Mining area information processing method and system

Abstract

The invention provides a mining area information processing method and system, and belongs to the technical field of information processing, and the method comprises the steps: carrying out the space-time alignment of the multi-modal target data of a mining area, obtaining a plurality of pieces of first mining area data, enabling any modal data to have the unique corresponding first mining area data, and enabling the first mining area data to be in one-to-one correspondence with the first mining area data; the space-time alignment means that the multi-modal target data are aligned in time and space; calculating correlation between the first mining area data; fusing the first mining area data of which the correlation is greater than a target threshold value to obtain mining area comprehensive information data; and generating mining area safety early warning information based on the mining area comprehensive information data. According to the mining area information processing method and system provided by the invention, the potential safety risk of the mining area can be accurately found.

Description

Technical Field

[0001] The present disclosure belongs to the technical field of information processing, and more particularly, relates to a method and system for processing mine area information. Background Art

[0002] In mine production operations, there are various complex production links and environmental factors, and ensuring safe production in the mine area has always been the core goal of the industry. Currently, there are rich and diverse data sources in the mine area, forming multi-modal data such as geological data, equipment operation data, environmental monitoring data, and video surveillance data. However, there are many deficiencies in the existing mine area information processing methods.

[0003] In the existing mine area information processing, there is a lack of effective integration and analysis of multi-modal data, making it difficult to discover potential safety risks in a timely and accurate manner. Summary of the Invention

[0004] The purpose of the present disclosure is to provide a method and system for processing mine area information to accurately discover potential safety risks in the mine area.

[0005] In the first aspect of the embodiments of the present disclosure, a method for processing mine area information is provided, including: Performing spatio-temporal alignment on multi-modal target data of the mine area to obtain first mine area data. The first mine area data includes multiple pieces, and any modal data has a uniquely corresponding first mine area data. Spatio-temporal alignment means that the multi-modal target data is aligned in terms of time and space; Calculating the correlation between the first mine area data; Fusing the first mine area data with a correlation greater than a target threshold to obtain mine area comprehensive information data; Generating mine area safety warning information based on the mine area comprehensive information data.

[0006] In the second aspect of the embodiments of the present disclosure, a system for processing mine area information is provided, including: A data alignment module for performing spatio-temporal alignment on multi-modal target data of the mine area to obtain first mine area data. The first mine area data includes multiple pieces, and any modal data has a uniquely corresponding first mine area data; A correlation calculation module for calculating the correlation between the first mine area data; A data fusion module for fusing the first mine area data with a correlation greater than a target threshold to obtain mine area comprehensive information data; A warning generation module for generating mine area safety warning information based on the mine area comprehensive information data.

[0007] In a third aspect of the embodiments of the present disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the above-mentioned mining area information processing method are implemented.

[0008] In a fourth aspect of the embodiments of the present disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above-mentioned mining area information processing method are implemented.

[0009] The beneficial effects of the mining area information processing method and system provided by the embodiments of the present disclosure are as follows: The embodiments of the present disclosure perform spatio-temporal alignment on multi-modal target data, solve the problems of differences in time and space of different modal data, make the data consistent and comparable under a unified spatio-temporal framework, and provide a solid foundation for subsequent accurate analysis. Calculating the correlation between the first mining area data can uncover potential internal connections between different data, helping to deeply understand the interaction mechanism of various factors in the mining area. By setting a target threshold to screen and fuse strongly correlated data, interference from irrelevant or weakly correlated data is avoided, and the obtained comprehensive mining area information data can more accurately reflect the actual situation of the mining area. The mining area safety warning information generated based on this has higher reliability and accuracy, can timely detect potential safety risks, enable the mining area staff to quickly take preventive and handling measures, thereby effectively ensuring the safe production of the mining area and reducing the possibility of accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present disclosure. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0011] Figure 1 It is a flowchart showing the mining area information processing method provided by an embodiment of the present disclosure; Figure 2 It is a block diagram showing the structure of the mining area information processing system provided by an embodiment of the present disclosure; Figure 3 It is a schematic block diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0012] In the following description, specific details such as specific system architectures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present disclosure. However, those skilled in the art should clearly understand that the present disclosure can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present disclosure.

[0013] To make the objectives, technical solutions, and advantages of the present disclosure clearer, the following will be described through specific embodiments with reference to the accompanying drawings.

[0014] Please refer to Figure 1 , Figure 1 which is a schematic flowchart of a method for processing mine area information provided by an embodiment of the present disclosure. The method includes: S101: Perform spatio-temporal alignment on the multi-modal target data of the mine area to obtain first mine area data. There are multiple pieces of first mine area data, and any piece of modal data has a uniquely corresponding first mine area data. Spatio-temporal alignment means aligning the multi-modal target data in terms of time and space.

