Regulation and control method for mining and irrigation of thermal reservoir of hydraulic ring liquid mining place

Abstract

The invention discloses a hydraulic ring liquid mining place thermal reservoir mining and irrigation regulation and control method, and relates to the technical field of mining and irrigation regulation and control, the method comprises the following steps: dividing a target geothermal reservoir into an upper water level space area and a lower water level space area, dividing the lower water level space area into four equal subspace areas, and dividing the subspace areas into four equal subspace areas; according to the obtained temperature, seepage and pressure data difference characteristics of each sub-region in the first lower water level sub-region set, differential marking of a first unstable marking sub-region and a first stable marking sub-region is carried out on the first lower water level sub-region set, a mining and irrigation position point is selected for mining and irrigation speed regulation and control, and a first mining and irrigation regulation and control result is obtained; according to the upper water level crack characteristic data to which the upper water level space area belongs, marking a second unstable sub-area and a second stable sub-area on a second lower water level sub-area set obtained through re-division, and selecting a mining and irrigation position point for mining and irrigation speed regulation and control to obtain a second mining and irrigation regulation and control result. The method can improve the extraction and irrigation regulation and control efficiency.

Description

Technical Field

[0001] This application relates to the field of extraction and irrigation regulation technology, and in particular to a method for extraction and irrigation regulation of hydrogeothermal reservoirs of liquid mineral deposits. Background Technology

[0002] By employing extraction-injection regulation technology, the water volume of geothermal reservoirs can be more accurately assessed and managed, ensuring a long-term stable energy supply and preventing resource depletion caused by over-exploitation. A rational extraction-injection strategy helps reduce environmental problems such as land subsidence and water pollution caused by extraction activities, thus protecting the ecological environment. This is of great significance for maintaining ecological balance and promoting sustainable economic and social development.

[0003] In traditional technologies, the detection of temperature, seepage, and pressure fields within the geothermal reservoir space during the extraction and irrigation control process is often carried out through integrated data detection. However, data dilution occurs in some blind areas, causing interference with the data in these blind areas. This results in deviations in the final detected integrated data, reducing the efficiency of extraction and irrigation control work and the timeliness of early warning. Summary of the Invention

[0004] To overcome the shortcomings of the prior art, this application provides a method for regulating the extraction and irrigation of hydrogeothermal reservoirs of liquid mineral deposits.

[0005] This application provides a method for regulating the production and irrigation of hydraulic-environmental liquid mineral geothermal reservoirs, the method comprising:

[0006] Step S1: Divide the target geothermal reservoir that needs resource extraction and irrigation into upper water level spatial region and lower water level spatial region, and divide the lower water level spatial region into four equal sub-regions to obtain the first lower water level sub-region set. Based on the differences in temperature, seepage and pressure data of each sub-region in the first lower water level sub-region set, distinguish the first non-stable labeled sub-region and the first stable labeled sub-region in the first lower water level sub-region set.

[0007] Step S2: The extraction and irrigation speed is controlled by using the boundary between every two adjacent sub-regions in the first unstable marked sub-region as the control point for re-irrigation and the boundary between every two adjacent sub-regions in the first stable marked sub-region as the control point for extraction, to obtain the first extraction and irrigation control result.

[0008] Step S3: Based on the fluctuating water levels between the upper and lower water level spatial regions and the upper water level crack feature data of the upper water level spatial region, preset new fluctuating crack feature data is performed, and the lower water level spatial region is re-divided to obtain a second lower water level sub-region set. The second lower water level sub-region set is marked with unstable and stable sub-regions to obtain a second unstable marked sub-region and a second stable marked sub-region. The overlapping area of ​​the first unstable marked sub-region and the second unstable marked sub-region is used as the control recharge location point, and the first stable marked sub-region and the second stable marked sub-region are used as the control extraction location point to control the extraction and irrigation speed, thereby obtaining the second extraction and irrigation control result.

[0009] Preferably, preset initial extraction characteristic parameters and initial reinjection characteristic parameters are obtained, and the reservoir spatial volume of the target geothermal reservoir requiring resource extraction and injection is detected to obtain spatial regional characteristics. The spatial regional characteristics are divided into upper water level spatial region and lower water level spatial region. The initial extraction characteristic parameters and initial reinjection characteristic parameters include extraction and reinjection speeds and extraction and reinjection location points.

[0010] If the velocity parameters in the initial extraction characteristic parameters and the initial reinjection characteristic parameters are the same, then according to the stable water level between the upper water level space region and the lower water level space region, the fracture characteristic data on the reservoir sidewall in the upper water level space region and the lower water level space region are detected respectively to obtain the upper water level fracture characteristic data and the lower water level fracture characteristic data.

