity based on the minority class sample, any minority class sample and a random number.
[0126] In the sample optimization device of the microseismic data set provided in the embodiment of the present application, the sample characteristics of the microseismic data set corresponding to the mining area are determined, and the minority class samples and the majority class samples in the microseismic data set are determined. Further, the total number of samples to be synthesized is determined based on the minority class samples and the majority class samples, and the proportion of the majority class samples in the neighboring samples of the minority class samples is determined and standardized. The number of target new samples to be generated can be determined based on the total number of samples and the proportion of the standardized majority class samples, so as to generate target new samples based on the minority class samples and the number of target new samples in the microseismic data set. By generating target new samples and adding the generated target new samples to the microseismic data set, an unbalanced microseismic data set can be optimized into a balanced microseismic data set, which helps to learn the characteristics of each category when using the optimized data set to predict the rock burst occurrence area and type, so as to improve the accuracy of the prediction, thereby improving the prediction performance of the model and improving the prediction efficiency of the model.
[0127] It should be noted that the above explanations and descriptions of the embodiment of the sample optimization method for a microseismic dataset are also applicable to the sample optimization device for a microseismic dataset of this embodiment, and will not be repeated here.
[0128] In order to implement the above embodiments, the present application also proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the method provided by the above embodiments.
[0129] In order to implement the above embodiments, the present application also proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.
[0130] The collection, storage, use, processing, transmission, provision and application of user personal information involved in this application are in compliance with relevant laws and regulations and do not violate public order and good morals.
[0131] It should be noted that personal information from users should be collected for legitimate and reasonable purposes and should not be shared or sold outside of these legitimate uses. In addition, such collection / sharing should be carried out after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign the agreement / authorization including authorization of relevant user information before the user uses the function. In addition, any necessary steps should be taken to protect and safeguard access to such personal information data and ensure that others who have access to personal information data comply with its privacy policy and procedures.
[0132] This application is expected to provide an implementation scheme for users to selectively block the use or access of personal information data. That is, this application is expected to provide hardware and / or software to prevent or block access to such personal information data. Once the personal information data is no longer needed, risks can be minimized by limiting data collection and deleting data. In addition, when applicable, personal identifiers are removed from such personal information to protect the privacy of users.
[0133] In the description of the aforementioned embodiments, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.
[0134] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0135] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.
[0136] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.
[0137] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or a combination thereof: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0138] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.
[0139] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.
[0140] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.
Claims
1. A sample optimization method for a microseismic dataset, characterized in that: The method comprises: Determine the reference sample characteristics of microseismic data based on reference mine information and reference microseismic events in the mining area; Monitor microseismic events in the mining area to determine a microseismic data set, and determine sample features corresponding to the microseismic data set based on mine information of the mining area, the microseismic data set and the reference features, wherein the sample features include at least sample quantity and sample spatiotemporal intensity features; Based on the number of samples and a set number threshold, determining minority class samples and majority class samples in the microseismic dataset; Determine the total number of samples to be synthesized, and based on the neighboring samples of the minority class samples, determine the proportion of the majority class samples in the neighboring samples; Standardizing the proportion of the majority class samples to obtain a standardized proportion of the majority class samples; The target number of new samples to be generated is determined according to the total number of samples and the proportion of the standardized majority class samples, and the target new samples are generated based on the minority class samples and the target number of new samples to optimize the microseismic dataset.
2. The method according to claim 1, characterized in that After determining the sample features corresponding to the microseismic data set, the method further includes: Based on the sample characteristics, determining the spatiotemporal strong characteristics of microseismic events in the microseismic data set; Based on the strong spatiotemporal features, determining a first strong spatiotemporal feature of the minority class sample and a second strong spatiotemporal feature of the majority class sample; Based on the first strong spatiotemporal characteristics and the second strong spatiotemporal characteristics, the danger of rock burst during the mining process is evaluated.
3. The method according to claim 2, characterized in that The strong spatiotemporal features include time features, space features and energy features, and the determining of the strong spatiotemporal features of the microseismic events in the microseismic data set includes: Determining coordinate information of the microseismic event, and normalizing the coordinate information to obtain normalized coordinates; Determine the energy corresponding to the microseismic event based on the sample feature, and project it onto space based on the energy and the normalized coordinates to obtain the spatial feature; Determining the time information of the microseismic event based on the sample feature to determine the corresponding relationship between the time information and the energy as the time feature; The energy information of the microseismic event is determined based on the sample characteristics to determine the energy distribution law of the microseismic event as the energy characteristic.
4. The method according to claim 1, characterized in that: The determining the total number of samples to be synthesized includes: Determining a first sample size of the minority class samples and a second sample size of the majority class samples; The total sample size is determined based on the first sample size, the second sample size, and a random number.
5. The method according to claim 4, characterized in that The determining, based on the neighboring samples of the minority class samples, the proportion of the majority class samples in the neighboring samples comprises: For any minority class sample, determine the neighboring samples of any minority class sample and the number of neighbors of the neighboring samples from the microseismic data set, and determine the number of majority class samples in the neighboring samples; A ratio operation is performed on the number of the nearest neighbors and the number of the majority class samples to determine the proportion of the majority class samples.
6. The method according to claim 5, characterized in that The standardization process of the proportion of the majority class samples to obtain a standardized proportion of the majority class samples includes: A ratio operation is performed on the proportion of the majority class samples and the sum of the proportions of the majority class samples of the number of minority samples, and the proportions of the majority class samples are standardized to obtain the standardized proportions of the majority class samples.
7. The method according to claim 6, characterized in that The step of determining the target number of new samples to be generated according to the total number of samples and the proportion of the standardized majority class samples includes: Based on the proportion of the standardized majority class samples and the total number of samples, the target number of new samples to be generated is determined.
8. The method according to claim 7, characterized in that The generating the target new sample based on the minority class sample and the number of the target new samples includes: Determine any minority class sample among the neighboring samples corresponding to the minority class sample; Based on the minority class samples, any minority class sample and a random number, the target number of new samples is generated.
9. A sample optimization device for a microseismic data set, characterized in that: The device comprises: A first determination module is used to determine reference sample characteristics of microseismic data based on reference mine information and reference microseismic events in the mining area; A second determination module is used to monitor microseismic events in the mining area, determine a microseismic data set, and determine sample features corresponding to the microseismic data set based on the mine information of the mining area, the microseismic data set and the reference features, wherein the sample features at least include the number of samples and the spatial and temporal strength features of the samples; A third determination module is used to determine the minority class samples and the majority class samples in the microseismic data set based on the sample quantity and a set quantity threshold; A fourth determination module is used to determine the total number of samples to be synthesized, and based on the neighboring samples of the minority class samples, determine the proportion of the majority class samples in the neighboring samples; A processing module, used for standardizing the proportion of the majority class samples to obtain a standardized proportion of the majority class samples; A generation module is used to determine the target number of new samples to be generated according to the total number of samples and the proportion of the standardized majority class samples, and to generate target new samples based on the minority class samples and the target number of new samples to optimize the microseismic dataset.
10. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 8.