Groundwater renewal capacity evaluation method and system based on multi-source data fusion

Abstract

The present application relates to the technical field of data processing, and more particularly to a groundwater renewal capacity evaluation method and system based on multi-source data fusion. The method comprises the following steps: obtaining a multi-source heterogeneous data set, processing geological heterogeneity through a spatial interpolation method, calculating the permeability value of each geological unit, and determining the permeability difference characteristics; predicting the spatio-temporal distribution of groundwater flow rate based on the permeability difference characteristics, combining with dynamic hydrological data, adjusting and calibrating the time sequence, and obtaining dynamic calibrated flow rate parameters; using the dynamic calibrated flow rate parameters, integrating the recharge rate and solute migration path information, calculating the stability parameters of hydrological cycle, and generating quantitative indicators of groundwater renewal rate and cycle efficiency; obtaining an optimized evaluation model, integrating multi-source data, calculating the comprehensive evaluation value of groundwater renewal capacity, and determining the final evaluation result of groundwater renewal capacity. The present application can accurately evaluate the groundwater renewal capacity and is suitable for the management and sustainable utilization of groundwater resources in karst areas.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for assessing groundwater renewal capacity based on multi-source data fusion. Background Technology

[0002] Groundwater is a crucial component of water resources, especially in karst regions. However, the complex geological characteristics of karst areas, with highly uneven spatial distribution of fissures, caves, and groundwater conduits, make it difficult to accurately capture the hydrological dynamics of groundwater flow. Traditional groundwater assessment methods typically rely on single hydrological data, such as water level and flow rate, lacking a comprehensive analysis of groundwater velocity and recharge rates in complex geological environments. Furthermore, existing technologies often face data heterogeneity issues when handling multi-source data fusion, making it difficult to effectively integrate data from different sources and time scales, thus affecting the accuracy and reliability of groundwater regeneration capacity assessments.

[0003] To address the aforementioned issues, this invention provides a groundwater regeneration capacity assessment method based on multi-source data fusion. By collecting heterogeneous data from multiple sources, such as geological structure data, dynamic hydrological data, and fracture distribution, it comprehensively analyzes factors such as the spatiotemporal distribution of groundwater, recharge rate, solute migration pathways, and surface hydrological changes. This results in a more accurate and dynamically responsive groundwater regeneration capacity assessment model. This model not only solves the problem of traditional methods neglecting geological heterogeneity and dynamic hydrological changes but also reflects the impact of sudden hydrological events on the groundwater system in real time, thereby improving the management efficiency and sustainability of groundwater resources. Summary of the Invention

[0004] This invention provides a method and system for assessing groundwater regeneration capacity based on multi-source data fusion, which is used to accurately assess the groundwater regeneration capacity of a target area.

[0005] In a first aspect, the present invention provides a method for assessing groundwater renewal capacity based on multi-source data fusion, comprising:

[0006] Step S1: Obtain a multi-source heterogeneous dataset, which includes geological structural data and dynamic monitoring data; based on the multi-source heterogeneous dataset, use spatial interpolation methods to process geological heterogeneity, identify and calculate the permeability values ​​of each geological unit, and determine the permeability difference characteristics between geological units;

[0007] Step S2: Based on the permeability difference characteristics, predict the spatiotemporal distribution of groundwater flow velocity to obtain preliminary flow velocity parameter estimates; further, combine the preliminary flow velocity parameter estimates with dynamic hydrological process data to perform temporal adjustment and calibration of surface factors to obtain dynamically calibrated flow velocity parameters that can reflect the dynamic influence of the surface.

[0008] Step S3: Using the dynamically calibrated flow velocity parameters, integrate the recharge rate and solute migration path information to calculate the stability parameters of the hydrological cycle, and generate quantitative indicators characterizing the groundwater renewal rate and cycle efficiency based on the stability parameters.

[0009] Step S4: Based on the quantitative indicators, iteratively optimize the calibration process of flow velocity parameters and recharge rate to obtain an optimized evaluation model; and integrate multi-source data based on the optimized evaluation model to calculate the comprehensive evaluation value of groundwater regeneration capacity, and finally determine the evaluation result of groundwater regeneration capacity.

[0010] As a preferred embodiment of the present invention, step S1, obtaining a multi-source heterogeneous dataset, includes:

[0011] Geological structural data is collected, and the distribution characteristics of fractures in the target area are extracted based on the collected data. Dynamic monitoring data is collected to obtain the time series information of groundwater level and flow. Based on the distribution characteristics of fractures and the water level and flow information, initial feature values ​​that can reflect the relationship between geology and hydrology are generated. Based on the initial feature values, the geological structural data and dynamic monitoring data are integrated to form a multi-source heterogeneous dataset, wherein the dataset simultaneously contains the spatial characteristics of fracture distribution and the time series characteristics of water level and flow.

[0012] As a preferred embodiment of the present invention, step S1, determining the permeability difference characteristics between geological units, includes:

[0013] Based on multi-source heterogeneous datasets, the spatial distribution boundaries of various geological units in the target area are identified;

[0014] For the identified geological units, spatial interpolation is used to process their uneven distribution; based on the processed geological unit distribution, the permeability value of each geological unit is calculated; the permeability values ​​are compared and analyzed to determine the permeability difference characteristics between geological units, wherein the difference characteristics are used to characterize the permeability variation law between different geological units.

[0015] As a preferred embodiment of the present invention, in step S2, based on the permeability difference characteristics, the spatiotemporal distribution of groundwater flow velocity is predicted to obtain preliminary flow velocity parameter estimates, including:

[0016] The permeability difference characteristics between geological units are obtained. If the permeability difference characteristics exceed a preset threshold, the spatiotemporal changes of groundwater flow velocity are predicted using a random forest algorithm to generate prediction results.