[0011] The crack inclination in the crack feature data of the lower water level is statistically analyzed in a centralized manner to obtain a first centralized crack inclination dataset. Based on the first centralized crack inclination dataset, the lower water level spatial region is divided into four equal sub-regions to obtain a first lower water level sub-region set.

[0012] Preferably, the first feature dataset is obtained by real-time detection of temperature, seepage, and pressure data in each sub-region of the first sub-region of the water level sub-region.

[0013] The first comprehensive comparison difference value is obtained by comparing and calculating the difference between the test feature datasets of two adjacent sub-regions in the first water level sub-region.

[0014] A preset warning judgment threshold is set. If the difference between the first comprehensive comparison is greater than or equal to the warning judgment threshold, the sub-regions corresponding to the first water level sub-regions are marked as unstable regions to obtain the first unstable marked sub-regions.

[0015] If the difference between the first comprehensive comparison and the warning judgment threshold is less than the threshold, the sub-regions in the corresponding first water level sub-regions will be marked as stable regions to obtain the first stable marked sub-regions.

[0016] Preferably, the boundary between every two adjacent sub-regions in the first unstable marked sub-region is taken as the control recharge location point, and the boundary between every two adjacent sub-regions in the first stable marked sub-region is taken as the control extraction location point, so as to obtain the control result of the first location point;

[0017] Historical irrigation and extraction characteristic data are obtained, and a matching correspondence table between irrigation and extraction rates and reservoir variation data is constructed based on the historical irrigation and extraction characteristic data to obtain a regulation matching database;

[0018] Based on the comprehensive data difference between adjacent sub-regions in the first unstable marked sub-region and the stable marked sub-region, the corresponding regulation and re-irrigation speed is matched from the regulation and matching database. Based on the comprehensive data difference between adjacent sub-regions in the first stable marked sub-region, the corresponding regulation and re-irrigation speed is matched from the regulation and matching database to obtain the first sampling and irrigation speed regulation result.

[0019] The first position point control result and the first irrigation speed control result are combined to form the first irrigation control result.

[0020] Preferably, if the velocity parameters in the initial extraction characteristic parameters and the initial reinjection characteristic parameters are different, then based on the changing water level between the upper water level space region and the lower water level space region, the characteristic data of the overburden fractures and the characteristic data of the lower water level fractures on the reservoir sidewall in the newly added upper water level space region in the lower water level space region are summarized to obtain the first newly added changing fracture characteristic data.

[0021] The crack inclination in the first newly added variable crack feature data is statistically analyzed to obtain a second centralized crack inclination dataset. Based on the second centralized crack inclination dataset, the water level spatial region is divided into four equal sub-regions to obtain the sub-region set of the water level to be measured.

[0022] Real-time detection of temperature, seepage, and pressure data in each sub-region of the sub-region of the target water level is used to obtain the second target feature dataset.

[0023] Preferably, the difference between the second set of crack tilt datasets and the first set of crack tilt datasets is calculated to obtain feature data training set one, and the difference between the second set of feature datasets to be tested and the first set of feature datasets to be tested is calculated to obtain feature data training set two.

[0024] The correlation and change influence trend characteristics between the feature data training set 1 and feature data training set 2 are statistically analyzed to obtain the data correlation influence characteristic coefficient between feature data training set 1 and feature data training set 2;

[0025] Based on the upper water level crack feature data, a second newly added variable crack feature data is preset, and the crack inclination in the second newly added variable crack feature data is centrally statistically analyzed to obtain a third centralized crack inclination dataset. Based on the third centralized crack inclination dataset, the lower water level spatial region is divided into four equal sub-spatial regions to obtain a second lower water level sub-region set.

[0026] Based on the crack tilt dataset and data association influence characteristic coefficients in the third set, the temperature, seepage, and pressure prediction data of each sub-region in the second lower water level sub-region set are obtained, and the preprocessed feature prediction dataset is output.

[0027] Preferably, based on the first comprehensive comparison difference, the second comprehensive comparison difference to which the preprocessed feature prediction dataset belongs is calculated, and based on the first unstable labeled sub-region and the first stable labeled sub-region, the second unstable labeled sub-region and the second stable labeled sub-region to which the second water level sub-region belongs are labeled.

[0028] The overlapping area between the second unstable labeled sub-region and the first unstable labeled sub-region is extracted as the control and reinjection location point, and the overlapping area between the second stable labeled sub-region and the first stable labeled sub-region is extracted as the control and extraction location point, thus obtaining the control result of the second location point;

[0029] Based on the first irrigation and drainage rate regulation results, the preprocessed feature prediction dataset, and the regulation matching database, the second irrigation and drainage rate regulation results are obtained.

[0030] The second position point control result and the second sampling and irrigation speed control result are combined to form the second sampling and irrigation control result.

