Method and device for identifying underground water level threshold value corresponding to ground surface deformation

Abstract

The invention discloses a method and a device for identifying an underground water level threshold value corresponding to ground surface deformation. The method comprises the following steps: S1, acquiring underground water levels of different aquifer groups and PS point settlement rates; and S2, according to the underground water levels of the different aquifer groups and the PS point settlement rates, time-space response characteristics of water level changes of the different aquifer groups in multiple time periods and surface deformation are quantified, and response thresholds of the water level changes of the different aquifer groups to the surface deformation are identified. By the adoption of the technical scheme, the space-time response characteristics between underground water level changes and earth surface deformation of different aquifer groups are quantitatively analyzed, the underground water level threshold value is recognized, and a scientific basis is provided for regional land subsidence regulation and control.

Description

Technical Field

[0001] The present invention belongs to the technical field of ground subsidence control, and in particular relates to a method and device for identifying a groundwater level threshold corresponding to surface deformation. Background Art

[0002] Land subsidence is a slow-changing geological disaster caused by compression of soil under the influence of natural and human factors, resulting in a decrease in regional ground elevation. The main cause of land subsidence is excessive groundwater exploitation, which leads to compaction of the aquifer system and thus land subsidence. At present, it is not possible to quantitatively analyze the spatiotemporal response characteristics between groundwater level changes and surface deformation in different aquifer groups, identify groundwater level thresholds, and provide a scientific basis for regional land subsidence regulation. Summary of the invention

[0003] The technical problem to be solved by the present invention is to provide a method and device for identifying a groundwater level threshold corresponding to surface deformation.

[0004] To achieve the above object, the present invention adopts the following technical solution:

[0005] A method for identifying a groundwater level threshold corresponding to surface deformation, comprising:

[0006] Step S1, obtaining the groundwater level of different aquifer groups and the settlement rate of the PS point;

[0007] Step S2: quantify the spatiotemporal response characteristics of water level changes in different aquifer groups and surface deformation in multiple time periods according to the groundwater levels of different aquifer groups and the settlement rates of the PS points, and identify the response thresholds of water level changes in different aquifer groups to surface deformation.

[0008] Preferably, step S2 comprises:

[0009] Step S21, taking the groundwater level of different aquifer groups as an independent variable and the settlement rate of the PS point as a dependent variable, using a random forest machine learning algorithm to obtain the predicted impact of the groundwater level of different aquifer groups on the surface deformation;

[0010] Step S22: According to the predicted impact of groundwater levels in different aquifer groups on surface deformation, the SHAP interpretable machine learning method is used to obtain the spatiotemporal response characteristics of water level changes in different aquifers and surface deformation in multiple time periods, and identify the groundwater level threshold.

[0011] Preferably, step S22 includes:

[0012] Step S221, calculating the Shapley value of the predicted impact of groundwater levels of different aquifer groups on surface deformation;

[0013] Step S222, sorting the importance of the influencing factors according to the Shapley value to obtain the spatiotemporal response characteristics of water level changes and surface deformation of different aquifers;

[0014] Step S223: Based on the Shapley value, using the single factor dependency of SHAP, identify the response thresholds of water level changes in different aquifers to surface deformation.

[0015] The present invention also provides a device for identifying a groundwater level threshold corresponding to surface deformation, comprising:

[0016] Acquisition module, used to obtain groundwater levels and settlement rates of PS points in different aquifer groups;

[0017] The identification module is used to quantify the spatiotemporal response characteristics of water level changes in different aquifer groups and surface deformation in multiple time periods according to the groundwater levels of different aquifer groups and the settlement rates of PS points, and to identify the response thresholds of water level changes in different aquifer groups to surface deformation.

[0018] Preferably, the identification module comprises:

[0019] The first processing unit is used to use the groundwater level of different aquifer groups as the independent variable and the settlement rate of the PS point as the dependent variable, and use the random forest machine learning algorithm to obtain the predicted impact of the groundwater level of different aquifer groups on the surface deformation;

[0020] The second processing unit is used to obtain the spatiotemporal response characteristics of water level changes in different aquifers and surface deformation in multiple time periods and identify the groundwater level threshold based on the predicted impact of groundwater levels in different aquifer groups on surface deformation through the SHAP interpretable machine learning method.

[0021] Preferably, the second processing unit comprises:

[0022] The first processing component is used to calculate the Shapley value of the predicted impact of groundwater levels on surface deformation in different aquifer groups;

[0023] The second processing component is used to sort the importance of influencing factors according to the Shapley value to obtain the spatiotemporal response characteristics of water level changes and surface deformation in different aquifers;

[0024] The third processing component is used to identify the response thresholds of water level changes in different aquifers to surface deformation based on the Shapley value and using the single factor dependency of SHAP.

