Mine geothermal risk checking and evaluation method and system based on multi-source data cooperation

Abstract

This invention discloses a method and system for verifying and assessing geothermal risks in mines based on multi-source data collaboration, belonging to the field of geothermal technology in mines. The method includes: acquiring multi-source environmental data and generating a benchmark score based on the multi-source environmental data; verifying the benchmark score based on historical scoring records to generate verification results; and, in response to the verification results, classifying the multi-source environmental data and historical scoring records into multiple assessment sample sets. After generating the benchmark score, this invention introduces time-weighted historical scoring records to calculate the verification score. By comparing the difference between the two, if the difference exceeds the allowable range, the relevant historical data is identified as verification data. Through this scoring verification and data classification mechanism, dynamic verification of the benchmark score is performed, identifying and correcting assessment biases caused by data anomalies, reducing defects in input data quality, and improving the accuracy of risk assessment.

Description

Technical Field

[0001] This invention belongs to the field of mine geothermal technology, specifically relating to a mine geothermal risk verification and assessment method and system based on multi-source data collaboration. Background Technology

[0002] With the continuous development of smart mine construction, precise monitoring and intelligent early warning of various geological risks in the mine production process are carried out to ensure personnel safety and improve mining efficiency. In mining operations, geothermal energy, as a geological environmental factor, may induce geological disasters and affect energy utilization due to its abnormal accumulation.

[0003] Existing technologies for geothermal risk assessment using multi-source data lack the ability to identify and process the quality of input data. Due to the complexity of the mining environment, commonly used detection sensors may malfunction or drift under harsh operating conditions, and data entry errors may occur due to human operational mistakes. During data fusion, all collected data are directly used for model training and risk inference, leading to erroneous data contaminating the assessment model and causing biases in the risk assessment results. At the same time, most existing assessment models use static or fixed thresholds for judgment and assessment, but the evolution of geothermal risk is a non-linear process generated by changes in factors such as mining activities and geological structures, which static models cannot capture. During data fusion, although multiple data such as temperature, stress, and microseismic data are nominally fused, most of them remain at the level of simple splicing or weighted averaging, with a shallow level of fusion of multi-source data, failing to detect potential risks caused by the combined effects of multiple factors in advance.

[0004] In view of this, the present invention provides a method and system for verifying and assessing geothermal risks in mines based on multi-source data collaboration. Summary of the Invention

[0005] The main objective of this invention is to provide a method for verifying and assessing the geothermal risk of mines based on multi-source data collaboration. This method can accurately judge the reliability of multi-source data and treat it as key data and verification data for unified assessment, ultimately obtaining an accurate assessment of the geothermal risk of the mine.

[0006] To achieve the above objectives, the technical solution adopted by this invention is as follows: a method for verifying and assessing geothermal risks in mines based on multi-source data collaboration, comprising the following steps:

[0007] Acquire multi-source environmental data and generate a benchmark score based on the multi-source environmental data; perform validation on the benchmark score based on historical score records to generate validation results;

[0008] In response to the verification results, multi-source environmental data and historical scoring records are classified into multiple assessment sample sets; a risk trend analysis model is generated based on the multiple assessment sample sets; and the current risk level is determined based on real-time environmental parameters and the risk trend analysis model.

[0009] The verification of the benchmark score based on historical scoring records to generate verification results includes: calculating a verification score based on historical scoring records and in combination with preset weights and scoring deviation correction values; comparing the difference between the benchmark score and the verification score with a preset allowable range to generate verification results.

[0010] Preferably, the multi-source environmental data includes: geothermal information, groundwater level, mining records, mining area structure, and geological structural parameters.

[0011] The weights of each environmental parameter's impact on geothermal risk were determined using the Analytic Hierarchy Process (AHP) and normalized after field verification, as follows: Geothermal Information Weights Groundwater level weight Mining record weight Mining area structural weight Geological structural parameter weights ;

[0012] All weights satisfy:

[0013]

[0014] in, Represents the parameter weight, its meaning is the first The preset impact weights of each environmental parameter on the total risk of geothermal energy; This represents the standardized parameter value, which means the first... The values ​​of each environmental parameter are standardized and range from 0 to 1. This indicates the number of parameters, which means the total number of environmental parameters involved in the calculation. In this embodiment... .

[0015] Basis for determination: Geothermal temperature is the most direct indicator of geothermal risk and is the core factor affecting geothermal risk, so it is given the highest weight; geological structural parameters are background influencing factors of geothermal risk and have a weaker direct effect on risk evolution, so they are given the lowest weight. This weight allocation is in line with the conventional practical logic of mine geothermal monitoring.

