Water quality management unit identification method based on forest ecological integrity

Abstract

The present application relates to the field of geographic information, and particularly relates to a water quality management unit identification method based on forest ecological integrity, which comprises the following steps: constructing a feature matrix according to forest ecological integrity data and water quality data of each forest region grid unit, obtaining the forest ecological integrity feature matrix and the water quality feature matrix of each grid unit; fusing the forest ecological integrity feature matrix and the water quality feature matrix of each grid unit, obtaining the dynamic coupling matrix of each grid unit; evaluating the dynamic influence of forest ecological integrity on water quality according to the dynamic coupling matrix of each grid unit, obtaining the dynamic influence evaluation value of each grid unit; identifying the water quality management unit according to the dynamic influence evaluation value, constructing a water quality management unit topology graph, quantifying the dynamic influence of forest ecological integrity on water quality, and improving the accuracy of water quality management unit identification.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of geographic information, in particular to a water quality management unit identification method and device based on forest ecological integrity, a computer device and a storage medium. BACKGROUND

[0002] Forest disturbance and management have significant impacts on water quality, and deforestation and forest fragmentation can lead to serious degradation of water quality. Therefore, quantitatively revealing the relationship between forest change and water quality is crucial for ensuring sustainable water environment protection through regulating forest management methods.

[0003] The species composition-stand structure-function framework based on ecosystem characteristics is widely used in current forest ecological integrity analysis, but it mainly focuses on two-dimensional vegetation index, and it is difficult to quantify the three-dimensional structure of forest and its correlation with water conservation, soil and water conservation, water purification and other functions. And in the assessment of watershed water quality, it usually relies on monitoring of physicochemical parameters and simulation of hydrological models, and the quantification of forest ecological structure is insufficient, often ignoring the regulation of forest on water quality, and it is difficult to accurately assess the impact of forest degradation on water quality. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a water quality management unit identification method and device based on forest ecological integrity, a computer device and a storage medium, which quantifies the dynamic impact of forest ecological integrity on water quality and improves the accuracy of water quality management unit identification.

[0005] In a first aspect, the present application provides a water quality management unit identification method based on forest ecological integrity, comprising the following steps:

[0006] Obtain multi-source data of forest area and dynamic coupling model, wherein the multi-source data includes forest ecological integrity data and water quality data of a plurality of grid cells; the dynamic coupling model includes a feature matrix construction module, a feature fusion module and an impact evaluation module;

[0007] Input the forest ecological integrity data and the water quality data of each grid cell into the feature matrix construction module for feature matrix construction, to obtain the forest ecological integrity feature matrix and the water quality feature matrix of each grid cell;

[0008] Input the forest ecological integrity feature matrix and the water quality feature matrix of each grid cell into the feature fusion module, and use a spatio-temporal attention fusion mechanism for feature fusion to obtain a dynamic coupling matrix of each grid cell;

[0009] input the dynamic coupling matrix of each grid unit into the influence evaluation module for dynamic influence evaluation of forest ecological integrity on water quality, to obtain a dynamic influence evaluation value of each grid unit;

[0010] water quality management unit identification is performed according to the dynamic influence evaluation value of each grid unit, and a water quality management unit topology graph is constructed.

[0011] In a second aspect, an embodiment of the present application provides a water quality management unit identification device based on forest ecological integrity, which comprises:

[0012] The data obtaining module is configured to obtain multi-source data of a forest area and a dynamic coupling model, wherein the multi-source data comprises forest ecological integrity data and water quality data of a plurality of grid units; and the dynamic coupling model comprises a feature matrix construction module, a feature fusion module, and an influence evaluation module.

[0013] The matrix construction module is configured to input the forest ecological integrity data and the water quality data of each grid unit into the feature matrix construction module for feature matrix construction, to obtain a forest ecological integrity feature matrix and a water quality feature matrix of each grid unit.

[0014] The matrix fusion module is configured to input the forest ecological integrity feature matrix and the water quality feature matrix of each grid unit into the feature fusion module, to perform feature fusion by using a spatio-temporal attention fusion mechanism, to obtain a dynamic coupling matrix of each grid unit.

[0015] The grid influence evaluation module is configured to input the dynamic coupling matrix of each grid unit into the influence evaluation module for dynamic influence evaluation of forest ecological integrity on water quality, to obtain a dynamic influence evaluation value of each grid unit.

[0016] The water quality management unit identification module is configured to perform water quality management unit identification according to the dynamic influence evaluation value of each grid unit, and to construct a water quality management unit topology graph.

[0017] In a third aspect, an embodiment of the present application provides a computer device, which comprises a processor, a memory, and a computer program stored in the memory and executable on the processor; when the computer program is executed by the processor, the steps of the water quality management unit identification method based on forest ecological integrity according to the first aspect are implemented.

[0018] In a fourth aspect, an embodiment of the present application provides a storage medium, which stores a computer program; when the computer program is executed by a processor, the steps of the water quality management unit identification method based on forest ecological integrity according to the first aspect are implemented.

[0019] In the embodiment of the present application, a water quality management unit identification method based on forest ecological integrity, a device, a computer device and a storage medium are provided, the dynamic influence of forest ecological integrity on water quality is quantified, and the accuracy of water quality management unit identification is improved.

[0020] In order to better understand and implement, the present application is described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0021] Figure 1 The flowchart of the water quality management unit identification method based on forest ecological integrity provided by the first embodiment of the present application is shown in the figure.

[0022] Figure 2 The flowchart of S2 in the water quality management unit identification method based on forest ecological integrity provided by the first embodiment of the present application is shown in the figure.

[0023] Figure 3 The flowchart of S2 in the water quality management unit identification method based on forest ecological integrity provided by the second embodiment of the present application is shown in the figure.

[0024] Figure 4 The flowchart of S24 in the water quality management unit identification method based on forest ecological integrity provided by the second embodiment of the present application is shown in the figure.

[0025] Figure 5 The flowchart of S24 in the water quality management unit identification method based on forest ecological integrity provided by the third embodiment of the present application is shown in the figure.

[0026] Figure 6 The flowchart of S24 in the water quality management unit identification method based on forest ecological integrity provided by the fourth embodiment of the present application is shown in the figure.

[0027] Figure 7 The flowchart of S3 in the water quality management unit identification method based on forest ecological integrity provided by the first embodiment of the present application is shown in the figure.

[0028] Figure 8 The flowchart of S4 in the water quality management unit identification method based on forest ecological integrity provided by the first embodiment of the present application is shown in the figure.

[0029] Figure 9 The flowchart of S5 in the water quality management unit identification method based on forest ecological integrity provided by the first embodiment of the present application is shown in the figure.

[0030] Figure 10 The structural diagram of the water quality management unit identification device based on forest ecological integrity provided by the fifth embodiment of the present application is shown in the figure.