[0017] Based on the prediction results, a spatiotemporal distribution characteristic of the flow velocity is formed to reflect the flow velocity differences between different geological units; according to the spatiotemporal distribution characteristic, a preliminary flow velocity parameter estimate is determined, wherein the preliminary flow velocity parameter estimate is used to characterize the flow pattern of groundwater between different geological units.

[0018] As a preferred embodiment of the present invention, in step S2, the dynamically calibrated flow velocity parameters are determined by combining the preliminary flow velocity parameter estimate with dynamic hydrological process data, including:

[0019] Acquire dynamic hydrological process data, wherein the data includes at least time-series information on rainfall and evaporation;

[0020] The preliminary velocity parameter estimate is combined with the dynamic hydrological process data to adjust the time series. Based on the adjusted time series data, the velocity parameter is calibrated, and the matching degree between the calibration result and the historical monitoring data is ensured through iterative optimization methods.

[0021] Based on the calibration results, the dynamically calibrated flow velocity parameters are determined, wherein the dynamically calibrated flow velocity parameters reflect the influence of surface factors on the flow velocity.

[0022] As a preferred embodiment of the present invention, step S3 includes:

[0023] The dynamically calibrated flow velocity parameters are obtained, and the replenishment rate and solute migration path information related to the flow velocity parameters are integrated to generate a feature set that can characterize the integrity of the hydrological process.

[0024] Based on the dynamically calibrated flow velocity parameters, recharge rate, and solute migration path information, stability parameters of the hydrological cycle are calculated, and a quantitative index of groundwater renewal capacity is generated based on the stability parameters. The quantitative index is used to characterize the renewal rate and circulation efficiency of the groundwater system.

[0025] As a preferred embodiment of the present invention, step S4, obtaining the optimized evaluation model, includes:

[0026] A quantitative index of the groundwater renewal capacity is obtained. Based on the quantitative index, the renewal rate and circulation efficiency of the groundwater system are evaluated. If the quantitative index is lower than a preset threshold, the parameter calibration process is adjusted through an iterative optimization method. Based on the adjusted parameters, the flow velocity parameters and recharge rate are updated. Based on the updated flow velocity parameters and recharge rate, an optimized evaluation model is generated.

[0027] As a preferred embodiment of the present invention, step S4, determining the final groundwater renewal capacity assessment result, includes:

[0028] Based on the optimized evaluation model, the features of multi-source data at different time scales are integrated to generate a unified feature set that can reflect short-term dynamic changes and long-term trends.

[0029] Based on the integrated multi-source data, a comprehensive assessment value of groundwater regeneration capacity is calculated to quantify the overall regeneration level of the groundwater system. Based on the comprehensive assessment value, the final assessment result of groundwater regeneration capacity is determined, wherein the assessment result is used to characterize the long-term regeneration capacity of the groundwater system.

[0030] Secondly, the present invention provides a groundwater regeneration capacity assessment system based on multi-source data fusion, for implementing the above-mentioned method, the system comprising:

[0031] The difference acquisition unit is used to acquire a multi-source heterogeneous dataset, which includes geological structural data and dynamic monitoring data; based on the multi-source heterogeneous dataset, a spatial interpolation method is used to process geological heterogeneity, identify and calculate the permeability values ​​of each geological unit, and determine the permeability difference characteristics between geological units;

[0032] The velocity calibration unit is used to predict the spatiotemporal distribution of groundwater velocity based on the permeability difference characteristics, and obtain preliminary velocity parameter estimates; further, the preliminary velocity parameter estimates are combined with dynamic hydrological process data to perform time-series adjustment and calibration of surface factors, and obtain dynamically calibrated velocity parameters that can reflect the dynamic influence of the surface.

[0033] The quantitative index unit is used to integrate the recharge rate and solute migration path information using the dynamically calibrated flow velocity parameters, calculate the stability parameters of the hydrological cycle, and generate quantitative indicators characterizing the groundwater renewal rate and cycle efficiency based on the stability parameters.

[0034] The evaluation and determination unit is used to iteratively optimize the calibration process of flow velocity parameters and recharge rate according to the quantitative indicators to obtain an optimized evaluation model; and to integrate multi-source data based on the optimized evaluation model to calculate the comprehensive evaluation value of groundwater regeneration capacity, and finally determine the evaluation result of groundwater regeneration capacity.

[0035] Thirdly, the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the above-described method.

[0036] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0037] This invention acquires multi-source heterogeneous datasets, including geological structural data and dynamic monitoring data, and uses spatial interpolation methods to handle the geological heterogeneity of karst areas. It identifies and calculates permeability differences, effectively solving the problem of spatially uneven distribution of fissures, caves, and groundwater channels in karst areas, providing accurate foundational data for subsequent flow velocity prediction. Based on permeability differences, a random forest algorithm is used to predict the spatiotemporal changes in flow velocity, generating preliminary flow velocity parameter estimates. These estimates are then combined with dynamic hydrological data such as rainfall and evaporation for time-series adjustment and calibration, ensuring that the flow velocity parameters reflect the dynamic influence of surface factors. This model accurately captures the short-term dynamic changes and long-term trends of the hydrological cycle. Furthermore, it integrates recharge rate and solute migration path information to calculate water... The system identifies stability parameters for groundwater circulation and generates quantitative indicators characterizing groundwater renewal rate and circulation efficiency. These indicators quantify the overall renewal level of the groundwater system and provide a basis for subsequent optimization. If the quantitative indicator falls below a preset threshold, the parameter calibration process is adjusted through iterative optimization to further optimize the evaluation model and ensure accurate and reliable evaluation results. By integrating multi-source data through the optimized evaluation model, a comprehensive evaluation value of groundwater renewal capacity is calculated, thus completing the final evaluation of groundwater renewal capacity. Through the cooperation of the above technical solutions, data consistency is ensured, the heterogeneity of multi-source data is overcome, and the system can dynamically respond to changes in geological characteristics and hydrological conditions in karst areas, providing strong support for the scientific management of groundwater resources in the region. Attached Figure Description

