Geomembrane anti-seepage performance testing and evaluating system based on artificial intelligence

Abstract

The invention relates to the field of geomembrane anti-seepage testing, in particular to a geomembrane anti-seepage performance testing and evaluating system based on artificial intelligence. According to the system, firstly, a moisture enrichment position point is selected from all position points of a geomembrane, and according to the distance between a target position point and the moisture enrichment position point and the water content difference between the target position point and the moisture enrichment position point at the same moment, the voltage data difference at the adjacent moment of the target position point is combined; obtaining the water flow influence degree of the target position point at each moment, and obtaining the adjustment collection frequency of the target position point in each time period according to the difference of the voltage data of the target position point at each moment and the difference of the water flow influence degree between the target position point and each position point except the moisture enrichment position point in time sequence; and acquiring the voltage of the target position point based on the adjusted acquisition frequency, and evaluating the anti-seepage performance of the geomembrane by using the acquired voltage data. The method can accurately evaluate the permeability of the geomembrane.

Description

Technical Field

[0001] This invention relates to the field of geomembrane seepage prevention testing, and specifically to an artificial intelligence-based geomembrane seepage prevention performance testing and evaluation system. Background Technology

[0002] During the production process, geomembranes may have defects such as pinholes, air bubbles, impurities, uneven thickness, and local weak points. In order to prevent unqualified or inferior geomembranes from entering the project site and to avoid pollutant leakage and water loss during the use of geomembranes, it is necessary to test and evaluate the seepage prevention performance of geomembranes to ensure environmental and engineering safety during the use of geomembranes.

[0003] In the process of testing the seepage prevention performance of geomembranes, the dual-electrode detection method mainly relies on increasing the water content of the geomembrane cover layer to improve conductivity. When geomembrane leakage occurs, it can manifest as abnormal electrical parameters such as voltage or current. Therefore, in related technologies, dual-electrode detection technology is usually used to collect voltage data at different locations on the geomembrane and use neural networks to analyze the characteristics of the voltage data at each location, thereby realizing the permeability test at different locations on the geomembrane. However, in the actual process of water spraying on the surface of the geomembrane, excessive water spraying will cause it to extend to the surrounding area, which will form a water flow phenomenon on the surface of the geomembrane. As a result, the voltage data collected at a fixed frequency without considering the water flow on the surface of the geomembrane cannot accurately assess its permeability performance. Summary of the Invention

[0004] To address the technical problem that existing methods, which fail to accurately assess the permeability performance of geomembranes by collecting voltage data at a fixed frequency without considering moisture flow on the geomembrane surface, aim to provide an artificial intelligence-based geomembrane seepage prevention performance testing and evaluation system. The specific technical solution adopted is as follows: This invention also proposes an artificial intelligence-based system for testing and evaluating the seepage prevention performance of geomembranes, the system comprising: The data acquisition module is used to acquire the water content and voltage data of each location point of the geomembrane at each time. The water flow impact analysis module is used to select water-rich locations from all locations based on the water content at each location; and to take any location other than the water-rich locations as the target location. Based on the distance between the target location and the water-rich locations, and the difference in water content between the target location and the water-rich locations at the same time, the module obtains the water flow diffusion coefficient of the target location at each time. Based on the difference in the water flow diffusion coefficient between the target location and the location to be measured at the same time, the module obtains the local water flow condition coefficient of the target location. Based on the difference in voltage data between the target location at each time and the next adjacent time, and the local water flow condition coefficient, the module obtains the degree of water flow impact of the target location at each time. The acquisition frequency adjustment module is used to obtain the voltage confidence of the target location point at each moment based on the difference in voltage data at each moment of the target location point and the temporal difference in the degree of water flow influence between the target location point and other location points except for the water-rich location point; and to adjust the voltage acquisition frequency of the target location point based on the voltage confidence of the target location point at each moment of each time period to obtain the adjusted acquisition frequency of the target location point in each time period. The seepage prevention performance evaluation module is used to collect voltage data at the target location point based on the adjusted acquisition frequency, and to evaluate the seepage prevention performance of the geomembrane using the collected voltage data.

[0005] Furthermore, the step of selecting water-rich location points from all location points includes: The average of the moisture content at each location point at all times is taken as the overall moisture content at each location point; The location point corresponding to the maximum value of the overall moisture content is taken as the moisture enrichment location point.

