A safety early warning method and system applied to a hydrogeological exploration scene

Abstract

The application relates to a safety warning method and system applied to a hydrogeological exploration scene. The method comprises the following steps: during a pumping test, drawing a drawdown value point of an observation hole to obtain a corresponding drawdown change curve, wherein the drawdown change curve comprises a drawdown turning point; taking the drawdown turning point as a fixed clustering center, performing clustering analysis on the drawdown value point to determine a bending curve in a middle region of the drawdown change curve; calculating an error between the bending curve and a standard curve; and performing safety warning based on the error. Through the technical scheme, safety warning of the pumping test process can be realized, normal pumping test is ensured, the accuracy of hydrogeological parameters is ensured, and the safety hidden danger in production operation is reduced.

Description

TECHNICAL FIELD

[0001] The present application generally relates to the field of safety warning, and in particular to a safety warning method and system applied to a hydrogeological exploration scene. BACKGROUND

[0002] Pumping test is a crucial link in hydrogeological exploration, which is the main method to determine the permeability coefficient, transmissibility coefficient, and water release coefficient of aquifer, and can also directly evaluate the allowable exploitation capacity of water source, so pumping test directly affects the accuracy of hydrogeological parameters. When the hydrogeological parameters are inaccurate, safety hazards may occur in subsequent production operations.

[0003] At present, the standardization of the operation of the operating personnel in the pumping test process is usually judged subjectively, and then the hydrogeological parameters are calculated directly according to the pumping test results. The patent document with publication number CN217718452U discloses a self-adaptive control device for pumping test, which can realize real-time monitoring of the water inflow and dynamic water level of the pumping hole during the pumping test process, and simultaneously draw the change curve of the water inflow and dynamic water level of the pumping hole with time in real time, so as to facilitate the operating personnel to analyze whether the pumping test is stable. However, the above-mentioned device can only draw the change curve, and whether the pumping test is stable is judged subjectively by human, which has low automation degree and large error, so that the calculated hydrogeological parameters have errors, which leads to safety hazards in subsequent production operations, so a real-time warning method is urgently needed to ensure the normal progress of pumping test. SUMMARY

[0004] In order to solve the above technical problems in the prior art, the present application provides a safety warning method and system applied to a hydrogeological exploration scene, to ensure the normal progress of pumping test and ensure the accurate calculation of hydrogeological parameters.

[0005] In the first aspect, the present application provides a safety warning method applied to a hydrogeological exploration scene, comprising: during the pumping test process, drawing the drawdown value points of the observation hole to obtain the corresponding drawdown change curve, wherein the drawdown change curve comprises a rapid growth segment and a slow growth segment, and a drawdown turning point between the rapid growth segment and the slow growth segment; taking the drawdown turning point as a fixed clustering center, performing clustering analysis on the drawdown value points to determine the bending curve in the middle region of the drawdown change curve, wherein the drawdown value change amount of the bending curve shows a trend of first increasing and then decreasing; calculating the error between the bending curve and a standard curve; and performing safety warning based on the error.

[0006] In some embodiments, the method for obtaining the drawdown turning point in the drawdown variation curve comprises: calculating the difference between the drawdown value at the current acquisition time and the drawdown value at the adjacent acquisition time before the current acquisition time as the drawdown value variation at the current acquisition time; calculating the difference between the drawdown value variation at the current acquisition time and the drawdown value variation at the adjacent acquisition time before the current acquisition time as the drawdown value variation increment; if the drawdown value variation increment is less than 0, the drawdown value point corresponding to the current acquisition time is taken as the drawdown turning point; if the drawdown value variation increment is greater than or equal to 0, the drawdown value variation increment of the adjacent acquisition time after the current acquisition time is calculated until the drawdown turning point in the drawdown variation curve is obtained.

[0007] In some embodiments, the method for clustering analysis of the drawdown value points with the drawdown turning point as the fixed clustering center to determine the curved curve in the middle region of the drawdown variation curve comprises: clustering analysis of the drawdown value points with the drawdown turning point as the fixed clustering center to obtain at least two clustering clusters; and taking the curved curve between all the drawdown value points in the clustering cluster corresponding to the fixed clustering center as the curved curve of the drawdown variation curve.