[0031] Figure 11 A structural schematic diagram of a computer device provided for a sixth embodiment of the present application is provided. DETAILED DESCRIPTION

[0032] The exemplary embodiments will be described in detail herein with reference to the attached drawings. In the following description, the same numbers are used to indicate the same or similar components. The embodiments described in the following exemplary embodiments do not represent all the embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present application.

[0033] The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the present application. As used in the present application and the appended claims, the singular forms "a," "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0034] It should be understood that although the terms first, second, third, etc. can be used herein to describe various information, these terms are not intended to denote a particular order of the information. These terms are used merely to distinguish one type of information from another. For example, a first information can be termed a second information, and similarly, a second information can be termed a first information, without departing from the scope of the present application. Depending on the context, the word "if' as used herein can be interpreted as "when" or "upon" or "in response to determining."

[0035] Reference will now be made to Figure 1 , Figure 1 A flowchart of a method for identifying a water quality management unit based on forest ecological integrity provided by a first embodiment of the present application is provided. The method comprises the following steps:

[0036] S1: Obtain multi-source data of a forest region and a dynamic coupling model.

[0037] The execution subject of the method for identifying a water quality management unit based on forest ecological integrity of the present application is an identification device (hereinafter referred to as an identification device) for identifying a water quality management unit based on forest ecological integrity. In an optional embodiment, the identification device can be a computer device, which can be a server or a server cluster formed by multiple computer devices.

[0038] In the embodiment, the identification device obtains multi-source data of the forest region, wherein the multi-source data includes forest ecological integrity data of a plurality of grid cells and water quality data, and the forest ecological integrity data includes species diversity parameters, stand structure parameters, ecological process parameters and soil property parameters.

[0039] Specifically, the species diversity parameters include species richness and Shannon-Wiener diversity index; the stand structure parameters include canopy density, leaf area index, vertical stratification structure and number density; the ecological process parameters include mortality rate, litter storage, litter thickness, litter nitrogen and phosphorus content and leaf water holding capacity; and the soil property parameters include soil bulk density, soil water content, soil erosion degree, total nitrogen content and total phosphorus content in soil. The water quality data are obtained by using a water quality sensor network and include pH value, water temperature parameter, five-day biochemical oxygen demand, dissolved oxygen, conductivity, permanganate index, nitrogen form data and phosphorus form data, wherein the nitrogen form data include total nitrogen, total dissolved nitrogen and nitrate nitrogen; and the phosphorus form data include total phosphorus, total dissolved phosphorus and particulate phosphorus.

[0040] The identification device obtains a dynamic coupling model, wherein the dynamic coupling model includes a feature matrix construction module, a feature fusion module and an influence evaluation module.

[0041] S2: inputting the forest ecological integrity data and the water quality data of each grid cell into the feature matrix construction module to perform feature matrix construction, to obtain forest ecological integrity feature matrix and water quality feature matrix of each grid cell.

[0042] In the embodiment, the identification device inputs the forest ecological integrity data and the water quality data of each grid cell into the feature matrix construction module to perform feature matrix construction, to obtain forest ecological integrity feature matrix and water quality feature matrix of each grid cell.

[0043] Please refer to Figure 2 , Figure 2 The flowchart of S2 in the water quality management unit identification method based on forest ecological integrity provided by the first embodiment of the present application is shown in FIG. 2, which includes steps S21-S23 and is specifically as follows:

[0044] S21: performing local voxel feature extraction on the forest ecological integrity data to obtain local voxel feature representation.

[0045] In the embodiment, the identification device performs local voxel feature extraction on the forest ecological integrity data according to a preset convolution kernel, specifically a 7×7×7 three-dimensional convolution kernel, to capture the nonlinear response of number density and soil erosion degree in the local voxel domain and obtain local voxel feature representation.

[0046] S22: Construct a forest ecological integrity heterogeneous graph based on the local voxel feature representation to obtain a forest ecological integrity heterogeneous graph; use the meta-path mapping method to construct a matrix based on the forest ecological integrity heterogeneous graph to obtain a forest ecological integrity heterogeneous graph matrix.

[0047] In this embodiment, the identification device constructs a forest ecological integrity heterogeneous graph based on the local voxel feature representation to obtain the forest ecological integrity heterogeneous graph. The forest ecological integrity heterogeneous graph includes several types of nodes and edges between these nodes, as described below:

[0048]

[0049] In the formula, These are nodes in the heterogeneity map of forest ecological integrity. Generate functions for node features. This is a local voxel feature. For K-means clustering function, This refers to the channel sub-feature within the local voxel features. To represent the fast point feature histogram calculation function, For spatial sub-features in local voxel features, For terrain gradient, For feature splicing operators, The first in the forest ecological integrity heterogeneity diagram i The type of node and the first j Edges between nodes of different types Construct functions for edges, The first in the forest ecological integrity heterogeneity diagram i The type of node and the first j Spatial distance between nodes of different types For the first i Connection radius of each type For the first i Node feature vectors of each type of node For the first i The type of node and the first j Similarity thresholds between nodes of different types For the first i Each type of node type identifier.

[0050] The identification device maps the meta-paths between the nodes corresponding to the forest ecological integrity data of each type in the forest ecological integrity heterogeneous graph and the nodes corresponding to the forest ecological integrity data of different types, and obtains a forest ecological integrity heterogeneous graph matrix, wherein the forest ecological integrity heterogeneous graph matrix includes label vectors of a plurality of meta-paths, as follows:

[0051]

[0052] In the formula, is a forest ecological integrity heterogeneous graph meta-path instance, is a specific meta-path, is a relationship type, is a node sequence, is a meta-path embedding, is a type code, is a meta-path type, is a path instance embedding, M is the total number of meta-paths, is the dimension of the meta-path embedding, is a meta-path embedding algorithm, is forest ecological integrity heterogeneous graph meta-path data.

[0053] S23: According to the label vectors of a plurality of meta-paths in the forest ecological integrity heterogeneous graph matrix, a gated aggregation feature matrix is constructed, and a gated aggregation feature matrix is obtained. The gated aggregation feature matrix is spatially three-dimensionally compressed to obtain a compressed feature matrix. According to the compressed feature matrix and the forest ecological integrity heterogeneous graph meta-path data, a matrix is constructed to obtain a forest ecological integrity feature matrix.

[0054] In this embodiment, the identification device constructs a gated aggregation feature matrix according to the label vectors of a plurality of meta-paths in the forest ecological integrity heterogeneous graph matrix and a preset gated aggregation feature algorithm, and obtains a gated aggregation feature matrix, wherein the gated aggregation feature matrix includes aggregation feature vectors of a plurality of nodes, and the gated aggregation feature algorithm is as follows:

[0055]

[0056] In the formula, is the embedding vector of the jth node in the ith meta-path, m l is a nonlinear activation function, is a max-pooling operator, is a meta-path transformation weight, is a neighbor node connected by the ith meta-path, m ​​​is a meta-path gate weight, is a sigmoid function, is a gate parameter vector, T is a transpose symbol, L is a number of graph convolution layers, is a gate aggregated feature matrix, is a tensor product.