[0038] Figure 1 This is a flowchart of the groundwater renewal capacity assessment method based on multi-source data fusion in an embodiment of the present invention;

[0039] Figure 2 This is a structural diagram of a groundwater renewal capacity assessment system based on multi-source data fusion in an embodiment of the present invention. Detailed Implementation

[0040] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0041] like Figure 1 The groundwater renewal capacity assessment method based on multi-source data fusion in this embodiment may specifically include:

[0042] Step S1: Obtain a multi-source heterogeneous dataset of the target area, which includes geological structural data and dynamic monitoring data; based on the multi-source heterogeneous dataset, use spatial interpolation methods to process geological heterogeneity, identify and calculate the permeability values ​​of each geological unit, and determine the permeability difference characteristics between geological units;

[0043] In step S1, obtaining the multi-source heterogeneous dataset of the target region includes:

[0044] Geological structural data is collected, and the distribution characteristics of fractures in the target area are extracted based on the collected data. Dynamic monitoring data is collected to obtain the time series information of groundwater level and flow. Based on the distribution characteristics of fractures and the water level and flow information, initial feature values ​​that can reflect the relationship between geology and hydrology are generated. Based on the initial feature values, the geological structural data and dynamic monitoring data are integrated to form a multi-source heterogeneous dataset, wherein the dataset simultaneously contains the spatial characteristics of fracture distribution and the time series characteristics of water level and flow.

[0045] Specifically, in areas with complex topography, due to long-term karst development, fissures, caves, and underground river systems are prevalent in the strata. These geological structures are extremely unevenly distributed, leading to significant differences in permeability between different geological units. Traditional groundwater research methods often rely on a single data source, such as hydrogeological parameters obtained solely from boreholes or dynamic data from water level monitoring. However, in complex environments, a single data source often fails to fully reflect the spatial heterogeneity and temporal dynamics of groundwater flow, resulting in large deviations in permeability parameters and insufficient reliability of the assessment results.

[0046] Therefore, in one embodiment, the process of acquiring a multi-source heterogeneous dataset of the target area can be specifically implemented as follows: Geological exploration is conducted in the target area to collect geological structural data. During the collection process, drilling sampling, geophysical exploration, and electromagnetic surveying are used to obtain information on rock strata structure and fracture location. Based on this, spatial analysis is performed on the collected data to identify the density, direction, and connectivity of fractures. The fracture density is calculated using the number of fractures per unit area. The orientation angle of the fractures is determined using vector analysis methods, and the connectivity between fractures is evaluated using graph theory methods. This forms a fracture distribution characteristic that can characterize the differences in geological units. Regarding dynamic monitoring, water level gauges and flow meters are installed in monitoring wells to collect data in real time. The data on changes in groundwater level and flow rate are recorded at a set frequency to form a time series of water level and flow rate, thereby reflecting the dynamic process of groundwater. Furthermore, the aforementioned fracture distribution characteristics are correlated with the water level and flow rate information. First, the fracture distribution characteristics are numerically converted into vector form; for example, fracture density and orientation angle constitute fracture spatial indicators. Simultaneously, the time series data of water level and flow rate are statistically processed to obtain indicators such as average water level and flow rate fluctuation amplitude. Then, the two types of data are fused using a weighted average method to form initial characteristic values ​​that reflect the interaction between geological structure and hydrological dynamics. The contribution of fracture density to the influence of water level can be reflected by empirical coefficient weighting to ensure the rationality and accuracy of the generated characteristic values.

[0047] Subsequently, the initial feature values ​​are normalized to map values ​​of different dimensions to a unified scale range, avoiding deviations caused by inconsistencies in dimensions during subsequent data integration. The normalized feature values ​​are then integrated with the original geological structure data and dynamic monitoring data to construct a composite dataset with both spatial raster layers and temporal curve layers. This aligns the spatial characteristics of fracture distribution with the temporal characteristics of water level and flow rate in the spatiotemporal dimensions, thereby forming a multi-source heterogeneous dataset that meets the analysis requirements. In large-scale regional applications, this process can also be extended to block integration, where the study area, i.e., the target area, is first divided into multiple sub-blocks, and a local dataset is generated independently within each sub-block. Then, the datasets are merged using a boundary matching algorithm to ensure the continuity and integrity of the overall dataset.

[0048] The above technical solution enables the multi-source heterogeneous dataset to maintain a detailed depiction of spatial distribution characteristics while taking into account the temporal evolution of hydrological dynamics. This provides a reliable data foundation for subsequent permeability calculations, flow velocity predictions, and quantitative assessments of groundwater renewal capacity, thereby improving the applicability and accuracy of the entire method under complex geological and dynamic hydrological conditions in the target area.

[0049] Further, in step S1, the permeability difference characteristics between geological units are determined, including:

[0050] Based on multi-source heterogeneous datasets, the spatial distribution boundaries of various geological units in the target area are identified;

[0051] For the identified geological units, spatial interpolation is used to process their uneven distribution; based on the processed geological unit distribution, the permeability value of each geological unit is calculated; the permeability values ​​are compared and analyzed to determine the permeability difference characteristics between geological units, wherein the difference characteristics are used to characterize the permeability variation law between different geological units.