[0008] In some embodiments, the method for clustering analysis of the drawdown value points with the drawdown turning point as the fixed clustering center to obtain at least two clustering clusters comprises: setting a preset clustering cluster number, the preset clustering cluster number being greater than or equal to 2; taking the drawdown turning point as the fixed clustering center and randomly selecting at least one initial clustering center from the drawdown value points other than the drawdown turning point, wherein the sum of the number of the initial clustering center and the fixed clustering center is equal to the preset clustering cluster number; for each drawdown value point, calculating the distance between the drawdown value point and each clustering center, and dividing the drawdown value point into the clustering cluster corresponding to the nearest clustering center, the clustering center including the initial clustering center and the fixed clustering center; for the clustering cluster corresponding to the initial clustering center, taking the mean value of all the drawdown value points as the updated initial clustering center, and for the clustering cluster corresponding to the fixed clustering center, the fixed clustering center remains unchanged; calculating the average offset of all the initial clustering centers before and after the update; cyclically dividing the clustering cluster, updating the initial clustering center, and calculating the average offset; in response to the average offset being less than a preset offset, stopping the cycle to obtain a plurality of clustering clusters, the number of the plurality of clustering clusters being equal to the preset clustering cluster number.

[0009] In some embodiments, the method for calculating the error between the curved curve and the standard curve comprises: multiple times of translating the curved curve, and after each translation, calculating the curve distance between the curved curve and the standard curve; and taking the minimum value of the curve distance as the error between the curved curve and the standard curve.

[0010] In some embodiments, the calculating the curve distance between the curved curve and the standard curve comprises: calculating distances between each drawdown value point in the curved curve and the drawdown turning point; calculating a confidence degree of each drawdown value point in the curved curve based on the distances, wherein the confidence degree is negatively correlated with the distance between the drawdown value point and the drawdown turning point; and calculating the curve distance between the curved curve and the standard curve based on the confidence degrees of the drawdown value points.

[0011] In some embodiments, the curve distance satisfies a relationship:

[0012]

[0013] wherein, is a number of drawdown value points in the curved curve is the ith drawdown value point in the curved curve is a standard curve, is a distance from the drawdown value point to the standard curve Q, is a confidence degree of the ith drawdown value point, and is the curve distance between the curved curve and the standard curve Q, wherein the confidence degree of the ith drawdown value point satisfies a relationship:

[0014]

[0015] wherein, and are distances between the ith drawdown value point and the jth drawdown value point and the drawdown turning point, respectively.

[0016] In some embodiments, the observation is performed on at least two observation holes, and the safety warning based on the error comprises: drawing drawdown value points of the observation holes to obtain a drawdown change curve corresponding to each observation hole; determining a curved curve corresponding to each drawdown change curve to calculate a corresponding error; performing weighted summation on all errors to obtain a comprehensive error, wherein the error corresponds to the observation hole one by one; and comparing the comprehensive error with a preset error threshold, and issuing a warning in response to the comprehensive error being greater than the preset error threshold.

[0017] In some embodiments, the performing weighted summation on all errors to obtain a comprehensive error comprises: calculating distances between the observation holes and the pumping hole; and assigning a weight coefficient to each error, wherein the weight coefficient satisfies a relationship:

[0018]

[0019] wherein, and​ Let K be the distance between observation well k and observation well n and the pumping well, respectively, and K be the total number of observation wells. The weighting coefficient is used to determine the error corresponding to observation hole k; based on the weighting coefficient, all errors are summed in a weighted manner to obtain the comprehensive error.

[0020] Secondly, the present invention also provides a safety early warning system for hydrogeological exploration scenarios, comprising: a processor; and a memory storing computer instructions for safety early warning in hydrogeological exploration scenarios, wherein when the computer instructions are executed by the processor, the safety early warning method for hydrogeological exploration scenarios is implemented.

[0021] The safety early warning system provided in this application embodiment, applied to hydrogeological exploration scenarios, plots the drawdown curve corresponding to the observation well during pumping tests. Using the drawdown inflection point in this curve as a fixed cluster center, it performs cluster analysis on all drawdown value points, adaptively and accurately determining the curved section in the central region, avoiding errors caused by manual designation. During the cluster analysis, the fixed cluster center is not updated, reducing the computational load and ensuring the timeliness of the safety early warning.

