Soil hydrological characteristic detection method and system

Abstract

The invention relates to the technical field of environment detection, in particular to a soil hydrological characteristic detection method and system.The method comprises the steps that a target detection area is divided, a soil sample material is collected, an in-situ state is kept and analyzed to obtain contained multi-dimensional parameters, and to-be-analyzed sample data is generated; establishing a characteristic analysis engine, inputting the to-be-analyzed sample data, and extracting basic hydrological data according to a preset basic analysis index; acquiring sample environment parameters, and generating a hydrological environment mapping set with the basic hydrological data; establishing a structural bidirectional network according to a characteristic analysis engine and the hydrological environment mapping set, generating a hydrological characteristic detection model, and calculating hydrological characteristic data of the soil sample material; and obtaining historical comparison data, performing arrangement analysis based on the time sequence, the target detection area and the hydrological characteristic data, and generating a hydrological characteristic detection report. According to the invention, the problems of insufficient precision, difficulty in dynamic change capture, non-uniform spatial distribution and the like in traditional soil hydrological characteristic detection are effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental detection, and in particular to a soil hydrological characteristic detection method and system. Background Art

[0002] Soil hydrological property detection technology has important applications in agriculture, environmental monitoring, engineering construction and other fields. Soil hydrological properties, such as moisture content, permeability, saturated hydraulic conductivity and porosity, directly affect decisions on soil water migration, irrigation efficiency, soil improvement and water resource management.

[0003] Traditional soil hydrological property detection methods usually rely on laboratory analysis and field sampling, which have problems such as complex operation, limited data accuracy and uneven spatial distribution. Existing technologies still have limitations in data fusion, spatiotemporal property analysis and soil hydrological behavior simulation, making it difficult to effectively deal with the interactive effects of multiple factors in complex environments.

[0004] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present disclosure, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art known to those skilled in the art. Summary of the invention

[0005] The present invention provides a soil hydrological characteristic detection method and system, which can effectively solve the problems in the background technology.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is: A soil hydrological property detection method, the method comprising: Divide the target detection area and collect soil sample materials, keep the soil sample materials in an in-situ state and analyze and obtain the multi-dimensional parameters contained therein, and generate sample data to be analyzed; Establishing a characteristic analysis engine, inputting the sample data to be analyzed into the characteristic analysis engine and extracting basic hydrological data according to preset basic analysis indicators; Acquire sample environmental parameters, and establish a multi-dimensional environmental interactive mapping with the basic hydrological data to generate a hydrological environment mapping set; Establishing a structural bidirectional network according to the characteristic analysis engine and the hydrological environment mapping set, generating a hydrological characteristic detection model based on the structural bidirectional network, and calculating the hydrological characteristic data of the soil sample material; Obtain historical comparison data, perform arrangement analysis based on the time sequence and the target detection area and the hydrological characteristic data, and generate a hydrological characteristic detection report.

[0007] Furthermore, a structural bidirectional network is established according to the characteristic analysis engine and the hydrological environment mapping set, including: Correcting the hydrological environment mapping set according to the spatiotemporal parameters, and generating a high-dimensional feature set using the spatiotemporal parameters as accompanying features; Based on the high-dimensional feature set, design and train a feature predictor and a feature evaluator; Adjusting the characteristic predictor according to the loss function feedback output by the characteristic evaluator, and generating hydrological prediction data that meets the soil hydrological environmental conditions based on the basic hydrological data; Data adversarial training is performed based on the hydrological prediction data, and the structural bidirectional network is generated through game optimization of the characteristic predictor and the characteristic evaluator.

[0008] Furthermore, data adversarial training is performed based on the hydrological prediction data, including: Inputting the high-dimensional feature set into the characteristic predictor to generate initial hydrological prediction data, and combining with the hydrological environment mapping set, analyzing the deviation of the initial hydrological prediction data in the core characteristics to obtain the deviation analysis result; generating a set of optimized hydrological prediction data according to the deviation analysis results and the characteristic predictor, wherein the optimized hydrological prediction data covers a temporal and spatial dynamic change pattern; Inputting the optimized hydrological prediction data into the characteristic evaluator, and combining the spatiotemporal parameters in the hydrological environment mapping set to detect and correct the hydrological characteristic data; A bidirectional optimization mechanism is established between the characteristic predictor and the characteristic evaluator, and the parameters of the characteristic predictor are adjusted by feedback of the detection correction result to generate the structural bidirectional network to detect the soil hydrological characteristics in the target area.

[0009] Furthermore, the soil hydrological characteristics in the target area are detected, including: Based on the optimized hydrological prediction data, key hydrological characteristic parameters in the target detection area are calculated, and an initial hydrological characteristic distribution matrix is ​​generated; Combined with the spatiotemporal parameters in the hydrological environment mapping set, the initial hydrological characteristic distribution matrix is ​​subjected to spatiotemporal analysis to extract the spatiotemporal dynamic characteristics of the hydrological characteristics in the target detection area; Conduct multi-dimensional coupling analysis on the key hydrological characteristic parameters and the spatiotemporal dynamic characteristics to construct a hydrological behavior model, which reflects the changing laws of soil hydrological characteristics under dynamic environmental conditions; Based on the hydrological behavior model, key hot spots of hydrological characteristics in the target detection area are identified, including high permeability areas, low water holding capacity areas, and high-frequency water evaporation areas; Generate the hydrological characteristic detection report in the key hotspot area, the hydrological characteristic detection report includes the spatiotemporal distribution diagram of key hydrological characteristic parameters, the dynamic change trend curve and the hydrological hotspot area analysis results.

[0010] Furthermore, a hydrological environment mapping set is generated, including: Performing data fusion on the sample environmental parameters and the basic hydrological data, calculating the permeability coefficient, and extracting the permeability weight of the permeability coefficient to the temporal and spatial changes; Performing multi-dimensional interactive calculation on the sample environmental parameters and the permeability coefficient in the target detection area to generate a multi-dimensional parameter interaction matrix, wherein the multi-dimensional parameter interaction matrix represents the influence of permeability behavior on the change of hydrological characteristics; According to the multidimensional parameter interaction matrix, combined with the spatiotemporal characteristics, a mapping relationship between the sample environmental parameters and basic hydrological data based on the permeability coefficient is established to generate an initial hydrological environment mapping set; The permeability characteristic data of the abnormal points in the initial hydrological environment mapping set are corrected and optimized to generate a hydrological environment mapping set, which includes hydrological characteristic distribution data and dynamic change rules dominated by permeability.