[0031] Compared with the prior art, the present invention has the following characteristics and beneficial effects:

[0032] By dividing the target geothermal reservoir into upper and lower water level spatial regions, preliminary real-time monitoring of temperature, seepage, and pressure data at the upper water level is conducted to facilitate subsequent early warning and adjustment of extraction and irrigation parameters. For example, predictive trend analysis is performed based on the changing trends of various data within the upper water level spatial region. Two scenarios are analyzed for data trend characteristics. The first scenario involves real-time monitoring of various data when the water level between the upper and lower water level spatial regions is stable under initially preset extraction and irrigation characteristic parameters. To avoid data dilution errors caused by the uniform monitoring of all data in traditional technologies, and to reduce missed detections in blind spots, the upper water level spatial region is divided into sub-regions to expand the detection range and improve data coverage. For example, the upper water level spatial region is divided into four... The system first divides the area into equal sub-regions, then compares the temperature, seepage, and pressure data of adjacent sub-regions to determine if any discrepancies are abnormal, and implements timely early warning and control measures. The second approach is when the water level between the upper and lower water level areas is fluctuating. In this case, the system needs to add and summarize the crack characteristic data of the lower water level area to re-divide the upper water level area into four equal sub-regions. Based on the comparison of data from each sub-region after the re-division, the system determines the locations for controlled extraction and recharge. The extraction and recharge rates are then adjusted based on the comprehensive data differences between adjacent sub-regions. These diverse extraction and recharge control methods improve the efficiency and timeliness of the extraction and recharge control work. Attached Figure Description

[0033] Figure 1 This is a flowchart illustrating the steps of a method for regulating the extraction and irrigation of hydrogeological liquid mineral reservoirs, which is the main feature of this embodiment. Detailed Implementation

[0034] The present invention will be further described in detail below with reference to the following embodiments.

[0035] Reference Figure 1 A method for regulating the production and irrigation of hydrogeothermal reservoirs in the hydraulic ring type, comprising the following steps:

[0036] Step S1: Divide the target geothermal reservoir requiring resource extraction and irrigation into upper water level spatial regions and lower water level spatial regions, and further divide the lower water level spatial region into four equal sub-regions to obtain the first lower water level sub-region set. Based on the differences in temperature, seepage, and pressure data of each sub-region in the first lower water level sub-region set, distinguish the first unstable labeled sub-region and the first stable labeled sub-region in the first lower water level sub-region set.

[0037] Step S2: The extraction and irrigation speed is controlled by using the boundary between every two adjacent sub-regions in the first unstable marked sub-region as the control point for re-irrigation and the boundary between every two adjacent sub-regions in the first stable marked sub-region as the control point for extraction, to obtain the first extraction and irrigation control result.

[0038] Step S3: Based on the fluctuating water levels between the upper and lower water level spatial regions and the upper water level crack characteristic data of the upper water level spatial region, preset new fluctuating crack characteristic data are made, and the lower water level spatial region is re-divided to obtain a second lower water level sub-region set. The second lower water level sub-region set is marked with unstable and stable sub-regions to obtain a second unstable marked sub-region and a second stable marked sub-region. The overlapping area of ​​the first unstable marked region and the second unstable marked region is used as the control and recharge location point, and the first stable marked region and the second stable marked region are used as the control and extraction location point to control the extraction and recharge rate, thereby obtaining the second extraction and recharge control result.

[0039] Specifically, by dividing the target geothermal reservoir into upper and lower water level spatial regions, preliminary real-time monitoring of temperature, seepage, and pressure data at the lower water level is conducted to facilitate subsequent early warning and adjustment of extraction and irrigation parameters. For example, predictive trend analysis is performed based on the changing trends of various data within the lower water level spatial region. Two scenarios are analyzed for data trend characteristics: First, under initially preset extraction and irrigation characteristic parameters, when the water level between the upper and lower water level spatial regions is stable, real-time monitoring of various data is performed. To avoid the data dilution error problem inherent in traditional unified data monitoring, i.e., to reduce missed detections in blind spots, the lower water level spatial region is divided into sub-regions to expand the detection range and improve data coverage. For example, the lower water level spatial region is further subdivided into... The system divides the area into four equal sub-regions. Then, it compares the temperature, seepage, and pressure data of adjacent sub-regions to determine if any discrepancies are abnormal and implements timely early warning and control measures. The second scenario involves a fluctuating water level between the upper and lower water level regions. In this case, the upper water level crack characteristic data of the upper water level region is used to add and summarize crack characteristic data, allowing for a re-division of the lower water level region into four equal sub-regions. Based on the comparison of data within each sub-region after the re-division, the locations for controlled extraction and recharge are determined. The extraction and recharge rates are then adjusted based on the comprehensive data differences between adjacent sub-regions. These diverse extraction and recharge control methods improve the efficiency and timeliness of early warning in extraction and recharge control work.