[0057] The recognition device performs spatial three-dimensional compression on the gate aggregated feature matrix according to a preset spatial compression algorithm, to obtain a compressed feature matrix, wherein the spatial compression algorithm is:

[0058]

[0059] In the formula, is a dense spatial feature tensor, is a nearest neighbor interpolation function, is a deformable offset, is a three-dimensional deformable convolution function, is average pooling, is a deep attention parameter, is a compressed feature matrix, is a channel broadcast multiplication symbol.

[0060] The recognition device performs matrix construction according to the compressed feature matrix and the forest ecological integrity heterogeneous graph meta-path data, to obtain a forest ecological integrity feature matrix, wherein the forest ecological integrity feature matrix is:

[0061]

[0062] In the formula, is a forest ecological integrity feature matrix, is a graph gate weight matrix, is an adjacency feature matrix constructed by the forest ecological integrity heterogeneous graph meta-path data, is a meta-path diagonal weight, is a sigmoid function, is a meta-path feature transformation function, is a feature concatenation operator, is an element multiplication symbol, is a local pooling function.

[0063] Please refer to Figure 3 , Figure 3 is a flowchart of S2 in the water quality treatment unit recognition method based on forest ecological integrity provided by the second embodiment of the present application, comprising steps S24-S27, and specifically as follows:

[0064] S24: performing water quality feature analysis on the water quality data to obtain water quality feature analysis data.

[0065] In this embodiment, the recognition device performs water quality feature analysis on the water quality data to obtain water quality feature analysis data, wherein the water quality feature analysis data includes a precipitation event dissolved oxygen sudden drop mode analysis sequence, a water quality parameter lag linkage response topology graph, and a multi-modal water quality credibility tensor, wherein the precipitation event dissolved oxygen sudden drop mode analysis sequence is used to quantify the risk probability or intensity of the dissolved oxygen concentration sudden drop caused by the precipitation impact; the water quality parameter lag linkage response topology graph is used to indicate the spatial migration path and time lag law of the pollutants; and the multi-modal water quality credibility tensor is used to reflect the credibility of the water quality monitoring data.

[0066] Referring to Figure 4 , Figure 4 The flowchart of S24 of the water quality management unit recognition method based on forest ecological integrity provided in the second embodiment of the present application includes steps S2401-S2405, and the details are as follows:

[0067] S2401: performing nonlinear fusion and three-dimensional reorganization on the pH value, water temperature parameter, and conductivity of each time point in the water quality data to obtain a three-dimensional hidden state feature matrix of each time point.

[0068] In this embodiment, the recognition device performs nonlinear fusion and three-dimensional reorganization on the pH value, water temperature parameter, and conductivity of each time point in the water quality data to obtain a three-dimensional hidden state feature matrix of each time point.

[0069] Specifically, the recognition device constructs a data matrix according to the pH value, water temperature parameter, and conductivity of each time point in the water quality data. The recognition device performs nonlinear fusion according to the data matrix and a preset nonlinear fusion algorithm by using weight transformation to obtain a nonlinear fusion matrix; and performs three-dimensional reorganization on the nonlinear fusion matrix to obtain a three-dimensional hidden state feature matrix, wherein the three-dimensional hidden state feature matrix contains a plurality of spatial response modes and a plurality of physicochemical channel features of a plurality of time points. The nonlinear fusion algorithm is as follows:

[0070]

[0071] In the formula, is a nonlinear fusion matrix, is a tangent curve function, is a nonlinear fusion weight matrix, is a feature bias vector.

[0072] S2402: Perform event correlation parameter calculation according to the three-dimensional hidden state feature matrix of each moment and the preset rain intensity mask matrix, and obtain an event correlation parameter set of each moment.

[0073] In the embodiment, the recognition device performs event correlation parameter calculation according to the three-dimensional hidden state feature matrix of each moment and the preset rain intensity mask matrix, and obtains an event correlation parameter set of each moment, wherein the event correlation parameter set includes correlation parameters of several events, and the correlation parameters of the events include acid dissolution event correlation parameters, temperature and density layering event correlation parameters, and salt oxygen transmission blocking event correlation parameters.

[0074] Specifically, the recognition device performs vector calculation according to the three-dimensional hidden state feature matrix, the preset rain intensity mask matrix, and a corresponding nonlinear activation function, and obtains a pH value vector, a water temperature parameter mutation vector, and a conductivity gradient vector, wherein the rain intensity mask matrix is constructed by a rainfall intensity sequence obtained by a weather radar, and the rain intensity mask matrix is:

[0075]

[0076] In the formula, is the rain intensity mask matrix, is is the sudden drop score data of the moment, is is the rainfall intensity of the moment, is the time difference from the starting point of the rainstorm to the current moment, is the rainfall intensity.

[0077] The recognition device obtains acid dissolution event correlation parameters, temperature and density layering event correlation parameters, and salt oxygen transmission blocking event correlation parameters according to the sudden drop score data in the rain intensity mask matrix, the pH value vector, the water temperature parameter mutation vector, the conductivity gradient vector, and an event correlation parameter algorithm, wherein the event correlation parameter algorithm is:

[0078]

[0079] In the formula, is the acid dissolution event correlation parameter, is the pH value vector, is the temperature and density layering event correlation parameter, is the water temperature parameter mutation vector, is the salt oxygen transmission blocking event correlation parameter, is the conductivity gradient vector.

[0080] S2403: Perform label vector transformation and matrix construction based on the event association parameter set at each time point to obtain the event label matrix at each time point; perform spatial weight distribution calculation based on the three-dimensional latent feature matrix, the event label matrix, and the preset spatial feature algorithm at each time point to obtain the spatial feature vector at each time point.

[0081] In this embodiment, the identification device performs tag vector conversion and matrix construction based on the event association parameter set at each time moment to obtain the event tag matrix at each time moment. The event tag matrix includes tag vectors of several events. The event tag vectors include acid dissolution event tag vectors, temperature-density stratification event tag vectors, and salt oxygen transport barrier event tag vectors.

[0082] Specifically, the identification device compares the associated parameters of the acid dissolution event, the temperature-density stratification event, and the salt-oxygen transport barrier event with their respective thresholds. Based on the comparison results, it constructs corresponding label vectors to obtain label vectors for the acid dissolution event, the temperature-density stratification event, and the salt-oxygen transport barrier event. The identification device combines the label vectors of the acid dissolution event, the temperature-density stratification event, and the salt-oxygen transport barrier event at the same time to obtain the event label matrix for each time point.

[0083] The recognition device calculates the spatial weight distribution based on the three-dimensional latent feature matrix, the event label matrix, and a preset spatial feature algorithm at each time step to obtain the spatial feature vector at each time step. The spatial feature algorithm is as follows:

[0084]

[0085] In the formula, For the first k Attention weight parameters at each time step. For normalized exponential functions, For the event projection matrix, For the first k The event label matrix at each moment, For the first k Spatial feature vector at each time step For the first k The moment of the first j Attention weight parameters for each spatial pattern, The third hidden state feature matrix is ​​the... k The moment of the first j The latent feature vectors of a spatial pattern.