[0015] In this embodiment, there are various types of data sources in the mine area, forming multi-modal data. For example, geological data (such as: formation structure data, rock property data, etc.), equipment operation data (such as: parameters such as equipment temperature, pressure, and rotation speed), environmental monitoring data (such as: gas concentration data, dust content data, humidity data, etc.), and video surveillance data (such as: personnel activity data, equipment status data, etc.). These different types of data constitute the multi-modal raw data. The multi-modal target data is the data obtained after preprocessing the multi-modal raw data of the mine area.

[0016] Data of different modalities have differences in time and space. For example, environmental monitoring data can be collected through different sensors, and sensor data can be collected at the second level, while video surveillance data is recorded in frames, and the installation positions of different devices are different, resulting in different spatial positions reflected by the data. Spatio-temporal alignment is to uniformly calibrate data of different modalities in the time and space dimensions so that they can be compared and analyzed in the same spatio-temporal coordinate system.

[0017] In this embodiment, the multi-modal target data can be arranged and matched in chronological order through a clock synchronization mechanism. The geographical location information can be marked for each target data point through a geographic information system, so that the multi-modal target data corresponds in space.

[0018] After spatio-temporal alignment, data of each modality can find its corresponding position in a unified spatio-temporal framework, forming the first mine area data, ensuring the consistency and comparability of the data in the spatio-temporal dimension.

[0019] S102: Calculate the correlation between the data of the first mining area.

[0020] In this embodiment, there are internal connections between different types of data of the first mining area. Calculating the correlation between different types of data of the first mining area can discover the potential relationships between these data. For example, the change in gas concentration is closely related to the operating state of ventilation equipment, and the abnormal temperature of the equipment is related to potential equipment failures. By analyzing the correlation, the interaction between various factors in the mining area can be understood.

[0021] In this embodiment, the correlation between the data of the first mining area can be calculated by the Pearson correlation coefficient or the Spearman correlation coefficient.

[0022] The Pearson correlation coefficient can measure the linear correlation degree between two variables. The correlation coefficient is obtained by calculating the ratio of the covariance of the two variables to the product of their standard deviations. Its value ranges from -1 to 1. The closer the absolute value is to 1, the stronger the correlation. The positive or negative sign indicates positive or negative correlation. The Spearman correlation coefficient can measure the monotonic relationship between variables and calculates the correlation based on the ranks of the data, regardless of the specific distribution form of the data. By calculating the correlation between the data of the first mining area, the degree of association between different data can be quantified.

[0023] S103: Integrate the data of the first mining area with a correlation greater than the target threshold to obtain the comprehensive information data of the mining area.

[0024] In this embodiment, the target threshold is a preset standard for screening out data with strong correlations. Setting the target threshold can ensure that the integrated data have a sufficiently close connection, so that the integrated data can more accurately reflect the actual situation of the mining area. If no threshold is set and all data are integrated, some irrelevant or weakly correlated data will be introduced, resulting in a decline in the quality of the comprehensive information data.

[0025] In this embodiment, for the data of the first mining area with a correlation greater than the target threshold, the weighted average method or the principal component analysis method can be used for integration.

[0026] The weighted average method assigns corresponding weights according to the importance or reliability of different data of the first mining area, and then performs weighted summation on the data to obtain the integration result. The principal component analysis method transforms the data of the first mining area into a set of uncorrelated principal components through linear transformation, and selects the principal components with larger variances as the integrated comprehensive information, thereby realizing data dimensionality reduction and information concentration. Through data integration, multiple related data can be integrated into a more representative comprehensive information data of the mining area.

[0027] S104: Generate mine safety warning information based on the comprehensive mine area information data.

[0028] In this embodiment, the comprehensive mine area information data can be used to establish a mine safety warning model through machine learning algorithms (such as decision trees, support vector machines, neural networks, etc.).

[0029] Taking the neural network as an example, it is trained with a large amount of historical comprehensive mine area information data to learn the feature patterns and rules in the data, as well as the corresponding relationships between these patterns and the mine safety status.

[0030] Input the obtained comprehensive mine area information data into the established mine safety warning model. The mine safety warning model evaluates and judges the current mine safety status based on the knowledge it has learned. If the model detects that some features or indicators in the data exceed the safety range, or conform to a preset dangerous pattern, corresponding mine safety warning information will be generated.

[0031] For example, when the model detects that the gas concentration continues to rise and the ventilation volume is insufficient, and at the same time the equipment temperature is abnormal, it is judged that there may be a risk of gas explosion, and a warning message such as "The gas concentration is too high, the ventilation is abnormal, the equipment temperature is abnormal, there is a risk of gas explosion, please take measures immediately" can be issued to timely remind the mine area staff to take corresponding preventive and treatment measures to ensure the safe production of the mine area.