[0016] Preferably, comparing the difference between the benchmark score and the verification score with a preset allowable range includes: if the difference is within the allowable range, the multi-source environmental data that generated the benchmark score is identified as critical data; if the difference exceeds the allowable range, the data that constitutes the historical scoring record is identified as verification data.

[0017] The allowable deviation range between the benchmark score and the verification score is set at ±5 points. The basis for this setting is that the typical measurement error of commonly used sensors for mine geothermal monitoring is ±3%, which translates to an allowable fluctuation range of ±5 points for the benchmark score (0 to 100 points), in line with the actual accuracy requirements of on-site monitoring.

[0018] Preferably, generating a risk trend analysis model based on multiple assessment sample sets includes: sorting the data in each assessment sample set to determine the dominant parameter and auxiliary parameters; combining the dominant parameter with the auxiliary parameters adjacent to the dominant parameter in the sorting to extract multiple sub-level risk trend values; and summarizing the sub-level risk trend values ​​to generate a risk trend analysis model.

[0019] Specifically, the risk trend analysis model is a multi-parameter coupled rule model that can be directly calculated without training. The specific construction steps are as follows: 1. Sort the parameters according to their contribution to geothermal risk. The sorting results are: temperature (dominant parameter) > groundwater level > mining progress > mining area structure > geological structure parameters.

[0020] 2. Take the dominant parameter in each evaluation sample set and form a parameter pair with one adjacent auxiliary parameter in the ranking to ensure the rationality of parameter coupling;

[0021] 3. Calculate the sub-level risk trend value. The calculation formula is: Sub-level risk trend value = dominant parameter × 0.7 + auxiliary parameter × 0.3, highlighting the core role of the dominant parameter;

[0022] 4. Summarize all sub-level risk trend values, assign weights to the correlation of risks according to each parameter, and sum them up to obtain the total risk trend value, forming a complete risk trend analysis model.

[0023] Preferably, determining the current risk level based on real-time environmental parameters and a risk trend analysis model includes: converting real-time environmental parameters into input parameters with the same format as the dominant parameters, substituting the input parameters into the risk trend analysis model for comparison to obtain an assessment result, and determining the current risk level based on the assessment result and a preset risk level standard.

[0024] Specifically, the real-time environmental parameters are first converted into input parameters with the same format as the primary and auxiliary parameters using the min-max standardization method. The standardization formula is: Standardized parameter value = (Original parameter value − Minimum value) / (Maximum value − Minimum value), with a value range of 0 to 1. These values ​​are then substituted into the risk trend analysis model to obtain the original assessment results. Then, the original assessment results are normalized into risk assessment values. The normalization formula is:

[0025]

[0026] in , ,make sure The risk level falls within the range of 0 to 100; the geothermal risk level is determined by the normalized risk assessment value. The geothermal environment is classified into four levels, with the following specific grading standards: 0-20 points: Level I / Safe, the geothermal environment is stable and there is no risk of geothermal disasters; 21-40 points: Level II / Lower risk, the geothermal environment is basically stable, but routine monitoring needs to be strengthened; 41-70 points: Level III / Higher risk, the geothermal environment is abnormal and early warning measures need to be activated; 71-100 points: Level IV / Dangerous, there is a potential for geothermal disasters, and operations need to be stopped immediately and preventive measures need to be taken.

[0027] The mine geothermal risk verification and assessment system based on multi-source data collaboration includes the following modules:

[0028] The environmental data acquisition module is used to acquire multi-source environmental data and real-time environmental parameters;

[0029] The scoring and verification module is used to generate a baseline score based on multi-source environmental data and to verify the baseline score based on historical scoring records in order to generate verification results.

[0030] The trend model generation module, configured in response to the verification results output by the scoring verification module, is used to classify multi-source environmental data and historical scoring records into multiple assessment sample sets and generate a risk trend analysis model based on the assessment sample sets.

[0031] The risk level determination module is used to determine the current risk level based on real-time environmental parameters and risk trend analysis models.

[0032] Preferably, the multi-source environmental data includes: geothermal information, groundwater level, mining records, mining area structure, and geological structural parameters.

[0033] The weights of each environmental parameter's impact on geothermal risk were determined using the Analytic Hierarchy Process (AHP) and normalized after field verification as follows: Geothermal Information Weights Groundwater level weight Mining record weight Mining area structural weight Geological structural parameter weights ;

[0034] All weights satisfy:

[0035]

[0036] in, Represents the parameter weight, its meaning is the first The preset impact weights of each environmental parameter on the total risk of geothermal energy; This represents the standardized parameter value, which means the first... The values ​​of each environmental parameter are standardized and range from 0 to 1. This indicates the number of parameters, which means the total number of environmental parameters involved in the calculation. In this embodiment... .