[0025] The present invention takes the groundwater level of different aquifer groups as the independent variable and the PS point settlement rate as the dependent variable, and adopts the random forest machine learning algorithm and the SHAP interpretable machine learning method to quantify the spatiotemporal response characteristics of the water level changes in different aquifer groups and the surface deformation in multiple time periods, and identifies the response thresholds of the water level changes in different aquifer groups to the surface deformation. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] 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 embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0027] Figure 1 This is a flow chart of a method for identifying a groundwater level threshold corresponding to surface deformation according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0030] Embodiment 1:

[0031] like Figure 1 As shown, an embodiment of the present invention provides a method for identifying a groundwater level threshold corresponding to surface deformation, comprising:

[0032] Step S1, obtaining the groundwater level of different aquifer groups and the settlement rate of the PS point;

[0033] Step S2: quantify the spatiotemporal response characteristics of water level changes in different aquifer groups and surface deformation in multiple time periods according to the groundwater levels of different aquifer groups and the settlement rates of the PS points, and identify the response thresholds of water level changes in different aquifer groups to surface deformation.

[0034] Preferably, step S2 comprises:

[0035] Step S21: Taking the groundwater level of different aquifer groups as an independent variable and the settlement rate of the PS point as a dependent variable, a random forest machine learning algorithm is used to obtain the predicted impact of the groundwater level of different aquifer groups on the surface deformation, specifically including:

[0036] S211. Use PS-InSAR to process SAR images and obtain surface deformation information.

[0037] PS-InSAR (Permanent Scatter Synthetic Aperture Radar Interferometry) is a technology used to monitor surface deformation. PS-InSAR technology is a differential interferometry technique for long-term SAR image data. It identifies points (permanent scatterers, PS) in the monitoring area that maintain stable reflectance characteristics in multiple phases as target points, uses PS points for differential modeling, removes the terrain residual phase and atmospheric delay phase, and then calculates the deformation information of PS points. Its basic principle is to use the N+1 registered single-view complex images covering the study area, select one of them as the main image, and the remaining N images as secondary images. The phase values ​​of the pixels corresponding to the PS points in the main image are subtracted to obtain N differential interferograms. Each pixel in each differential interferogram contains five phase components

[30] :

[0038]

[0039] in, It is the flat ground phase; is the terrain phase; is the deformation phase; is the atmospheric delay phase; The noise phase is caused by system noise and other errors. The SRTM 30m spatial resolution DEM data is selected to remove the terrain phase, and the atmospheric delay phase and noise phase are removed by time series filtering.

[0040] S212. Based on ArcGIS, the surface deformation data and groundwater level data that need to be input into the random forest (RF) regression model are preprocessed.

[0041] The available groundwater level data are usually point features or contour features. The water level data based on contour lines are interpolated into surface data using ArcGIS Kriging interpolation method. Due to the differences in geological structures, the depth of groundwater level is not directly correlated with surface deformation, so it is necessary to calculate the change in groundwater level data for each layer as model input data.

[0042] S213: Using the above preprocessed data, construct an RF regression model.

[0043] The water level of different aquifers is used as the characteristic variable (independent variable), the deformation rate of PS point is used as the target variable (dependent variable), 70% of the data is selected as the training set, and 30% of the data is selected as the test set. Random sampling with replacement is performed from the training data to create multiple sample sets, each of which is used to train a decision tree. For each decision tree, during the process of building the tree, a feature subset (usually the square root of the number of features) is randomly selected for each node split. The selection index of the split point uses the minimum mean square error (MSE), and the calculation formula is as follows:

[0044]

[0045] Among them, y i is the actual value, is the model prediction value, and N is the number of samples. The decision tree selects the split point that minimizes the MSE at each node to achieve the best splitting effect. Repeat the above process to create a large number of decision trees, each of which is trained on a different sample set and feature subset. Each tree is trained independently and learns different features of the data. For new data, predictions are made separately by each decision tree in the forest. The prediction results of all trees are averaged to obtain the predicted target variable result:

[0046]

[0047] Where T is the total number of trees, The final prediction result is obtained by taking the average value, which is the deformation rate of the PS point in this example.

[0048] S214. Use the characteristic importance index of the model to evaluate the contribution of water level changes in different aquifers to surface deformation, and obtain the impact of water level changes in different aquifers on surface deformation.