[0040] The specific step S1 includes the following sub-steps:

[0041] The preset initial extraction characteristic parameters and initial reinjection characteristic parameters are obtained. The reservoir spatial volume of the target geothermal reservoir that needs resource extraction and injection is detected to obtain spatial regional characteristics. The spatial regional characteristics are divided into upper water level spatial region and lower water level spatial region. The initial extraction characteristic parameters and initial reinjection characteristic parameters include extraction and reinjection rates as well as extraction and reinjection location points.

[0042] If the velocity parameters in the initial extraction characteristic parameters and the initial reinjection characteristic parameters are the same, then based on the stable water levels between the upper and lower water level spatial regions, the fracture characteristic data on the reservoir sidewalls in the upper and lower water level spatial regions are detected respectively to obtain the upper water level fracture characteristic data and the lower water level fracture characteristic data.

[0043] A centralized statistical analysis of the crack inclination in the crack feature data of the groundwater level was performed to obtain the first centralized crack inclination dataset. Based on the first centralized crack inclination dataset, the groundwater level spatial region was divided into four equal sub-regions to obtain the first groundwater level sub-region set.

[0044] Real-time monitoring of temperature, seepage, and pressure data in each sub-region of the first sub-region of the water level is used to obtain the first dataset of features to be measured.

[0045] The first comprehensive comparison difference value is obtained by comparing and calculating the difference between the feature datasets of two adjacent sub-regions in the first water level sub-region.

[0046] A preset warning judgment threshold is set. If there is a first comprehensive comparison difference that is greater than or equal to the warning judgment threshold, the sub-regions in the corresponding first water level sub-region set are marked as unstable regions to obtain the first unstable marked sub-region.

[0047] If the difference between the first comprehensive comparison and the warning judgment threshold is less than the threshold, the sub-regions in the corresponding first water level sub-regions will be marked as stable regions to obtain the first stable marked sub-regions.

[0048] Specifically, this includes initial extraction and recharge characteristic parameters (referring to the selection of initial extraction and recharge locations and the determination of extraction and recharge rates by extraction and recharge technicians based on data such as permeability, porosity, and reservoir pressure collected from the target geothermal reservoir), spatial regional characteristics (such as using a 3D seismic exploration system: artificially stimulating seismic waves, receiving reflected wave signals from underground rock strata, and using computer processing to reconstruct the underground 3D geological structure, accurately depicting reservoir boundaries, thickness, and spatial distribution), and upper and lower water level fracture characteristic data (if the velocity parameters in the initial extraction and recharge characteristic parameters are the same, it indicates that extraction and recharge are in dynamic equilibrium, and the liquid mineral water level is stable. At this time, real-time monitoring of temperature, seepage, and pressure data in the upper water level spatial area can be performed to prioritize the lower water level spatial area). The changes in internal temperature, seepage, and pressure under the influence of extraction and irrigation are analyzed to determine whether timely extraction and irrigation control is necessary to balance the target geothermal reservoir. Since temperature, seepage, and pressure changes in the reservoir below the water level typically precede those in the caprock above the water level, extraction and irrigation control of geothermal reservoirs should prioritize timely control based on the temperature, seepage, and pressure fields in the space below the water level, rather than on the overall temperature, seepage, and pressure fields of the upper and lower water level areas, as this reduces timeliness and efficiency. Fracture characteristic data can be obtained through a microseismic monitoring system: during extraction and irrigation, surface or downhole geophone arrays are deployed to monitor microseismic events (frequency greater than 1 Hz) generated by rock fractures in real time. The spatiotemporal distribution of microseismic events can reflect the propagation direction and length of artificial fractures and their communication with natural fractures, indirectly inferring fracture connectivity and seepage paths.The crack feature data includes crack opening (width), crack inclination (angle), crack connectivity, and other feature data. The first set of crack inclination datasets (if the cracks exhibit a curved trend, the cracks are segmented into several segments (straight trend segments, such as l1, l2, l3, ln) for inclination statistics (using the vertical direction relative to the ground surface as a reference), and then a concentrated inclination angle is calculated (if l3, l4, and li are all concentrated between 25° and 35°, they are classified as concentrated inclination angles of 30°, and 30° is one type of concentrated crack inclination angle in the first set of crack inclination data, and so on) for statistical division (crack segments with continuous relationships, such as l3, l4, and li being continuous segments)). The first upper water level sub-region set (based on the first set of crack inclination datasets (statistical analysis of crack feature data on the sidewalls within the lower water level spatial region, to classify cracks in different side locations) is also included. The tilted concentrated data is divided into regions. For example, if the tilted concentrated data of connected cracks are 10°, 20°, 30°, 45°, and 60°, then four equal sub-regions are divided. If there are multiple regions with different crack concentrated data, they are merged (the condition is that the difference between the crack concentrated data of the merged regions is the smallest, and they can be merged into one sub-region). For example, the regions belonging to 10° and 20° are merged, and the concentrated data is 15°, which is divided equally with the sub-regions belonging to 30°, 45°, and 60°. If they are q1, q2, q3, and q4 respectively, then they are the first upper water level sub-region set. The first feature dataset to be measured (temperature: such as through a fiber optic grating (FBG) sensor - reflecting temperature changes through changes in the Bragg wavelength in the optical fiber, resistant to electromagnetic interference, suitable for long-term burial; seepage: such as through a pore water pressure gauge - indirectly reflecting the seepage state, and calculating the permeability coefficient by combining Darcy's law).Pressure: (e.g., via a vibrating wire pressure sensor), First comprehensive comparison difference (e.g., if q1 and q2 are adjacent, q2 and q3 are adjacent, q3 and q4 are adjacent, q1 and q4 are adjacent; if the temperature detection data of q1 is w1, the seepage coefficient is s1, and the pressure detection data is f1, and the temperature detection data of q2 is w2, the seepage coefficient is s2, and the pressure detection data is f2, then the temperature difference between q1 and q2 is w1-w2=W1, the seepage coefficient difference is s1-s2=S1, and the pressure difference is f1-f2=F1. Based on the variation amplitude of temperature, seepage, and pressure in historical detection data, the ratio W:S:F is statistically calculated. Based on W:S:F, W1, S1, and F1 are uniformly converted, and the converted values ​​are summed to obtain the first comprehensive comparison difference, and so on). Preset. The warning judgment threshold (obtained from empirical data in historical monitoring data, serving as a critical threshold for determining whether a dynamic equilibrium is reached), the first unstable marked sub-region (if the first comprehensive comparison difference is greater than or equal to the warning judgment threshold, it indicates that the sub-regions in the corresponding first upper water level sub-regions are in an unbalanced state, and adjacent sub-regions cannot achieve self-balancing adjustment, requiring external force, i.e., the control and treatment of sampling and irrigation, and the range of subsequent sampling and irrigation location point control is selected from the sub-regions in the corresponding first upper water level sub-regions, and the region is marked, thus obtaining the first unstable marked sub-region), and the first stable marked sub-region (if the first comprehensive comparison difference is less than the warning judgment threshold, it indicates that the region can achieve self-dynamic balance adjustment with the data between adjacent sub-regions, and the region is marked, thus obtaining the first stable marked sub-region).