[0037] Preferably, the benchmark score is verified based on historical scoring records to generate a verification result: based on historical scoring records and combined with preset weights and scoring deviation correction values, a verification score is calculated, and the difference between the benchmark score and the verification score is compared with a preset allowable range to generate a verification result.

[0038] Specifically, the score deviation correction value Determined according to the following rule: Let the difference in scores between two adjacent time points in the historical scoring sequence be denoted as . ,in Rate the current historical data. For the historical score of the previous moment, if The score indicates that the historical score fluctuations were minor and meaningless; the correction value is... ;like A score of 1 indicates significant fluctuations in historical scores, requiring bias correction. The correction value is... Verification score The verification is based on time-weighted historical scores and score deviation correction values. The synthesis, specifically the calculation formula is as follows:

[0039]

[0040] in, For the latest historical rating, Rate the newest historical stock. Earlier historical data is used for scoring; time weighting is determined according to the time decay principle, with recent historical data better reflecting current geothermal risk trends, therefore it is given higher weight; the allowable deviation range between the benchmark score and the verification score is set at ±5 points. A score of ≤5 indicates that the current multi-source environmental data is consistent with historical trends and the data is valid; if A score greater than 5 indicates that the current data is abnormal or that the geothermal risk trend has changed significantly, requiring the data correction process to begin.

[0041] Preferably, the trend model generation module is further configured as follows:

[0042] If the verification results indicate that the difference between the benchmark score and the verification score is within the acceptable range, then the multi-source environmental data that generated the benchmark score will be identified as critical data.

[0043] If the verification result indicates that the difference exceeds the allowable range, the data constituting the historical scoring record will be determined as verification data.

[0044] The data correction process employs exponential smoothing to smooth the verification data, with a fixed smoothing factor. This value balances the stability of geothermal risk trends with response sensitivity: Values ​​that are too small will lead to a lag in trend response and an inability to capture changes in risk in a timely manner; Taking values ​​that are too large will result in poor smoothing and an inability to filter out data noise. The values ​​used have been verified in the field and are suitable for dynamic scenarios in mine geothermal monitoring. The specific formula for exponential smoothing is:

[0045]

[0046]

[0047] in For time points Smoothed score For time points The original historical rating, For time points The smoothed score.

[0048] Preferably, the trend model generation module is further configured to: sort the data in each evaluation sample set to determine the dominant parameter and auxiliary parameter, combine the dominant parameter with the auxiliary parameter adjacent to the dominant parameter in the sorting to extract multiple sub-level risk trend values, and summarize the sub-level risk trend values ​​to generate a risk trend analysis model.

[0049] Specifically, the parameters are ranked according to their contribution to geothermal risk, and the ranking is as follows: temperature (dominant parameter) > groundwater level > mining progress > mining area structure > geological structural parameters. The dominant parameter in each assessment sample set is paired with one adjacent auxiliary parameter in the ranking to form a parameter pair. The sub-level risk trend value = dominant parameter × 0.7 + auxiliary parameter × 0.3. All sub-level risk trend values ​​are summarized, and weights are assigned according to the correlation between each parameter and risk. The weighted sum is then used to obtain the total risk trend value, forming a risk trend analysis model.

[0050] Beneficial effects

[0051] 1. After generating the benchmark score, this invention introduces time-weighted historical score records to calculate the verification score. By comparing the difference between the two, if the difference exceeds the allowable range, the relevant historical data is identified as verification data. Through this score verification and data classification mechanism, the benchmark score is dynamically verified, the assessment bias caused by data anomalies is identified and corrected, the defects in the quality of input data are reduced, and the accuracy of risk assessment is improved.

[0052] 2. This invention categorizes the classified data into multiple evaluation sample sets containing key data and verification data. Within each evaluation sample set, dominant and auxiliary parameters are determined, and sub-level risk trend values ​​are extracted based on their combination to generate a risk trend analysis model for static numerical evaluation. Furthermore, by constructing the risk trend analysis model, the interaction relationships between key risk factors are explored to reveal the evolutionary patterns of risk. This elevates risk assessment from static judgment to dynamic prediction.

