Reservoir dam leakage intelligent detection device

Abstract

The invention discloses a reservoir dam leakage intelligent detection device, and the device comprises a soil hardness detection module which is used for determining a detection point according to the axis of a dam and the type of a dam body filling material; the grid division module is used for dividing the corresponding region into grid structures with associated sizes through the soil hardness mean value of the same region; the detection strategy matching module is used for determining a leakage detection strategy corresponding to the grid structure according to the soil hardness; the leakage detection module is used for detecting the leakage flow velocity of each grid in the grid structure according to the determined leakage detection strategy and determining the leakage state of each grid according to the leakage flow velocity; and the prediction module is used for predicting the leakage state of the grid according to the historical leakage data to obtain the duration required for deterioration of the leakage state of the corresponding region. According to the method, grids are reasonably divided according to the hardness of soil, so that detection is more detailed and comprehensive, potential leakage points can be accurately found, and the precision and reliability of leakage detection are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of reservoir leakage detection, and in particular to an intelligent detection device for reservoir dam leakage. Background Art

[0002] As critical water conservancy infrastructure, the safety of reservoirs and dams is directly linked to the safety of life and property downstream and the stability of the ecological environment. Leakage is a common safety hazard in dams. If not detected and addressed promptly, it can damage the dam structure and even lead to serious accidents such as dam failure.

[0003] Currently, conventional methods for detecting reservoir and dam leakage mainly include electrical, seismic, and temperature methods. However, these methods have certain limitations in practical application. Existing detection technologies are ineffective for detecting deep leakage channels and leakage in concrete and bedrock. Deep leakage channels may be hidden deep inside the dam, making it difficult for conventional methods to effectively penetrate and locate them. The structures of concrete and bedrock are complex, and the physical properties of different areas vary greatly. Existing detection systems have difficulty accurately identifying signs of leakage, resulting in the inability to promptly detect potential leakage risks, posing a hidden danger to the safe operation of the dam. In addition, most existing detection systems can only detect current leakage conditions and lack the ability to predict future leakage conditions, making it impossible to take preventive measures in advance. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent detection device for reservoir dam leakage, which can solve the technical problem that existing leakage detection technology is difficult to accurately identify leakage signs.

[0005] One aspect of the present invention provides an intelligent device for detecting leakage in a reservoir dam, the device comprising: a soil hardness detection module for determining detection points based on the dam axis and the type of dam body filling material, and measuring the soil hardness at the detection points using a soil hardness detection instrument to obtain the soil hardness at each detection point; A grid division module is used to compare the soil hardness of different detection points, divide the detection points where the difference in soil hardness is less than a preset threshold into a region, and divide the corresponding region into a grid structure of associated size based on the average soil hardness of the same region, wherein the soil hardness uniquely corresponds to the grid size; a detection strategy matching module, configured to determine a leakage detection strategy corresponding to the grid structure according to the soil hardness, wherein the soil hardness uniquely corresponds to the leakage detection strategy; A leakage detection module, configured to detect the leakage flow rate of each grid in the grid structure according to a determined leakage detection strategy, and determine the leakage state of each grid according to the leakage flow rate; A prediction module is used to predict the leakage status of the grid based on historical leakage data to obtain the time required for the leakage status of the corresponding area to deteriorate, wherein the historical leakage data includes the frequency of occurrence of historical leakage points, surrounding water quality information, surrounding soil information, depth, and temperature.

[0006] Optionally, determining the detection point position according to the dam axis and the type of dam filling material includes: Along the dam axis, the dam body is divided into multiple longitudinal inspection lines at fixed intervals; Obtain all dam filling material types corresponding to the measured dam, and set up transverse detection lines at the junctions of different dam filling materials; The intersection of the longitudinal detection line and the transverse detection line is set as a detection point.

[0007] Optionally, comparing the soil hardness of different detection points and dividing the detection points where the difference in soil hardness is less than a preset threshold into one area includes: Sequentially extract a detection point that needs to be compared as an anchor point, and determine the detection points adjacent to the anchor point; Determine whether the determined detection point has been divided into the same area as the anchor point. If so, continue to extract the next detection point to be compared as the anchor point; If not, calculate the soil hardness difference between the anchor point and the adjacent detection points, divide the adjacent detection points whose soil hardness difference is less than the preset threshold and the anchor point into one area, and divide the adjacent detection points whose soil hardness difference is greater than or equal to the preset threshold into a separate area.