[0032] It can be concluded from the above that in this embodiment, the spatio-temporal alignment of multi-modal target data is carried out to solve the problems of differences in time and space between different modal data, making the data consistent and comparable under a unified spatio-temporal framework, providing a solid foundation for subsequent accurate analysis. Calculating the correlation between the data in the first mine area can uncover the potential internal connections between different data, helping to deeply understand the interaction mechanism of various factors in the mine area. By setting target thresholds to screen and fuse strongly correlated data, the interference of irrelevant or weakly correlated data is avoided, and the obtained comprehensive mine area information data can more accurately reflect the actual situation of the mine area. The mine safety warning information generated based on this has higher reliability and accuracy, can timely discover potential safety risks, enable the mine area staff to quickly take preventive and treatment measures, thereby effectively ensuring the safe production of the mine area and reducing the possibility of accidents.

[0033] In an embodiment of the present disclosure, spatio-temporal alignment is performed on the multi-modal target data of the mine area to obtain the first mine area data, including: Process the multi-modal target data to obtain the first feature vector data. There are multiple pieces of the first feature vector data, and each modal data has a uniquely corresponding first feature vector data; Based on the timestamp information of the multi-modal target data, the first eigenvector data is roughly aligned in the time dimension to obtain the first initial mining area data; Spatio-temporal features are extracted from the first eigenvector data to obtain the first spatio-temporal feature data; Based on the first initial mining area data and the first spatio-temporal feature data, fine alignment is performed to obtain the first mining area data.

[0034] In this embodiment, the multi-modal target data contains various types of information, and its original form is relatively complex and not convenient for direct spatio-temporal alignment operations. Processing these data into the first eigenvector data is to abstract the data into a more representative and computable feature form. Each modality of data has its unique features, and through appropriate feature extraction and transformation methods, it can be transformed into the uniquely corresponding first eigenvector data, which is convenient for analysis and alignment in a unified feature space.

[0035] An adaptive noise filtering algorithm can be used to remove noise interference in the multi-modal target data, and a normalization method is used to unify the data into a specific value range to eliminate the influence of data dimension differences on subsequent processing. For example, for device temperature data and pressure data, they are normalized to the interval [0, 1].

[0036] In this embodiment, the timestamp is the key information for recording the data acquisition time. There are differences in the data acquisition times of different modalities. Using the timestamp information, the first eigenvector data can be initially matched and aligned in the time dimension. The purpose of rough alignment is to roughly arrange the data in chronological order, so that different modalities of data have a basic correspondence in time, providing a basis for subsequent fine alignment.

[0037] Exemplarily, the environmental monitoring data is collected once per second, and the video surveillance data is collected one frame every two seconds. By comparing their timestamps, the environmental monitoring eigenvector data and the video surveillance eigenvector data within the same time period are associated to form the first initial mining area data.

[0038] In this embodiment, although the first eigenvector data has been preliminarily processed, it still contains some information irrelevant to spatio-temporal alignment. Through spatio-temporal feature extraction, key features related to time and space can be mined from the first eigenvector data, such as the change trend, periodicity, and spatial distribution law of the data. These spatio-temporal features can more accurately reflect the characteristics of the data in the spatio-temporal dimension and can provide a more accurate basis for fine alignment.

[0039] In this embodiment, the first spatio-temporal feature data provides detailed information about the data in terms of space and time. By using this information to fine-tune the first initial mining area data, the data of different modalities can achieve a more accurate correspondence in time and space, and finally accurate first mining area data can be obtained.

[0040] Alignment in the time dimension: Extract features from the first spatio-temporal feature data to obtain the periodic features corresponding to the first spatio-temporal feature data; Based on the periodic features, obtain the phase difference corresponding to the first spatio-temporal feature data; Adjust the first initial mining area data in terms of time based on the periodic features and phase difference corresponding to the first spatio-temporal feature data.

[0041] Exemplarily, for the first spatio-temporal feature data with periodic changes, such as the periodic operation data of equipment or the periodic fluctuation data of environmental parameters, extract its main periodic features through methods such as Fourier transform. Compare the periodic features of different modality data to find the periodic differences and phase differences between them. For example, for the power data and temperature data of equipment operation, it is found that the period of the power data is 60 minutes, the period of the temperature data is 62 minutes, and there is a certain phase difference. According to the periodic features and phase information, fine-tune the first initial mining area data in terms of time. The method of linear interpolation can be used to insert or delete some data points in the data sequence to make the periods and phases of different modality data match as much as possible. For example, translate and interpolate the temperature data sequence appropriately to make its period and phase consistent with the power data.

[0042] Alignment in the space dimension: Based on the spatial distribution of the first spatio-temporal feature data, obtain the deviation between different first spatio-temporal feature data; Adjust the first initial mining area data in terms of time based on the deviation between different first spatio-temporal feature data.