[0011] Further, the permeability coefficient is calculated, including: Based on the sample environmental parameters and the basic hydrological data, the physical properties of the soil are decomposed to obtain characteristic parameters related to the permeability coefficient to form a parameter characteristic set; According to the physical statistical analysis method, the influence weight of each characteristic parameter in the parameter characteristic set on the penetration behavior is calculated, and a penetration characteristic weight matrix is ​​constructed based on the influence weight; Based on the permeability characteristic weight matrix, the target detection area is divided into a plurality of sub-areas, and an equivalent permeability field is generated in each of the sub-areas, wherein the equivalent permeability field characterizes the change trend and local characteristics of the influencing weights in different areas; Perform spatial coupling calculation according to the equivalent penetration field, integrate the influence weights of the sub-areas, and generate the spatial distribution penetration sub-coefficient of the entire target detection area; The key area where the permeability behavior occurs is identified, and the permeability coefficient is integrated and generated according to the spatial distribution of the permeability sub-coefficients in the key area.

[0012] Furthermore, a feature analysis engine is established, including: Analyze the sample data to be analyzed, screen out hydrological related parameters related to hydrological characteristics, the hydrological related parameters include soil moisture content, permeability, saturated hydraulic conductivity and porosity, and form a hydrological related parameter set; Based on the hydrological related parameter set, analyzing the association rules between the parameters and constructing a parameter association matrix; Calculate the cumulative value of the hydrological characteristics distribution of the target detection area by the hydrological related parameters according to the parameter association matrix, and generate a cumulative distribution table; Based on the hydrological related parameter set, the parameter association matrix and the cumulative allocation table, a characteristic analysis engine is established, and the characteristic analysis engine is used to simulate the changes in hydrological characteristics in the target detection area, including the water migration trend and the changes in the infiltration behavior.

[0013] Further, performing an arrangement analysis based on the time sequence and the target detection area and the hydrological characteristic data includes: Arrange the historical hydrological characteristic data and the sample data to be analyzed in the target detection area according to the time sequence, and reconstruct the time sequence in sections according to the fluctuation amplitude of the hydrological characteristics to form a time node set; Based on the time node set, analyzing the characteristic change trend of the hydrological characteristic data, the characteristic change trend including the permeability change rate, the phase characteristics of the water migration direction and the fluctuation law of the evaporation equilibrium point; Based on the hydrological characteristic data, identifying the mutual influence between permeability, water content and saturated hydraulic conductivity at the time nodes, and extracting the coordinated change characteristics of the parameters at different time nodes; Identifying a hydrological response pattern within the target detection area according to the parameter coordinated variation characteristics and the characteristic variation trend; The hydrological characteristic detection report is generated by integrating the results of the permutation analysis according to the hydrological response mode.

[0014] A soil hydrological characteristic detection system, the system comprising: The data acquisition module divides the target detection area and collects soil sample materials, maintains the in-situ state of the soil sample materials and analyzes and obtains the multi-dimensional parameters contained therein to generate sample data to be analyzed; The characteristic analysis module establishes a characteristic analysis engine, inputs the sample data to be analyzed into the characteristic analysis engine and extracts basic hydrological data according to the preset basic analysis indicators; Environmental mapping module, which obtains sample environmental parameters and establishes multi-dimensional environmental interactive mapping with basic hydrological data to generate a hydrological environmental mapping set; A network detection module, which establishes a structural bidirectional network according to the characteristic analysis engine and the hydrological environment mapping set, generates a hydrological characteristic detection model based on the structural bidirectional network, and calculates the hydrological characteristic data of the soil sample material; The report generation module obtains historical comparison data, performs arrangement analysis based on the time series and target detection area and hydrological characteristic data, and generates a hydrological characteristic detection report.

[0015] Furthermore, the network detection module includes: The spatiotemporal correction unit corrects the hydrological environment mapping set according to the spatiotemporal parameters and generates a high-dimensional feature set using the spatiotemporal parameters as accompanying features; The tool training unit designs and trains feature predictors and feature evaluators based on high-dimensional feature sets; A data prediction unit adjusts the characteristic predictor according to the loss function feedback output by the characteristic evaluator, and generates hydrological prediction data that meets the soil hydrological environmental conditions based on the basic hydrological data; The adversarial optimization unit performs data adversarial training based on hydrological prediction data and generates a structural bidirectional network through game optimization of the characteristic predictor and the characteristic evaluator.

[0016] The technical solution of the present invention can achieve the following technical effects: Through the in-depth integration of soil sample data and environmental parameters, and based on the spatiotemporal correction of the structural bidirectional network and permeability coefficient, the problems of insufficient accuracy, difficulty in capturing dynamic changes and uneven spatial distribution in traditional soil hydrological characteristics detection are solved. The changing patterns of soil hydrological characteristics in complex environments are simulated and predicted more accurately, and efficient monitoring and dynamic analysis of the hydrological characteristics of the target detection area are achieved.

[0017] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 It is a schematic diagram of a process of detecting soil hydrological characteristics; Figure 2 It is a structural schematic diagram of a bidirectional network; Figure 3 Schematic diagram of the process for generating a hydrological environment mapping set; Figure 4 It is a structural diagram of the characteristic analysis engine; Figure 5 Schematic diagram of the process for generating a hydrological characteristics test report. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0022] Embodiment 1; like Figure 1 As shown, the present application provides a method for detecting soil hydrological characteristics, the method comprising: S10: Divide the target detection area and collect soil sample materials, keep the soil sample materials in an in-situ state and analyze and obtain the multi-dimensional parameters contained therein, and generate sample data to be analyzed; S20: Establishing a characteristic analysis engine, inputting the sample data to be analyzed into the characteristic analysis engine and extracting basic hydrological data according to preset basic analysis indicators; S30: Obtain sample environmental parameters, and establish a multi-dimensional environmental interactive mapping with basic hydrological data to generate a hydrological environment mapping set; S40: establishing a structural bidirectional network according to the characteristic analysis engine and the hydrological environment mapping set, generating a hydrological characteristic detection model based on the structural bidirectional network, and calculating hydrological characteristic data of the soil sample material; S50: Obtain historical comparison data, perform arrangement analysis based on the time series and target detection area and hydrological characteristic data, and generate a hydrological characteristic detection report.