[0049] The specific step S2 includes the following sub-steps:

[0050] The boundary between every two adjacent sub-regions in the first unstable marked sub-region is taken as the control and reinjection location point, and the boundary between every two adjacent sub-regions in the first stable marked sub-region is taken as the control and extraction location point, thus obtaining the control result of the first location point.

[0051] Historical extraction and irrigation characteristic data are obtained. Based on the historical extraction and irrigation characteristic data, a matching correspondence table between extraction and irrigation rates and reservoir variation data is constructed to obtain the regulation matching database.

[0052] Based on the comprehensive data difference between adjacent sub-regions in the first unstable marked sub-region and the stable marked sub-region, the corresponding regulation and re-irrigation speed is matched from the regulation and matching database. Based on the comprehensive data difference between adjacent sub-regions in the first stable marked sub-region, the corresponding regulation and re-irrigation speed is matched from the regulation and matching database to obtain the first sampling and irrigation speed regulation result.

[0053] The first position point control result and the first sampling and irrigation speed control result are combined to form the first sampling and irrigation control result.

[0054] Specifically, for example, the first location point control result (if the number of the first unstable marker sub-regions is one: q1, then q1 is used as the control recharge location point z1 (the center location of q1), where q2, q3, and q4 are the first stable marker sub-regions, the boundary between q2 and q3 (such as any location point on the boundary line) is used as one of the control sampling locations z2, and the boundary between a3 and q4 is used as another control sampling location z3, i.e., the first location point control result. If the first unstable... If there are two adjacent marked sub-regions, q1 and q2, then the boundary between q1 and q2 is designated as the control recharge location, and the boundary between q3 and q4 is designated as the control extraction location. If the first unstable marked sub-regions are q1 and q3, then q1 and q3 are designated as control recharge locations, and q2 and q4 are designated as control extraction locations. If there are three first unstable marked sub-regions, q1, q2, and q3, then the boundary between q1 and q2 is designated as one of the control sub-regions. The recharge location is determined by using the boundary between q2 and q3 as another control recharge location and q4 as the control extraction location. A control matching database is established (e.g., historical extraction and recharge characteristic data, including extraction and recharge control rate parameter records, if V), and reservoir variation data (e.g., comprehensive comparison difference, if B). A matching table is created between the two, i.e., the control matching database: V(v1, v2, vn) - B(b1, b2, bn)). The first extraction and recharge rate control result (i.e., based on the first comprehensive comparison difference (if...)) is used. (For example, if the number of the first unstable marker sub-regions is two and they are adjacent: q1 and q2, if the comprehensive comparison difference between q1 and q4 is bi, vi is obtained by matching V(v1, v2, vn) - B(b1, b2, bn), and the comprehensive comparison difference between q2 and q3 is br, vr is obtained. Then vi and vr are averaged to vp, and so on, to adjust the re-irrigation rate parameters of q3 and q4, which is the result of the first sampling and irrigation rate adjustment).