[0043] Exemplarily, the method of coordinate transformation can be used to translate, rotate or scale the spatial coordinates of different modality data to make their spatial distribution patterns coincide as much as possible. For example, translate the coordinates of the equipment failure occurrence location data according to the set step length to align it with the clustering center of the geological structure data.

[0044] It can be concluded from the above that in this embodiment, the multi-modal target data is first processed into the first feature vector data for subsequent operations. Coarse alignment is performed based on the time stamp to make the data of different modalities have a preliminary correspondence in time, laying a foundation for fine alignment. Spatio-temporal feature extraction mines the spatio-temporal characteristics of the data, providing a basis for precise alignment. Finally, fine alignment is performed by combining the coarse alignment result and the spatio-temporal feature data, greatly improving the alignment accuracy of the data in the spatio-temporal dimension and making the first mining area data more accurate and reliable.

[0045] In one embodiment of the present disclosure, calculating the correlation between the first mining area data includes: Calculating the correlation between the first mining area data based on the first formula; The first formula is:

[0046] Wherein, represents the correlation between the first mining area data sample and the first mining area data sample ; represents the weight parameter that balances the mutual information and the cosine similarity; represents the first mining area data sample and the first mining area data sample 's mutual information; represents the dynamic weight of the i-th data; represents the first mining area data sample and the first mining area data sample 's cosine similarity.

[0047] In this embodiment, the correlation between the first mining area data is calculated based on the mutual information and the cosine similarity.

[0048] Suppose we have two first mining area data samples and , where n is the data dimension.

[0049] First, calculate the mutual information of X and Y . The mutual information is used to measure the degree of dependence between two random variables. In the mining area data scenario, the mutual information can capture the internal correlation between different data samples. For example, there is a complex dependence relationship between the gas concentration and the operation status of the ventilation equipment, and the mutual information can quantify the intensity of this dependence.

[0050] The formula for the mutual information is:

[0051] Wherein, represents the joint probability distribution of X = x and Y = y; and represent the marginal probability distributions of X = x and Y = y, respectively.

[0052] The mutual information reflects the summation over all possible combinations of data values, thus comprehensively considering the mutual influence between data. When differs significantly from , the value of the mutual information is large, indicating a high degree of dependence between the two variables.

[0053] Calculate the cosine similarity between X and Y , which is used to measure the similarity in direction between two vectors. In mining area data, the cosine similarity can determine the similarity degree of different data samples in the feature space. For example, the temperature and power data of equipment have similarity in the changing trend, and the cosine similarity can quantify the similarity degree of this trend.

[0054] The formula for cosine similarity is:

[0055] Then, introduce the data dynamic weight , and the data dynamic weight reflects the change of the importance of the data in the i-th dimension at different times or in different environments. It can be dynamically adjusted according to factors such as the change frequency of the data and the degree of association with other key data. For example, for the gas concentration data, the data dynamic weight can be increased during the peak period of mining operations, while the data dynamic weight can be decreased during the maintenance period.

[0056] Finally, calculate the correlation between the data of the first mining area by integrating mutual information and cosine similarity, which is expressed as:

[0057] Among them, , can be adjusted according to the actual data characteristics and application requirements. When it is close to 1, it focuses more on the variable dependence relationship reflected by mutual information; when is close to 0, it emphasizes more on the vector direction similarity reflected by cosine similarity. and respectively represent the data of X and Y in the i-th dimension.

[0058] It can be concluded from the above that compared with the traditional correlation calculation method, this embodiment not only considers the statistical dependence relationship and vector similarity of the data, but also introduces the data dynamic weight, and can more flexibly and accurately reflect the complex correlation between the multi-modal data in the mining area, adapting to the dynamic changes of the mining area environment and data characteristics.

[0059] In an embodiment of the present disclosure, fuse the first mining area data with a correlation greater than the target threshold to obtain the mining area comprehensive information data, including: Regard the first mining area data with a correlation greater than the target threshold as the second mining area data; Assign weights to the second mining area data based on the importance level of the second mining area data; Perform weighted fusion on the weights of each second mining area data to obtain the mining area comprehensive information data.

[0060] In this embodiment, not all the first mining area data play an equally important role in forming accurate comprehensive mining area information. By setting a target threshold, the first mining area data with a correlation greater than this threshold are selected as the second mining area data. Correlation reflects the degree of internal connection between different data. A high correlation means that these data have strong relevance and complementarity when describing the phenomena in the mining area. Selecting these data can avoid the interference of irrelevant or weakly related data in the subsequent fusion process, making the fusion result better reflect the actual situation of the mining area.