[0023] Specifically, during the implementation process, the distribution, topography and hydrological conditions of the soil are first analyzed through remote sensing technology, geographic information system (GIS) or on-site investigation, and a suitable target detection area is selected. For example, in agricultural soil monitoring, the target area may be a specific area of ​​farmland. After the target detection area is divided, several typical sub-areas are selected for sampling. The collection of soil samples requires the use of professional soil samplers, and the sampling depth is generally 20-50 cm to obtain representative samples. The soil is kept in situ during sampling to avoid water evaporation or damage to the soil structure; the collected soil samples are treated in situ to maintain the natural state of the soil such as moisture and gas composition, and the physical and chemical properties of the soil samples are analyzed. Conventional instruments such as gas chromatographs and mass spectrometers are used to measure multidimensional parameters such as soil moisture content, porosity, permeability, saturated hydraulic conductivity, soil texture, etc. The analyzed data are input into the computer through the data acquisition system to generate sample data to be analyzed; the preset analysis model and algorithm are used to input the sample data to be analyzed, and the basic water content is extracted from it using the characteristic analysis engine. The characteristic analysis engine extracts basic hydrological data such as soil permeability, water conductivity, and water retention capacity according to the set basic analysis indicators. The extraction process is to input the sample hydrological data into the analysis engine, and extract characteristic parameters related to soil hydrological behavior through statistical methods such as multiple regression analysis and principal component analysis (PCA); obtain sample environmental parameters (such as temperature, humidity, air pressure, precipitation, etc.), which will affect the hydrological behavior of the soil. Through data fusion technology, these environmental parameters are analyzed with the basic hydrological data extracted from the characteristic analysis engine in a multi-dimensional interactive manner, and the sample environmental parameters and hydrological data are standardized to ensure the comparability of the data. Machine learning algorithms (such as support vector machines SVM or neural networks) are used to establish hydrological environmental interactive mapping, identify the changing laws of hydrological characteristics under different environmental conditions, and the generated hydrological environmental mapping set includes the influence of environmental parameters on the distribution and dynamic changes of hydrological characteristics; using the hydrological environmental mapping set and the characteristic analysis engine, a structural bidirectional network is constructed, and the structural bidirectional network is trained multiple times through data input and reverse optimization.The purpose is to generate an accurate hydrological characteristic detection model that can predict and simulate the changes in the hydrological characteristics of the soil in the target detection area. The structural two-way network first inputs the environmental mapping set and basic hydrological data, optimizes the network parameters through the back propagation algorithm, and generates the hydrological characteristic detection model of the target area based on the trained structural two-way network; obtain historical hydrological data, including past precipitation, soil moisture content, etc., combine the historical data with the current sample data to be analyzed, arrange them according to the time series, analyze the fluctuation amplitude of the hydrological characteristic data, determine the time node, and reconstruct them in segments in combination with environmental changes, extract the key stages of hydrological characteristic changes, such as infiltration change rate, soil moisture migration direction, evaporation equilibrium point, etc., arrange and reconstruct the hydrological characteristics of the target area through time series analysis, and through the time series arrangement of these data, the change law of hydrological characteristics over time, as well as potential periodic or sudden change trends can be identified; combine the hydrological characteristic data of the target area with historical comparison data, and analyze to obtain a hydrological characteristic detection report. The content of the hydrological characteristic detection report includes a spatiotemporal distribution map, a dynamic change trend curve, and an analysis of hydrological hotspot areas.

[0024] The technical solution of the present invention solves the problems of insufficient accuracy, difficulty in capturing dynamic changes and uneven spatial distribution in traditional soil hydrological characteristics detection, more accurately simulates and predicts the changing patterns of soil hydrological characteristics in complex environments, and realizes efficient monitoring and dynamic analysis of the hydrological characteristics of the target detection area.

[0025] Further, if Figure 2 As shown, a structural bidirectional network is established based on the characteristic analysis engine and the hydrological environment mapping set, including: The hydrological environment mapping set is corrected according to the spatiotemporal parameters, and the spatiotemporal parameters are used as accompanying features to generate a high-dimensional feature set; Based on high-dimensional feature sets, feature predictors and feature evaluators are designed and trained; The characteristic predictor is adjusted according to the loss function feedback output by the characteristic evaluator, and hydrological prediction data that meets the soil hydrological environmental conditions is generated based on the basic hydrological data; Data adversarial training is performed based on hydrological prediction data, and a structural bidirectional network is generated through game optimization of characteristic predictor and characteristic evaluator.

[0026] As a preferred embodiment of the above, the hydrological environment mapping set generated according to the basic hydrological data and the sample environmental data is subjected to spatiotemporal correction. Specifically, based on the temporal and spatial characteristics of the collected soil samples (such as the sampling time, the geographical coordinates of the sampling location, etc.), the spatial distribution and temporal changes in the mapping set are corrected through the geographic information system (GIS) technology and the time series analysis method. The corrected hydrological environment mapping set can more accurately reflect the dynamic evolution of soil hydrological characteristics. After the correction is completed, a high-dimensional feature set is generated with spatiotemporal parameters as accompanying features (such as the spatiotemporal changes of soil temperature and humidity). The high-dimensional feature set contains the characteristics related to the hydrological environment. The characteristic predictor and characteristic evaluator are designed and constructed based on the high-dimensional feature set. The characteristic predictor uses a deep learning algorithm (such as a multi-layer perceptron for training) to extract and predict the evolution law of soil hydrological characteristics from the high-dimensional feature set. The characteristic evaluator uses a loss function to evaluate the accuracy of the prediction results output by the predictor. The characteristic evaluator continuously adjusts the parameters of the characteristic predictor by comparing it with the actual soil hydrological characteristic data, so that the characteristic predictor can more accurately reflect the hydrological characteristics of the soil; feedback is given through the loss function output by the characteristic evaluator to adjust the characteristic predictor. The goal of the adjustment is to optimize the performance of the characteristic predictor so that it can generate hydrological prediction data that meets the soil hydrological environmental conditions based on basic hydrological data (such as soil type, moisture content, permeability, etc.). These hydrological prediction data will be used to predict the changes in hydrological characteristics under different soil types and environmental conditions, such as predicting soil permeability, water retention capacity, infiltration rate, etc.; according to the generated hydrological prediction data, data adversarial training is performed. Data adversarial training is an optimization method based on generative adversarial networks (GANs). The performance of the model is optimized through the game optimization process of the characteristic predictor and the characteristic evaluator. Specifically, the characteristic predictor generates prediction data, and the characteristic evaluator adjusts the parameters of the predictor through evaluation loss feedback. Through this game process, the characteristic predictor and the characteristic evaluator continuously compete and optimize with each other, and finally generate an efficient and accurate structural bidirectional network; the structural bidirectional network after data adversarial training and optimization can accurately simulate the changes in soil hydrological characteristics, especially for the performance of soil hydrological characteristics under different time and space conditions. The structural bidirectional network can not only be used to detect the current soil hydrological characteristics, but also predict possible changes in hydrological characteristics in the future, thereby providing scientific decision-making support for agricultural irrigation, soil improvement, flood warning, etc.