[0086] S2404: Perform spatiotemporal convolutional co-mapping on the event label matrix at each time step to obtain the channel gating signal at each time step; perform feature aggregation based on the three-dimensional latent feature matrix, the channel gating signal, and the preset spatiotemporal feature aggregation algorithm at each time step to obtain the spatiotemporal aggregated features at each time step.

[0087] In this embodiment, the identification device performs spatiotemporal convolutional co-mapping based on the event label matrix to obtain a channel gating signal, wherein the channel gating signal is:

[0088]

[0089] In the formula, For the first k The channel gating signal at each moment, This is the event gating matrix.

[0090] The recognition device obtains spatiotemporal aggregated features at each time step based on the three-dimensional latent feature matrix, channel gating signals, and a preset spatiotemporal feature aggregation algorithm. The spatiotemporal feature aggregation algorithm is as follows:

[0091]

[0092] In the formula, For the first k The spatiotemporal aggregation characteristics at each moment, For spatiotemporal convolution kernels, The third hidden state feature matrix is ​​the... k + τ At the [time]th moment j Latent feature vectors of spatial patterns τ The time step is the lag step.

[0093] S2405: Based on the spatial feature vectors, spatiotemporal aggregation features, event label vectors, and the preset warning vector algorithm at each time point, the warning vectors at each time point are obtained; based on the warning vectors at each time point, a sequence is constructed to obtain the dissolved oxygen drop pattern analysis sequence for precipitation events.

[0094] In this embodiment, the identification device obtains the warning vector for each moment based on the spatial feature vector, spatiotemporal aggregation feature, event tag vector, and a preset warning vector algorithm. A sequence is constructed based on the warning vectors for each moment to obtain the dissolved oxygen drop pattern analysis sequence for precipitation events. The warning vector algorithm is as follows:

[0095]

[0096] In the formula, For the first k The warning vector at each moment, To fuse the projection matrix, For feature bias, This is the layer normalization function.

[0097] Please see Figure 5 , Figure 5 The flowchart of step S24 in the water quality management unit identification method based on forest ecological integrity provided in the third embodiment of this application includes steps S2411 to S2413, as follows:

[0098] S2411: Encode the nitrogen and phosphorus speciation data at each time point in the water quality data to obtain nitrogen speciation coding sequences and phosphorus speciation coding sequences.

[0099] In this embodiment, the identification device encodes the nitrogen and phosphorus speciation data at various times in the water quality data to obtain nitrogen speciation encoding sequences and phosphorus speciation encoding sequences, wherein the nitrogen speciation encoding sequences and phosphorus speciation encoding sequences reflect the migration forms of particulate and dissolved states.

[0100] S2412: Input the nitrogen form coding sequence and phosphorus form coding sequence into the bidirectional gated loop unit to calculate the effect state feature and obtain the effect state feature sequence; perform pollutant effect classification analysis based on the effect state feature sequence and the preset pollutant effect classification algorithm to obtain the pollutant effect category sequence.

[0101] In this embodiment, the identification device inputs the nitrogen form encoding sequence and the phosphorus form encoding sequence into a bidirectional gated recurrent unit for effect-state feature calculation to obtain the effect-state feature sequence. Specifically, the identification device concatenates the nitrogen form encoding sequence and the phosphorus form encoding sequence, performs forward hidden state calculation and backward hidden state calculation based on the concatenated sequence and the bidirectional gated recurrent unit to obtain the forward hidden state sequence and the backward hidden state sequence, and performs effect-state feature calculation based on the forward hidden state sequence and the backward hidden state sequence to obtain the effect-state feature sequence.

[0102] The identification device performs pollutant effect classification analysis according to a preset pollutant effect classification algorithm to obtain a pollutant effect category sequence. This sequence includes effect category vectors at various time points, and each vector includes a synchronous effect vector, a lag effect vector, and an antagonistic effect vector. The pollutant effect classification algorithm is as follows:

[0103]

[0104] In the formula, The first of the pollutant effect category sequence t The effect category vector at each time step. is a classification weight, is an effect state feature vector of an effect state feature sequence of a t moment.

[0105] S2413: performing pollutant multi-time sequence coupling analysis according to the pollutant effect category sequence, and constructing a water quality lag linkage response topology graph.

[0106] In the embodiment, the identification device performs pollutant multi-time sequence coupling analysis according to the pollutant effect category sequence, and constructs a water quality lag linkage response topology graph, wherein the water quality lag linkage response topology graph includes a node set, an edge set and a topology weight, the node set includes node vectors of each pollutant form of a corresponding node at each moment, and the edge set includes edge vectors between each pollutant form of each node, as follows:

[0107]

[0108] wherein, is a node set, is a node vector of a i pollutant form of a p node, is a nitrogen form coding sequence, is a phosphorus form coding sequence, is a form component of a i node, the form component is a numerical quantity directly separated out by multiplying total nitrogen concentration obtained based on nitrogen form data by a preset organic nitrogen proportion coefficient or multiplying total phosphorus concentration obtained based on phosphorus form data by a preset dissolved state balance coefficient, is a hydrodynamic adjustment parameter of a i node, is an effect state feature vector of a i node, is a node fusion function, is a hyperbolic tangent function, is an edge set, is an edge vector between a i pollutant form of a p node and a j pollutant form of a l node, is a time node index, is a maximum time lag window, is an association strength between a i pollutant form of a p node and a j pollutant form of a l node, For the correlation threshold, The source projection matrix, For the target projection matrix, For the first i The node of the first p Node vectors representing pollutant morphologies For the first j The node of the first l Node vectors representing pollutant morphologies The attenuation coefficient is... For the first i The node and the first j The difference in hysteresis characteristics between nodes For topological weights, This is the weighted sharpening coefficient. For morphological correlation similarity function, It is a time-delay decay function. For the first i A set of neighboring nodes for a given node, wherein the set of neighboring nodes includes several other neighboring nodes, which are all other nodes that may migrate pollutants within a preset time window based on the various pollutant forms of the node. m This is the index of a neighbor node in the set of neighbor nodes.

[0109] Please see Figure 6 , Figure 6 The flowchart of step S24 in the water quality management unit identification method based on forest ecological integrity provided in the fourth embodiment of this application includes steps S2421 to S2424, as follows:

[0110] S2421: Sequences are constructed for the biochemical oxygen demand and initial permanganate index at each time point in the water quality data to obtain the initial biochemical oxygen demand sequence and the initial permanganate index sequence.

[0111] In this embodiment, the identification device constructs sequences of the biochemical oxygen demand (BOD) and the initial permanganate index at various times in the water quality data to obtain the initial BOD sequence and the initial permanganate index sequence.