[0052] Specifically, based on the acquired multi-source heterogeneous dataset, the spatial distribution boundaries of various geological units in the target area are identified by combining geological structural data with dynamic monitoring data. Cluster analysis of fracture distribution characteristics and water level / discharge information allows for grouping of data points, distinguishing between fracture-dense and non-dense areas, and generating a preliminary spatial distribution map to provide a basic grid for subsequent interpolation. Subsequently, based on the geological units identified by the spatial distribution boundaries, spatial interpolation methods are used to address the uneven fracture development and highly heterogeneous spatial distribution in the target area. Kriging interpolation is preferred as the core algorithm. First, a spatial autocorrelation model is constructed by calculating the semi-variogram, and then weighted estimations are performed on unknown points based on this model, allowing for continuous spatial representation of geological attributes. In data-dense areas, the range parameter of the variogram can be set to a smaller value to enhance local adaptability, while in data-sparse areas, the range is appropriately expanded to ensure estimation accuracy. The interpolated geological unit distribution eliminates errors caused by uneven sampling and improves the continuity and rationality of geological boundary delineation.

[0053] Darcy's law is used to calculate the permeability values ​​of each geological unit. The parameters of flow rate, cross-sectional area, and hydraulic gradient are all provided by interpolated distribution data, ensuring that the permeability calculation reflects the local hydrodynamic characteristics. By comparing and analyzing the permeability values ​​obtained from each unit, the permeability difference characteristics between different units can be extracted. For example, statistical measures such as ratio, slope, or standard deviation can be used to characterize the variation law of permeability between units, thereby identifying areas with high variability or potential flow barriers. These differences not only reveal the spatial heterogeneity between geological units at the static level, but can also serve as input parameters for subsequent velocity prediction models. Combined with machine learning algorithms such as random forest models, they can be used to optimize the initial velocity estimation and further improve the accuracy of groundwater renewal capacity assessment.

[0054] The above technical solution, by interpolating and calculating the fracture distribution and hydrological information in multi-source data within a unified spatial framework, realizes the logical transformation from raw data to permeability difference characteristics, thereby ensuring the scientificity and reliability of the assessment process, and providing solid data support for groundwater flow velocity prediction and hydrological cycle stability analysis in the overall technical solution.

[0055] Step S2: Based on the permeability difference characteristics, predict the spatiotemporal distribution of groundwater flow velocity to obtain preliminary flow velocity parameter estimates; further, combine the preliminary flow velocity parameter estimates with dynamic hydrological process data to perform temporal adjustment and calibration of surface factors to obtain dynamically calibrated flow velocity parameters that can reflect the dynamic influence of the surface.

[0056] In step S2, based on the permeability difference characteristics, the spatiotemporal distribution of groundwater flow velocity is predicted to obtain preliminary flow velocity parameter estimates, including:

[0057] The permeability difference characteristics between geological units are obtained. If the permeability difference characteristics exceed a preset threshold, the spatiotemporal changes of groundwater flow velocity are predicted using a random forest algorithm to generate prediction results.

[0058] Based on the prediction results, a spatiotemporal distribution characteristic of the flow velocity is formed to reflect the flow velocity differences between different geological units; according to the spatiotemporal distribution characteristic, a preliminary flow velocity parameter estimate is determined, wherein the preliminary flow velocity parameter estimate is used to characterize the flow pattern of groundwater between different geological units.

[0059] Specifically, in one embodiment, based on the predicted flow velocity parameters, a preliminary flow velocity parameter estimate is obtained. The permeability difference feature obtained in the above steps is compared with a preset threshold. When the permeability difference feature is detected to exceed the preset threshold, it is considered that there is significant heterogeneity between geological units, which can easily lead to drastic changes in groundwater flow velocity in space and time. At this time, the prediction process based on the random forest algorithm is triggered. The random forest algorithm, as an ensemble learning method, extracts features from multi-source heterogeneous datasets and inputs fracture distribution, water level and flow rate, geological coordinates and dynamic monitoring information into the model. During the training process, the model generates multiple sub-samples using bootstrap sampling and randomly selects features at the nodes of each decision tree for splitting, thereby forming differentiated fitting paths between different decision trees. Finally, the spatiotemporal prediction of groundwater flow velocity is achieved by integrating the outputs of the majority trees. This can capture the strong nonlinear relationship in the groundwater system of the target area and effectively improve the robustness of the prediction.

[0060] The predicted results are spatialized and a spatiotemporal distribution feature map of flow velocity is generated through gridding. Spatial interpolation is then used to smooth the discrete data to ensure the continuity of the overall distribution and the rationality of the visualization. In the generated spatiotemporal distribution feature, each grid point corresponds to a flow velocity value, which includes both spatial location attributes and a time series correspondence, thus achieving a unified expression of multi-source heterogeneous data in the flow velocity dimension. Based on this spatiotemporal distribution feature, the average flow velocity parameter and its degree of variation are calculated to determine the preliminary flow velocity parameter estimate. This estimate reflects the variation pattern of groundwater flow velocity between different geological units.

[0061] The above technical solution was used to obtain the causal relationship between the permeability difference characteristics and the random forest prediction results, and a correspondence was established between the prediction output and the preliminary velocity parameter estimate, thus laying a reliable foundation for subsequent dynamic calibration and comprehensive assessment of groundwater renewal capacity.