[0022] Furthermore, when calculating the error between the curved curve and the standard curve, the confidence level of each drop point in the curved curve is first assigned based on the distance between each drop point and the drop inflection point. Then, the confidence level of each drop point is used in the error calculation, which improves the accuracy of the error calculation and thus improves the accuracy of the safety warning.

[0023] Furthermore, during the pumping test, at least two observation wells are observed, and the drawdown curve corresponding to each observation well is plotted. The error corresponding to each observation well is then calculated. A weighting coefficient is assigned to each error based on the distance between the observation well and the pumping well, and all errors are weighted and summed to obtain the comprehensive error. The final safety warning result obtained based on the comprehensive error avoids false warnings caused by data observation errors of a single observation well, thereby improving the accuracy of the safety warning. Attached Figure Description

[0024] The above and other objects, features, and advantages of exemplary embodiments of this application will become readily understood by reading the following detailed description with reference to the accompanying drawings. In the drawings, several embodiments of this application are illustrated by way of example and not limitation, and the same or corresponding reference numerals denote the same or corresponding parts, wherein:

[0025] Figure 1 This is a flowchart of a safety early warning method applied in a hydrogeological exploration scenario according to an embodiment of this application;

[0026] Figure 2 is a flow chart of the cluster analysis according to the embodiments of the present application;

[0027] Figure 3 is a schematic block diagram of a safety warning system applied to a hydrogeological exploration scene according to an embodiment of the present application. DETAILED DESCRIPTION

[0028] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0029] It should be understood that when the terms "first", "second", etc. are used in the claims, the specification and the drawings of the present application, they are only used to distinguish different objects, and are not used to describe a specific sequence. The terms "comprise" and "include" used in the specification and claims of the present application indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or sets thereof.

[0030] According to a first aspect of the present application, the present application provides a safety warning method applied to a hydrogeological exploration scene, applied to the process of pumping test. In the process of pumping test, usually one pumping hole and at least one observation hole are set, wherein the distances between different observation holes and the pumping hole are different. Please refer to Figure 1 the flow chart of the safety warning method applied to the hydrogeological exploration scene provided by the preferred embodiments of the present application. According to different needs, the order of steps in the flow chart can be changed, and some steps can be omitted.

[0031] S11, in the process of pumping test, the drawdown value points of the observation hole are plotted to obtain the corresponding drawdown change curve, wherein the drawdown change curve includes a rapid growth segment and a slow growth segment, and a drawdown turning point between the rapid growth segment and the slow growth segment.

[0032] In an optional embodiment, multiple collection time points are set in the process of pumping test, and the time interval between adjacent collection time points gradually increases as the pumping test proceeds, that is, the collection time points are distributed from dense to sparse. Exemplarily, the collection time points in the process of pumping test can be set as: 1 minute, 3 minutes, 5 minutes, 10 minutes, 20 minutes, 30 minutes, 45 minutes, 60 minutes, 75 minutes after the start of pumping test.

[0033] The drawdown value of any observation hole is obtained at each collection time, and the drawdown value refers to the drawdown depth of the water level between the current collection time and the adjacent collection time.

[0034] In the early stage of the pumping test, the drawdown value changes rapidly, the slope of the curve gradually increases, and corresponds to the rapid growth section of the curve. With the progress of the pumping test, the observed water level in the observation hole gradually approaches the groundwater level line, the drawdown value between adjacent collection times gradually tends to be stable, and the slope of the curve gradually decreases, which corresponds to the slow growth section of the curve. Therefore, the drawdown turning point is the drawdown value point where the change trend changes from gradually increasing to gradually decreasing, which is an important node in the drawdown change curve. Therefore, the method for obtaining the drawdown turning point in the drawdown change curve is introduced in detail, which includes: calculating the difference between the drawdown value between the current collection time and the adjacent collection time above the current collection time as the drawdown value change of the current collection time; calculating the difference between the drawdown value change of the current collection time and the adjacent collection time above the current collection time as the drawdown value change increment; if the drawdown value change increment is less than 0, the drawdown value point corresponding to the current collection time is taken as the drawdown turning point; if the drawdown value change increment is greater than or equal to 0, the drawdown value change increment of the next adjacent collection time of the current collection time is calculated until the drawdown turning point of the drawdown change curve is obtained.