[0112] S2422: Perform composite noise modeling and signal separation on the initial biochemical oxygen demand sequence to obtain the real biochemical oxygen demand sequence; perform asymmetric dilated convolution processing on the real biochemical oxygen demand sequence to obtain a reliable biochemical oxygen demand sequence.

[0113] In this embodiment, the identification device performs composite noise modeling and signal separation on the initial biochemical oxygen demand (BOD) sequence to obtain the true BOD sequence; the true BOD sequence is then subjected to asymmetric dilated convolution processing to obtain a reliable BOD sequence, wherein the true BOD sequence is:

[0114]

[0115] In the formula, The first in the true biochemical oxygen demand sequence t The true biochemical oxygen demand vector at any given moment. The first in the initial biochemical oxygen demand sequence t The initial biochemical oxygen demand vector at each time point. The aging factor is... For the first t Electrode aging index at a given time point Historical decay factor This is the temperature fluctuation coefficient. It serves as the periodic base.

[0116] The reliable biochemical oxygen demand sequence is as follows:

[0117]

[0118] In the formula, For the first t A reliable biochemical oxygen demand vector at each moment. It is a first-order difference operator. The mutation threshold, For convolution kernel weights, b It is an expansion factor.

[0119] S2423: Anomaly mask generation is performed on the initial permanganate index sequence to obtain a corrected permanganate index sequence; a credibility index matrix is ​​constructed based on the credible biochemical oxygen demand sequence and the corrected permanganate index sequence to obtain a credibility index matrix.

[0120] In this embodiment, the identification device generates an anomaly mask on the initial permanganate index sequence to obtain a corrected permanganate index sequence; the identification device constructs a credibility index matrix based on the reliable biochemical oxygen demand sequence and the corrected permanganate index sequence to obtain a credibility index matrix, wherein the credibility index matrix is:

[0121]

[0122] In the formula, This is a credibility index matrix. For the first Na credible biochemical oxygen demand vector of the time point, a corrected permanganate index vector of the time point. N a corrected permanganate index vector of the time point.

[0123] S2424: forward attention calculation and backward attention calculation are performed according to the credibility index matrix and a preset bidirectional attention gate algorithm, to obtain a forward attention vector and a backward attention vector of each time point; deep fusion and tensor generation are performed according to the forward attention vector and the backward attention vector of each time point, to obtain a multi-modal water quality credibility tensor.

[0124] In this embodiment, the recognition device performs forward attention calculation and backward attention calculation according to the credibility index matrix and a preset bidirectional attention gate algorithm, to obtain a forward attention vector and a backward attention vector of each time point, wherein the bidirectional attention gate algorithm is:

[0125]

[0126] wherein, a forward attention vector of the time point, t a forward attention vector of the time point, a credibility index of the time point, t a credibility index of the time point, a credibility accumulation matrix constructed by accumulating credibility indexes of a historical time window (1: t a credibility accumulation matrix constructed by accumulating credibility indexes of a future time window (1: t N a dimension parameter, a forward hidden state of the time point, a forward hidden state of the time point, t a backward attention vector of the time point, a similarity scale, t a backward hidden state of the time point, a sparse activation function, a max-pooling function, t a Euclidean norm. The recognition device performs deep fusion and tensor generation according to the forward attention vector and the backward attention vector of each time point, to obtain a multi-modal water quality credibility tensor, wherein the multi-modal water quality credibility tensor is:

[0127]

[0128]

[0129] wherein, ​​​​​is a depth fusion vector, is a fusion matrix, is a forward gate vector, is a backward gate vector, is a multi-modal water quality credibility tensor, is a time mapping matrix, is a spectrum mapping matrix, is a 1-modulus product, is a 2-modulus product.

[0130] S25: obtaining a hydrological topology node matrix; performing spatial feature calculation according to the hydrological topology node matrix and a precipitation event dissolved oxygen shock reduction mode analysis sequence to obtain a spatial feature vector; and performing spatiotemporal dynamic coding according to the water quality parameter lag linkage response topology graph and the multi-modal water quality credibility tensor to obtain a spatiotemporal dynamic coding matrix.

[0131] In this embodiment, the recognition device obtains a hydrological topology node matrix, wherein the hydrological topology node matrix is an adjacency matrix generated according to a DEM confluence path.

[0132] The recognition device performs spatial feature calculation according to the hydrological topology node matrix, a peak early warning vector in the precipitation event dissolved oxygen shock reduction mode analysis sequence, and a preset spatial feature algorithm to obtain a spatial feature vector, so as to realize convolution alignment, wherein the spatial feature algorithm is:

[0133]

[0134] wherein, F is a spatial feature vector, is RELU an activation function, is a peak early warning vector, is a river network density feature, specifically, the river network density feature vector is obtained by feature extraction based on the hydrological topology node matrix.

[0135] The recognition device performs spatiotemporal dynamic coding according to the water quality parameter lag linkage response topology graph and the multi-modal water quality credibility tensor to obtain a spatiotemporal dynamic coding matrix, wherein the spatiotemporal dynamic coding algorithm is:

[0136]

[0137] wherein, is a spatiotemporal dynamic coding matrix, is a dynamic coding function, is a topology structure coding function, is a topology weight fusion symbol, is a credibility transformation function.

[0138] S26: Based on the spatial feature vector, spatiotemporal dynamic coding matrix, and analysis sequence of dissolved oxygen drop pattern in precipitation events, feature fusion is performed to obtain a fused node feature matrix; based on the fused node feature matrix and a preset graph attention algorithm, adjacent node features are aggregated to obtain an aggregated node feature matrix.

[0139] In this embodiment, the identification device performs feature fusion based on the spatial feature vector, the spatiotemporal dynamic coding matrix, the peak warning vector in the dissolved oxygen sudden drop pattern analysis sequence of the precipitation event, and a preset feature fusion algorithm to obtain a fused node feature matrix. The fused node feature matrix includes node features of several nodes, and the feature fusion algorithm is as follows:

[0140]

[0141] In the formula, To fuse the node feature matrix, For tensor vectorization operators, This is the peak warning vector.

[0142] The recognition device performs adjacent node feature aggregation on the fused node feature matrix based on the fused node feature matrix and a preset graph attention algorithm to obtain an aggregated node feature matrix. The graph attention algorithm is as follows:

[0143]

[0144] In the formula, For the first i The node and the first j Attention weight coefficients among neighboring nodes This is the transpose of the attention mechanism weight vector. For the first i The node characteristics of each node. For the first j The node characteristics of each neighboring node, To query the transformation matrix, The key-value transformation matrix, For the first i The set of adjacent nodes of a node. For the first i The node aggregation characteristics of each node. For multi-head splicing operators, M For the number of heads to focus on, For the first m The first i The node and the first j Attention weight coefficients among neighboring nodes For the firstm Head value transformation matrix.