[0062] Further, in step S2, the dynamically calibrated velocity parameters are determined by combining the preliminary velocity parameter estimates with dynamic hydrological process data, including:

[0063] Acquire dynamic hydrological process data, wherein the data includes at least time-series information on rainfall and evaporation;

[0064] The preliminary velocity parameter estimate is combined with the dynamic hydrological process data to adjust the time series. Based on the adjusted time series data, the velocity parameter is calibrated, and the matching degree between the calibration result and the historical monitoring data is ensured through iterative optimization methods.

[0065] Based on the calibration results, the dynamically calibrated flow velocity parameters are determined, wherein the dynamically calibrated flow velocity parameters reflect the influence of surface factors on the flow velocity.

[0066] Specifically, in one embodiment, during the process of determining dynamically calibrated flow velocity parameters by combining preliminary flow velocity parameter estimates with dynamic hydrological process data, on-site automatic weather stations or other monitoring equipment are used to collect rainfall and evaporation information of the target area in real time and record it in time series form, thereby forming a dynamic hydrological process dataset. To ensure the consistency of data in terms of units, the collected raw data is standardized so that it can be matched with the previously obtained preliminary flow velocity parameter estimates within the same analytical framework. The preliminary flow velocity parameter estimates are used as a baseline sequence and aligned with the aforementioned rainfall and evaporation time series data. The influence of surface hydrological changes on groundwater flow velocity is superimposed on the flow velocity baseline sequence through time series adjustment methods. Preferably, smoothing filtering or weighted moving average methods are used to reflect the flow velocity caused by increased rainfall. Acceleration and enhanced evaporation slow down the flow velocity, generating new adjusted time-series data that accurately reflects the impact of surface hydrological factors on groundwater dynamics. The adjusted time-series data is then fitted to the flow velocity parameters using numerical optimization methods. The least squares method is preferred, combined with iterative optimization techniques. By continuously minimizing the sum of squared errors between predicted values ​​and historical monitoring data, the parameters gradually converge to the optimal solution. During the iteration process, the consistency with the observed data is checked in real time until the preset accuracy requirements are met, ensuring the reliability of the calibration results. Key indicators such as average flow velocity and coefficient of variation are extracted from the calibrated results to quantify the influence of surface factors on groundwater flow velocity. These extracted indicators are used as dynamically calibrated flow velocity parameters, serving as inputs for subsequent recharge rate integration and hydrological cycle stability assessment.

[0067] The above technical solution integrates preliminary velocity estimation with dynamic hydrological process data within a unified time-series framework. By combining the adjusted data sequence with iterative calibration methods, the resulting dynamically calibrated velocity parameters not only inherit the spatial distribution patterns of the preliminary predictions but also incorporate the temporal effects of surface dynamic factors such as rainfall and evaporation. This provides more accurate and dynamically responsive basic parameters for the overall assessment of groundwater renewal capacity.

[0068] Step S3: Using the dynamically calibrated flow velocity parameters, integrate the recharge rate and solute migration path information to calculate the stability parameters of the hydrological cycle, and generate quantitative indicators characterizing the groundwater renewal rate and cycle efficiency based on these stability parameters; specifically including:

[0069] The dynamically calibrated flow velocity parameters are obtained, and the replenishment rate and solute migration path information related to the flow velocity parameters are integrated to generate a feature set that can characterize the integrity of the hydrological process.

[0070] Based on the dynamically calibrated flow velocity parameters, recharge rate, and solute migration path information, stability parameters of the hydrological cycle are calculated, and a quantitative index of groundwater renewal capacity is generated based on the stability parameters. The quantitative index is used to characterize the renewal rate and circulation efficiency of the groundwater system.

[0071] Specifically, in one embodiment, the corrected flow velocity parameters are extracted from the above dynamic calibration steps. These flow parameters have comprehensively considered the influence of surface hydrological factors such as rainfall and evaporation, and can more realistically reflect the dynamic changes of groundwater in the complex environment of karst areas. The recharge rate information and solute migration path information, which are closely related to the flow velocity parameters, are integrated. The recharge rate data comes from multi-source heterogeneous datasets and is extracted based on permeability difference features, while the solute migration path information is obtained through spatial interpolation and geological heterogeneity processing. Through this integration, a characteristic flow velocity parameter corresponding to the flow velocity parameters can be formed on a spatiotemporal scale. The collection of data links flow velocity, recharge, and migration paths within a unified framework, ensuring that subsequent calculations comprehensively reflect the integrity of the groundwater cycle. Using dynamically calibrated flow velocity parameters as a benchmark, the average groundwater cycle time is calculated in conjunction with path length information. The cycle time is then adjusted using the recharge rate, and the variability coefficient of the migration path is introduced to correct the cycle's stability. This yields stability parameters that quantify the cycle's robustness. This not only reflects the inherent coupling relationship between different input variables but also, through error correction and weighting, ensures that the stability parameters can dynamically respond to changes in geological heterogeneity and hydrological conditions.

[0072] Based on the stability parameters, a quantitative index of groundwater renewal capacity is generated. The quantitative index characterizes the overall renewal level and circulation efficiency of the groundwater system by mapping the renewal rate and circulation efficiency to quantifiable values. The renewal rate is determined by the functional relationship between the stability parameters and the efficiency factor, and the efficiency factor is the ratio of the recharge rate to the flow velocity parameter, thus forming a logical closed loop between input and output.

[0073] The above technical solution couples the dynamically calibrated flow velocity parameters, recharge rate, and solute migration path information within a unified computational framework, forming a correspondence between parameters and indicators. This allows the aforementioned quantitative indicators to intuitively reflect the renewal rate and circulation efficiency of the groundwater system in the target area, thus providing reliable data support and methodological assurance for the accurate assessment of groundwater renewal capacity.