[0055] The specific step S3 includes the following sub-steps:

[0056] If the velocity parameters in the initial extraction characteristic parameters and the initial reinjection characteristic parameters are different, then based on the changing water levels between the upper and lower water level spatial regions, the characteristic data of the overburden fractures and the characteristic data of the lower water level fractures on the reservoir sidewalls in the newly added upper water level spatial region in the lower water level spatial region are summarized to obtain the first newly added changing fracture characteristic data.

[0057] The crack inclination in the first newly added variable crack feature data is statistically analyzed to obtain the second centralized crack inclination dataset. Based on the second centralized crack inclination dataset, the groundwater level spatial region is divided into four equal sub-regions to obtain the groundwater level sub-region set to be measured.

[0058] Real-time detection of temperature, seepage, and pressure data in each sub-region of the sub-region of the target water level is used to obtain the second target feature dataset.

[0059] The difference between the crack tilt dataset in the second set and the crack tilt dataset in the first set is used to obtain the first feature data training set. The difference between the second feature dataset to be tested and the first feature dataset to be tested is used to obtain the second feature data training set.

[0060] The correlation and change influence trend characteristics between statistical feature data training set 1 and feature data training set 2 are analyzed, and the data correlation influence feature coefficients between feature data training set 1 and feature data training set 2 are obtained.

[0061] Based on the upper water level crack feature data, a second set of newly added variable crack feature data is preset. The crack inclination in the second set of newly added variable crack feature data is statistically analyzed to obtain a third set of concentrated crack inclination datasets. Based on the third set of concentrated crack inclination datasets, the lower water level spatial region is divided into four equal sub-regions to obtain the second lower water level sub-region set.

[0062] Based on the crack tilt dataset and data association influence characteristic coefficients in the third set, the predicted temperature, seepage, and pressure data of each sub-region in the second water level sub-region set are obtained, and the preprocessed feature prediction dataset is output.

[0063] Based on the first comprehensive comparison difference, the second comprehensive comparison difference to which the preprocessed feature prediction dataset belongs is calculated. Based on the first unstable labeled sub-region and the first stable labeled sub-region, the second unstable labeled sub-region and the second stable labeled sub-region to which the second water level sub-region belongs are marked.

[0064] The overlapping area between the second unstable labeled sub-region and the first unstable labeled sub-region is extracted as the control and reinjection location point, and the overlapping area between the second stable labeled sub-region and the first stable labeled sub-region is extracted as the control and extraction location point, thus obtaining the control result of the second location point.

[0065] Based on the first irrigation rate regulation result, the preprocessed feature prediction dataset, and the regulation matching database, the second irrigation rate regulation result is obtained.

[0066] The second position point control result and the second sampling and irrigation speed control result are combined to form the second sampling and irrigation control result.