[0145] S27: Adopting the hyperspherical projection method, the precipitation intensity mapping is performed on the aggregation node feature matrix to obtain a precipitation intensity matrix; according to the precipitation intensity matrix, the litter nitrogen and phosphorus content data in the forest ecological integrity data, and a preset litter nitrogen and phosphorus release kinetics model, a water quality feature matrix is constructed to obtain the water quality feature matrix.

[0146] In the embodiment, the identification device adopts the hyperspherical projection method, and according to a preset precipitation intensity mapping algorithm, the precipitation intensity mapping is performed on the aggregation node feature matrix to obtain a precipitation intensity matrix, wherein the precipitation intensity matrix includes precipitation intensity vectors of each node at each time, and the precipitation intensity mapping algorithm is:

[0147]

[0148] In the formula, is a precipitation intensity vector of the i th node at the j th time, t is a precipitation intensity vector of the i th node at the j th time, i is a precipitation intensity vector of the i th node at the j th time, is a time-dependent intensity scaling factor, is a time-dependent ball center vector, is a node aggregation feature of the i th node, j is a node aggregation feature of the i th node, is a time-dependent radius scalar, N is a total number of nodes.

[0149] According to the precipitation intensity matrix, the litter nitrogen and phosphorus content data in the forest ecological integrity data, and a preset litter nitrogen and phosphorus release kinetics model, according to a preset enzyme kinetics-hydrodynamics coupling algorithm, the identification device performs nutrient salt pulse flux and node cumulative flux calculation, and according to the obtained nutrient salt pulse flux and node cumulative flux, constructs a water quality feature matrix to obtain an initial water quality feature matrix, wherein the enzyme kinetics-hydrodynamics coupling algorithm is:

[0150]

[0151] In the formula, is a nutrient salt pulse flux of the i th node at the j th time, t is a nutrient salt pulse flux of the i th node at the j th time, i is a nutrient salt pulse flux of the i th node at the j th time, is a nitrogen release index, is a terrain damping coefficient, is a phosphorus release index, is the i th node at the j th time in the litter nitrogen data matrix, t is the i th node at the j th time in the litter nitrogen data matrix, i is the i th node at the j th time in the litter nitrogen data matrix, is the i th node at the j th time in the litter nitrogen data matrix, for the ith node, i topographic gradient modulus of the node, for the maximum enzymatic rate, for the Michaelis constant, for the ith node, t for the ith node, i for the ith node, for the ith node, for the ith node, i for the node cumulative flux of the ith node, for the time step, for the surface runoff vector, for the soil vector.

[0152] The identification device performs convergence weighted pooling on the initial water quality feature matrix to obtain a water quality feature matrix, where the water quality feature matrix is:

[0153]

[0154] In the formula, is the pooled feature matrix, D is the degree matrix, A is the hydrological topology adjacency matrix, is the convergence weight matrix, is the initial water quality feature matrix, is the feature transformation matrix, is the multi-layer perception function, is the first feature fusion matrix, is the second feature fusion matrix, is the max-pooling function, is the average-pooling function, is the water quality feature matrix, Q is the water quality enhanced feature matrix after feature enhancement on the initial water quality feature matrix.

[0155] S3: input the forest ecological integrity feature matrix and the water quality feature matrix of each grid cell into the feature fusion module, and perform feature fusion by using a spatio-temporal attention fusion mechanism to obtain a dynamic coupling matrix of each grid cell.

[0156] In this embodiment, the identification device inputs the forest ecological integrity feature matrix and the water quality feature matrix of each grid cell into the feature fusion module, and performs feature fusion by using a spatio-temporal attention fusion mechanism to obtain a dynamic coupling matrix of each grid cell.

[0157] Please refer to Figure 7 , Figure 7A flowchart of S3 in the water quality management unit identification method based on forest ecological integrity provided by the first embodiment of the present application is shown in FIG. 3, which includes steps S31-S33 and is specifically as follows.

[0158] S31: Obtain a terrain weighted mask; perform Hadamard product operation on the forest ecological integrity feature matrix and the terrain weighted mask to construct a terrain-enhanced forest ecological field.

[0159] In this embodiment, the identification device obtains a terrain weighted mask, which is dynamically generated by coupling three layers of calculation of erosion sensitive area calibration, root anchoring weight quantification, and canopy interception correction, using a soil erosion degree gradient matrix constructed based on a terrain gradient tensor and a land cover type.

[0160] The identification device performs Hadamard product operation on the forest ecological integrity feature matrix and the terrain weighted mask to construct a terrain-enhanced forest ecological field, wherein the terrain-enhanced forest ecological field is:

[0161]

[0162] In the formula, is the terrain-enhanced forest ecological field, is the terrain weighted mask, is a terrain gradient enhancement coefficient, is a Sobel spatial gradient operator.

[0163] S32: Perform hydrological response modeling on the water quality feature matrix to obtain a precipitation time lag sequence; and obtain a time-varying water quality decay feature matrix according to the water quality feature matrix, the precipitation time lag sequence, and a preset time-varying water quality decay feature algorithm.

[0164] In this embodiment, the identification device performs hydrological response modeling on the water quality feature matrix to obtain a precipitation time lag sequence; and obtains a time-varying water quality decay feature matrix according to the water quality feature matrix, the precipitation time lag sequence, and a preset time-varying water quality decay feature algorithm, wherein the time-varying water quality decay feature algorithm is:

[0165]

[0166] In the formula, is the time-varying water quality decay feature matrix, is a pollutant degradation coefficient, is the precipitation time lag sequence, is a time-varying gradient weight, is a time gradient matrix.

[0167] S33: Perform feature fusion according to the terrain-enhanced forest ecological field, the time-varying water quality attenuation feature matrix, and a preset spatiotemporal attention fusion algorithm to obtain a fused feature matrix as the dynamic coupling matrix.

[0168] In this embodiment, the identification device performs feature fusion according to the terrain-enhanced forest ecological field, the time-varying water quality attenuation feature matrix, and a preset spatiotemporal attention fusion algorithm to obtain a fused feature matrix as the dynamic coupling matrix, wherein the spatiotemporal attention fusion algorithm is:

[0169]

[0170] In the formula, is a fused feature matrix, is a Laplacian operator, is a curvature mapping function, is a graph attention network, is a temporal convolution network, is a cross-domain gating function constructed using a spatial dominant factor constructed based on the terrain-enhanced forest ecological field.

[0171] S4: Input the dynamic coupling matrix of each grid cell into the influence evaluation module to perform dynamic influence evaluation of forest ecological integrity on water quality, and obtain a dynamic influence evaluation value of each grid cell.

[0172] In this embodiment, the identification device inputs the dynamic coupling matrix of each grid cell into the influence evaluation module to perform dynamic influence evaluation of forest ecological integrity on water quality, and obtains a dynamic influence evaluation value of each grid cell, wherein the dynamic influence evaluation value reflects the dynamic influence of forest ecological integrity on water quality.