[0027] Furthermore, data adversarial training based on hydrological prediction data includes: Input the high-dimensional feature set into the characteristic predictor to generate initial hydrological prediction data, and combine it with the hydrological environment mapping set to analyze the deviation of the initial hydrological prediction data in the core characteristics and obtain the deviation analysis results; Generate a set of optimized hydrological prediction data based on the deviation analysis results and the characteristic predictor, the optimized hydrological prediction data covers the dynamic change pattern of time and space; The optimized hydrological prediction data is input into the characteristic evaluator, and the hydrological characteristic data is tested and corrected in combination with the spatiotemporal parameters in the hydrological environment mapping set; A bidirectional optimization mechanism is established between the characteristic predictor and the characteristic evaluator. Through the feedback of the detection and correction results, the parameters of the characteristic predictor are adjusted to generate a structural bidirectional network to detect the soil hydrological characteristics in the target area.

[0028] As a preferred embodiment of the above embodiment, a high-dimensional feature set is input into the characteristic predictor, and the characteristic predictor generates initial hydrological prediction data based on a deep learning algorithm. The initial hydrological prediction data represents a preliminary guess of the soil hydrological characteristics of the target area before optimization. The generated initial hydrological prediction data is compared with the actual hydrological environment mapping set to analyze the deviation of the initial prediction data in core characteristics (such as permeability, moisture content, saturated hydraulic conductivity, etc.). The deviation analysis evaluates the gap between the predicted data and the actual data through error calculation (such as mean square error MSE, mean absolute error MAE, etc.); according to the deviation analysis results, the characteristic predictor generates a set of optimized hydrological prediction data through model adjustment and optimization, and the optimized hydrological prediction data covers the temporal and spatial dynamic change patterns of the soil hydrological characteristics in the target detection area. The optimization process adjusts the weights and parameters in the characteristic predictor to make the prediction results more in line with the actual environmental conditions. For example, the characteristic predictor will consider the hydrological changes at different time points and different spatial locations to generate prediction data that can accurately describe the changes in soil hydrological characteristics; the optimized The hydrological prediction data is input into the characteristic evaluator for further detection and correction. The characteristic evaluator evaluates the optimized prediction data according to the spatiotemporal parameters (such as temperature, humidity, soil type, terrain, etc.) in the hydrological environment mapping set, determines whether the prediction data has deviations, and corrects it according to the spatiotemporal parameters. The evaluation process calculates the gap between the predicted data and the actual observed data by applying the loss function, and fine-tunes the predicted data. A two-way optimization mechanism is established between the characteristic predictor and the characteristic evaluator. Through the feedback loop, the characteristic evaluator feeds back the results of its detection and correction to the characteristic predictor to adjust the parameters and structure of the predictor. The characteristic predictor adjusts its parameters based on these feedbacks to optimize its ability to predict changes in soil hydrological characteristics. The characteristic predictor and the characteristic evaluator continuously interact with each other in the two-way optimization process to optimize the prediction results, and finally generate a structural two-way network to predict soil hydrological characteristics. It can also make real-time adjustments according to different environmental change patterns. In the target detection area, the structural two-way network can accurately detect and predict hydrological characteristics based on hydrological environmental characteristic data at different times and spaces.

[0029] Furthermore, the soil hydrological characteristics in the target area are tested, including: Based on the optimized hydrological prediction data, the key hydrological characteristic parameters in the target detection area are calculated, and the initial hydrological characteristic distribution matrix is ​​generated; Combined with the spatiotemporal parameters in the hydrological environment mapping set, the initial hydrological characteristic distribution matrix is ​​spatiotemporally analyzed to extract the spatiotemporal dynamic characteristics of the hydrological characteristics in the target detection area; The key hydrological characteristic parameters are multi-dimensionally coupled with the spatiotemporal dynamic characteristics to construct a hydrological behavior model, which reflects the changing laws of soil hydrological characteristics under dynamic environmental conditions. Based on the hydrological behavior model, identify the key hot spots of hydrological characteristics in the target detection area, including high permeability areas, low water holding capacity areas and high-frequency water evaporation areas; Generate a hydrological characteristic detection report in the key hotspot area, which includes the spatiotemporal distribution map of key hydrological characteristic parameters, dynamic change trend curve and analysis results of hydrological hotspot areas.