[0049] The importance of a feature is measured by calculating the reduction in the mean square error between the predicted output and the actual output of the input feature during model training. j It can be expressed as:

[0050]

[0051] Where, j is the jth aquifer, S j is the set of nodes that are split using feature j in decision tree t, ΔMSE s,t Represents the reduction in MSE caused by splitting s.

[0052] Step S22: According to the predicted impact of groundwater levels of different aquifer groups on surface deformation, the SHAP interpretable machine learning method is used to obtain the spatiotemporal response characteristics of water level changes in different aquifers and surface deformation in multiple time periods, and identify the groundwater level threshold, specifically including:

[0053] Step S221, calculate the Shapley value of the predicted impact of groundwater levels of different aquifer groups on surface deformation, and further quantify the contribution and impact of water level changes of different aquifers on surface deformation.

[0054] Based on the data after RF model training in S213, the water levels of different aquifers are used as the prediction variables, i.e., the input features of the model, and the sedimentation rate at the PS point is used as the target variable, i.e., the predicted value of the model. The model input feature set is divided into several subsets, each feature subset represents a different feature combination. For each feature, the marginal contribution in all possible feature combinations is calculated. The calculation process is that for each feature combination, assuming the impact of adding or removing a certain feature, the incremental contribution of the feature to the model output is calculated. The Shapley value of the feature is obtained by calculating the average marginal contribution of all feature combinations. The Shapley values ​​of each feature are added to obtain the explanation of the model prediction value. The SHAP method regards the contribution of each feature as a linear sum model, so that the predicted value can be decomposed into the sum of the contributions of each feature.

[0055] Given a feature i, the calculation formula for its Shapley value is:

[0056]

[0057] Where: N is the full set of features, S is the feature subset excluding feature i, f(S) represents the model prediction value when only the feature subset S is included. f(S∪{i}) represents the model prediction value after adding feature i, |S| and (|N|-|S|-1)! represent the factorial of the subset size to ensure that the marginal contributions of all feature combinations are weighted fairly.

[0058] The SHAP method decomposes the model prediction value f(x) into the sum of the Shapley values ​​of each feature:

[0059]

[0060] Where f(x) is the predicted value of the model, is the constant that explains the model, i.e., the predicted mean of all training samples, is the Shapley value of the ith feature, indicating the marginal contribution of the prediction variable to the prediction value of the target variable. By calculating the absolute value of the Shapley value of each feature, the importance of the feature is quantified, that is, the contribution of the water level changes of different aquifers to the surface deformation. The larger the absolute value of the Shapley value of the feature, the more significant the impact of the feature on the model output. At the same time, the positive and negative Shapley values ​​of the feature are used to quantify the impact of the feature on the predicted value. When the Shapley value of the feature is positive, it means that the feature has a positive impact on the model output, that is, the increase of the feature value will increase the predicted value of the model, and the corresponding aquifer water level value in the example becomes larger, that is, the water level rises, the deformation rate increases, and the surface uplift phenomenon is more obvious; on the contrary, when the Shapley value of the feature is negative, it means that the feature has a negative impact on the model output, that is, the decrease of the feature value will reduce the predicted value of the model, and the corresponding aquifer water level value in the example becomes smaller, that is, the water level drops, the deformation rate decreases, and the surface settlement phenomenon is more obvious.

[0061] Step S222: Sort the importance of the influencing factors according to the Shapley value to obtain the spatial response characteristics of water level changes and surface deformation of different aquifers.

[0062] First, based on a single sample, the feature corresponding to the maximum absolute value of the Shapley value is selected, that is, the aquifer water level corresponding to the maximum absolute value of the Shapley value of the PS point, and then the spatial position of the sample, that is, the longitude and latitude of the PS, is extracted. By traversing each sample point, the feature with the greatest impact on each sample point in space can be obtained, that is, the aquifer water level corresponding to the greatest impact on the surface deformation in space. Using the above steps, the spatial response characteristics of different aquifer water level changes and surface deformation are obtained.

[0063] Step S223: Analyze the threshold effect of water level changes in different aquifers on surface deformation based on the aggregation characteristics of the characteristic values ​​of water level changes in different aquifers and their Shapley values.

[0064] By analyzing the changes in Shapley values ​​corresponding to water level changes in different aquifers, we find the characteristic interval corresponding to when the Shapley value changes from small to large and then maintains equilibrium, which is used as the response threshold range.

[0065] Embodiment 2:

[0066] The embodiment of the present invention further provides a device for identifying a groundwater level threshold corresponding to surface deformation, comprising:

[0067] Acquisition module, used to obtain groundwater levels and settlement rates of PS points in different aquifer groups;

[0068] The identification module is used to quantify the spatiotemporal response characteristics of water level changes in different aquifer groups and surface deformation in multiple time periods according to the groundwater levels of different aquifer groups and the settlement rates of PS points, and to identify the response thresholds of water level changes in different aquifer groups to surface deformation.