[0006] Furthermore, the water flow diffusion coefficient at the target location point at each time step includes: Take any time as the target time, use the distance between the target location and the water-rich location as the numerator, and use the difference in water content between the water-rich location and the target location at the target time and the sum of the preset first adjustment parameter as the denominator. Then, normalize the comparison values ​​to obtain the water flow diffusion coefficient of the target location at the target time.

[0007] Furthermore, the local flow condition coefficient at the target location point includes: The average value of the water flow diffusion coefficient at all locations except the water enrichment location at each time moment is taken as the overall water flow diffusion coefficient at each time moment. The difference between the water flow diffusion coefficient at the target location point at each time moment and the overall water flow diffusion coefficient is taken as the diffusion coefficient difference value at the target location point at each time moment. The average value of the diffusion coefficient difference at all times of the target location is normalized to obtain the local flow condition coefficient of the target location.

[0008] Furthermore, the degree of influence of water flow at the target location point at each moment includes: The absolute value of the difference between the voltage data of the target location point at each time and the next adjacent time is used as the voltage change of the target location point at each time. The voltage data of the target location point at each time is used as the numerator, and the sum of the voltage change of the target location point at each time and the preset second adjustment parameter is used as the denominator. The ratio is used as the water flow interference coefficient of the target location point at each time. The water flow interference coefficient and the local water flow condition coefficient of the target location point at each time moment are combined and normalized to obtain the degree of water flow influence of the target location point at each time moment.

[0009] Furthermore, obtaining the voltage confidence level of the target location point at each time step includes: The average voltage data of the target location at all times is taken as the overall voltage value of the target location. The absolute value of the difference between the voltage data of the target location point at each time moment and the overall voltage value of the target location point is taken as the voltage deviation value of the target location point at each time moment. According to the time sequence, the sequence of the degree of water flow influence of each location point other than the water-rich location point at all times is taken as the sequence of the degree of water flow influence of each location point other than the water-rich location point. The average value of the sequence of water flow influence at all locations except for the water-rich locations is used as the reference sequence; Based on the difference between the water flow influence degree sequence at the target location and the reference sequence, the water flow influence similarity at the target location is obtained; The similarity of the water flow influence at the target location point is used as the numerator, and the sum of the voltage deviation value of the target location point at each time moment and the preset third adjustment parameter is used as the denominator. The values ​​are then compared and normalized with negative correlation to obtain the voltage confidence level of the target location point at each time moment.

[0010] Furthermore, the similarity of the water flow influence at the target location point includes: The absolute value of the cosine similarity between the sequence of water flow influence at the target location and the reference sequence is taken as the water flow influence similarity at the target location.

[0011] Furthermore, the adjustment of the acquisition frequency for obtaining the target location point in each time period includes: The average voltage confidence of the target location point at all times within each time period is negatively correlated and normalized to obtain the frequency adjustment weight of the target location point in each time period. Based on the frequency adjustment weight, the frequency of voltage acquisition at the target location point in each time period is adjusted to obtain the adjusted acquisition frequency of the target location point in each time period.

[0012] Further, the step of adjusting the frequency of voltage acquisition at the target location point in each time period based on the frequency adjustment weight, to obtain the adjusted acquisition frequency of the target location point in each time period, includes: The product of the frequency adjustment weight of the target location point in each time period and the standard voltage acquisition frequency is used as the frequency adjustment amount of the target location point in each time period. The sum of the standard voltage acquisition frequency and the frequency adjustment amount is used as the adjusted acquisition frequency for the target location point in each time period.

[0013] Furthermore, the evaluation of the impermeability of the geomembrane includes: The voltage data of the target location point within the corresponding time period, which is collected using the adjusted acquisition frequency, is input into a binary classification convolutional neural network, and the binary classification convolutional neural network outputs whether the geomembrane has a seepage phenomenon at the target location point.