[0030] As a preferred embodiment of the above-mentioned embodiment, the soil hydrological characteristics in the target detection area are calculated according to the optimized hydrological prediction data, and key hydrological characteristic parameters are identified, including soil moisture content, permeability, saturated hydraulic conductivity, porosity, etc. The optimized hydrological prediction data is processed to generate an initial hydrological characteristic distribution matrix, wherein each matrix element represents the value of the hydrological characteristic parameter corresponding to a specific spatial position; the initial hydrological characteristic distribution matrix is ​​combined with the spatiotemporal parameters in the hydrological environment mapping set to perform spatiotemporal analysis, the spatiotemporal parameters include environmental change factors at different time and space scales, such as temperature, humidity, soil type, topography, etc. Through spatiotemporal analysis, the changing trends and patterns of the hydrological characteristic parameters at different time periods and spatial positions are revealed, and the spatiotemporal dynamic characteristics of the hydrological characteristics in the target detection area are extracted, which are specifically manifested in the laws of water migration, changes in permeability, fluctuations in the evaporation equilibrium point, etc.; the key hydrological characteristic parameters are multi-dimensionally coupled with the extracted spatiotemporal dynamic characteristics, and the coupling analysis associates different hydrological characteristic parameters with their spatiotemporal dynamic characteristics through a mathematical model, and further constructs a hydrological behavior model. The hydrological behavior model is responsible for reflecting the dynamic change law of soil hydrological characteristics under changing environmental conditions. It is established based on the mutual relationship and spatiotemporal characteristics between parameters. Based on the hydrological behavior model, the distribution of hydrological characteristics in the target detection area is analyzed to identify the key hot spots. Specifically, the key hot spots include high permeability areas (areas with strong permeability and good drainage), low water holding capacity areas (areas with weak water holding capacity and easy to dry up), and high-frequency water evaporation areas (areas with fast evaporation rate and drastic humidity changes). These areas have important hydrological characteristics and can affect the hydrological cycle and soil moisture conditions of the entire area. According to the identified key hot spots of hydrological characteristics, a hydrological characteristic detection report is generated. The report includes a spatiotemporal distribution map of key hydrological characteristic parameters, showing the spatial distribution and temporal change trend of each key parameter; a dynamic change trend curve, showing the dynamic trend of each hydrological characteristic parameter over time; and the analysis results of hydrological hot spots, highlighting the hydrological characteristics of different hot spots, such as high permeability areas, low water holding capacity areas and high-frequency water evaporation areas.

[0031] Further, if Figure 3 As shown, a hydrological environment mapping set is generated, including: The sample environmental parameters are fused with the basic hydrological data to calculate the permeability coefficient and extract the permeability weight of the permeability coefficient to the temporal and spatial changes; Perform multi-dimensional interactive calculations on sample environmental parameters and permeability coefficients in the target detection area to generate a multi-dimensional parameter interaction matrix that characterizes the influence of permeability behavior on changes in hydrological characteristics. According to the multi-dimensional parameter interaction matrix, combined with the spatiotemporal characteristics, the mapping relationship between the sample environmental parameters and the basic hydrological data based on the permeability coefficient is established to generate the initial hydrological environment mapping set; The permeability characteristic data of the abnormal points in the initial hydrological environment mapping set are corrected and optimized to generate a hydrological environment mapping set, which includes the hydrological characteristic distribution data and dynamic change rules dominated by permeability.

[0032] As a preferred embodiment of the above, the sample environmental parameters (such as soil temperature, humidity, pressure, porosity, etc.) in the target detection area are fused with basic hydrological data (such as soil moisture content, saturated hydraulic conductivity, permeability, etc.), and the two types of data are integrated to obtain a description of soil hydrological characteristics. The permeability coefficient is calculated by weighted average method. The calculation of the permeability coefficient reflects the ability of soil to flow water under different environmental conditions; the influence weight of the permeability coefficient on spatiotemporal changes is extracted, and based on the relationship between soil permeability behavior and environmental conditions (such as temperature, humidity, pressure, etc.), the variation law of the permeability coefficient at different time and space scales and its influence on hydrological characteristics are calculated by regression analysis, and the influence weight shows the dynamic characteristics of the permeability coefficient changing with time and space; multidimensional interactive calculation is performed between the sample environmental parameters and the permeability coefficient, and the relationship between different environmental parameters (such as soil type, porosity, temperature and humidity, etc.) and the permeability coefficient is processed by multivariate statistical analysis. Through interactive calculation, a multidimensional parameter interaction matrix is ​​generated, which reflects the spatiotemporal variation law of the permeability coefficient and also reveals how environmental factors work together. The changes that affect soil hydrological characteristics, each element in the matrix represents the contribution of the permeability coefficient to the changes in hydrological characteristics under specific environmental conditions; based on the generated multidimensional parameter interaction matrix, combined with spatiotemporal characteristics (such as seasonal changes, precipitation conditions, temperature fluctuations, etc.), a mapping relationship between sample environmental parameters and basic hydrological data is established. This mapping relationship is used to describe the dynamic interaction between permeability coefficient and hydrological characteristics. For example, considering the influence of permeability coefficient on water migration, infiltration rate and water retention capacity under different seasons or climatic conditions, combined with these spatiotemporal characteristics, an initial hydrological environment mapping set is generated. The initial hydrological environment mapping set describes the changing trends of permeability coefficient and hydrological characteristics in different time and space dimensions; the abnormal points in the initial hydrological environment mapping set are corrected and optimized. The abnormal points may be caused by data noise, measurement errors or sudden changes in the external environment. They need to be processed through data cleaning, outlier detection and correction algorithms (such as interpolation method, weighted average method, etc.). After optimization, an accurate hydrological environment mapping set is generated, which includes the distribution data of hydrological characteristics dominated by permeability and their dynamic change laws.

[0033] Further, the permeability coefficient is calculated, including: Based on sample environmental parameters and basic hydrological data, the physical properties of the soil are decomposed to obtain characteristic parameters related to the permeability coefficient and form a parameter characteristic set; According to the physical statistical analysis method, the influence weight of each characteristic parameter in the parameter characteristic set on the penetration behavior is calculated, and the penetration characteristic weight matrix is ​​constructed based on the influence weight; Based on the permeability characteristic weight matrix, the target detection area is divided into multiple sub-areas, and an equivalent permeability field is generated in each sub-area. The equivalent permeability field represents the change trend and local characteristics of the influence weights in different areas. Perform spatial coupling calculation based on the equivalent permeability field, integrate the influence weights of the sub-regions, and generate the spatial distribution permeability sub-coefficient of the entire target detection area; The key areas where infiltration behavior occurs are identified, and the permeability coefficient is generated by integration based on the spatial distribution of permeability sub-coefficients in the key areas.