[0008] Optionally, determining the leakage detection strategy corresponding to the grid structure according to the soil hardness includes: Obtaining a preset soil hardness level definition table, and matching the soil hardness with the preset soil hardness level definition table, wherein the preset soil hardness level definition table includes soil hardness levels and leakage detection strategies; If the soil hardness is low, the leakage flow rate of the grid structure is detected using the power distribution method; If the soil hardness is of a high level, the seismic method is used to detect the leakage flow rate of the grid structure.

[0009] Optionally, determining the leakage state of each grid according to the leakage flow rate includes: When the leakage velocity is less than 0.1m / s, it is judged as no leakage or slight leakage; When the leakage velocity is between 0.1m / s and 0.5m / s, it is judged as moderate leakage; When the leakage flow rate is greater than 0.5m / s, it is judged as a serious leakage.

[0010] Optionally, the prediction of the leakage state of the grid based on historical leakage data and the time required for the leakage state of the corresponding area to deteriorate are calculated by the following formula: Among them: t is the time required for the current leakage state to deteriorate to the next level state; T is the transition time required for two leakage states; f is the frequency of leakage at the detection point; is the weight factor of the leakage frequency; d is the depth of the detection point in the dam body; is the weight factor of the depth; x is the suspended solids concentration in the water of the same detection point at the current time node; is the suspended solids concentration in the water when the same detection point is in the first level of leakage state; is the suspended solids concentration in the water when the same detection point is in the second level of leakage state; y is the soil hardness at the same detection point at the current time node; is the soil hardness when the same detection point is in the first level of leakage state; is the soil hardness when the same detection point is in the second level of leakage state; z is the temperature at the same detection point at the current time node; is the temperature when the same detection point is in the first level of leakage state; is the temperature when the same detection point is in the second level of leakage state.

[0011] Optionally, the time required for the leakage state of the region to deteriorate can also be calculated through a neural network model, wherein the neural network model can be any one of the following: a convolutional neural network, a recurrent neural network.

[0012] Optionally, the training process of the neural network model is implemented by the following steps: Obtain historical leakage data at multiple locations on the same dam at different time points; The historical leakage data of the same point at different time nodes are used as input data, and the time difference corresponding to the time node is used as output data, and are input into the initial neural network model for training; Until the model converges, the neural network model training is completed.

[0013] Optionally, it also includes: The early warning module is electrically connected to the leakage detection module and the prediction module respectively, and issues an early warning signal when the determined or predicted leakage state reaches a preset early warning threshold; wherein, the early warning signal includes a sound alarm, flashing lights or sending a notification message to a preset terminal.

[0014] Optionally, it also includes: Environmental monitoring module, used to monitor the environmental parameters around the reservoir dam in real time, including water level, rainfall, and temperature; The environmental monitoring module is electrically connected to the leakage detection module and the prediction module respectively, and uses the monitored environmental parameters as auxiliary information to participate in leakage status judgment and prediction.

[0015] By first detecting the hardness of the soil and dividing it into regions, the present invention can adopt the most appropriate detection method for regions with different soil characteristics, thereby improving the detection effect of deep leakage channels and leakage in concrete and bedrock, and solving the problem of poor detection effect of existing technologies. Reasonable division of the grid according to the hardness of the soil makes the detection more detailed and comprehensive, and can accurately detect potential leakage points, thereby improving the accuracy and reliability of leakage detection. In addition, a prediction module has been added, which can predict the leakage status based on historical leakage data, detect potential leakage risks in advance, and provide a basis for taking preventive measures. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present invention. The same reference symbols are used throughout the drawings to represent the same components. In the drawings: Figure 1 The figure shows a structural block diagram of an intelligent device for detecting leakage of a reservoir dam provided in the first embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0018] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element. Example

[0019] The first embodiment of the present invention further provides an intelligent device for detecting leakage of a reservoir dam, specifically, Figure 1The structural block diagram of the intelligent detection device for reservoir dam leakage is shown in FIG. Figure 1 As shown, the intelligent detection device 100 for reservoir dam leakage includes a soil hardness detection module 101, a grid division module 102, a detection strategy matching module 103, a leakage detection module 104, and a prediction module 105, wherein: The soil hardness detection module 101 is used to determine the detection points according to the dam axis and the type of dam filling material, and to measure the soil hardness at the detection points using a soil hardness detection instrument to obtain the soil hardness of each detection point; Specifically, soil hardness testing instruments can be either geological radar or a static cone penetration tester. With a geological radar, soil hardness is inferred by analyzing the intensity and frequency of the reflected wave; with a static cone penetration tester, soil hardness is determined by measuring the resistance experienced by the probe when pressed into the soil.