[0173] The influence evaluation module includes a feature coupling module and a fully connected module, please refer to Figure 8 , Figure 8 FIG. 4 is a flowchart of S4 in the water quality management unit identification method based on forest ecological integrity provided in the first embodiment of the present application, including steps S41-S42, which are as follows:

[0174] S41: Input the dynamic coupling matrix into the feature coupling module to perform parameter extraction, obtain a scale parameter and a translation parameter of the dynamic coupling matrix, perform feature decoupling on the dynamic coupling matrix according to the scale parameter and the translation parameter to obtain a feature decoupled matrix, and perform feature compression on the feature decoupled matrix to obtain a compressed feature matrix.

[0175] The feature coupling module adopts a reversible normalized flow model. In this embodiment, the identification device inputs the dynamic coupling matrix into the feature coupling module, splits the dynamic coupling matrix into two components, and generates scale parameters and translation parameters by using a three-layer convolutional network to obtain the scale parameters and the translation parameters of the dynamic coupling matrix.

[0176] According to the scale parameters and the translation parameters, the identification device performs feature decoupling on the dynamic coupling matrix by using a Householder orthogonal transformation method to obtain a feature decoupling matrix. The identification device performs feature compression on the feature decoupling matrix by using a positive definite constraint linear layer to obtain a compressed feature matrix.

[0177] S42: inputting the compressed feature matrix into the fully connected module for nonlinear dimension reduction to obtain a nonlinear dimension reduction vector; and obtaining a dynamic impact evaluation value according to the nonlinear dimension reduction vector and a preset dynamic impact evaluation algorithm of forest ecological integrity on water quality.

[0178] In this embodiment, the identification device inputs the compressed feature matrix into the fully connected module for nonlinear dimension reduction to obtain a nonlinear dimension reduction vector. Specifically, the identification device inputs the compressed feature matrix into a residual block layer of the fully connected module, extracts high-order ecological correlations by using a GeLU activation function, and outputs a hidden state vector: the identification device inputs the hidden state vector into an affine transformation layer of the fully connected module, compresses the hidden state to a single scalar by using a trainable weight matrix, and obtains a nonlinear dimension reduction vector.

[0179] The identification device obtains a dynamic impact evaluation value according to the nonlinear dimension reduction vector and a preset dynamic impact evaluation algorithm of forest ecological integrity on water quality, wherein the dynamic impact evaluation algorithm of forest ecological integrity on water quality is:

[0180]

[0181] In the formula, FQ is a dynamic impact evaluation value, is a nonlinear dimension reduction vector.

[0182] The identification device performs water quality management unit identification according to the dynamic impact evaluation values of each grid unit to construct a water quality management unit topology graph.

[0183] S5: performing water quality management unit identification according to the dynamic impact evaluation values of each grid unit to construct a water quality management unit topology graph.

[0184] In the embodiment, the identification device identifies the water quality management unit according to the dynamic influence evaluation value of each grid unit, constructs a water quality management unit topology graph, quantifies the dynamic influence of forest ecological integrity on water quality, and improves the accuracy of water quality management unit identification.

[0185] Referring to Figure 9 , Figure 9 The flowchart of S5 in the water quality management unit identification method based on forest ecological integrity provided in the first embodiment of the application is shown in the figure, which includes steps S51-S52, and the details are as follows.

[0186] S51: A time-space clustering method is used to perform clustering analysis according to the preset spatial neighborhood radius, time neighborhood, minimum point number, and the dynamic influence evaluation value of each grid unit, and a plurality of clustering clusters are obtained.

[0187] In the embodiment, the identification device uses a T-DBSCAN time-space clustering method to perform clustering analysis according to the preset spatial neighborhood radius, time neighborhood, minimum point number, and the dynamic influence evaluation value of each grid unit, and a plurality of clustering clusters are obtained, wherein the clustering clusters include a plurality of grid units.

[0188] S52: According to the dynamic influence evaluation value of the plurality of grid units in each clustering cluster, a target grid unit is extracted from the plurality of grid units as a water quality management unit, and a water quality management unit topology graph is constructed.

[0189] In the embodiment, the identification device extracts a target grid unit from the plurality of grid units as a water quality management unit according to the dynamic influence evaluation value of the plurality of grid units in each clustering cluster, and constructs a water quality management unit topology graph. Specifically, the identification device calculates the difference of the dynamic influence evaluation value of the plurality of grid units in each clustering cluster, i.e., calculates the average value of the dynamic influence evaluation value of all grid units in the same clustering cluster, subtracts the dynamic influence evaluation value of each grid unit from the average value, and obtains the difference of the dynamic influence evaluation value of each clustering cluster. If all grid units in the clustering cluster are less than a preset deviation constraint value, the clustering cluster is taken as a target clustering cluster, and the soil consistency of each grid unit in the target clustering cluster is obtained. If the soil consistency is greater than a preset soil consistency proportion threshold value, the grid unit is extracted as a target grid unit.

[0190] Referring to Figure 10 , Figure 10A structural schematic diagram of a water quality management unit identification device based on forest ecological integrity provided by the fifth embodiment of the present application is provided. The device can realize all or part of the water quality management unit identification device based on forest ecological integrity through software, hardware or a combination of both. The water quality management unit identification device 10 based on forest ecological integrity comprises:

[0191] A data obtaining module 101 is configured to obtain multi-source data of a forest area and a dynamic coupling model, wherein the multi-source data comprises forest ecological integrity data and water quality data of a plurality of grid cells; and the dynamic coupling model comprises a feature matrix construction module, a feature fusion module and an influence evaluation module.

[0192] A matrix construction module 102 is configured to input the forest ecological integrity data and the water quality data of each grid cell into the feature matrix construction module to construct a feature matrix, thereby obtaining a forest ecological integrity feature matrix and a water quality feature matrix of each grid cell.

[0193] A matrix fusion module 103 is configured to input the forest ecological integrity feature matrix and the water quality feature matrix of each grid cell into the feature fusion module, and perform feature fusion using a spatiotemporal attention fusion mechanism to obtain a dynamic coupling matrix of each grid cell.

[0194] A grid influence evaluation module 104 is configured to input the dynamic coupling matrix of each grid cell into the influence evaluation module to evaluate the dynamic influence of forest ecological integrity on water quality, thereby obtaining a dynamic influence evaluation value of each grid cell.

[0195] A water quality management unit identification module 105 is configured to identify a water quality management unit according to the dynamic influence evaluation value of each grid cell, and construct a water quality management unit topology graph.