[0014] The present invention has the following beneficial effects: This invention addresses the limitation of existing methods that fail to accurately assess the permeability of geomembranes by collecting voltage data at a fixed frequency without considering surface moisture flow. Therefore, it selects moisture-rich locations from all available points, using these locations as a benchmark. The invention employs a water diffusion coefficient to reflect the likelihood of moisture accumulation affecting the target location at each moment, and further utilizes a local flow condition coefficient to reflect the moisture content variation at the target location due to the flow of accumulated water. Based on the difference in voltage data between each moment and the next adjacent moment, and combined with the local flow condition coefficient, the invention accurately analyzes the degree of water flow influence at each moment. The invention also uses voltage confidence levels to reflect the confidence level of the voltage data collected at each moment in assessing the geomembrane's permeability. Based on these voltage confidence levels, the invention adjusts the voltage collection frequency at the target location, and uses the voltage data collected at the adjusted frequency to accurately assess the geomembrane's impermeability. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a block diagram of an artificial intelligence-based geomembrane seepage prevention performance testing and evaluation system provided in one embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an artificial intelligence-based geomembrane seepage prevention performance testing and evaluation system proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following description, in conjunction with the accompanying drawings, details a specific scheme for a geomembrane seepage prevention performance testing and evaluation system based on artificial intelligence provided by the present invention.

[0020] Please see Figure 1 The diagram illustrates a block diagram of an artificial intelligence-based geomembrane seepage prevention performance testing and evaluation system according to an embodiment of the present invention. The system includes: a data acquisition module 101, a water flow influence analysis module 102, a data acquisition frequency adjustment module 103, and a seepage prevention performance evaluation module 104.

[0021] The data acquisition module 101 is used to acquire the water content and voltage data of each location point of the geomembrane at each time.

[0022] In this embodiment of the invention, a certain amount of water is first sprayed onto the surface of the laid geomembrane to increase the conductivity of the geomembrane surface. Then, multiple location points are evenly set on the surface of the geomembrane, and the voltage data of each location point of the geomembrane at each time moment is collected using dual-electrode detection technology. The dual-electrode detection technology is a well-known technique in the art and will not be described in detail here.

[0023] Since the flow of moisture on the surface of the geomembrane can affect the accuracy of its permeability assessment, this embodiment of the invention also requires the use of a moisture sensor to collect the moisture content at each location of the geomembrane at each time. The moisture content and voltage data are collected at the same frequency. In one embodiment of the invention, the data collection frequency is set to 10Hz, but this is not limited here.

[0024] It should be noted that different types of data have different dimensions. Therefore, the embodiments of the present invention also need to standardize the collected different types of data to eliminate the influence of dimensions. Data standardization is a technical means well known to those skilled in the art and will not be elaborated here.

[0025] The water flow impact analysis module 102 is used to select water-rich locations from all locations based on the water content at each location point; and to take any location other than the water-rich locations as the target location point. Based on the distance between the target location point and the water-rich locations, and the difference in water content between the target location point and the water-rich locations at the same time, the water flow diffusion coefficient of the target location point at each time point is obtained; based on the difference in the water flow diffusion coefficient between the target location point and the location to be measured at the same time point, the local water flow condition coefficient of the target location point is obtained; and based on the difference in voltage data between the target location point at each time point and the next adjacent time point, and the local water flow condition coefficient, the degree of water flow impact of the target location point at each time point is obtained.

[0026] In the test of geomembrane seepage prevention performance, the two-electrode method applies different potentials to the top and bottom of the geomembrane and uses its insulation properties to locate the seepage point. If there is a seepage, the electric field at the hole is conductive. When the moisture content of the cover layer changes, such as excessive watering causing water flow, it will destroy the uniformity of conductivity and cause the electric potential field to be distorted. It will form an abnormally low resistance path in the water flow area, causing the signal to drift or false alarm, masking the electrical abrupt change characteristics of the real seepage point.

[0027] Since the flow conditions at the geomembrane testing location are mainly influenced by changes in conductivity caused by the flow of moisture in the cover layer, these changes occur in stages, which can be divided into an initial scouring stage and a steady-state diffusion stage. Because the measurement process is based on the conductivity of the cover layer moisture, the moisture content at the time of measurement is fundamental to ensuring conductivity. Furthermore, the topography of the geomembrane laying location also plays a role. For example, as the bottom isolation layer of a sanitary landfill, its uneven bottom can lead to significant moisture flow during excessive water spraying from the cover layer. This moisture flow typically occurs from areas of high moisture concentration to other areas. Therefore, this embodiment of the invention first selects moisture-concentrated locations from all locations based on their moisture content. Subsequently, these moisture-concentrated locations can be used as a benchmark to evaluate the permeability of the geomembrane at other locations.

[0028] Preferably, in one embodiment of the present invention, the method for obtaining the water enrichment location points specifically includes: The average water content at each location point at all times is taken as the overall water content of each location point. The higher the overall water content at a location point, the more water it contains. Therefore, the location point corresponding to the maximum overall water content can be taken as the water-rich location point.