[0020] Determining the softness and hardness of the soil enables targeted treatment of leakage detection in different areas of the reservoir dam, which is conducive to improving the accuracy of leakage detection of the reservoir dam.

[0021] The grid division module 102 is connected to the soil hardness detection module 101 and is used to compare the soil hardness of different detection points, and divide the detection points where the difference in soil hardness is less than a preset threshold into a region. The corresponding region is divided into a grid structure of associated size based on the average soil hardness of the same region, wherein the soil hardness uniquely corresponds to the grid size; The preset threshold is set based on actual engineering experience and detection accuracy requirements and is not limited here. Soil hardness and grid size are negatively correlated, and the mapping between the two can be determined based on actual construction experience and is not limited here. Using the mean soil hardness value to determine the grid size for the corresponding area allows for targeted regional leakage detection. Specifically, since hard soil is not conducive to detection, using a smaller grid size in hard soil areas allows for more accurate leakage detection. On the other hand, using a larger grid size in soft soil areas allows for easier detection, accurate results, and low resistance, enabling faster detection in soft soil areas, thus improving the accuracy and efficiency of reservoir and dam seepage detection. For example, for areas with uniform soil and minimal hardness variations, a grid size of 5m×5m can be used; for areas with complex soil and significant hardness variations, a grid size of 2m×2m can be used. This rational grid division allows for more comprehensive and detailed detection of dam leakage.

[0022] The detection strategy matching module 103 is connected to the grid division module 102 and is used to determine the leakage detection strategy corresponding to the grid structure according to the softness and hardness of the soil, wherein the softness and hardness of the soil uniquely correspond to the leakage detection strategy; Based on the generated grid structure, a corresponding type of leakage detection strategy is assigned to each grid structure. Leakage detection strategies are conventional reservoir and dam leakage detection methods, including but not limited to electrical methods, seismic methods, and temperature methods. Regions with different soil hardness have different adaptability to detection methods. For example, electrical methods may be more effective in areas with softer soil because soft soil has a higher moisture content and better conductivity, and electrical methods can more clearly detect electrical changes caused by leakage. For areas with harder soil, seismic methods may be more advantageous because the propagation characteristics of seismic waves in hard soil are more stable, which can more accurately identify leakage channels. The most appropriate detection method is selected for each area based on pre-set rules or algorithms.

[0023] The leakage detection module 104 is connected to the detection strategy matching module 103 and is used to detect the leakage flow rate of each grid in the grid structure according to the determined leakage detection strategy, and determine the leakage state of each grid according to the leakage flow rate; Leakage status includes no leakage, slight leakage, moderate leakage, and severe leakage. In particular, when measuring the leakage flow rate of each grid, the water quality information, soil information, temperature, depth, and frequency of leakage at each grid are also collected.

[0024] When the electrical method is used to detect the leakage flow rate, the leakage flow rate is calculated by measuring the change in the potential difference between the electrodes; when the temperature method is used to detect the leakage flow rate, the leakage flow rate is indirectly calculated by monitoring the change in the temperature field.

[0025] The prediction module 105 is connected to the leakage detection module 104 and is used to predict the leakage status of the grid based on historical leakage data to obtain the time required for the leakage status of the corresponding area to deteriorate, wherein the historical leakage data includes the frequency of occurrence of historical leakage points, surrounding water quality information, surrounding soil information, depth, and temperature.

[0026] By predicting the length of time required for the leakage condition to worsen, the staff is provided with the time and extent to which subsequent treatment measures can be prepared, which is conducive to the targeted treatment of reservoir dam leakage detection.

[0027] By first detecting the hardness of the soil and dividing it into regions, the present invention can adopt the most appropriate detection method for regions with different soil characteristics, thereby improving the detection effect of deep leakage channels and leakage in concrete and bedrock, and solving the problem of poor detection effect of existing technologies. Reasonable division of the grid according to the hardness of the soil makes the detection more detailed and comprehensive, and can accurately detect potential leakage points, thereby improving the accuracy and reliability of leakage detection. In addition, a prediction module has been added, which can predict the leakage status based on historical leakage data, detect potential leakage risks in advance, and provide a basis for taking preventive measures.