[0034] As a preferred embodiment of the above embodiment, the physical properties of the soil are decomposed according to sample environmental parameters (such as soil temperature, humidity, porosity, etc.) and basic hydrological data (such as moisture content, permeability, saturated hydraulic conductivity, etc.), and characteristic parameters related to the permeability coefficient, such as soil porosity, particle size, soil type, pressure, temperature, etc., are extracted through soil physics analysis to form a parameter characteristic set; Physical statistical analysis methods (such as regression analysis, principal component analysis (PCA) or variance analysis, etc.) are used to quantitatively analyze the influence of each parameter in the parameter characteristic set on the infiltration behavior. By analyzing the correlation between each characteristic parameter and soil permeability, the influence weight of each parameter on the infiltration behavior is calculated, and a permeability characteristic weight matrix is ​​constructed. The permeability characteristic weight matrix reflects the contribution of different physical parameters to the permeability coefficient; based on the permeability characteristic weight matrix, the target detection area is divided into multiple sub-areas. The division of each sub-area is based on different soil physical properties (such as different soil types, humidity, temperature, etc.). In each sub-area, an equivalent infiltration field is generated to represent the local characteristics of the infiltration behavior in the area. The generation of the equivalent permeability field reflects the change of soil permeability under a specific physical environment by modeling and calculating the local characteristics; the local characteristics of the equivalent permeability field are used to perform spatial coupling calculations, and the spatial distribution permeability sub-coefficient of the entire target detection area is generated by integrating the influence weights of each sub-area. The physical differences of each sub-area and the permeability characteristics of the soil in different areas are considered to obtain the overall permeability coefficient distribution of the area; the key areas where permeability occurs are identified through the calculation results of the spatial distribution permeability sub-coefficient. The key areas are usually high permeability areas, low water holding capacity areas or high-frequency water evaporation areas, etc. The final permeability coefficient is generated based on the distributed permeability sub-coefficients of the key areas.

[0035] Further, if Figure 4 As shown, a feature analysis engine is established, including: Analyze the sample data to be analyzed, screen out hydrological parameters related to hydrological characteristics, including soil moisture content, permeability, saturated hydraulic conductivity and porosity, and form a set of hydrological parameters; Based on the set of hydrological related parameters, the association rules between the parameters are analyzed and the parameter association matrix is ​​constructed; According to the parameter association matrix, the cumulative value of the hydrological characteristics distribution of the hydrological related parameters in the target detection area is calculated to generate a cumulative distribution table; Based on the set of hydrological related parameters, parameter association matrix and cumulative distribution table, a characteristic analysis engine is established. The characteristic analysis engine is used to simulate the changes in hydrological characteristics in the target detection area, including water migration trends and changes in infiltration behavior.

[0036] As a preferred embodiment of the above, the sample data to be analyzed are collected and analyzed, and the hydrological related parameters closely related to the hydrological characteristics are screened out. Common hydrological related parameters include soil moisture content, permeability, saturated hydraulic conductivity and porosity, etc., which reflect the soil's water retention capacity, permeability and water flow dynamic characteristics, and are integrated to form a set of hydrological related parameters; after obtaining the set of hydrological related parameters, statistical methods (such as correlation analysis, principal component analysis, etc.) are used to analyze the relationship between these parameters, and a parameter association matrix is ​​constructed by conducting an in-depth analysis of the association rules between the parameters. The parameter association matrix describes the mutual influence and dependence between the various hydrological related parameters, and can reveal which parameters have an impact on the change of hydrological characteristics and which parameters have synergistic or antagonistic relationships; according to the constructed parameter association matrix, the cumulative value of the distribution of hydrological related parameters to the hydrological characteristics of the target detection area is calculated, and the cumulative value of the distribution represents The overall influence of each hydrological related parameter in the target detection area is calculated, reflecting the contribution of each hydrological related parameter to the regional hydrological characteristics. According to the calculation results of the distribution cumulative value, a cumulative distribution table is generated. The cumulative distribution table lists the influence weights of different hydrological related parameters in different regions, as well as their distribution proportions in the overall hydrological characteristics. Based on the set of hydrological related parameters, parameter association matrix and cumulative distribution table, a characteristic analysis engine is established to simulate the changes in hydrological characteristics in the target detection area, including changes in water migration trends and infiltration behaviors. The characteristic analysis engine simulates the changes in hydrological characteristics under different environmental conditions through finite element analysis. Through the characteristic analysis engine, the changes in hydrological characteristics in the target detection area are simulated, such as analyzing the migration trend of water in the soil, determining the distribution of water in different soil layers, simulating changes in infiltration behavior, and studying the fluctuations in soil permeability under different wet conditions.

[0037] Further, if Figure 5As shown, the permutation analysis is performed based on the time series and target detection area and hydrological characteristics data, including: The historical hydrological characteristic data and the sample data to be analyzed in the target detection area are arranged in time sequence, and the time sequence is segmented and reconstructed according to the fluctuation amplitude of the hydrological characteristics to form a time node set; Based on the time node set, the characteristic change trend of the hydrological characteristic data is analyzed, including the phased characteristics of the permeability change rate, the direction of water migration, and the fluctuation law of the evaporation equilibrium point; Based on the hydrological characteristic data, the mutual influence between permeability, water content and saturated hydraulic conductivity at different time nodes is identified, and the coordinated variation characteristics of parameters at different time nodes are extracted; Identify the hydrological response pattern in the target detection area based on the co-variation characteristics of parameters and the trend of characteristic changes; The results of the permutation analysis are integrated according to the hydrological response pattern to generate a hydrological characteristic detection report.

[0038] As a preferred embodiment of the above, the historical hydrological characteristic data and the sample data to be analyzed in the target detection area are arranged in time sequence. The hydrological characteristic data include parameters such as soil moisture content, permeability, and saturated hydraulic conductivity. The arrangement of the time sequence allows the data to be presented in time order, so that the law and trend of its change over time can be analyzed. According to the fluctuation amplitude of the hydrological characteristics, the time series is segmented and reconstructed to divide the time node set. These time nodes represent the change points of the hydrological characteristics in different periods. After determining the time node set, the change trend of the hydrological characteristic data at each time node is analyzed. The characteristic change trend mainly includes three aspects: the permeability change rate, the stage characteristics of the water migration direction, and the fluctuation law of the evaporation equilibrium point. Through the analysis of these trends, the dynamic changes of the hydrological characteristics in the target detection area over time can be revealed. For example, the permeability change rate reflects how the permeability of the soil changes over time, the water migration direction reveals the flow trend of water in the soil, and the fluctuation of the evaporation equilibrium point can reflect the regular changes of water evaporation. At each time node, the mutual influence relationship between the permeability, moisture content and saturated hydraulic conductivity is further identified, and the interaction between these parameters and their coordinated change characteristics are revealed through statistical analysis methods. Extract the coordinated variation characteristics of parameters at different time nodes, and identify which hydrological characteristic parameters have an important influence on the moisture status and infiltration behavior of the soil. For example, the permeability of the soil may change with the increase of moisture content, while the saturated hydraulic conductivity may be affected by the pore structure of the soil. Based on the extracted coordinated variation characteristics of parameters and the aforementioned trend of hydrological characteristic changes, identify the hydrological response pattern in the target detection area. The hydrological response pattern describes the typical response pattern of soil hydrological characteristics under different time and environmental conditions. By analyzing the pattern, the changes in hydrological characteristics under different soil conditions and their responses to environmental factors can be predicted, such as certain Some regions may show strong water mobility, while other regions may show relatively stable water retention characteristics; based on the identified hydrological response pattern, the results of the permutation analysis are integrated to generate a hydrological characteristic detection report. The report includes: spatiotemporal distribution diagrams of key hydrological characteristic parameters, dynamic change trend curves, and hydrological hotspot area analysis results. The spatiotemporal distribution diagram can intuitively display the distribution of hydrological characteristics under different time and space conditions, and the dynamic change trend curve shows the specific trend of hydrological characteristics changing over time. The hydrological hotspot area analysis results point out areas in specific areas where soil hydrological characteristics are particularly sensitive or change dramatically.