[0053] 3. This invention integrates geothermal information, groundwater level, mining records and other types of multi-source environmental data, and inputs real-time environmental parameters into the constructed risk trend analysis model to determine the current risk level. This enables a comprehensive consideration of multiple factors affecting geothermal risk, allowing risk assessment to more realistically reflect the complex geological and production conditions of the mine. Attached Figure Description

[0054] Figure 1 This is a flowchart of the method of the present invention;

[0055] Figure 2 This is a system module diagram of the present invention. Detailed Implementation

[0056] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0057] Example 1

[0058] Please see Figure 1 As shown, this embodiment discloses a method for verifying and assessing geothermal risks in mines based on multi-source data collaboration. By judging the reliability of multi-source environmental data, distinguishing and processing key data and verification data, the method achieves geothermal risk assessment in mines.

[0059] The specific steps of this method are as follows:

[0060] Data acquisition and standardization: Acquire multi-source environmental data from the mine. This multi-source environmental data includes geothermal information such as rock mass temperature, water inflow temperature and flow rate; groundwater levels such as real-time water level monitoring values ​​of each aquifer; mining records such as mining depth, mining progress and blasting operation logs; mining area structure such as three-dimensional spatial data of roadway layout and goaf distribution; and geological structural parameters such as fault location, rock strata properties and thermal conductivity.

[0061] After acquisition, the acquired multi-source environmental data undergoes a standardized processing procedure. This procedure aims to convert raw data from different sources and in different formats into standardized data with the same units, numerical range, and time reference, thus providing a data foundation for subsequent unified calculations.

[0062] The specific standardized formula is as follows: =(original parameter value − minimum value) / (maximum value − minimum value), where These are the standardized parameter values, ranging from 0 to 1.

[0063] Furthermore, a benchmark score is generated: standardized multi-source environmental data is input into a pre-set geothermal risk assessment system. Then, based on the predetermined influence weights of each environmental parameter on geothermal risk, a comprehensive quantitative calculation is performed to generate risk parameters. The risk parameters can characterize the comprehensive geothermal risk status jointly determined by the current multi-source environmental data. These risk parameters are determined as the benchmark score, i.e., the initial risk assessment score, to serve as the basis for subsequent verification and comparison.

[0064] The weights of each environmental parameter's impact on geothermal risk were determined using the Analytic Hierarchy Process (AHP) and normalized after field verification, as follows: Geothermal Information Weights Groundwater level weight Mining record weight Mining area structural weight Geological structural parameter weights ;

[0065] All weights satisfy:

[0066]

[0067] The specific mathematical formulas for the geothermal risk assessment system are defined as follows:

[0068] The input is a set of standardized environmental parameter values. Examples include normalized rock mass temperature and water inflow rate.

[0069] The output is risk parameters. .

[0070]

[0071] Make the benchmark score It falls within the range of 0 to 100;

[0072] In the formula, This represents the risk parameter, which is a quantitative value representing the overall risk of geothermal energy.

[0073] Represents the parameter weight, its meaning is the first The preset weights of the impact of each environmental parameter on the total geothermal risk; the specific values ​​are: , , , , ;

[0074] This represents the standardized parameter value, which means the first... The values ​​of each environmental parameter are standardized and range from 0 to 1.

[0075] This indicates the number of parameters, which means the total number of environmental parameters involved in the calculation. In this embodiment... .

[0076] Further, determine the risk level and benchmark score: based on the generated benchmark score and in comparison with the preset risk level standards, determine the risk level corresponding to the benchmark score;

[0077] Specifically, geothermal risks are divided into several discrete levels and mapped using pre-established risk level standards. The geothermal risk level is then determined based on the normalized risk assessment value. (Compared to benchmark score) The scores are all within the same range of 0 to 100 points and are divided into four levels. The specific grading standards are as follows: 0 to 20 points: Level I (Safe); 21 to 40 points: Level II (Lower Risk); 41 to 70 points: Level III (Higher Risk); 71 to 100 points: Level IV (Dangerous). A fixed score is pre-assigned to each risk level as its baseline score, which provides a stable reference point for subsequent verification calculations.

[0078] Further, generate verification scores: read historical scoring records associated with the current risk level, arrange them in chronological order according to their timestamps to form a comparison data of time series data, analyze and verify historical trends, and then set corresponding weights and scoring deviation correction values ​​based on the timestamps of each record in the comparison data.