[0028] Optionally, the detection points are determined by the dam axis and the type of dam filling material, including: Along the dam axis, the dam body is divided into multiple longitudinal inspection lines at fixed intervals; Obtain all dam filling material types corresponding to the measured dam, and set up transverse detection lines at the junctions of different dam filling materials; The intersection of the longitudinal detection line and the transverse detection line is set as a detection point.

[0029] The determination of detection points realizes the balanced treatment of leakage detection in various areas of the reservoir dam, and improves the accuracy of the reservoir dam detection effect.

[0030] Preferably, comparing the soil hardness of different detection points and dividing the detection points where the difference in soil hardness is less than a preset threshold into a region includes: Sequentially extract a detection point that needs to be compared as an anchor point, and determine the detection points adjacent to the anchor point; Determine whether the determined detection point has been divided into the same area as the anchor point. If so, continue to extract the next detection point to be compared as the anchor point; If not, calculate the soil hardness difference between the anchor point and the adjacent detection points, divide the adjacent detection points whose soil hardness difference is less than the preset threshold and the anchor point into one area, and divide the adjacent detection points whose soil hardness difference is greater than or equal to the preset threshold into a separate area.

[0031] Areas where the difference in soil hardness is less than a preset threshold are classified as a single zone. This threshold can be set based on actual engineering experience and test accuracy requirements. For example, if the preset threshold is set at 10% of the soil hardness value, then adjacent areas will be classified as the same zone if the soil hardness values ​​differ by no more than 10%. In this way, the dam can be divided into multiple areas with similar soil hardness characteristics, allowing for the subsequent application of appropriate testing methods for different areas.

[0032] Preferably, determining the leakage detection strategy corresponding to the grid structure according to the soil hardness includes: Obtaining a preset soil hardness level definition table, and matching the soil hardness with the preset soil hardness level definition table, wherein the preset soil hardness level definition table includes soil hardness levels and leakage detection strategies; If the soil hardness is low, the leakage flow rate of the grid structure is detected using the power distribution method; If the soil hardness is of a high level, the seismic method is used to detect the leakage flow rate of the grid structure. Preferably, determining the leakage state of each grid by the leakage flow rate includes: When the leakage velocity is less than 0.1m / s, it is judged as no leakage or slight leakage; When the leakage velocity is between 0.1m / s and 0.5m / s, it is judged as moderate leakage; When the leakage flow rate is greater than 0.5m / s, it is judged as a serious leakage.

[0033] According to the leakage velocity data of each grid, it is classified into the corresponding leakage state, so as to intuitively display the leakage situation of each area of ​​the dam.

[0034] Preferably, the leakage state of the grid is predicted based on historical leakage data, and the time required for the leakage state of the corresponding area to deteriorate is calculated by the following formula: Where: t is the time required for the current leakage state to deteriorate to the next level; T is the transition time required for two leakage states; f is the frequency of leakage at the detection point; is the weight factor of leakage frequency; d is the depth of the detection point in the dam body; is the weight factor of depth; x is the concentration of suspended solids in water at the same detection point at the current time node; It is the concentration of suspended solids in water when the same detection point is in the first-level leakage state; is the concentration of suspended solids in water at the same detection point when it is in the second-level leakage state; y is the soil hardness at the same detection point at the current time node; It is the soil hardness at the same testing point when it is in the first level leakage state; is the soil hardness of the same detection point when it is in the second-level leakage state; z is the temperature of the same detection point at the current time node; It is the temperature when the same detection point is in the first level leakage state; It is the temperature when the same detection point is in the second level leakage state; The above formula is used to summarize leakage patterns and trends, providing a scientific basis for formulating maintenance and repair plans.

[0035] Preferably, the time required for the leakage state of a region to deteriorate can also be determined through a neural network model, wherein the neural network model can be any one of the following: a convolutional neural network, a recurrent neural network.

[0036] The types of neural network models are not limited to the items listed above, and can also be other neural network models, which are not limited here.

[0037] Preferably, the training process of the neural network model is achieved through the following steps: Obtain historical leakage data at multiple locations on the same dam at different time points; The historical leakage data of the same point at different time nodes are used as input data, and the time difference corresponding to the time node is used as output data, and are input into the initial neural network model for training; Until the model converges, the neural network model training is completed.

[0038] By comprehensively storing and analyzing historical leakage data from different locations and different periods, and using neural network models to summarize leakage patterns and trends, a scientific basis is provided for formulating maintenance and repair plans, realizing intelligent leakage detection of reservoir dams, thereby improving the accuracy of leakage detection.