[0029] During the diffusion and flow of water from the water-rich location to other locations, the greater the distance between the water-rich location and other locations, and the smaller the difference in water content between them, the greater the influence of the flow conditions after the sprayed water enrichment on other locations. Therefore, any location other than the water-rich location can be taken as the target location. Based on the distance between the target location and the water-rich location, and the difference in water content between them at the same time, the water flow diffusion coefficient of the target location at each time can be obtained. The water flow diffusion coefficient reflects the possibility that the target location will be affected by the flow diffusion phenomenon after water enrichment at each time. Subsequently, the degree of influence of water flow on the target location of the geomembrane at each time can be accurately analyzed based on the water flow diffusion coefficient.

[0030] Preferably, in one embodiment of the present invention, the method for obtaining the water flow diffusion coefficient at the target location point at each time step specifically includes: Using any given time as the target time, the distance between the target location and the water-rich location is used as the numerator, and the difference in water content between the water-rich location and the target location at the target time is used as the denominator. The values ​​are then normalized, and the calculation results are limited to a certain range. Within the range, the water flow diffusion coefficient at the target location point at the target time can be obtained.

[0031] In one embodiment of the present invention, the normalization process can be specifically, for example, maximum and minimum value normalization. Furthermore, the normalization in subsequent steps can all adopt maximum and minimum value normalization. In other embodiments of the present invention, other normalization methods can be selected according to the specific range of values, or activation functions and hyperbolic tangent functions can be used to implement the normalization process. These will not be elaborated or limited further. Moreover, the same method can also be used for normalization in subsequent steps.

[0032] As an example, in one embodiment of the present invention, the expression for the water diffusion coefficient at the target location point at the target time can be specifically as follows:

[0033] in, This represents the water diffusion coefficient at the target location point at the target time. This indicates the distance between the target location and the location of water enrichment. This indicates the water content at the target time at the location of water enrichment. This indicates the water content at the target location point at the target time. This represents the normalization function, used for normalization processing; This indicates the preset first adjustment parameter, used to prevent the denominator from being 0. The range of values ​​is In one embodiment of the present invention, the following is used: Set to 0.01, The specific values ​​can be set by the implementer according to the specific implementation scenario, and are not limited here.

[0034] in, The larger the value, the greater the likelihood that the target location will be affected by the flow of water after it has accumulated at the target time.

[0035] The same method described above can be used to obtain the water diffusion coefficient of the target location at each time. Correspondingly, for the target location, the influence of enriched water flow will gradually decrease as time progresses after the detection begins. Therefore, the local water flow condition coefficient of the target location can be obtained based on the difference in water diffusion coefficient between the target location and the test location at the same time. The local water flow condition coefficient reflects the water content change of the target location due to the influence of enriched water flow. Subsequently, the degree of influence of water flow on the target location of the geomembrane at each time can be accurately analyzed based on the local water flow condition coefficient.

[0036] Preferably, in one embodiment of the present invention, the method for obtaining the local flow condition coefficient at the target location specifically includes:

[0037] The average water flow diffusion coefficient of all locations except the water-rich locations at each time step is taken as the overall water flow diffusion coefficient at each time step. The difference between the water flow diffusion coefficient of the target location at each time step and the overall water flow diffusion coefficient is taken as the diffusion coefficient difference value of the target location at each time step. The larger the diffusion coefficient difference value at a certain time step, the greater the difference between the water flow diffusion coefficient of the target location at that time step and the overall level of the water flow diffusion coefficient of all locations except the water-rich locations at that time step.

[0038] Furthermore, the average value of the diffusion coefficient difference at all times can be normalized, limiting the calculation results to... Within the range, the local flow condition coefficient at the target location point is obtained.

[0039] As an example, in one embodiment of the present invention, the expression for the local flow condition coefficient at the target location point can be specifically as follows:

[0040] in, This represents the local flow condition coefficient at the target location point; Indicates the target location point at the th The water flow diffusion coefficient at a given time; Indicates the first The overall water flow diffusion coefficient at a given moment; Indicates the target location point at the th The difference in diffusion coefficient at each time point; Indicates the number of all moments; This represents the normalization function, used for normalization processing.