[0079] The weights are determined according to the time decay principle: the most recent historical score is 0.6, the previous score is 0.3, and the earliest score is 0.1; the score deviation correction value... Determined according to the following rule: Let the difference in scores between two adjacent time points in the historical scoring sequence be denoted as . ,in Rate the current historical data. For the historical score of the previous moment, if Points, correction value ;like Points, correction value Then, based on the set weights, scoring deviation correction values, and the benchmark score determined in the "Determine Risk Level and Benchmark Score" step, a weighted calculation is performed to generate the verification score. This score reflects the verification result of the current risk status based on historical data trends. The specific calculation formula is as follows:

[0080]

[0081] For the latest historical rating, Rate the newest historical stock. This is an earlier historical rating;

[0082] Specifically, the weights can be calculated based on the time decay coefficient of historical data. For example, by applying exponential decay calculation to the time interval between each historical record and the present, the more recent records are given higher weights, thereby reflecting the time value of the data, that is, reflecting the decrease in the timeliness of information.

[0083] The rating deviation correction value is set to filter out meaningless small fluctuations in historical data. Its generation process is as follows: calculate the difference between each historical score and its previous historical score, then compare the absolute value of the difference with the preset rating threshold to determine whether the difference between consecutive historical scores is a significant fluctuation that needs attention, and decide whether to generate a non-zero rating deviation correction value based on this.

[0084] Specifically, a non-zero scoring deviation correction value is generated only when the absolute value of the difference exceeds the threshold to adjust the verification calculation; otherwise, the correction value is zero. Then, a weighted calculation is performed based on the set weights, the scoring deviation correction value, and the benchmark score determined in the "Determine Risk Level and Benchmark Score" step to generate a verification score. This score reflects the verification result of the current risk status in combination with historical data trends.

[0085] Further, data screening and classification: After verification, the difference between the verification score and the benchmark score is calculated, and it is determined whether the difference falls within the preset allowable range to define the acceptability of the difference between the verification score and the benchmark score, and to determine the validity of the current data accordingly; the allowable deviation range is set at ±5 points, based on the fact that the typical measurement error of commonly used sensors for mine geothermal monitoring is ±3%, which is converted into an allowable fluctuation range of ±5 points for the benchmark score (0 to 100 points);

[0086] When the difference is within the allowable range, the benchmark score is deemed valid, and the current batch of multi-source environmental data that generated the benchmark score is identified as key data. This situation indicates that the current real-time monitoring data shows consistency with historical trends.

[0087] When the difference exceeds the set allowable range, the reference data designated to enter the data correction process indicates that the current score is abnormal or the overall trend has changed significantly. Then, the reference data generated in the "Generate Verification Score" step is determined as verification data, and the subsequent data correction process is initiated.

[0088] Furthermore, a data correction process is executed: In this process, a preset data smoothing procedure is applied to the historical scoring record sequence identified as verification data to eliminate random noise in the historical scoring sequence; specifically, exponential smoothing is used for smoothing, with a fixed smoothing factor. This value balances the stability of geothermal risk trends with response sensitivity.

[0089] Specifically, by calculating the average value within the sliding time window or by using an exponentially weighted average, a smooth data sequence reflecting the true trend of change can be generated. Then, the smoothed data sequence is weighted to obtain a single aggregated value, which forms a unified verification parameter to comprehensively reflect the overall correction trend of historical data.

[0090] The weighted processing process is as follows: each data point in the sequence is assigned a weight, that is, a weight factor is assigned according to its position in the sequence or its deviation from the sequence mean, etc. Then, the weighted average of the entire sequence is calculated, which is the verification parameter. Based on the generated verification parameter and the smoothed verification data, the verification result is calculated and output, which represents the correction to the trend of the original historical data.

[0091] The specific mathematical formula for the data smoothing process is defined as follows:

[0092] The input is the original historical rating sequence. ,in For at a certain point in time The rating;

[0093] The output is a smoothed score sequence. ;

[0094]

[0095]

[0096] In the formula, This represents the smoothed score, meaning the score at a given time point. A smoothed score that incorporates historical information;

[0097] This represents the original historical score, which means the score at a given point in time. The original historical rating;

[0098] This represents the score after smoothing from the previous time point, meaning that at that time... Smoothed score value;

[0099] The smoothing factor is a coefficient between 0 and 1 that determines the weight of the current observation in the smoothing result. In this embodiment... ;

[0100] This represents a time point index, which means the sequence number of a time point in the historical scoring sequence;

[0101] This represents the total length of the sequence, which is the total number of data points in the historical scoring sequence.