[0039] Preferably, it also includes: The early warning module is electrically connected to the leakage detection module and the prediction module, and issues an early warning signal when the determined or predicted leakage state reaches a preset early warning threshold; wherein the early warning signal includes a sound alarm, flashing lights, or sending a notification message to a preset terminal.

[0040] The early warning module can issue an early warning signal in time when the leakage status reaches the warning threshold, ensuring that relevant personnel can take measures quickly to ensure the safe operation of the dam.

[0041] Preferably, it also includes: Environmental monitoring module, used to monitor the environmental parameters around the reservoir dam in real time, including water level, rainfall, and temperature; The environmental monitoring module is electrically connected to the leakage detection module and the prediction module respectively, and uses the monitored environmental parameters as auxiliary information to participate in leakage status judgment and prediction.

[0042] The environmental monitoring module can monitor the surrounding environmental parameters in real time and use them as auxiliary information to participate in leakage status judgment and prediction, further improving the accuracy and reliability of the system.

[0043] The working principle of the device is as follows: The soil hardness testing module is activated to conduct a comprehensive inspection of the reservoir dam according to the pre-set inspection route and inspection points. During the inspection process, the soil hardness data of each inspection point is recorded and transmitted to the regional division module.

[0044] After receiving the soil hardness data, the gridding module divides the dam into zones based on a preset threshold. Areas with soil hardness differences below the threshold are merged into a single zone, and each zone is assigned a unique identifier. A grid structure of corresponding size is also set for each zone, and the results are displayed on a graphical interface for easy viewing by the operator.

[0045] The detection strategy matching module queries pre-defined detection method selection rules based on soil hardness and assigns the most appropriate detection method to each grid structure. For example, electrical detection is assigned to areas with softer soils, while seismic detection is assigned to areas with harder soils.

[0046] After receiving the detection method and grid information, the leakage detection module uses the corresponding detection method to perform leakage velocity testing on each grid. During the testing process, data is collected and processed in real time to obtain the leakage velocity value for each grid. The leakage status of each grid is determined based on the pre-defined correspondence between the leakage velocity range and the leakage status. The results are displayed graphically, using different colors to represent different leakage statuses, visually demonstrating the leakage status of each area of ​​the dam. The results are also stored in the data storage module.

[0047] The data collection submodule of the prediction module collects historical leakage data from the data storage module, and the model construction submodule constructs a leakage status prediction model based on the collected data. The prediction submodule uses the constructed model to predict the current leakage status and transmits the prediction results to the early warning module.

[0048] The data analysis module regularly analyzes the data stored in the data storage module, summarizing patterns and trends in dam leakage. Based on these results, it develops appropriate maintenance and repair plans to promptly address any leaks and ensure the safe operation of the reservoir and dam. Furthermore, the detection system undergoes regular maintenance and calibration to ensure its accuracy and reliability.

[0049] When the early warning module receives information from the leakage status determination module or the prediction module that the warning threshold has been reached, it issues an early warning signal. Relevant personnel take timely measures based on the early warning signal, such as on-site inspection and maintenance.

[0050] The environmental monitoring module monitors the water level, rainfall, temperature and other environmental parameters around the reservoir dam in real time, and transmits the monitoring data to the leakage status judgment module and prediction module as auxiliary information to participate in leakage status judgment and prediction.

[0051] It should be noted that the serial numbers of the embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0052] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method.

[0053] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An intelligent detection device for reservoir dam leakage, characterized in that: The device comprises: The soil hardness detection module is used to determine the detection points according to the dam axis and the type of dam filling material, and use the soil hardness detection instrument to measure at the detection points to obtain the soil hardness of each detection point; A grid division module is used to compare the soil hardness of different detection points, divide the detection points where the difference in soil hardness is less than a preset threshold into a region, and divide the corresponding region into a grid structure of associated size based on the average soil hardness of the same region, wherein the soil hardness uniquely corresponds to the grid size; a detection strategy matching module, configured to determine a leakage detection strategy corresponding to the grid structure according to the soil hardness, wherein the soil hardness uniquely corresponds to the leakage detection strategy; A leakage detection module, configured to detect the leakage flow rate of each grid in the grid structure according to a determined leakage detection strategy, and determine the leakage state of each grid according to the leakage flow rate; A prediction module is used to predict the leakage status of the grid based on historical leakage data to obtain the time required for the leakage status of the corresponding area to deteriorate, wherein the historical leakage data includes the frequency of occurrence of historical leakage points, surrounding water quality information, surrounding soil information, depth, and temperature.