[0035] In other optional embodiments, the drawdown turning point in each drawdown change curve can also be directly specified by a person.

[0036] In this way, the drawdown change curve of each observation hole is obtained, and the drawdown turning point of each drawdown change curve is determined, which provides a data basis for subsequent acquisition of the bending curve of the drawdown change curve.

[0037] S12, taking the drawdown turning point as a fixed clustering center, performing clustering analysis on the drawdown value points to determine the bending curve of the middle region of the drawdown change curve, wherein the drawdown value change of the bending curve shows a trend of first increasing and then decreasing.

[0038] In one embodiment, the depth drop turning points are taken as fixed clustering centers, and the depth drop value points are subjected to clustering analysis to obtain at least two clustering clusters; and in the depth drop change curve, a curve section between all depth value points in the clustering cluster corresponding to the fixed clustering center is taken as a bending curve of the depth drop change curve.

[0039] In the bending curve, there is at least one depth drop value point, and the number of the depth drop value points in the bending curve is negatively correlated with the number of the clustering clusters. The bending curve is a part of the depth drop change curve in which the depth drop value change trend is relatively stable and the change characteristics are obvious, and the error can be improved in accuracy by calculating the error according to the bending curve.

[0040] In an optional embodiment, the positions of the fixed clustering centers are fixed and unchanged in the clustering analysis process, and only the initial clustering centers are updated. Please refer to Figure 2 The clustering analysis process of the embodiment of the application is shown in the flowchart. The depth drop turning points are taken as fixed clustering centers, and the depth drop value points are subjected to clustering analysis to obtain at least two clustering clusters, which includes: setting a preset clustering cluster number, wherein the preset clustering cluster number is greater than or equal to 2; taking the depth drop turning points as fixed clustering centers, and randomly selecting at least one initial clustering center from the depth drop value points except the depth drop turning points, wherein the sum of the number of the initial clustering centers and the fixed clustering centers is equal to the preset clustering cluster number; for each depth drop value point, calculating the distance between the depth drop value point and each clustering center, and dividing the depth drop value point into the clustering cluster corresponding to the nearest clustering center, wherein the clustering centers include the initial clustering centers and the fixed clustering centers, and each clustering center corresponds to a clustering cluster; judging whether the clustering center is a fixed clustering center or an initial clustering center, and in response to the clustering center being an initial clustering center, updating the clustering center; in response to the clustering center being a fixed clustering center, not updating the clustering center; calculating the average offset of all clustering centers before and after updating; cyclically dividing the clustering clusters, updating the initial clustering centers, and calculating the average offset; and in response to the average offset being less than a preset offset, stopping the cycle to obtain a plurality of clustering clusters.

[0041] The number of the plurality of clustering clusters is equal to the preset clustering cluster number; the preset offset is 1; and the distance between the depth drop value point and the clustering center is the Euclidean distance between two points.

[0042] The updated clustering centers are all initial clustering centers, and the updating process of the clustering center includes: taking the mean value of all depth drop value points in the clustering cluster corresponding to the clustering center as the updated initial clustering center. The updated initial clustering center satisfies the relationship:

[0043]

[0044] wherein, denotes the number of depth value points in the cluster corresponding to the initial cluster center C, and denote the collection time and the depth value of the depth value point a in the cluster corresponding to the initial cluster center C, is the updated initial cluster center corresponding to the initial cluster center C.

[0045] wherein, for each cluster center, the Euclidean distance of the cluster center before and after the update is calculated as the offset distance of the cluster center, and the average of the offset distances of all cluster centers in each update process is taken as the average offset. It can be understood that the offset distance of the fixed cluster center is 0 in each update process.

[0046] In this way, by setting the fixed cluster center and the cluster analysis, the bending curve of the set region in the middle part of each depth variation curve is adaptively obtained.

[0047] S13, calculating the error between the bending curve and the standard curve.

[0048] In an optional embodiment, the calculation of the error between the bending curve and the standard curve comprises: translating the bending curve multiple times, and after each translation, calculating the curve distance between the bending curve and the standard curve; taking the minimum value of the curve distance as the error between the bending curve and the standard curve.

[0049] wherein, the bending curve comprises at least one depth value point, the process of translating the bending curve is the process of translating all depth value points in the bending curve along the horizontal and vertical coordinates, and the bending curve after translation is denoted as wherein, is the translation amount of the bending curve along the horizontal coordinate, is the translation amount of the bending curve along the horizontal coordinate.