[0102] Furthermore, data grouping and parameter determination: The key data in the "data filtering and classification" step, as well as the verification data in the "execute data correction process" step, are grouped according to their respective data sources to generate multiple evaluation sample sets for analyzing parameters in specific physical domains;

[0103] Then, for each data item in the assessment sample set, internal ranking is performed based on its prior influence or importance level on geothermal risk. After ranking, the parameter ranked first in each assessment sample set is determined as the dominant parameter, and the remaining parameters in the assessment sample set that have an auxiliary or secondary influence on geothermal risk, other than the dominant parameter, are determined as auxiliary parameters. The specific ranking results are: temperature (dominant parameter) > groundwater level > mining progress > mining area structure > geological structure parameters.

[0104] Furthermore, a set of risk trend rules is constructed: for each assessment sample set, the dominant parameter is combined with one or more auxiliary parameters that are adjacent to it in the ranking to form a parameter pair; for each parameter pair, a specific analysis and processing logic is applied, and then based on the preset physical correlation or statistical law;

[0105] Once the specific numerical relationship between the dominant parameter and the auxiliary parameter is determined, the degree of contribution of both to the risk under this relationship is calculated to obtain the sub-level risk trend value; the calculation formula is: sub-level risk trend value = dominant parameter × 0.7 + auxiliary parameter × 0.3.

[0106] By repeating the above process for all predetermined parameter combinations in all evaluation sample sets, all calculated sub-level risk trend values ​​and their corresponding parameter combination conditions are summarized to form a risk trend rule set.

[0107] Furthermore, a real-time risk assessment is conducted: The currently monitored real-time environmental parameters are acquired and converted into input parameters consistent with the format of the primary and secondary parameters. Specifically, the min-max normalization method is used, with the following formula: =(original parameter value − minimum value) / (maximum value − minimum value); then these input parameters are matched and queried in the risk trend rule set;

[0108] Based on the parameter combination conditions, one or more rules in the set that are closest to the current input parameters are found. Then, the sub-level risk trend values ​​contained in the matched rules are extracted as the evaluation result. This evaluation result quantifies the current state of the real-time parameter combination in the geothermal risk evolution trend.

[0109] Furthermore, the final risk level is output: Before outputting the final conclusion, the assessment results obtained in the "conduct real-time risk assessment" step are input into a predefined normalization process; the assessment results, which may be multidimensional or non-standardized values, are mapped to a standardized single scoring interval to achieve comparability of results; then, they are converted into risk judgment values, and then compared with the risk level standards defined in the "determine risk level and benchmark score" step to obtain the final risk level; by judging the numerical interval in which it falls, the corresponding current risk level is output as the final assessment conclusion of the current geothermal risk status of the mine.

[0110] The specific mathematical formula for the normalization process is defined as follows:

[0111] Input: Original evaluation results ;

[0112] Output: Risk assessment value ;

[0113]

[0114] In the formula, This represents the risk assessment value, which is a standardized score used for the final risk level determination after normalization, and its value ranges from 0 to 100.

[0115] This indicates the evaluation result, which means the original risk trend quantification value obtained by matching from the risk trend rule set;

[0116] This represents the maximum expected value, which signifies the evaluation result. The maximum value that may occur theoretically or in historical data, in this embodiment ;

[0117] This represents the minimum expected value, which signifies the evaluation result. The minimum value that may occur theoretically or in historical data, in this embodiment .

[0118] Example 2

[0119] Please see Figure 2 As shown in the figure, this embodiment discloses a mine geothermal risk verification and assessment system based on multi-source data collaboration. Through dynamic analysis and modeling of historical and real-time data, it can assess and warn of mine geothermal risks.

[0120] In its specific implementation, the system can be deployed on a central monitoring server, cloud platform, or dedicated embedded computing device in the mining area. The system interacts with sensors deployed throughout the mine, such as ground temperature sensors, water level gauges, and microseismic monitors, as well as mining information management systems such as production scheduling systems and geological information databases, through means such as industrial Ethernet and wireless sensor networks.

[0121] The system specifically includes the following modules:

[0122] The environmental data acquisition module is responsible for acquiring all the data required for risk assessment. It collects information from multiple data sources periodically or as instructed to form multi-source environmental data.

[0123] The multi-source environmental data specifically includes: geothermal information obtained through geothermal sensor arrays, groundwater level data obtained through water level monitoring equipment, mining records retrieved from the production management system, mining area structure information extracted from design drawings or three-dimensional geological models, and geological structural parameters obtained from geological exploration reports.

[0124] The weights of each environmental parameter's impact on geothermal risk were determined using the Analytic Hierarchy Process (AHP) and normalized after field verification, as follows: Geothermal Information Weights Groundwater level weight Mining record weight Mining area structural weight Geological structural parameter weights ;

[0125] All weights satisfy:

[0126]

[0127] It is also responsible for collecting various environmental parameters that reflect the current mine status in real time and at high frequency, which are used as real-time environmental parameters for subsequent modules to make immediate risk assessments.