[0074] Step S4: Based on the quantitative indicators, iteratively optimize the calibration process of flow velocity parameters and recharge rate to obtain an optimized evaluation model; and integrate multi-source data based on the optimized evaluation model to calculate the comprehensive evaluation value of groundwater regeneration capacity, and finally determine the evaluation result of groundwater regeneration capacity.

[0075] In step S4, the optimized evaluation model is obtained, including:

[0076] A quantitative index of the groundwater renewal capacity is obtained. Based on the quantitative index, the renewal rate and circulation efficiency of the groundwater system are evaluated. If the quantitative index is lower than a preset threshold, the parameter calibration process is adjusted through an iterative optimization method. Based on the adjusted parameters, the flow velocity parameters and recharge rate are updated. Based on the updated flow velocity parameters and recharge rate, an optimized evaluation model is generated.

[0077] Specifically, in one embodiment, a quantitative index of groundwater renewal capacity generated in the preceding steps is obtained. This quantitative index is calculated by integrating dynamically calibrated flow velocity parameters, recharge rate, and solute migration path information, and can reflect the renewal rate and circulation efficiency of the groundwater system under specific geological conditions. When the quantitative index is detected to be lower than a preset threshold, it indicates that the current model does not adequately characterize the groundwater renewal capacity, and there are parameter biases or geological heterogeneity effects that have not been fully captured. Therefore, it is necessary to trigger an iterative optimization process. In the iterative optimization process, the gradient descent algorithm is preferably used. The difference between the quantitative index and the threshold is defined as a loss function, and the gradient of the parameters with respect to the loss is calculated in each iteration. The core variables such as flow velocity parameters and recharge rate are updated according to the gradient information until the loss converges, thereby gradually approaching the optimal solution.

[0078] In scenarios involving significant fluctuations in dynamic monitoring data, the optimization process can also incorporate Monte Carlo simulations to generate parameter distribution samples and apply genetic algorithms to simulate natural selection mechanisms. Through selection, crossover, and mutation operations, the optimal parameter subset is iteratively selected, enabling the optimization to adapt to the uncertainties of multi-source heterogeneous data and the complexity of geological conditions in the target area. The optimized parameters are directly applied to velocity prediction and recharge rate calculation. The velocity parameters are estimated by obtaining the average spatiotemporal distribution of the parameters through a retrained random forest model, while the recharge rate is recalculated based on temporal relationships using surface factors such as rainfall and evaporation, and an optimization factor is applied to ensure consistency with the updated velocity distribution. Using the updated velocity parameters and recharge rate as inputs, an optimized evaluation model is constructed. This model typically employs spatial interpolation methods such as Kriging interpolation to generate a continuous permeability field and further integrates multi-source heterogeneous data to form a complete hydrological cycle description framework.

[0079] The aforementioned technical solution establishes a correspondence between the initial and corrected parameters, gradually aligning the predicted results with the measured monitoring data. The optimized model's output not only reduces biases caused by geological heterogeneity and data uncertainty but also maintains stability and foresight across different time scales. Therefore, the optimized assessment model significantly improves the accuracy and reliability of groundwater renewal capacity assessment, providing a solid technical guarantee for groundwater resource management and sustainable utilization in target areas.

[0080] Further, in step S4, the final groundwater renewal capacity assessment results are determined, including:

[0081] Based on the optimized evaluation model, the features of multi-source data at different time scales are integrated to generate a unified feature set that can reflect short-term dynamic changes and long-term trends.

[0082] Based on the integrated multi-source data, a comprehensive assessment value of groundwater regeneration capacity is calculated to quantify the overall regeneration level of the groundwater system. Based on the comprehensive assessment value, the final assessment result of groundwater regeneration capacity is determined, wherein the assessment result is used to characterize the long-term regeneration capacity of the groundwater system.

[0083] Specifically, in one embodiment, multi-source data is integrated to determine the final groundwater renewal capacity assessment result. This involves extracting the integrated parameter set from the assessment model obtained through iterative optimization. This model already includes permeability differences between geological units, dynamically calibrated flow velocity parameters, and quantitative indicators of groundwater renewal capacity. Geological structural information, dynamic monitoring records, and fracture distribution and water level / flow data from the multi-source heterogeneous dataset are integrated according to different time scales. The short-term scale primarily reflects the immediate impact of sudden rainfall or evaporation events on the groundwater system, while the long-term scale reflects the cumulative effect of geological structural stability and multi-year hydrological trends. During the integration process, spatial interpolation methods are used to correct the uneven distribution of geological heterogeneity to ensure that permeability characteristics remain spatially continuous. Furthermore, a random forest algorithm is triggered to predict flow when a preset threshold is exceeded. The spatiotemporal variation of velocity allows the prediction results to be integrated with information on recharge rate and solute migration pathways, thereby generating a feature set across time scales. To express dynamic monitoring and long-term trends within the same framework, a time-series adjustment method is used to combine preliminary velocity parameters with surface factors such as rainfall and evaporation. A mapping mechanism is then used to uniformly map the dynamically calibrated velocity parameters to different time scales, supporting real-time monitoring at the daily scale while also reflecting trend analysis at the annual scale. This ultimately forms a unified feature representation that possesses both short-term response capability and long-term stability. Based on this feature set, a comprehensive assessment value of groundwater renewal capacity is calculated. This calculation is typically achieved by integrating recharge rate and solute migration pathway information and applying a weighted average. The weighting coefficients are set based on the correlation of different time scales with the circulation process to ensure that the calculation results can balance the impact of short-term disturbances and long-term evolution.