2. The intelligent detection device for reservoir dam leakage according to claim 1, characterized in that: Determining the detection points based on the dam axis and the type of dam filling material includes: Along the dam axis, the dam body is divided into multiple longitudinal inspection lines at fixed intervals; Obtain all dam filling material types corresponding to the measured dam, and set up transverse detection lines at the junctions of different dam filling materials; The intersection of the longitudinal detection line and the transverse detection line is set as a detection point.

3. The intelligent detection device for reservoir dam leakage according to claim 1, characterized in that: The step of comparing the soil hardness of different detection points and dividing the detection points where the difference in soil hardness is less than a preset threshold into a region includes: Sequentially extract a detection point that needs to be compared as an anchor point, and determine the detection points adjacent to the anchor point; Determine whether the determined detection point has been divided into the same area as the anchor point. If so, continue to extract the next detection point to be compared as the anchor point; If not, calculate the soil hardness difference between the anchor point and the adjacent detection points, divide the adjacent detection points whose soil hardness difference is less than the preset threshold and the anchor point into one area, and divide the adjacent detection points whose soil hardness difference is greater than or equal to the preset threshold into a separate area.

4. The intelligent detection device for reservoir dam leakage according to claim 1, characterized in that: The leakage detection strategy corresponding to the grid structure is determined according to the soil hardness, including: Obtaining a preset soil hardness level definition table, and matching the soil hardness with the preset soil hardness level definition table, wherein the preset soil hardness level definition table includes soil hardness levels and leakage detection strategies; If the soil hardness is low, the leakage flow rate of the grid structure is detected using the power distribution method; If the soil hardness is of a high level, the seismic method is used to detect the leakage flow rate of the grid structure.

5. The intelligent detection device for reservoir dam leakage according to claim 1, characterized in that: The determining the leakage state of each grid according to the leakage flow rate includes: When the leakage velocity is less than 0.1m / s, it is judged as no leakage or slight leakage; When the leakage velocity is between 0.1m / s and 0.5m / s, it is judged as moderate leakage; When the leakage flow rate is greater than 0.5m / s, it is judged as a serious leakage.

6. The intelligent detection device for reservoir dam leakage according to claim 1, characterized in that: The leakage state of the grid is predicted based on the historical leakage data, and the time required for the leakage state of the corresponding area to deteriorate is calculated by the following formula: Where: t is the time required for the current leakage state to deteriorate to the next level; T is the transition time required for two leakage states; f is the frequency of leakage at the detection point; is the weight factor of leakage frequency; d is the depth of the detection point in the dam body; is the weight factor of depth; x is the concentration of suspended solids in water at the same detection point at the current time node; It is the concentration of suspended solids in water when the same detection point is in the first-level leakage state; is the concentration of suspended solids in water at the same detection point when it is in the second-level leakage state; y is the soil hardness at the same detection point at the current time node; It is the soil hardness at the same testing point when it is in the first level leakage state; is the soil hardness of the same detection point when it is in the second-level leakage state; z is the temperature of the same detection point at the current time node; It is the temperature when the same detection point is in the first level leakage state; It is the temperature when the same detection point is in the second level leakage state.

7. The intelligent detection device for reservoir dam leakage according to claim 1, characterized in that: The time required for the leakage state of a region to deteriorate can also be calculated through a neural network model, wherein the neural network model can be any one of the following: a convolutional neural network and a recurrent neural network.

8. The intelligent detection device for reservoir dam leakage according to claim 7, characterized in that: The training process of the neural network model is achieved through the following steps: Obtain historical leakage data at multiple locations on the same dam at different time points; The historical leakage data of the same point at different time nodes are used as input data, and the time difference corresponding to the time node is used as output data, and are input into the initial neural network model for training; Until the model converges, the neural network model training is completed.

9. The intelligent detection device for reservoir dam leakage according to claim 1, characterized in that: Also includes: The early warning module is electrically connected to the leakage detection module and the prediction module respectively, and issues an early warning signal when the determined or predicted leakage state reaches a preset early warning threshold; wherein, the early warning signal includes a sound alarm, flashing lights or sending a notification message to a preset terminal.

10. The intelligent detection device for reservoir dam leakage according to claim 1, characterized in that: Also includes: Environmental monitoring module, used to monitor the environmental parameters around the reservoir dam in real time, including water level, rainfall, and temperature; The environmental monitoring module is electrically connected to the leakage detection module and the prediction module respectively, and uses the monitored environmental parameters as auxiliary information to participate in leakage status judgment and prediction.

Images