[0041] Since the local flow condition coefficient at the target location is obtained by analyzing the diffusion and accumulation performance throughout the entire flow stage, but for artificial intelligence in intelligent data processing, which involves feature extraction and classification quantification of all raw data, whether the geomembrane at a single location has permeated, or the manifestation of permeation, is more reflected in the voltage data obtained by the dual-electrode detection method, which is more time-sensitive. Therefore, regarding the process of enrichment flow influence at a single geomembrane location, its location is affected by topography; the higher the water content, the more obvious the overflow condition, and the slower the rate of water content reduction in a short period of time. At this time, the geomembrane covering layer... The smaller the infiltration rate, the smaller the voltage data fluctuation. Therefore, based on the difference in voltage data between the target location point at each moment and the next adjacent moment, and combined with the local water flow condition coefficient of the target location point, the degree of water flow influence at each moment can be obtained. The degree of water flow influence reflects the extent to which the target location point is affected by water flow at each moment. The greater the degree of water flow influence at a certain moment, the greater the degree of water flow influence on the voltage data of the target location point at that moment. Subsequently, based on the temporal differences in the degree of water flow influence between each location point, the voltage confidence level of the target location point at each moment can be accurately analyzed, which facilitates the effective adjustment of the voltage data acquisition frequency.

[0042] Preferably, in one embodiment of the present invention, the method for obtaining the degree of influence of water flow at the target location point at each moment specifically includes: The absolute value of the difference between the voltage data of the target location point at each time and the next adjacent time is used as the voltage change of the target location point at each time. The voltage data of the target location point at each time is used as the numerator, and the sum of the voltage change of the target location point at each time and the preset second adjustment parameter is used as the denominator. The ratio is used as the water flow interference coefficient of the target location point at each time. The larger the water flow interference coefficient at a certain time, the greater the influence of the overflow flow condition exhibited by the target location point at that time, and the stronger the influence on the target location point at that time.

[0043] It should be noted that there is no adjacent next moment for the last moment of the target location. Therefore, the average of the voltage changes of the target location at the last moment can be used as the voltage change of the target location at the last moment.

[0044] Then, the flow disturbance coefficient and the local flow condition coefficient of the target location at each time point are combined and normalized to limit the calculation results to... Within the range, the extent of water flow influence at the target location point at each moment can be obtained.

[0045] In embodiments of the present invention, the sum or product of the water flow interference coefficient and the local water flow condition coefficient of the target location point at each time moment can be calculated to achieve the integration of the two, which is not limited here. Furthermore, the same method can be used to integrate two or more data in subsequent steps.

[0046] As an example, in one embodiment of the present invention, the expression for the degree of influence of the water flow at the target location point at each moment can be specifically as follows:

[0047] in, Indicates the target location point at the th The degree of influence of water flow at any given moment; This represents the local flow condition coefficient at the target location point; Indicates the target location point at the th Voltage data at each moment; Indicates the target location point at the th Voltage data at time n, where the first... The moment is the first The next time immediately following the previous time; Indicates the target location point at the th The voltage change at each moment; Indicates the target location point at the th The water flow disturbance coefficient at each moment; This represents the normalization function, used for normalization processing; This indicates a preset second adjustment parameter, used to prevent the denominator from being 0. The range of values ​​is In one embodiment of the present invention, the following is used: Set to 0.01, The specific values ​​can be set by the implementer according to the specific implementation scenario, and are not limited here.

[0048] The frequency adjustment module 103 is used to obtain the voltage confidence of the target location at each moment based on the difference in voltage data of the target location at each moment, and the temporal difference in the degree of water flow influence between the target location and other locations except for the water-rich location; and to adjust the voltage acquisition frequency of the target location at each moment within each time period based on the voltage confidence of the target location at each moment, thereby obtaining the adjusted acquisition frequency of the target location at each time period.

[0049] Because the changes in the moisture content of the cover layer in the geomembrane test area during the initial scouring and steady-state diffusion phases are both prolonged, the fluctuations in the acquired voltage data can lead to insufficient accuracy in using artificial intelligence to assess the geomembrane's seepage prevention performance. Therefore, considering the actual voltage data fluctuations, the leakage situation indicated by the voltage data collected at the target location point is quantified by measuring the change in the degree of water flow influence at that point over time. This improves the ability of the subsequent neural network to distinguish the influence of enriched water flow at different leakage conditions. Thus, the voltage confidence level of the target location point at each moment can be obtained by first considering the differences in voltage data at the target location point at each moment, as well as the temporal differences in the degree of water flow influence between the target location point and other locations except for the water-rich location point. The voltage confidence level reflects the confidence level of the voltage data collected at the target location point at each moment in assessing the geomembrane's permeability performance. Subsequently, based on the voltage confidence level, the frequency of voltage data acquisition at the target location point can be adjusted to improve the accuracy of the assessment of the geomembrane's permeability performance at the target location point.