[0084] When the quantitative indicators or comprehensive evaluation values ​​are lower than the threshold, the model parameters are iterated and corrected again until the evaluation values ​​meet the preset stability requirements. The comprehensive evaluation values ​​are compared with the threshold. If they are higher than the threshold, it is determined that the groundwater system has a strong long-term renewal capacity. If they are lower than the threshold, it indicates that there is a risk of insufficient renewal, and further optimization or management measures are needed. The final groundwater renewal capacity evaluation results can comprehensively reflect the overall renewal level and circulation efficiency of the groundwater system in the target area under multi-source data fusion. It not only reveals the dynamic response of the system to short-term hydrological events, but also accurately depicts the long-term evolution trend, thus providing a reliable basis for the scientific management and sustainable utilization of water resources.

[0085] This invention also provides a groundwater regeneration capacity assessment system based on multi-source data fusion, used to implement the above-mentioned method, such as... Figure 2 As shown, the system includes:

[0086] The difference acquisition unit is used to acquire a multi-source heterogeneous dataset, which includes geological structural data and dynamic monitoring data; based on the multi-source heterogeneous dataset, a spatial interpolation method is used to process geological heterogeneity, identify and calculate the permeability values ​​of each geological unit, and determine the permeability difference characteristics between geological units;

[0087] The velocity calibration unit is used to predict the spatiotemporal distribution of groundwater velocity based on the permeability difference characteristics, and obtain preliminary velocity parameter estimates; further, the preliminary velocity parameter estimates are combined with dynamic hydrological process data to perform time-series adjustment and calibration of surface factors, and obtain dynamically calibrated velocity parameters that can reflect the dynamic influence of the surface.

[0088] The quantitative index unit is used to integrate the recharge rate and solute migration path information using the dynamically calibrated flow velocity parameters, calculate the stability parameters of the hydrological cycle, and generate quantitative indicators characterizing the groundwater renewal rate and cycle efficiency based on the stability parameters.

[0089] The evaluation and determination unit is used to iteratively optimize the calibration process of flow velocity parameters and recharge rate according to the quantitative indicators to obtain an optimized evaluation model; and to integrate multi-source data based on the optimized evaluation model to calculate the comprehensive evaluation value of groundwater regeneration capacity, and finally determine the evaluation result of groundwater regeneration capacity.

[0090] The present invention also provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the above-described method.

[0091] In summary, this invention acquires multi-source heterogeneous datasets, including geological structural data and dynamic monitoring data, and uses spatial interpolation methods to handle the geological heterogeneity of karst areas. It identifies and calculates permeability difference characteristics, effectively solving the problem of spatially uneven distribution of fissures, caves, and groundwater channels in karst areas, providing accurate basic data for subsequent flow velocity prediction. Based on permeability difference characteristics, a random forest algorithm is used to predict the spatiotemporal changes in flow velocity, thereby generating preliminary flow velocity parameter estimates. These estimates are then combined with dynamic hydrological data such as rainfall and evaporation for time-series adjustment and calibration, ensuring that the flow velocity parameters reflect the dynamic influence of surface factors. This model can accurately capture the short-term dynamic changes and long-term trends of the hydrological cycle. Furthermore, it integrates recharge rate and solute migration path information to calculate... The system identifies stability parameters for the hydrological cycle and generates quantitative indicators characterizing groundwater renewal rate and cycle efficiency. These indicators quantify the overall renewal level of the groundwater system and provide a basis for subsequent optimization. If the quantitative indicator falls below a preset threshold, the parameter calibration process is adjusted through iterative optimization to further optimize the evaluation model and ensure accurate and reliable evaluation results. By integrating multi-source data through the optimized evaluation model, a comprehensive evaluation value of groundwater renewal capacity is calculated, thus completing the final evaluation of groundwater renewal capacity. The cooperation between the above technical solutions ensures data consistency, overcomes the heterogeneity of multi-source data, and can dynamically respond to changes in geological characteristics and hydrological conditions in karst areas, providing strong support for the scientific management of groundwater resources in the region.

[0092] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for assessing groundwater renewal capacity based on multi-source data fusion, characterized in that, include: Step S1: Obtain a multi-source heterogeneous dataset, which includes geological structural data and dynamic monitoring data; Based on the multi-source heterogeneous dataset, spatial interpolation method is used to process geological heterogeneity, identify and calculate the permeability values ​​of each geological unit, and determine the permeability difference characteristics between geological units. Step S2: Based on the permeability difference characteristics, predict the spatiotemporal distribution of groundwater flow velocity to obtain preliminary flow velocity parameter estimates; further, combine the preliminary flow velocity parameter estimates with dynamic hydrological process data to perform time-series adjustment and calibration on the preliminary flow velocity parameter estimates to obtain dynamically calibrated flow velocity parameters that can reflect the dynamic influence of the land surface. Step S3: Using the dynamically calibrated flow velocity parameters, integrate the recharge rate and solute migration path information to calculate the stability parameters of the hydrological cycle, and generate quantitative indicators characterizing the groundwater renewal rate and cycle efficiency based on the stability parameters. Step S4: Based on the quantitative indicators, iteratively optimize the calibration process of flow velocity parameters and recharge rate to obtain an optimized evaluation model; and integrate multi-source data based on the optimized evaluation model to calculate the comprehensive evaluation value of groundwater regeneration capacity, and finally determine the evaluation result of groundwater regeneration capacity.