[0128] The scoring and verification module generates a preliminary score from the acquired data and verifies its reliability. It receives multi-source environmental data from the environmental data acquisition module and performs comprehensive quantitative processing on this data in accordance with the geothermal risk assessment system to generate an initial benchmark score.

[0129] Specifically, the multi-source environmental data is first subjected to min-max standardization (formula: =(Original parameter value − Minimum value) / (Maximum value − Minimum value)), and then calculate the benchmark score according to the determined weights, the formula is:

[0130]

[0131] Make the benchmark score It falls within the range of 0 to 100;

[0132] The benchmark score initially reflects the risk level indicated by the current multi-source data. Then, historical score records are retrieved from the historical database and combined with preset weights and score deviation correction values ​​that are dynamically adjusted based on expert experience or machine learning models to verify the accuracy of the benchmark score.

[0133] The weights are determined according to the time decay principle: the most recent historical score is 0.6, the previous score is 0.3, and the earliest score is 0.1; the score deviation correction value... Determined according to the following rule: Let the difference in scores between two adjacent time points in the historical scoring sequence be denoted as . ,,like Points, correction value ;like Points, correction value ;

[0134] The verification score is then calculated, representing the theoretical risk score predicted based on historical trends. The calculation formula is as follows:

[0135]

[0136] For the latest historical rating, Rate the newest historical stock. The system first assigns an earlier historical score; then it compares the baseline score with the verification score, calculates the difference between the two, and compares this difference with a preset tolerance range, such as ±5 points, to generate a verification result. The verification result is then output to the trend model generation module as the basis for its subsequent operations.

[0137] The trend model generation module responds to the verification results output by the scoring verification module, classifies the data and builds an analysis model, and classifies the data according to the verification results to form multiple evaluation sample sets;

[0138] If the verification results indicate that the difference between the baseline score and the verification score is within the acceptable range, such as If the score is ≤5, it indicates that the multi-source environmental data obtained is reliable and representative. Therefore, the multi-source environmental data of the benchmark score is identified as key data and included in the sample set.

[0139] If the difference exceeds the allowable range, such as | - A score greater than 5 indicates anomalies or disturbances in the current data, while historical data better reflects stable trends. The comparison data is then identified as verification data and incorporated into another sample set. At this point, the data correction process is initiated, using exponential smoothing to smooth the verification data, with a smoothing factor... The smoothing formula is:

[0140]

[0141]

[0142] Then, based on the multiple assessment sample sets obtained from the classification, a risk trend analysis model is generated. For each assessment sample set, the data parameters within the set are sorted according to their contribution or correlation to the risk. The sorting results are: temperature (dominant parameter) > groundwater level > mining progress > mining area structure > geological structure parameters, thereby determining the dominant parameter and auxiliary parameter.

[0143] The identified dominant parameters are combined with the auxiliary parameters adjacent to them in the sorting list, and the coupling relationship between them is analyzed to extract multiple sub-level risk trend values. The calculation formula is: sub-level risk trend value = dominant parameter × 0.7 + auxiliary parameter × 0.3. Each sub-level risk trend value represents a specific risk evolution law. Then, through weighting and fusion, all extracted sub-level risk trend values ​​are summarized to construct a risk trend analysis model.

[0144] The risk level determination module is responsible for determining the current risk based on real-time data and the built model. It receives real-time environmental parameters provided by the environmental data acquisition module and risk trend analysis models generated by the trend model generation module.

[0145] During the judgment process, real-time environmental parameters are preprocessed and converted into input parameters consistent with the format of the dominant parameters in the risk trend analysis model. Specifically, the min-max standardization method is used, with the following formula: =(original parameter value − minimum value) / (maximum value − minimum value), to ensure the consistency and comparability of data format;

[0146] The input parameters are then substituted into the risk trend analysis model for calculation or comparison to obtain quantitative assessment results. These results are then compared with preset risk level standards, such as mapping the assessment results to a threshold table of levels like "0-20 points: Level I / Safe", "21-40 points: Level II / Lower Risk", "41-70 points: Level III / Higher Risk", and "71-100 points: Level IV / Dangerous". This determines the current risk level, which can be displayed on a large screen in the monitoring center, trigger audible and visual alarms, or automatically execute emergency plans such as ventilation and cooling.