[0050] Preferably, in one embodiment of the present invention, the method for obtaining the voltage confidence level of the target location point at each time step specifically includes: First, the average voltage data of the target location point at all times is taken as the overall voltage value of the target location point. The absolute value of the difference between the voltage data of the target location point at each time and the overall voltage value of the target location point is taken as the voltage deviation value of the target location point at each time.

[0051] Then, according to the time sequence, the sequence of water flow influence at all times for each location point other than the water-rich location point is taken as the water flow influence sequence for each location point other than the water-rich location point. The average value of the water flow influence sequence for all location points other than the water-rich location point is taken as the reference sequence. The average value of the water flow influence sequence for all location points other than the water-rich location point is calculated by averaging the elements of all water flow influence sequences at the same position to obtain the reference sequence.

[0052] The similarity of water flow influence at the target location is obtained based on the difference between the sequence of water flow influence at the target location and the reference sequence.

[0053] Preferably, in one embodiment of the present invention, the method for obtaining the similarity of the water flow influence at the target location specifically includes: The absolute value of the cosine similarity between the sequence of water flow influence at the target location and the reference sequence is taken as the water flow influence similarity at the target location.

[0054] Finally, the similarity of the water flow influence at the target location point is used as the numerator, and the sum of the voltage deviation value of the target location point at each time moment and the preset third adjustment parameter is used as the denominator. The values ​​are then normalized by negative correlation, and the calculation results are limited to... Within the range, the voltage confidence level of the target location point at each time moment is obtained.

[0055] In one embodiment of the present invention, a negative exponential function with the natural constant e as the base can be used to achieve the normalization of negative correlation. Furthermore, the normalization of negative correlation in subsequent steps can all be performed using a negative exponential function with the natural constant e as the base.

[0056] As an example, in one embodiment of the present invention, the expression for the voltage confidence level of the target location point at each time step can be specifically as follows:

[0057] in, Indicates the target location point at the th Voltage confidence at each moment; This indicates the similarity of the water flow influence at the target location point; Indicates the target location point at the th Voltage data at each moment; This represents the overall voltage value at the target location point; Indicates the target location point at the th Voltage deviation value at each moment; Represented by natural constant An exponential function with base 0 is used for normalization of negative correlations; This indicates a preset third adjustment parameter, used to prevent the denominator from being 0. The range of values ​​is In one embodiment of the present invention, the following is used: Set to 0.01, The specific values ​​can be set by the implementer according to the specific implementation scenario, and are not limited here.

[0058] in, This is used to measure the relative relationship between the temporal variation of the impact of water flow at a target location and the anomalies in voltage data. The larger the value, the greater the impact of the overflowing water flow on the anomaly, and the lower the confidence level of the voltage data at the target location.

[0059] After obtaining the voltage confidence level of the target location at each moment, the voltage acquisition frequency of the target location can be adjusted based on the voltage confidence level at each moment within each time period. This yields the adjusted acquisition frequency of the target location at each time period. Subsequently, voltage data of the target location can be acquired based on the adjusted acquisition frequency of each time period, increasing the amount of voltage data acquired and improving the proportion of fuzzy judgments in the dataset of the artificial intelligence algorithm. This enhances the algorithm's ability to judge this part of the situation, thereby accurately evaluating the permeability performance of the geomembrane at the target location. The length of each time period is typically 1 to 5 minutes. In one embodiment of the present invention, the length of the time period is set to 2 minutes. The specific length of the time period can also be set by the implementer according to the specific implementation scenario and is not limited here.

[0060] Preferably, in one embodiment of the present invention, the method for obtaining the adjusted acquisition frequency of the target location point in each time period specifically includes:

[0061] The lower the overall voltage confidence level within a time period, the greater the need to adjust the voltage data acquisition frequency. Therefore, the average voltage confidence level of the target location point at all times within each time period can be negatively correlated and normalized to limit the calculation results to... Within the range, the frequency adjustment weight of the target location point in each time period is obtained.