2. The method as described in claim 1, characterized in that, In step S1, a multi-source heterogeneous dataset is obtained, including: Geological structural data is collected, and the distribution characteristics of fractures in the target area are extracted based on the collected data. Dynamic monitoring data is collected to obtain the time series information of groundwater level and flow. Based on the distribution characteristics of fractures and the water level and flow information, initial feature values ​​that can reflect the relationship between geology and hydrology are generated. Based on the initial feature values, the geological structural data and dynamic monitoring data are integrated to form a multi-source heterogeneous dataset, wherein the dataset simultaneously contains the spatial characteristics of fracture distribution and the time series characteristics of water level and flow.

3. The method as described in claim 1, characterized in that, In step S1, the permeability difference characteristics between geological units are determined, including: Based on multi-source heterogeneous datasets, the spatial distribution boundaries of various geological units in the target area are identified; For the identified geological units, spatial interpolation is used to process their uneven distribution; based on the processed geological unit distribution, the permeability value of each geological unit is calculated; the permeability values ​​are compared and analyzed to determine the permeability difference characteristics between geological units, wherein the difference characteristics are used to characterize the permeability variation law between different geological units.

4. The method as described in claim 1, characterized in that, In step S2, based on the permeability difference characteristics, the spatiotemporal distribution of groundwater flow velocity is predicted to obtain preliminary flow velocity parameter estimates, including: The permeability difference characteristics between geological units are obtained. If the permeability difference characteristics exceed a preset threshold, the spatiotemporal changes of groundwater flow velocity are predicted using a random forest algorithm to generate prediction results. Based on the prediction results, a spatiotemporal distribution characteristic of the flow velocity is formed to reflect the flow velocity differences between different geological units; according to the spatiotemporal distribution characteristic, a preliminary flow velocity parameter estimate is determined, wherein the preliminary flow velocity parameter estimate is used to characterize the flow pattern of groundwater between different geological units.

5. The method as described in claim 4, characterized in that, In step S2, the dynamically calibrated velocity parameters are determined by combining the preliminary velocity parameter estimates with dynamic hydrological process data, including: Acquire dynamic hydrological process data, which includes at least time series information on rainfall and evaporation; combine the preliminary estimated flow velocity parameters with the dynamic hydrological process data to adjust the time series; calibrate the flow velocity parameters based on the adjusted time series data; ensure the matching degree between the calibration results and historical monitoring data through iterative optimization methods; determine the dynamically calibrated flow velocity parameters based on the calibration results, wherein the dynamically calibrated flow velocity parameters reflect the influence of surface factors on flow velocity.

6. The method as described in claim 1, characterized in that, Step S3 includes: The dynamically calibrated flow velocity parameters are obtained, and the replenishment rate and solute migration path information related to the flow velocity parameters are integrated to generate a feature set that can characterize the integrity of the hydrological process. Based on the dynamically calibrated flow velocity parameters, recharge rate, and solute migration path information, stability parameters of the hydrological cycle are calculated, and a quantitative index of groundwater renewal capacity is generated based on the stability parameters. The quantitative index is used to characterize the renewal rate and circulation efficiency of the groundwater system.

7. The method as described in claim 1, characterized in that, In step S4, the optimized evaluation model is obtained, including: A quantitative index of the groundwater renewal capacity is obtained. The renewal rate and circulation efficiency of the groundwater system are evaluated based on the quantitative index. If the quantitative index is lower than a preset threshold, the flow velocity parameters and recharge rate calibration process are adjusted through an iterative optimization method. The flow velocity parameters and recharge rate are updated according to the adjusted flow velocity parameters and recharge rate. An optimized evaluation model is generated according to the updated flow velocity parameters and recharge rate.

8. The method as described in claim 7, characterized in that, In step S4, the final groundwater renewal capacity assessment results are determined, including: Based on the optimized evaluation model, the features of multi-source data at different time scales are integrated to generate a unified feature set that can reflect short-term dynamic changes and long-term trends. Based on the integrated multi-source data, a comprehensive assessment value of groundwater regeneration capacity is calculated to quantify the overall regeneration level of the groundwater system. Based on the comprehensive assessment value, the final assessment result of groundwater regeneration capacity is determined, wherein the assessment result is used to characterize the long-term regeneration capacity of the groundwater system.

9. A groundwater renewal capacity assessment system based on multi-source data fusion, used to implement the method as described in any one of claims 1-8, characterized in that, The system includes: The difference acquisition unit is used to acquire a multi-source heterogeneous dataset, which includes geological structural data and dynamic monitoring data; based on the multi-source heterogeneous dataset, a spatial interpolation method is used to process geological heterogeneity, identify and calculate the permeability values ​​of each geological unit, and determine the permeability difference characteristics between geological units; The velocity calibration unit is used to predict the spatiotemporal distribution of groundwater velocity based on the permeability difference characteristics, and obtain preliminary velocity parameter estimates; further, the preliminary velocity parameter estimates are combined with dynamic hydrological process data to perform time-series adjustment and calibration on the preliminary velocity parameter estimates, and obtain dynamically calibrated velocity parameters that can reflect the dynamic influence of the land surface. The quantitative index unit is used to integrate the recharge rate and solute migration path information using the dynamically calibrated flow velocity parameters, calculate the stability parameters of the hydrological cycle, and generate quantitative indicators characterizing the groundwater renewal rate and cycle efficiency based on the stability parameters. The evaluation and determination unit is used to iteratively optimize the calibration process of flow velocity parameters and recharge rate according to the quantitative indicators to obtain an optimized evaluation model; and to integrate multi-source data based on the optimized evaluation model to calculate the comprehensive evaluation value of groundwater regeneration capacity, and finally determine the evaluation result of groundwater regeneration capacity.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the method as described in any one of claims 1-8.

Images