[0039] Embodiment 2: Based on the same inventive concept as the soil hydrological characteristic detection method in the aforementioned embodiment, the present invention also provides a soil hydrological characteristic detection system, the system comprising: The data acquisition module divides the target detection area and collects soil sample materials, maintains the in-situ state of the soil sample materials and analyzes and obtains the multi-dimensional parameters contained therein to generate sample data to be analyzed; The characteristic analysis module establishes a characteristic analysis engine, inputs the sample data to be analyzed into the characteristic analysis engine and extracts basic hydrological data according to the preset basic analysis indicators; Environmental mapping module, which obtains sample environmental parameters and establishes multi-dimensional environmental interactive mapping with basic hydrological data to generate a hydrological environmental mapping set; A network detection module, which establishes a structural bidirectional network according to the characteristic analysis engine and the hydrological environment mapping set, generates a hydrological characteristic detection model based on the structural bidirectional network, and calculates the hydrological characteristic data of the soil sample material; The report generation module obtains historical comparison data, performs arrangement analysis based on the time series and target detection area and hydrological characteristic data, and generates a hydrological characteristic detection report.

[0040] The above-mentioned adjustment system in the present invention can effectively implement a soil hydrological characteristic detection method, and the technical effects that can be achieved are as described in the above-mentioned embodiments, which will not be repeated here.

[0041] Specifically, the network detection module includes: The spatiotemporal correction unit corrects the hydrological environment mapping set according to the spatiotemporal parameters and generates a high-dimensional feature set using the spatiotemporal parameters as accompanying features; The tool training unit designs and trains feature predictors and feature evaluators based on high-dimensional feature sets; A data prediction unit adjusts the characteristic predictor according to the loss function feedback output by the characteristic evaluator, and generates hydrological prediction data that meets the soil hydrological environmental conditions based on the basic hydrological data; The adversarial optimization unit performs data adversarial training based on hydrological prediction data and generates a structural bidirectional network through game optimization of the characteristic predictor and the characteristic evaluator.

[0042] Similarly, the above-mentioned optimization schemes for the system can also respectively achieve the optimization effects corresponding to the method in Example 1, which will not be repeated here.

[0043] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application as defined therein, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method for detecting soil hydrological characteristics, characterized in that: The method comprises: Divide the target detection area and collect soil sample materials, keep the soil sample materials in an in-situ state and analyze and obtain the multi-dimensional parameters contained therein, and generate sample data to be analyzed; Establishing a characteristic analysis engine, inputting the sample data to be analyzed into the characteristic analysis engine and extracting basic hydrological data according to preset basic analysis indicators; Acquire sample environmental parameters, and establish a multi-dimensional environmental interactive mapping with the basic hydrological data to generate a hydrological environment mapping set; Establishing a structural bidirectional network according to the characteristic analysis engine and the hydrological environment mapping set, generating a hydrological characteristic detection model based on the structural bidirectional network, and calculating the hydrological characteristic data of the soil sample material; Obtain historical comparison data, perform arrangement analysis based on the time sequence and the target detection area and the hydrological characteristic data, and generate a hydrological characteristic detection report.

2. A soil hydrological characteristics detection method according to claim 1, characterized in that: A structural bidirectional network is established according to the characteristic analysis engine and the hydrological environment mapping set, including: Correcting the hydrological environment mapping set according to the spatiotemporal parameters, and generating a high-dimensional feature set using the spatiotemporal parameters as accompanying features; Based on the high-dimensional feature set, design and train a feature predictor and a feature evaluator; Adjusting the characteristic predictor according to the loss function feedback output by the characteristic evaluator, and generating hydrological prediction data that meets the soil hydrological environmental conditions based on the basic hydrological data; Data adversarial training is performed based on the hydrological prediction data, and the structural bidirectional network is generated through game optimization of the characteristic predictor and the characteristic evaluator.

3. A soil hydrological characteristics detection method according to claim 2, characterized in that: Data adversarial training is performed based on the hydrological prediction data, including: Inputting the high-dimensional feature set into the characteristic predictor to generate initial hydrological prediction data, and combining with the hydrological environment mapping set, analyzing the deviation of the initial hydrological prediction data in the core characteristics to obtain the deviation analysis result; generating a set of optimized hydrological prediction data according to the deviation analysis results and the characteristic predictor, wherein the optimized hydrological prediction data covers a temporal and spatial dynamic change pattern; Inputting the optimized hydrological prediction data into the characteristic evaluator, and combining the spatiotemporal parameters in the hydrological environment mapping set to detect and correct the hydrological characteristic data; A bidirectional optimization mechanism is established between the characteristic predictor and the characteristic evaluator, and the parameters of the characteristic predictor are adjusted by feedback of the detection correction result to generate the structural bidirectional network to detect the soil hydrological characteristics in the target area.