[0062] As an example, in one embodiment of the present invention, the expression for adjusting the frequency weight of the target location point in each time period can be specifically as follows:

[0063] in, Indicates the target location point at the th Frequency adjustment weights for each time period; Indicates the target location point at the th The first time period Voltage confidence at each moment; This indicates the number of moments contained in each time period, where all time periods contain the same number of moments. Represented by natural constant An exponential function with base 1 is used for normalization of negative correlations.

[0064] Furthermore, the frequency of voltage acquisition at the target location point in each time period can be adjusted based on the frequency adjustment weight, thereby obtaining the adjusted acquisition frequency of the target location point in each time period.

[0065] Preferably, in one embodiment of the present invention, the method for obtaining the adjusted acquisition frequency of the target location point in each time period further includes: The product of the frequency adjustment weight of the target location point in each time period and the standard voltage acquisition frequency is used as the frequency adjustment amount of the target location point in each time period. In order to accurately test and evaluate the permeability performance of the geomembrane, it is necessary to increase the acquisition frequency of voltage data. Therefore, the sum of the standard voltage acquisition frequency and the frequency adjustment amount can be used as the adjustment acquisition frequency of the target location point in each time period. The standard voltage acquisition frequency is a built-in parameter of the sensor and is a known value.

[0066] As an example, in one embodiment of the present invention, the expression for adjusting the sampling frequency of the target location point in each time period can be specifically as follows:

[0067] in, Indicates the target location point at the th Adjust the sampling frequency for each time period; Indicates the standard voltage sampling frequency; Indicates the target location point at the th Frequency adjustment weights for each time period; Indicates the target location point at the th Frequency adjustment amount for each time period.

[0068] The seepage prevention performance evaluation module 104 is used to collect voltage data at target locations based on an adjusted acquisition frequency, and to evaluate the seepage prevention performance of the geomembrane using the collected voltage data.

[0069] After obtaining the adjusted acquisition frequency of the target location point in each time period, the voltage of the target location point can be acquired based on the adjusted acquisition frequency, and the acquired voltage data can be used to accurately evaluate the seepage prevention performance of the geomembrane.

[0070] Preferably, in one embodiment of the present invention, the method for evaluating the impermeability of the geomembrane specifically includes: Voltage data of the target location point within the corresponding time period, collected using an adjusted acquisition frequency, is input into a binary convolutional neural network. The binary convolutional neural network then outputs whether the geomembrane has permeated at the target location point. The binary convolutional neural network outputs two results: yes or no, representing whether the geomembrane has permeated or not at the target location point, respectively.

[0071] The same method described above can be used to test and evaluate the permeability of the geomembrane at every location except for locations with high water concentration.

[0072] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0073] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A geomembrane seepage prevention performance testing and evaluation system based on artificial intelligence, characterized in that, The system includes: The data acquisition module is used to acquire the water content and voltage data of each location point of the geomembrane at each time. The water flow impact analysis module is used to select water-rich locations from all locations based on the water content at each location; and to take any location other than the water-rich locations as the target location. Based on the distance between the target location and the water-rich locations, and the difference in water content between the target location and the water-rich locations at the same time, the module obtains the water flow diffusion coefficient of the target location at each time. Based on the difference in the water flow diffusion coefficient between the target location and the location to be measured at the same time, the module obtains the local water flow condition coefficient of the target location. Based on the difference in voltage data between the target location at each time and the next adjacent time, and the local water flow condition coefficient, the module obtains the degree of water flow impact of the target location at each time. The acquisition frequency adjustment module is used to obtain the voltage confidence of the target location point at each moment based on the difference in voltage data at each moment of the target location point and the temporal difference in the degree of water flow influence between the target location point and other location points except for the water-rich location point; and to adjust the voltage acquisition frequency of the target location point based on the voltage confidence of the target location point at each moment of each time period to obtain the adjusted acquisition frequency of the target location point in each time period. The seepage prevention performance evaluation module is used to collect the voltage at the target location point based on the adjusted acquisition frequency, and to evaluate the seepage prevention performance of the geomembrane using the collected voltage data.

2. The artificial intelligence-based geomembrane seepage prevention performance testing and evaluation system according to claim 1, characterized in that, The selection of water-rich locations from all locations includes: The average of the moisture content at each location point at all times is taken as the overall moisture content at each location point; The location point corresponding to the maximum value of the overall moisture content is taken as the moisture enrichment location point.