[0061] Exemplarily, in a mining area, the correlation between gas concentration data and the operating status data of ventilation equipment is relatively high. When the correlation is greater than the target threshold, these two types of data will be selected as the second mining area data because they jointly reflect the ventilation safety status of the mining area.

[0062] In this embodiment, different second mining area data play different roles in the comprehensive mining area information, that is, they have different importance levels. In order to fully reflect this difference in the fusion process, it is necessary to assign corresponding weights to each second mining area data according to the importance level of the data. The evaluation of the importance level can be based on various factors, such as the degree of influence of the data on mining area safety, the accuracy and reliability of the data, and the key degree of the mining area characteristics reflected by the data.

[0063] Exemplarily, for mining area safety, gas concentration data is directly related to risks such as explosion, and its importance is relatively high, so a relatively high weight can be assigned; while some secondary environmental monitoring data, such as air humidity data, have a relatively small direct impact on safety and can be assigned a relatively low weight.

[0064] In this embodiment, weighted fusion can comprehensively consider different data and their importance. By multiplying each second mining area data by its corresponding weight and then adding the results, the comprehensive mining area information data is obtained. Thus, the final comprehensive information can more comprehensively and accurately reflect the overall situation of the mining area.

[0065] In this embodiment, the weights of each second mining area data can be weighted and fused based on the second formula. The second formula is:

[0066] where represents the comprehensive mining area information data obtained after fusion, N represents the number of second mining area data, Represents the importance weight of the j-th second mining area data. This importance weight can be assigned based on the importance level of the data for the comprehensive information of the mining area and can be determined according to the degree of influence of the data on aspects such as mining area safety and production efficiency. For example, the weight of gas concentration data is relatively high, while the weight of environmental humidity data is relatively low. Represents the value of the j-th second mining area data. Represents the adjustment coefficient, with a value range of [0, 1], which is used to control the degree of influence of the correlation between data on the fusion result and can be adjusted according to the actual situation. Represents the correlation coefficient between the j-th and f-th second mining area data, which can be obtained through the above-mentioned correlation calculation method. The correlation coefficient reflects the degree of association between different data. If is relatively large, it indicates that these two data have a relatively large mutual influence. Represents the reliability coefficient of the j-th second mining area data, with a value range of [0, 1]. The reliability coefficient is used to measure the accuracy and credibility of the data and can be determined according to factors such as the accuracy of the data acquisition device and the stability of data transmission. For example, the reliability coefficient of data collected by high-precision sensors is relatively high, while the reliability coefficient of data affected by more interference is relatively low.

[0067] In this embodiment, the second formula is different from the traditional formula. The second formula introduces the correlation coefficient between data. In the mining area environment, different types of data are often interrelated and interact with each other. For example, the gas concentration is closely related to the operating state of the ventilation equipment. Considering the correlation between data can more accurately reflect this interaction and make the fusion result more in line with the actual situation. By introducing the reliability coefficient, the quality of the data is considered. In practical applications, the reliability of different data sources may vary. Incorporating reliability into the fusion calculation can improve the accuracy and credibility of the fusion result. For example, for data with relatively low reliability, its influence on the final result will be correspondingly reduced during fusion. The adjustment coefficient allows users to flexibly adjust the degree of influence of the correlation between data on the fusion result according to actual needs and data characteristics, making the formula more adaptable and flexible.

[0068] In an embodiment of the present disclosure, it further includes: Performing data analysis on the multi-modal raw data to obtain missing values and outliers in the multi-modal raw data; Filling the missing values in the multi-modal raw data based on the interpolation method and correcting the outliers in the multi-modal raw data based on historical data to obtain multi-modal target data.

[0069] In this embodiment, during the processes of collecting and transmitting multi-modal raw data, missing values and outliers may occur due to reasons such as equipment failures, communication interruptions, and environmental interferences. Missing values can lead to incomplete data and affect the accuracy of subsequent analyses; outliers are caused by measurement errors or special events, and if not processed, they may mislead the results of data analysis. Therefore, it is necessary to conduct comprehensive data analysis on the multi-modal raw data to accurately identify the missing values and outliers among them.

[0070] For missing values, they can be identified by checking for null values and specific missing value identifiers in the data records. For outliers, a method based on standard deviation can be used. If the deviation of a data point from the mean exceeds a certain multiple of the standard deviation, it is considered an outlier.

[0071] In multi-modal raw data, different types of data often have a certain degree of continuity and correlation. The interpolation method can utilize these characteristics to reasonably fill in the missing values, making the data more complete and facilitating subsequent analysis and processing.

[0072] Using historical data to correct outliers can make the data more in line with the actual situation and reduce the impact of outliers on subsequent analyses. Historical data contains the normal change patterns and rules of the data. By analyzing historical data, information such as the normal value range and change trend of the data can be understood, thereby determining the causes of outliers and correcting them.