[0050] In an application scenario, the depth turning point is the center point of the bending curve, and the depth value change feature reflected by the depth value point closer to the depth turning point is more obvious, so in order to improve the accuracy of error calculation, different confidence levels are assigned to the depth value points in the bending curve, which is described in detail as follows. The calculation of the curve distance between the bending curve and the standard curve comprises: calculating the distance between each depth value point in the bending curve and the depth turning point; calculating the confidence level of each depth value point in the bending curve based on the distance, wherein the confidence level is negatively correlated with the distance between the depth value point and the depth turning point; calculating the curve distance between the bending curve and the standard curve based on the confidence level of each depth value point.

[0051] wherein, taking the i-th drawdown value point in the curved curve as an example, the confidence of the drawdown value point satisfies the relationship:

[0052]

[0053] wherein, and are distances between the i-th drawdown value point and the j-th drawdown value point in the curved curve and the drawdown turning point, respectively, Num is a number of all drawdown value points in the curved curve, is the confidence of the i-th drawdown value point.

[0054] wherein, the curve distance satisfies the relationship:

[0055]

[0056] wherein, is a number of drawdown value points in the curved curve, is the i-th drawdown value point in the curved curve, Q is a standard curve, is a distance from the drawdown value point to the standard curve Q, is the confidence of the i-th drawdown value point, is a curve distance between the curved curve and the standard curve Q. The translation process of the curved curve is constrained by using an optimization algorithm, constantly changing and , until the minimum value of the curve distance is obtained, and the minimum value of the curve distance is taken as the error between the curved curve and the standard curve. The optimization algorithm can be an existing optimization algorithm such as simulated annealing algorithm, ant colony algorithm, whale algorithm, etc., which is not limited in the present application. In this way, the curved curve with relatively stable drawdown value change trend and obvious change characteristics in the drawdown change curve is used for error calculation, thereby improving the accuracy of error calculation. S14, performing safety warning based on the error.

[0057] In an optional embodiment, one or more observation holes are observed, a drawdown change curve of each observation hole is drawn to calculate errors corresponding to the observation holes, and the errors corresponding to the observation holes are directly compared with a preset error threshold. If at least one error is greater than the preset error threshold, a warning is issued. The preset error threshold is 0.3.

[0058] S14, performing safety warning based on the error.

[0059] In an optional embodiment, one or more observation holes are observed, a drawdown change curve of each observation hole is drawn to calculate errors corresponding to the observation holes, and the errors corresponding to the observation holes are directly compared with a preset error threshold. If at least one error is greater than the preset error threshold, a warning is issued. The preset error threshold is 0.3.

[0060] ​​​In another optional embodiment, the at least two observation holes are observed, and the safety warning based on the error comprises: drawing the drawdown value points of the observation holes to obtain a drawdown change curve corresponding to each observation hole; determining a bending curve corresponding to each drawdown change curve to calculate a corresponding error; performing weighted summation on all errors to obtain a comprehensive error, wherein the error corresponds to the observation hole one by one; and comparing the comprehensive error with a preset error threshold, and in response to the comprehensive error being greater than the preset error threshold, issuing a warning. The preset error threshold is 0.3.

[0061] wherein the weighted summation on all errors to obtain a comprehensive error comprises: calculating the distance between the observation hole and the pumping hole, and assigning a weight coefficient to each error, wherein the weight coefficient satisfies the relationship:

[0062]

[0063] wherein, and respectively, are the distances between the observation hole k and the observation hole n and the pumping hole, K is the number of all observation holes, is the weight coefficient of the error corresponding to the observation hole k; the weight coefficient of each drawdown change curve is calculated in the same way, the value range of the weight coefficient is [0, 1], and the sum of the weight coefficients of all drawdown change curves is 1; based on the weight coefficient, the weighted summation is performed on all errors, and the comprehensive error is obtained. When the comprehensive error is greater than the preset error threshold, a warning is issued to remind the staff to standardize the operation; when the comprehensive error is less than or equal to the preset error threshold, it indicates that the pumping test is in a normal operation state.