3. The geomembrane seepage prevention performance testing and evaluation system based on artificial intelligence according to claim 1, characterized in that, The water flow diffusion coefficient at the target location point at each time moment includes: Take any time as the target time, use the distance between the target location and the water-rich location as the numerator, and use the difference in water content between the water-rich location and the target location at the target time and the sum of the preset first adjustment parameter as the denominator. Then, normalize the comparison values ​​to obtain the water flow diffusion coefficient of the target location at the target time.

4. The geomembrane seepage prevention performance testing and evaluation system based on artificial intelligence according to claim 1, characterized in that, The local flow condition coefficients at the target location point include: The average value of the water flow diffusion coefficient at all locations except the water enrichment location at each time moment is taken as the overall water flow diffusion coefficient at each time moment. The difference between the water flow diffusion coefficient at the target location point at each time moment and the overall water flow diffusion coefficient is taken as the diffusion coefficient difference value at the target location point at each time moment. The average value of the diffusion coefficient difference at all times of the target location is normalized to obtain the local flow condition coefficient of the target location.

5. The geomembrane seepage prevention performance testing and evaluation system based on artificial intelligence according to claim 1, characterized in that, The degree of water flow influence at the target location point at each time moment includes: The absolute value of the difference between the voltage data of the target location point at each time and the next adjacent time is used as the voltage change of the target location point at each time. The voltage data of the target location point at each time is used as the numerator, and the sum of the voltage change of the target location point at each time and the preset second adjustment parameter is used as the denominator. The ratio is used as the water flow interference coefficient of the target location point at each time. The water flow interference coefficient and the local water flow condition coefficient of the target location point at each time moment are combined and normalized to obtain the degree of water flow influence of the target location point at each time moment.

6. The geomembrane seepage prevention performance testing and evaluation system based on artificial intelligence according to claim 1, characterized in that, The voltage confidence level of the target location point at each time step includes: The average voltage data of the target location at all times is taken as the overall voltage value of the target location. The absolute value of the difference between the voltage data of the target location point at each time moment and the overall voltage value of the target location point is taken as the voltage deviation value of the target location point at each time moment. According to the time sequence, the sequence of the degree of water flow influence of each location point other than the water-rich location point at all times is taken as the sequence of the degree of water flow influence of each location point other than the water-rich location point. The average value of the sequence of water flow influence at all locations except for the water-rich locations is used as the reference sequence; Based on the difference between the water flow influence degree sequence at the target location and the reference sequence, the water flow influence similarity at the target location is obtained; The similarity of the water flow influence at the target location point is used as the numerator, and the sum of the voltage deviation value of the target location point at each time moment and the preset third adjustment parameter is used as the denominator. The values ​​are then compared and normalized with negative correlation to obtain the voltage confidence level of the target location point at each time moment.

7. The geomembrane seepage prevention performance testing and evaluation system based on artificial intelligence according to claim 6, characterized in that, The similarity of the water flow influence at the target location point includes: The absolute value of the cosine similarity between the sequence of water flow influence at the target location and the reference sequence is taken as the water flow influence similarity at the target location.

8. The geomembrane seepage prevention performance testing and evaluation system based on artificial intelligence according to claim 1, characterized in that, The adjustment of the acquisition frequency of the target location point in each time period includes: The average voltage confidence of the target location point at all times within each time period is negatively correlated and normalized to obtain the frequency adjustment weight of the target location point in each time period. Based on the frequency adjustment weight, the frequency of voltage acquisition at the target location point in each time period is adjusted to obtain the adjusted acquisition frequency of the target location point in each time period.

9. The artificial intelligence-based geomembrane seepage prevention performance testing and evaluation system according to claim 8, characterized in that, The step of adjusting the frequency of voltage acquisition at the target location point in each time period based on the frequency adjustment weight, to obtain the adjusted acquisition frequency of the target location point in each time period, includes: The product of the frequency adjustment weight of the target location point in each time period and the standard voltage acquisition frequency is used as the frequency adjustment amount of the target location point in each time period. The sum of the standard voltage acquisition frequency and the frequency adjustment amount is used as the adjusted acquisition frequency for the target location point in each time period.

10. The artificial intelligence-based geomembrane seepage prevention performance testing and evaluation system according to claim 1, characterized in that, The evaluation of the impermeability of the geomembrane includes: The voltage data of the target location point within the corresponding time period, which is collected using the adjusted acquisition frequency, is input into a binary classification convolutional neural network, and the binary classification convolutional neural network outputs whether the geomembrane has a seepage phenomenon at the target location point.

Images