[0073] The mean of the historical data can be calculated, and the outliers can be corrected according to the calculated mean. For example, if a data point is determined to be an outlier, it can be corrected to the mean of the historical data.

[0074] From the above, it can be concluded that in this embodiment, the missing values and outliers in the multi-modal raw data are identified through data analysis, and then the interpolation method and the method based on historical data are respectively used to process the missing values and outliers. Finally, more complete and accurate multi-modal target data is obtained, providing a good foundation for subsequent spatio-temporal alignment and data analysis.

[0075] In an embodiment of the present disclosure, it further includes: Determining a target threshold based on the number of missing values and outliers.

[0076] In this embodiment, missing values ​​mean incomplete data information, which may cause the data to fail to fully and accurately reflect the actual situation of the mining area. Outliers are data points that deviate from the normal data pattern, which will distort the distribution of the data and destroy the stability and regularity of the data. When calculating the correlation between the data of the first mining area, data quality plays a key role. The existence of missing values ​​and outliers will interfere with the calculation results of the correlation, causing the correlation between the data that originally had an intrinsic connection to be underestimated or overestimated. For example, if there are multiple missing values ​​in a data sequence, when calculating its correlation with other data sequences, due to the influence of missing values, the true relationship between the two cannot be accurately captured, resulting in inaccurate correlation calculation.

[0077] The target threshold is dynamically determined based on the number of missing values ​​and outliers, making the data fusion process adaptive. The screening criteria can be flexibly adjusted according to the actual quality of the data to ensure that relatively ideal comprehensive information data of the mining area can be obtained under different data quality conditions.

[0078] In one embodiment of the present disclosure, determining a target threshold based on the number of missing values ​​and outliers includes: Calculate the proportion of missing values ​​and outliers in multimodal target data; In response to the proportion of missing values ​​and outliers in the multimodal target data being greater than a first ratio, using the first threshold as a target threshold; In response to the proportion of missing values ​​and outliers in the multimodal target data being less than or equal to the first ratio and greater than the second ratio, using the second threshold as the target threshold; In response to the proportion of missing values ​​and outliers in the multimodal target data being less than the second ratio, the third threshold is used as the target threshold.

[0079] In this embodiment, the proportion of missing values ​​and outliers in the multimodal target data is calculated. Missing values ​​will lead to incomplete data information, and outliers will interfere with the normal distribution and regularity of the data. By calculating the proportion, the amount of data with problems in the data can be quantified, thereby evaluating the overall quality of the data. The higher the proportion, the worse the data quality; the lower the proportion, the better the data quality.

[0080] When the proportion of missing values ​​and outliers is greater than the first ratio, it indicates that the data quality is poor. At this time, if a higher threshold is used for data fusion, a large amount of relevant data will be excluded due to problems in the data, making the comprehensive information data of the mining area obtained after fusion incomplete. Therefore, the first threshold is used as the target threshold, and the first threshold is relatively small, which can reduce the screening criteria for data fusion and retain as much data as possible for fusion to make up for the information loss caused by insufficient data quality.

[0081] When the proportion of missing values and outliers is less than or equal to the first ratio and greater than the second ratio, the data quality is at a medium level. At this time, the second threshold is used as the target threshold. The second threshold is between the first threshold and the third threshold, which can ensure the fusion of a certain number of relevant data and exclude some weakly relevant data to a certain extent, so that the fusion result achieves a balance between information integrity and accuracy.

[0082] When the proportion of missing values and outliers is less than the second ratio, it indicates that the data quality is high. At this time, a higher third threshold can be used as the target threshold. The higher threshold can screen out more strongly relevant data for fusion, thereby improving the accuracy and reliability of the integrated mine area information data after fusion and reducing the interference of irrelevant or weakly relevant data.

[0083] It can be concluded from the above that in this embodiment, the data quality is evaluated by calculating the proportion of missing values and outliers, and the target threshold is dynamically adjusted according to different quality levels, realizing the adaptive adjustment of the data fusion screening criteria, which helps to obtain high-quality integrated mine area information data under different data quality conditions.

[0084] Corresponding to the mine area information processing method in the above embodiment, Figure 2 is a structural block diagram of a mine area information processing system provided by an embodiment of the present disclosure. For the sake of illustration, only the parts related to the embodiments of the present disclosure are shown. Refer to Figure 2 The mine area information processing system 20 includes: a data alignment module 21, a correlation calculation module 22, a data fusion module 23, and an early warning generation module 24.

[0085] Among them, the data alignment module 21 is used to perform spatio-temporal alignment on the multi-modal target data of the mine area to obtain the first mine area data. There are multiple pieces of the first mine area data, and each modal data has a unique corresponding first mine area data. Spatio-temporal alignment means that the multi-modal target data is aligned in time and space; The correlation calculation module 22 is used to calculate the correlation between the first mine area data; The data fusion module 23 is used to fuse the first mine area data with a correlation greater than the target threshold to obtain the integrated mine area information data; The early warning generation module 24 is used to generate mine area safety warning information based on the integrated mine area information data.