[0064] In this way, the safety warning is realized according to the error of the drawdown change curve corresponding to each observation hole in the pumping test process, the pumping test is ensured to be carried out normally, and the accuracy of the hydrogeological parameters is ensured, thereby reducing the safety hidden danger in the subsequent mining operation.

[0065] The safety warning system provided by the embodiment of the application is applied to the hydrogeological exploration scene, the drawdown change curve corresponding to the observation hole is drawn in the pumping test, the drawdown turning point in the drawdown change curve is taken as a fixed clustering center, clustering analysis is performed on all drawdown value points, the bending curve part of the middle region is accurately determined adaptively, and errors caused by manual designation are avoided; in the clustering analysis process, the fixed clustering center is not updated, the clustering analysis calculation amount is reduced, and the timeliness of the safety warning is ensured.

[0066] According to a second aspect of the present application, the present application also provides a safety warning system applied to a hydrogeological exploration scene, comprising a memory and a processor, the memory stores computer executable instructions, the computer executable instructions, when executed by the processor, implement the safety warning method according to the first aspect of the present application.

[0067] Figure 3 is a schematic block diagram of a safety warning system applied to a hydrogeological exploration scene according to an embodiment of the present application. The device 40 comprises a processor and a memory, the memory stores computer program instructions, when the computer program instructions are executed by the processor, a safety warning method applied to a hydrogeological exploration scene according to the first aspect of the present application is implemented. The device also includes a communication bus and a communication interface and other components familiar to those skilled in the art, the settings and functions of which are known in the art, and therefore will not be described here.

[0068] In the present application, the aforementioned readable memory can be any tangible medium containing or storing programs, which can be used or combined with an instruction execution system, device or instrument. For example, the computer readable storage medium can be any appropriate magnetic storage medium or magneto-optical storage medium, such as resistive random access memory RRAM (Resistive Random Access Memory), dynamic random access memory DRAM (Dynamic Random Access Memory), static random access memory SRAM (Static Random-Access Memory), enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), high bandwidth memory HBM (High-Bandwidth Memory), hybrid memory cube HMC (Hybrid Memory Cube) and the like, or any other medium that can be used to store the required information and can be accessed by an application, module or both. Any such computer storage medium can be part of the device or accessible or connectable to the device. Any application or module described in the present application can be implemented using computer readable / executable instructions that can be stored or otherwise held by such computer readable medium.

[0069] The technical features of the above-described embodiments can be combined in any manner. In order to make the description concise, all possible combinations of the technical features in the above-described embodiments are not described, however, as long as the combinations of the technical features do not exist contradictions, it should be considered that they are within the scope of the present application.

[0070] The above-described embodiments are merely illustrative of several embodiments of the present application, which are described in more detail and in a specific manner, but should not be construed as limiting the scope of the patent application. It should be noted that for those of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A safety early warning method applied in hydrogeological exploration scenarios, characterized in that: During the pumping test, the drawdown values ​​of the observation wells are plotted to obtain the corresponding drawdown curves. The drawdown curves include a rapid growth segment and a slow growth segment, as well as the drawdown inflection point between the rapid growth segment and the slow growth segment. Using the inflection point of the depth drop as a fixed cluster center, cluster analysis is performed on the depth drop value points to determine the curved curve in the middle region of the depth drop change curve. Among them, the depth drop value change of the curved curve shows a trend of first increasing and then decreasing. Calculate the error between the bending curve and the standard curve; Safety warnings based on errors; Using the inflection point of depth drop as a fixed cluster center, cluster analysis is performed on the depth drop value points to determine the curved curve in the middle region of the depth drop change curve. This includes: using the inflection point of depth drop as a fixed cluster center, performing cluster analysis on the depth drop value points to obtain at least two clusters; in the depth drop change curve, the curved portion between all depth value points in the cluster corresponding to the fixed cluster center is taken as the curved curve of the depth drop change curve. The curved curve includes at least one point of depth of drop, and the number of points of depth of drop in the curved curve is negatively correlated with the number of clusters; the curved curve is the part of the depth of drop change curve with a relatively stable trend and obvious change characteristics. Using the inflection point of depth descent as the fixed cluster center, cluster analysis is performed on the depth descent points to obtain at least two clusters. This includes: setting a preset number of clusters, which is greater than or equal to 2; using the inflection point of depth descent as the fixed cluster center, randomly selecting at least one initial cluster center from the depth descent points other than the inflection point, wherein the sum of the number of initial cluster centers and the number of fixed cluster centers equals the preset number of clusters; for each depth descent point, calculating the distance between the depth descent point and each cluster center, and assigning the depth descent point to the cluster corresponding to the nearest cluster center. Cluster centers include initial cluster centers and fixed cluster centers; each cluster center corresponds to a cluster; for the cluster corresponding to the initial cluster center, the mean of all descent points is used as the updated initial cluster center; for the cluster corresponding to the fixed cluster center, the fixed cluster center remains unchanged; the average offset of all initial cluster centers before and after the update is calculated; the clusters are cyclically divided, the initial cluster centers are updated, and the average offset is calculated; in response to the average offset being less than a preset offset, the loop stops, resulting in multiple clusters, the number of which is equal to the preset number of clusters.