[0147] Through the collaborative work of the above modules, this embodiment ensures the reliability of data analysis by cross-validating the benchmark score with historical score records, and improves the accuracy and foresight of risk assessment by using a dynamically generated risk trend analysis model.

[0148] The above are merely preferred embodiments of this application and are not intended to limit this application. For those skilled in the art, this application can have various modifications and variations. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for verifying and assessing geothermal risks in mines based on multi-source data collaboration, characterized in that, Includes the following steps: Acquire multi-source environmental data and generate a benchmark score based on the multi-source environmental data; perform validation on the benchmark score based on historical score records to generate validation results; In response to the verification results, multi-source environmental data and historical scoring records are classified into multiple evaluation sample sets; A risk trend analysis model is generated based on multiple assessment sample sets. The current risk level is determined based on real-time environmental parameters and a risk trend analysis model. The verification of the benchmark score based on historical scoring records to generate verification results includes: calculating a verification score based on historical scoring records and in combination with preset weights and scoring deviation correction values; comparing the difference between the benchmark score and the verification score with a preset allowable range to generate verification results.

2. The method for verifying and assessing geothermal risks in mines based on multi-source data collaboration as described in claim 1, characterized in that, Multi-source environmental data includes: geothermal information, groundwater level, mining records, mining area structure, and geological structural parameters.

3. The method for verifying and assessing geothermal risks in mines based on multi-source data collaboration as described in claim 1, characterized in that, Comparing the difference between the benchmark score and the verification score with a preset tolerance range includes: if the difference is within the tolerance range, the multi-source environmental data that generated the benchmark score is identified as critical data; if the difference exceeds the tolerance range, the data that constitutes the historical scoring record is identified as verification data.

4. The method for verifying and assessing geothermal risks in mines based on multi-source data collaboration as described in claim 1, characterized in that, Based on multiple assessment sample sets, the risk trend analysis model is generated by: sorting the data in each assessment sample set to determine the dominant and auxiliary parameters; combining the dominant parameter with the auxiliary parameters adjacent to the dominant parameter in the sorting to extract multiple sub-level risk trend values; and summarizing the sub-level risk trend values ​​to generate the risk trend analysis model.

5. The method for verifying and assessing geothermal risks in mines based on multi-source data collaboration as described in claim 1, characterized in that, Determining the current risk level based on real-time environmental parameters and a risk trend analysis model involves: converting real-time environmental parameters into input parameters with the same format as the dominant parameters; substituting the input parameters into the risk trend analysis model for comparison to obtain the assessment results; and determining the current risk level based on the assessment results and the preset risk level standards.

6. A mine geothermal risk verification and assessment system based on multi-source data collaboration, characterized in that, Includes the following modules: The environmental data acquisition module is used to acquire multi-source environmental data and real-time environmental parameters; The scoring and verification module is used to generate a baseline score based on multi-source environmental data and to verify the baseline score based on historical scoring records in order to generate verification results. The trend model generation module, configured in response to the verification results output by the scoring verification module, is used to classify multi-source environmental data and historical scoring records into multiple assessment sample sets and generate a risk trend analysis model based on the assessment sample sets. The risk level determination module is used to determine the current risk level based on real-time environmental parameters and risk trend analysis models.

7. The mine geothermal risk verification and assessment system based on multi-source data collaboration according to claim 6, characterized in that, Multi-source environmental data includes: geothermal information, groundwater level, mining records, mining area structure, and geological structural parameters.

8. The mine geothermal risk verification and assessment system based on multi-source data collaboration according to claim 6, characterized in that, The benchmark score is validated based on historical scoring records to generate validation results: Based on historical scoring records and combined with preset weights and scoring deviation correction values, a verification score is calculated. The difference between the benchmark score and the verification score is compared with a preset allowable range to generate validation results.

9. The mine geothermal risk verification and assessment system based on multi-source data collaboration according to claim 6, characterized in that, The trend model generation module is further configured as follows: If the verification results indicate that the difference between the benchmark score and the verification score is within the acceptable range, then the multi-source environmental data that generated the benchmark score will be identified as critical data. If the verification result indicates that the difference exceeds the allowable range, the data constituting the historical scoring record will be determined as verification data.

10. The mine geothermal risk verification and assessment system based on multi-source data collaboration according to claim 6, characterized in that, The trend model generation module is further configured to: sort the data in each evaluation sample set to determine the dominant and auxiliary parameters, combine the dominant parameter with the auxiliary parameters adjacent to the dominant parameter in the sorting to extract multiple sub-level risk trend values, and summarize the sub-level risk trend values ​​to generate a risk trend analysis model.

Images