4. A soil hydrological characteristics detection method according to claim 3, characterized in that: Detect soil hydrological characteristics within the target area, including: Based on the optimized hydrological prediction data, key hydrological characteristic parameters in the target detection area are calculated, and an initial hydrological characteristic distribution matrix is ​​generated; Combined with the spatiotemporal parameters in the hydrological environment mapping set, the initial hydrological characteristic distribution matrix is ​​subjected to spatiotemporal analysis to extract the spatiotemporal dynamic characteristics of the hydrological characteristics in the target detection area; Conduct multi-dimensional coupling analysis on the key hydrological characteristic parameters and the spatiotemporal dynamic characteristics to construct a hydrological behavior model, which reflects the changing laws of soil hydrological characteristics under dynamic environmental conditions; Based on the hydrological behavior model, identifying key hot spots of hydrological characteristics in the target detection area, including high permeability areas, low water holding capacity areas and high-frequency water evaporation areas; Generate the hydrological characteristic detection report in the key hotspot area, the hydrological characteristic detection report includes the spatiotemporal distribution diagram of key hydrological characteristic parameters, the dynamic change trend curve and the hydrological hotspot area analysis results.

5. A soil hydrological characteristics detection method according to claim 1, characterized in that: Generate a hydrological environment map set including: Performing data fusion on the sample environmental parameters and the basic hydrological data, calculating the permeability coefficient, and extracting the permeability weight of the permeability coefficient to the temporal and spatial changes; Performing multi-dimensional interactive calculation on the sample environmental parameters and the permeability coefficient in the target detection area to generate a multi-dimensional parameter interaction matrix, wherein the multi-dimensional parameter interaction matrix represents the influence of permeability behavior on the change of hydrological characteristics; According to the multidimensional parameter interaction matrix, combined with the spatiotemporal characteristics, a mapping relationship between the sample environmental parameters and basic hydrological data based on the permeability coefficient is established to generate an initial hydrological environment mapping set; The permeability characteristic data of the abnormal points in the initial hydrological environment mapping set are corrected and optimized to generate a hydrological environment mapping set, which includes hydrological characteristic distribution data and dynamic change rules dominated by permeability.

6. A soil hydrological characteristic detection method according to claim 5, characterized in that: Calculates permeability coefficients, including: Based on the sample environmental parameters and the basic hydrological data, the physical properties of the soil are decomposed to obtain characteristic parameters related to the permeability coefficient to form a parameter characteristic set; According to the physical statistical analysis method, the influence weight of each characteristic parameter in the parameter characteristic set on the penetration behavior is calculated, and a penetration characteristic weight matrix is ​​constructed based on the influence weight; Based on the permeability characteristic weight matrix, the target detection area is divided into a plurality of sub-areas, and an equivalent permeability field is generated in each of the sub-areas, wherein the equivalent permeability field characterizes the change trend and local characteristics of the influencing weights in different areas; Perform spatial coupling calculation according to the equivalent penetration field, integrate the influence weights of the sub-areas, and generate the spatial distribution penetration sub-coefficient of the entire target detection area; The key area where the permeability behavior occurs is identified, and the permeability coefficient is integrated and generated according to the spatial distribution of the permeability sub-coefficients in the key area.

7. A soil hydrological characteristics detection method according to claim 1, characterized in that: Build a feature analysis engine, including: Analyze the sample data to be analyzed, screen out hydrological related parameters related to hydrological characteristics, the hydrological related parameters include soil moisture content, permeability, saturated hydraulic conductivity and porosity, and form a hydrological related parameter set; Based on the hydrological related parameter set, analyzing the association rules between the parameters and constructing a parameter association matrix; Calculate the cumulative value of the hydrological characteristics distribution of the target detection area by the hydrological related parameters according to the parameter association matrix, and generate a cumulative distribution table; Based on the hydrological related parameter set, the parameter association matrix and the cumulative allocation table, a characteristic analysis engine is established, and the characteristic analysis engine is used to simulate the changes in hydrological characteristics in the target detection area, including the water migration trend and the changes in the infiltration behavior.

8. A soil hydrological characteristic detection method according to claim 1, characterized in that: Performing an arrangement analysis based on the time series and the target detection area and the hydrological characteristic data includes: Arrange the historical hydrological characteristic data and the sample data to be analyzed in the target detection area according to the time sequence, and reconstruct the time sequence in sections according to the fluctuation amplitude of the hydrological characteristics to form a time node set; Based on the time node set, analyzing the characteristic change trend of the hydrological characteristic data, the characteristic change trend including the permeability change rate, the phase characteristics of the water migration direction and the fluctuation law of the evaporation equilibrium point; Based on the hydrological characteristic data, identifying the mutual influence between permeability, water content and saturated hydraulic conductivity at the time nodes, and extracting the coordinated change characteristics of the parameters at different time nodes; Identifying a hydrological response pattern within the target detection area according to the parameter coordinated variation characteristics and the characteristic variation trend; The hydrological characteristic detection report is generated by integrating the results of the permutation analysis according to the hydrological response mode.

9. A soil hydrological characteristics detection system, characterized in that: The system comprises: The data acquisition module divides the target detection area and collects soil sample materials, maintains the in-situ state of the soil sample materials and analyzes and obtains the multi-dimensional parameters contained therein to generate sample data to be analyzed; The characteristic analysis module establishes a characteristic analysis engine, inputs the sample data to be analyzed into the characteristic analysis engine and extracts basic hydrological data according to the preset basic analysis indicators; Environmental mapping module, which obtains sample environmental parameters and establishes multi-dimensional environmental interactive mapping with basic hydrological data to generate a hydrological environmental mapping set; A network detection module, which establishes a structural bidirectional network according to the characteristic analysis engine and the hydrological environment mapping set, generates a hydrological characteristic detection model based on the structural bidirectional network, and calculates the hydrological characteristic data of the soil sample material; The report generation module obtains historical comparison data, performs arrangement analysis based on the time series and target detection area and hydrological characteristic data, and generates a hydrological characteristic detection report.

10. The soil hydrological characteristic detection system according to claim 9, characterized in that: The network detection module comprises: The spatiotemporal correction unit corrects the hydrological environment mapping set according to the spatiotemporal parameters and generates a high-dimensional feature set using the spatiotemporal parameters as accompanying features; The tool training unit designs and trains feature predictors and feature evaluators based on high-dimensional feature sets; A data prediction unit adjusts the characteristic predictor according to the loss function feedback output by the characteristic evaluator, and generates hydrological prediction data that meets the soil hydrological environmental conditions based on the basic hydrological data; The adversarial optimization unit performs data adversarial training based on hydrological prediction data and generates a structural bidirectional network through game optimization of the characteristic predictor and the characteristic evaluator.

Images