[0086] In an embodiment of the present disclosure, the data alignment module 21 is specifically used for: Processing the multi-modal target data to obtain the first feature vector data. There are multiple pieces of the first feature vector data, and each modal data has a unique corresponding first feature vector data; Based on the timestamp information of the multi-modal target data, the first feature vector data is roughly aligned in the time dimension to obtain the first initial mining area data; Extract spatio-temporal features from the first feature vector data to obtain the first spatio-temporal feature data; Based on the first initial mining area data and the first spatio-temporal feature data, perform fine alignment to obtain the first mining area data.

[0087] In an embodiment of the present disclosure, the correlation calculation module 22 is specifically configured to: Calculate the correlation between the first mining area data based on the first formula; The first formula is:

[0088] Wherein, represents the correlation between the first mining area data sample and the first mining area data sample , represents the weight parameter that balances mutual information and cosine similarity, represents the mutual information of the first mining area data sample and the first mining area data sample , represents the dynamic weight of the i-th data, represents the cosine similarity of the first mining area data sample and the first mining area data sample .

[0089] In an embodiment of the present disclosure, the data fusion module 23 is specifically configured to: Take the first mining area data with a correlation greater than the target threshold as the second mining area data; Assign weights to the second mining area data based on the importance level of the second mining area data; Perform weighted fusion on the weights of each second mining area data to obtain the mining area comprehensive information data.

[0090] In an embodiment of the present disclosure, the mining area information processing system 20 further includes: a data processing module; the data processing module is specifically configured to: Perform data analysis on the multi-modal raw data to obtain the missing values and outliers in the multi-modal raw data; Fill in the missing values in the multi-modal raw data based on the interpolation method, and correct the outliers in the multi-modal raw data based on the historical data to obtain the multi-modal target data.

[0091] In an embodiment of the present disclosure, the data processing module is specifically further configured to: Determine the target threshold based on the number of missing values and outliers.

[0092] In one embodiment of the present disclosure, the data processing module is further specifically configured to: Calculate the proportions of missing values and outliers in the multi-modal target data; In response to the proportions of missing values and outliers in the multi-modal target data being greater than a first ratio, use the first threshold as the target threshold; In response to the proportions of missing values and outliers in the multi-modal target data being less than or equal to the first ratio and greater than a second ratio, use the second threshold as the target threshold; In response to the proportions of missing values and outliers in the multi-modal target data being less than the second ratio, use the third threshold as the target threshold; The first ratio is greater than the second ratio, the first threshold is less than the second threshold, and the second threshold is less than the third threshold.

[0093] Refer to Figure 3 , Figure 3 , which is a schematic block diagram of an electronic device provided by an embodiment of the present disclosure. As Figure 3 shown, the electronic device 300 in this embodiment may include: one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The above-mentioned processors 301, input devices 302, output devices 303, and memories 304 communicate with each other through a communication bus 305. The memory 304 is used to store computer programs, and the computer programs include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. Among them, the processor 301 is configured to call the program instructions to execute the functions of each module in the above system embodiments, such as Figure 2 the functions of the modules 21 to 24 shown.

[0094] It should be understood that in the embodiments of the present disclosure, the so-called processor 301 may be a central processing unit (CPU), and this processor may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or this processor may also be any conventional processor, etc.

[0095] The input device 302 may include a touchpad, a fingerprint acquisition sensor (for acquiring the fingerprint information and the orientation information of the fingerprint of the user), a microphone, etc., and the output device 303 may include a display (such as an LCD), a speaker, etc.

[0096] The memory 304 may include a read-only memory and a random access memory, and provide instructions and data to the processor 301. A part of the memory 304 may further include a non-volatile random access memory. For example, the memory 304 may also store information about the device type.

[0097] In a specific implementation, the processor 301, the input device 302, and the output device 303 described in the embodiments of the present disclosure may implement the implementation manners described in the first and second embodiments of the mining area information processing method provided by the embodiments of the present disclosure, and may also implement the implementation manner of the electronic device described in the embodiments of the present disclosure, which will not be elaborated herein.

[0098] In another embodiment of the present disclosure, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and the computer program includes program instructions. When the program instructions are executed by a processor, all or part of the processes in the methods of the above embodiments are implemented. It may also be completed by instructing relevant hardware through the computer program. The computer program may be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above method embodiments may be implemented. Among them, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or system that can carry the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disc, a computer memory, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0099] The computer-readable storage medium may be an internal storage unit of the electronic device in any of the foregoing embodiments, such as the hard disk or memory of the electronic device. The computer-readable storage medium may also be an external storage device of the electronic device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device. Further, the computer-readable storage medium may include both an internal storage unit and an external storage device of the electronic device. The computer-readable storage medium is used to store the computer program and other programs and data required by the electronic device. The computer-readable storage medium may also be used to temporarily store the data that has been output or will be output.