[0067] Specifically, the first newly added variable crack feature data (the velocity parameters in the initial extraction feature parameters and the initial reinjection feature parameters are different, so the liquid mineral water level is decreasing or increasing. If we take the cover as an example, the corresponding liquid mineral water level is increasing, so the crack feature data of the sidewall in the lower water level space area increases (if the number of cracks increases, the crack inclination needs to be re-collected and statistically analyzed, and the subsequent sub-regions need to be re-divided). The increased cover crack inclination data (obtained by statistical analysis in this way) and the first concentrated crack inclination dataset are summarized to obtain the first newly added variable crack feature data (based on the first concentrated crack inclination dataset, and so on)), the lower water level sub-region set to be measured, the second feature dataset to be measured (based on the first lower water level sub-region set and the first feature dataset to be measured, and so on), and the feature data training set one (such as the crack inclination data being subtracted from the corresponding region values ​​(if they correspond to the same region, then data processing is performed directly).If the regions are not the same, the crack inclination data of the two sub-regions with the largest overlapping area are selected for subtraction. If the first feature data training set is t1, t2, tn, then the second feature data training set (based on the sub-regions of the first feature data training set, the numerical difference is calculated between the second test feature dataset (including temperature, seepage coefficient, and pressure, and the data is uniformly converted according to W:S:F) and the first test feature dataset, if the difference is T1, T2, Tn) is calculated. The data correlation influence feature coefficients (the values ​​of the first feature data training set t1, t2, tn are used as the values ​​on the x-axis, and the values ​​of the second feature data training set t1, t2, tn are used as the values ​​on the x-axis). T1, T2, and Tn are used as values ​​on the y-axis. A trend curve of the correlation and change between them is constructed, and the slope (average value) is calculated, which is the data correlation and change characteristic coefficient. Preset second newly added variable crack feature data (i.e., the simulated height of water level rise is selected. Based on the simulated height, crack feature data within the corresponding simulated height range is extracted from the upper water level crack feature data. Then, based on the first newly added variable crack feature data, and so on, the second newly added variable crack feature data is obtained. Third set of crack tilt dataset and second lower water level sub-region set (based on the first set of crack tilt dataset and the first lower water level sub-region set). Domain set, and so on. If the second upper water level sub-region set is Q1, Q2, Q3, Q4, the preprocessed feature prediction dataset (e.g., multiplying the concentrated fracture inclination data of each sub-region in the third concentrated fracture inclination dataset with the data association influence feature coefficient to obtain the comprehensive change values ​​of temperature, seepage coefficient, and pressure of each sub-region, and then converting the temperature, seepage coefficient, and pressure data of each sub-region through W:S:F, thus obtaining the preprocessed feature prediction dataset), the second comprehensive comparison difference (based on the first comprehensive comparison difference, and so on), the second unstable labeled sub-region and the second stable labeled sub-region (based on the first The results of the second location point control are as follows: (If the second unstable sub-region is Q1 and the second stable sub-region is Q2, Q3, or Q4, then the overlapping area of ​​Q1 with q1 and q2 is extracted. If it is u1, it is used as the control and re-irrigation location point. If the overlapping area of ​​the second stable sub-regions Q2, Q3, and Q4 with q3 and q4 is extracted, and if it is u2 or u3, then u2 and u3 are both used as control and extraction location points. u1, u2, and u3 are the second location point control results.) The results of the second extraction and irrigation rate control are as follows (based on the first extraction and irrigation rate control results, and so on).

[0068] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for regulating the production and irrigation of hydrogeological liquid mineral geothermal reservoirs, characterized in that, Includes the following steps: Step S1: Divide the target geothermal reservoir that needs resource extraction and irrigation into upper water level spatial region and lower water level spatial region, and divide the lower water level spatial region into four equal sub-regions to obtain the first lower water level sub-region set. Based on the differences in temperature, seepage and pressure data of each sub-region in the first lower water level sub-region set, distinguish the first non-stable labeled sub-region and the first stable labeled sub-region in the first lower water level sub-region set. Step S2: The extraction and irrigation speed is controlled by using the boundary between every two adjacent sub-regions in the first unstable marked sub-region as the control point for re-irrigation and the boundary between every two adjacent sub-regions in the first stable marked sub-region as the control point for extraction, to obtain the first extraction and irrigation control result. Step S3: Based on the fluctuating water levels between the upper and lower water level spatial regions and the upper water level crack feature data of the upper water level spatial region, preset new fluctuating crack feature data is performed, and the lower water level spatial region is re-divided to obtain a second lower water level sub-region set. The second lower water level sub-region set is marked with unstable and stable sub-regions to obtain a second unstable marked sub-region and a second stable marked sub-region. The overlapping area of ​​the first unstable marked sub-region and the second unstable marked sub-region is used as the control recharge location point, and the first stable marked sub-region and the second stable marked sub-region are used as the control extraction location point to control the extraction and irrigation speed, thereby obtaining the second extraction and irrigation control result.

2. The method for regulating the production and irrigation of hydrogeological liquid mineral reservoirs according to claim 1, characterized in that, Step S1 includes: The preset initial extraction characteristic parameters and initial reinjection characteristic parameters are obtained. The reservoir spatial volume of the target geothermal reservoir that needs resource extraction and injection is detected to obtain spatial regional characteristics. The spatial regional characteristics are divided into upper water level spatial region and lower water level spatial region. The initial extraction characteristic parameters and initial reinjection characteristic parameters include extraction and reinjection speed and extraction and reinjection location points. If the velocity parameters in the initial extraction characteristic parameters and the initial reinjection characteristic parameters are the same, then according to the stable water level between the upper water level space region and the lower water level space region, the fracture characteristic data on the reservoir sidewall in the upper water level space region and the lower water level space region are detected respectively to obtain the upper water level fracture characteristic data and the lower water level fracture characteristic data. The crack inclination in the crack feature data of the lower water level is statistically analyzed in a centralized manner to obtain a first centralized crack inclination dataset. Based on the first centralized crack inclination dataset, the lower water level spatial region is divided into four equal sub-regions to obtain a first lower water level sub-region set.