2. The safety early warning method applied in hydrogeological exploration scenarios as described in claim 1, characterized in that, The method for obtaining the inflection point in the depth change curve includes: Calculate the difference in depth of view between the current acquisition time and the previous adjacent acquisition time as the change in depth of view at the current acquisition time; Calculate the difference between the current acquisition time and the previous adjacent acquisition time as the increment of the depth change; If the increment of the change in the depth value is less than 0, then the depth value point corresponding to the current acquisition time is taken as the depth inflection point. If the increment of the depth change is greater than or equal to 0, then calculate the increment of the depth change at the next adjacent acquisition time from the current acquisition time until the inflection point of the depth change curve is obtained.

3. The safety early warning method applied in hydrogeological exploration scenarios as described in claim 1, characterized in that, The calculation of the error between the bending curve and the standard curve includes: The curved curve is translated multiple times, and the distance between the curved curve and the standard curve is calculated after each translation. The minimum value of the curve distance is taken as the error between the curved curve and the standard curve.

4. The safety early warning method applied in hydrogeological exploration scenarios as described in claim 3, characterized in that, The calculation of the curve distance between the curved curve and the standard curve includes: Calculate the distance between each depth point in the curved curve and the depth inflection point; The confidence level of each depth point in the curvature is calculated based on the distance, wherein the confidence level is negatively correlated with the distance between the depth point and the depth inflection point; The curve distance between the curved curve and the standard curve is calculated based on the confidence level of each drawdown point.

5. The safety early warning method for hydrogeological exploration scenarios as described in claim 4, characterized in that, The distance between the curves satisfies the following relationship: in, The curved curve The number of points with moderate depth of precipitation For curved curves The i-th point of descent, where Q is the standard curve. For the depth value point The distance to the standard curve Q, Let i be the confidence level for the i-th descent value point. For curved curves The curve distance between the curve and the standard curve Q, where the confidence level of the i-th depth point. Satisfying the relation: in, and These are the distances from the i-th and j-th depth points to the depth inflection point, respectively. The curved curve The number of points with moderate depth.

6. The safety early warning method for hydrogeological exploration scenarios as described in claim 1, characterized in that, Observations are made through at least two observation holes, and the safety warning based on the error includes: Plot the drawdown points of the observation holes to obtain the drawdown variation curve for each observation hole; Determine the curvature corresponding to each drawdown curve to calculate the corresponding error; All errors are weighted and summed to obtain a comprehensive error, wherein each error corresponds one-to-one with an observation hole; The overall error is compared with a preset error threshold, and an early warning is issued in response to the overall error being greater than the preset error threshold.

7. A safety early warning method for hydrogeological exploration scenarios as described in claim 6, characterized in that, The weighted summation of all errors to obtain the comprehensive error includes: Calculate the distance between the observation hole and the pumping hole, and assign a weighting coefficient to each error, wherein the weighting coefficient satisfies the following relationship: in, and Let K be the distance between observation well k and observation well n and the pumping well, respectively, and K be the total number of observation wells. The weighting coefficient for the error corresponding to observation hole k; The combined error is obtained by weighting and summing all errors based on the weighting coefficients.

8. A safety early warning system applied in hydrogeological exploration scenarios, characterized in that, The system includes: Processor; and A memory storing computer instructions for safety warning in hydrogeological exploration scenarios, which, when executed by the processor, implement the safety warning method according to any one of claims 1-7.

Images