[0100] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this disclosure.

[0101] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described electronic devices and units can refer to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0102] In several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed couplings or direct couplings or communication connections to each other can be indirect couplings or communication connections through some interfaces or units, and can also be electrical, mechanical, or other forms of connection.

[0103] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiments of this disclosure.

[0104] In addition, the functional units in each embodiment of this disclosure can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0105] The above is only the specific implementation manner of this disclosure, but the protection scope of this disclosure is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or substitutions within the technical scope disclosed by this disclosure, and these modifications or substitutions should all be covered within the protection scope of this disclosure. Therefore, the protection scope of this disclosure should be subject to the protection scope of the claims.

Claims

1. A mining area information processing method, characterized in that: include: Performing spatiotemporal alignment on the multimodal target data of the mining area to obtain first mining area data, wherein the first mining area data includes a plurality of first mining area data, and any modal data has a unique corresponding first mining area data, and the spatiotemporal alignment indicates that the multimodal target data is aligned in time and space; Calculate the correlation between the data of the first mining area; The first mining area data with correlation greater than the target threshold are integrated to obtain mining area comprehensive information data; Mine area safety warning information is generated based on the mine area comprehensive information data.

2. The mining area information processing method according to claim 1, characterized in that: The step of performing spatiotemporal alignment on the multimodal target data of the mining area to obtain first mining area data includes: Processing the multimodal target data to obtain first feature vector data, wherein the first feature vector data includes a plurality of first feature vector data, and any modal data has a unique corresponding first feature vector data; Based on the timestamp information of the multimodal target data, the first feature vector data is roughly aligned in the time dimension to obtain the first initial mining area data; Performing spatiotemporal feature extraction on the first feature vector data to obtain first spatiotemporal feature data; Fine alignment is performed based on the first initial mining area data and the first spatiotemporal feature data to obtain first mining area data.

3. The mining area information processing method according to claim 1, characterized in that: The calculating the correlation between the first mining area data includes: Calculate the correlation between the first mining area data based on a first formula; The first formula is: in, Represents the first mining area data sample And the first mining area data sample The correlation between represents the weight parameter that balances mutual information and cosine similarity, Represents the first mining area data sample And the first mining area data sample The mutual information of represents the dynamic weight of the i-th data, Represents the first mining area data sample And the first mining area data sample The cosine similarity of .

4. The mining area information processing method according to claim 1, characterized in that: The fusing of the first mining area data whose correlation is greater than the target threshold to obtain the mining area comprehensive information data includes: The first mining area data with correlation greater than the target threshold is used as the second mining area data; assigning a weight to the second mining area data based on a level of importance of the second mining area data; The weight of each second mining area data is weighted and fused to obtain the comprehensive information data of the mining area.

5. The mining area information processing method according to claim 1, characterized in that: Also includes: Perform data analysis on the multimodal raw data to obtain missing values ​​and outliers in the multimodal raw data; The missing values ​​in the multimodal original data are filled based on the interpolation method, and the abnormal values ​​in the multimodal original data are corrected based on the historical data to obtain the multimodal target data.

6. The mining area information processing method according to claim 5, characterized in that: Also includes: The target threshold is determined based on the number of missing values ​​and outliers.

7. The mining area information processing method according to claim 6, characterized in that: The determining the target threshold based on the number of missing values ​​and outliers comprises: Calculate the proportion of missing values ​​and outliers in the multimodal target data; In response to the proportion of missing values ​​and outliers in the multimodal target data being greater than a first ratio, using the first threshold as a target threshold; In response to the proportion of missing values ​​and outliers in the multimodal target data being less than or equal to the first ratio and greater than the second ratio, using the second threshold as the target threshold; In response to the proportion of missing values ​​and outliers in the multimodal target data being less than a second ratio, using a third threshold as a target threshold; The first ratio is greater than the second ratio, the first threshold is less than the second threshold, and the second threshold is less than a third threshold.

8. A mining area information processing system, characterized in that: include: A data alignment module is used to perform spatiotemporal alignment on the multimodal target data of the mining area to obtain first mining area data, wherein the first mining area data includes a plurality of first mining area data, and any modal data has a unique corresponding first mining area data; A correlation calculation module, used for calculating the correlation between the data of the first mining area; A data fusion module, used for fusing the first mining area data with correlation greater than a target threshold to obtain comprehensive information data of the mining area; The warning generation module is used to generate mining area safety warning information based on the mining area comprehensive information data.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images