3. The method for regulating the production and irrigation of hydrogeological liquid mineral reservoirs according to claim 2, characterized in that, Step S1 also includes: Real-time detection of temperature, seepage, and pressure data in each sub-region of the first sub-region of the water level sub-region yields the first dataset of features to be measured. The first comprehensive comparison difference value is obtained by comparing and calculating the difference between the test feature datasets of two adjacent sub-regions in the first water level sub-region. A preset warning judgment threshold is set. If the difference between the first comprehensive comparison is greater than or equal to the warning judgment threshold, the sub-regions corresponding to the first water level sub-regions are marked as unstable regions to obtain the first unstable marked sub-regions. If the difference between the first comprehensive comparison and the warning judgment threshold is less than the threshold, the sub-regions in the corresponding first water level sub-regions will be marked as stable regions to obtain the first stable marked sub-regions.

4. The method for regulating the production and irrigation of hydrogeological liquid mineral reservoirs according to claim 3, characterized in that, Step S2 includes: The boundary between every two adjacent sub-regions in the first unstable marked sub-region is taken as the control recharge location point, and the boundary between every two adjacent sub-regions in the first stable marked sub-region is taken as the control extraction location point, thus obtaining the control result of the first location point. Historical irrigation and extraction characteristic data are obtained, and a matching correspondence table between irrigation and extraction rates and reservoir variation data is constructed based on the historical irrigation and extraction characteristic data to obtain a regulation matching database; Based on the comprehensive data difference between adjacent sub-regions in the first unstable marked sub-region and the stable marked sub-region, the corresponding regulation and re-irrigation speed is matched from the regulation and matching database. Based on the comprehensive data difference between adjacent sub-regions in the first stable marked sub-region, the corresponding regulation and re-irrigation speed is matched from the regulation and matching database to obtain the first sampling and irrigation speed regulation result. The first position point control result and the first irrigation speed control result are combined to form the first irrigation control result.

5. A method for regulating the production and irrigation of hydrogeological liquid mineral geothermal reservoirs according to claim 4, characterized in that, Step S3 includes: If the velocity parameters in the initial extraction characteristic parameters and the initial reinjection characteristic parameters are different, then based on the changing water level between the upper water level space region and the lower water level space region, the characteristic data of the overburden fractures and the characteristic data of the lower water level fractures on the reservoir sidewall in the newly added upper water level space region in the lower water level space region are summarized to obtain the first newly added changing fracture characteristic data. The crack inclination in the first newly added variable crack feature data is statistically analyzed to obtain a second centralized crack inclination dataset. Based on the second centralized crack inclination dataset, the water level spatial region is divided into four equal sub-regions to obtain the sub-region set of the water level to be measured. Real-time detection of temperature, seepage, and pressure data in each sub-region of the sub-region of the target water level is used to obtain the second target feature dataset.

6. A method for regulating the production and irrigation of hydrogeological liquid mineral reservoirs according to claim 5, characterized in that, Step S3 also includes: The difference between the second set of crack tilt datasets and the first set of crack tilt datasets is calculated to obtain feature data training set one. The difference between the second set of feature datasets to be tested and the first set of feature datasets to be tested is calculated to obtain feature data training set two. The correlation and change influence trend characteristics between the feature data training set 1 and feature data training set 2 are statistically analyzed to obtain the data correlation influence characteristic coefficient between feature data training set 1 and feature data training set 2; Based on the upper water level crack feature data, a second newly added variable crack feature data is preset, and the crack inclination in the second newly added variable crack feature data is centrally statistically analyzed to obtain a third centralized crack inclination dataset. Based on the third centralized crack inclination dataset, the lower water level spatial region is divided into four equal sub-spatial regions to obtain a second lower water level sub-region set. Based on the crack tilt dataset and data association influence characteristic coefficients in the third set, the temperature, seepage, and pressure prediction data of each sub-region in the second lower water level sub-region set are obtained, and the preprocessed feature prediction dataset is output.

7. A method for regulating the production and irrigation of hydrogeothermal reservoirs of hydraulic and environmental types according to claim 6, characterized in that, Step S3 also includes: Based on the first comprehensive comparison difference, the second comprehensive comparison difference to which the preprocessed feature prediction dataset belongs is calculated. Based on the first unstable labeled sub-region and the first stable labeled sub-region, the second unstable labeled sub-region and the second stable labeled sub-region to which the second water level sub-region belongs are marked. The overlapping area between the second unstable labeled sub-region and the first unstable labeled sub-region is extracted as the control and reinjection location point, and the overlapping area between the second stable labeled sub-region and the first stable labeled sub-region is extracted as the control and extraction location point, thus obtaining the control result of the second location point; Based on the first irrigation and drainage rate regulation results, the preprocessed feature prediction dataset, and the regulation matching database, the second irrigation and drainage rate regulation results are obtained. The second position point control result and the second sampling and irrigation speed control result are combined to form the second sampling and irrigation control result.

Images