[0069] As an implementation of the embodiment of the present invention, the identification module includes:

[0070] The first processing unit is used to use the groundwater level of different aquifer groups as the independent variable and the settlement rate of the PS point as the dependent variable, and use the random forest machine learning algorithm to obtain the predicted impact of the groundwater level of different aquifer groups on the surface deformation;

[0071] The second processing unit is used to obtain the spatiotemporal response characteristics of water level changes in different aquifers and surface deformation in multiple time periods and identify the groundwater level threshold based on the predicted impact of groundwater levels in different aquifer groups on surface deformation through the SHAP interpretable machine learning method.

[0072] As an implementation manner of the embodiment of the present invention, the second processing unit includes:

[0073] The first processing component is used to calculate the Shapley value of the predicted impact of groundwater levels on surface deformation in different aquifer groups;

[0074] The second processing component is used to sort the importance of influencing factors according to the Shapley value to obtain the spatiotemporal response characteristics of water level changes and surface deformation in different aquifers;

[0075] The third processing component is used to identify the response thresholds of water level changes in different aquifers to surface deformation based on the Shapley value and using the single factor dependency of SHAP.

[0076] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for identifying groundwater level thresholds corresponding to surface deformation, characterized in that: include: Step S1, obtaining the groundwater level of different aquifer groups and the settlement rate of the PS point; Step S2: quantify the spatiotemporal response characteristics of water level changes in different aquifer groups and surface deformation in multiple time periods according to the groundwater levels of different aquifer groups and the settlement rates of the PS points, and identify the response thresholds of water level changes in different aquifer groups to surface deformation.

2. The method for identifying the groundwater level threshold corresponding to the occurrence of surface deformation according to claim 1, characterized in that: Step S2 includes: Step S21, taking the groundwater level of different aquifer groups as an independent variable and the settlement rate of the PS point as a dependent variable, using a random forest machine learning algorithm to obtain the predicted impact of the groundwater level of different aquifer groups on the surface deformation; Step S22: According to the predicted impact of groundwater levels in different aquifer groups on surface deformation, the SHAP interpretable machine learning method is used to obtain the spatiotemporal response characteristics of water level changes in different aquifers and surface deformation in multiple time periods, and identify the groundwater level threshold.

3. The method for identifying the groundwater level threshold corresponding to the occurrence of surface deformation according to claim 2, characterized in that: Step S22 includes: Step S221, calculating the Shapley value of the predicted impact of groundwater levels of different aquifer groups on surface deformation; Step S222, sorting the importance of the influencing factors according to the Shapley value to obtain the spatiotemporal response characteristics of water level changes and surface deformation of different aquifers; Step S223: Based on the Shapley value, using the single factor dependency of SHAP, identify the response thresholds of water level changes in different aquifers to surface deformation.

4. A device for identifying groundwater level thresholds corresponding to surface deformation, characterized in that: include: Acquisition module, used to obtain groundwater levels and settlement rates of PS points in different aquifer groups; The identification module is used to quantify the spatiotemporal response characteristics of water level changes in different aquifer groups and surface deformation in multiple time periods according to the groundwater levels of different aquifer groups and the settlement rates of PS points, and to identify the response thresholds of water level changes in different aquifer groups to surface deformation.

5. The device for identifying groundwater level threshold corresponding to surface deformation according to claim 4, characterized in that: The identification module includes: The first processing unit is used to use the groundwater level of different aquifer groups as the independent variable and the settlement rate of the PS point as the dependent variable, and use the random forest machine learning algorithm to obtain the predicted impact of the groundwater level of different aquifer groups on the surface deformation; The second processing unit is used to obtain the spatiotemporal response characteristics of water level changes in different aquifers and surface deformation in multiple time periods and identify the groundwater level threshold based on the predicted impact of groundwater levels in different aquifer groups on surface deformation through the SHAP interpretable machine learning method.

6. The device for identifying groundwater level threshold corresponding to surface deformation according to claim 5, characterized in that: The second processing unit comprises: The first processing component is used to calculate the Shapley value of the predicted impact of groundwater levels on surface deformation in different aquifer groups; The second processing component is used to sort the importance of influencing factors according to the Shapley value to obtain the spatiotemporal response characteristics of water level changes and surface deformation in different aquifers; The third processing component is used to identify the response thresholds of water level changes in different aquifers to surface deformation based on the Shapley value and using the single factor dependency of SHAP.

Images