[0196] In the embodiment, the data obtaining module is used to obtain multi-source data of a forest area and a dynamic coupling model, wherein the multi-source data comprises forest ecological integrity data and water quality data of a plurality of grid cells; the dynamic coupling model comprises a feature matrix construction module, a feature fusion module, and an influence evaluation module; the forest ecological integrity data and the water quality data of each grid cell are input into the feature matrix construction module to construct a feature matrix, so as to obtain a forest ecological integrity feature matrix and a water quality feature matrix of each grid cell; the forest ecological integrity feature matrix and the water quality feature matrix of each grid cell are input into the feature fusion module to perform feature fusion by using a space-time attention fusion mechanism, so as to obtain a dynamic coupling matrix of each grid cell; the dynamic coupling matrix of each grid cell is input into the influence evaluation module to evaluate the dynamic influence of forest ecological integrity on water quality, so as to obtain a dynamic influence evaluation value of each grid cell; a water quality management unit identification module is used to identify a water quality management unit according to the dynamic influence evaluation value of each grid cell, and a water quality management unit topology is constructed. The dynamic influence of forest ecological integrity on water quality is quantified, and the accuracy of water quality management unit identification is improved.

[0197] Reference is made to Figure 11 , Figure 11 The computer device provided in the sixth embodiment of the present application is shown in a structural schematic diagram, and the computer device 11 comprises a processor 111, a memory 112, and a computer program 113 stored in the memory 112 and capable of running on the processor 111; the computer device can store a plurality of instructions, the instructions are suitable for being loaded by the processor 111 and performing the steps of the first to fourth embodiments, and the specific execution process can be referred to the specific description of the first to fourth embodiments, which will not be repeated here.

[0198] The processor 111 can include one or more processing cores. The processor 111 connects various parts within the server through various interfaces and lines, executes various functions and processes data of the water quality management unit identification device 10 based on forest ecological integrity by running or executing instructions, programs, code sets or instruction sets stored in the memory 112, and calling data in the memory 112. Optionally, the processor 111 can be implemented in at least one of a hardware form of a digital signal processing (DSP), a field-programmable gate array (FPGA), and a programmable logic array (PLA). The processor 111 can be integrated with one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the touch display screen; and the modem is used for processing wireless communication. It can be understood that the above-mentioned modem can also not be integrated into the processor 111, but can be realized by a separate chip.

[0199] The memory 112 can include a random access memory (RAM) and a read-only memory (ROM). Optionally, the memory 112 includes a non-transitory computer-readable storage medium. The memory 112 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 112 can include a program storage area and a data storage area, wherein the program storage area can store instructions for implementing an operating system, instructions for at least one function (such as touch instructions), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area can store data involved in the above-mentioned various method embodiments, etc. The memory 112 can also be at least one storage device located away from the above-mentioned processor 111. The embodiments of the present application also provide a storage medium, which can store a plurality of instructions. The instructions are suitable for being loaded and executed by the processor to execute the method steps of the first to fourth embodiments described above. The specific execution process can be referred to the specific description of the first to fourth embodiments, which will not be described here.

[0200] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional units and modules is taken as an example, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit or module in the embodiment can be integrated in one processing unit, or each unit can exist physically independently, or two or more units can be integrated in one unit, and the integrated unit can be realized in the form of hardware or in the form of software functional unit. In addition, the specific names of each functional unit or module are only for convenient distinction, and do not limit the protection scope of the present application. The specific working process of the units and modules in the above system can refer to the corresponding process in the foregoing method embodiments, and will not be described here. In the above embodiments, the description of each embodiment has its own emphasis, and the parts not described or recorded in a certain embodiment can be referred to the relevant description of other embodiments.

[0201] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware or in a combination of computer software and electronic hardware. Whether the functions are executed in hardware or software depends on the specific application and design constraints of the algorithm. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0202] In the embodiments of the present application, it should be understood that the disclosed apparatus / terminal device and method can be implemented in other manners. For example, the described apparatus / terminal device embodiments are merely schematic. For example, the division of the modules or units is merely logical function division. There can be another division manner in actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be electrical, mechanical or in other forms.

[0203] The integrated module / unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such an understanding, the present application implements all or part of the processes in the above-described method embodiments, and can also be completed by a computer program instructing related hardware. The computer program can be stored in a computer readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. The computer program includes computer program code. The computer program code can be in the form of source code, object code, executable file or some intermediate form.

[0204] The present application is not limited to the above-described embodiments. If various modifications or changes are made to the present application without departing from the spirit and scope of the present application, the present application is intended to include these modifications and changes.

Claims

1. A method for identifying a water quality management unit based on forest ecological integrity, characterized in that, The method comprises the following steps: obtaining multi-source data of a forest area and a dynamic coupling model, wherein the multi-source data comprises forest ecological integrity data and water quality data of a plurality of grid cells; the dynamic coupling model comprises a feature matrix construction module, a feature fusion module and an influence evaluation module; inputting the forest ecological integrity data and the water quality data of each grid cell into the feature matrix construction module for feature matrix construction, to obtain a forest ecological integrity feature matrix of each grid cell; performing water quality feature analysis on the water quality data to obtain water quality feature analysis data, wherein the water quality feature analysis data comprises a dissolved oxygen sudden drop mode analysis sequence of a precipitation event, a water quality parameter lag linkage response topology graph and a multi-modal water quality credibility tensor; obtaining a hydrological topology node matrix; performing spatial feature calculation according to the hydrological topology node matrix and the dissolved oxygen sudden drop mode analysis sequence of the precipitation event, to obtain a spatial feature vector; performing spatio-temporal dynamic coding according to the water quality parameter lag linkage response topology graph and the multi-modal water quality credibility tensor, to obtain a spatio-temporal dynamic coding matrix; performing feature fusion according to the spatial feature vector, the spatio-temporal dynamic coding matrix and the dissolved oxygen sudden drop mode analysis sequence of the precipitation event, to obtain a fusion node feature matrix; performing adjacent node feature aggregation on the fusion node feature matrix according to a preset graph attention algorithm, to obtain an aggregated node feature matrix; performing precipitation intensity mapping on the aggregated node feature matrix by using a hyperspherical projection method, to obtain a precipitation intensity matrix; performing water quality feature matrix construction according to the precipitation intensity matrix, litter nitrogen and phosphorus content data in the forest ecological integrity data and a preset litter nitrogen and phosphorus release kinetics model, to obtain a water quality feature matrix of each grid cell; inputting the forest ecological integrity feature matrix and the water quality feature matrix of each grid cell into the feature fusion module, and performing feature fusion by using a spatio-temporal attention fusion mechanism, to obtain a dynamic coupling matrix of each grid cell; inputting the dynamic coupling matrix of each grid cell into the influence evaluation module for dynamic influence evaluation of forest ecological integrity on water quality, to obtain a dynamic influence evaluation value of each grid cell; performing water quality management unit identification according to the dynamic influence evaluation value of each grid cell, to construct a water quality management unit topology graph. 2.The forest-ecological-integrity-based water quality management unit identification method of claim 1, wherein, The method for inputting the forest ecological integrity data and the water quality data of each grid cell into the feature matrix construction module for feature matrix construction, to obtain a forest ecological integrity feature matrix and a water quality feature matrix of each grid cell, comprises the following steps: performing local voxel feature extraction on the forest ecological integrity data, to obtain a local voxel feature representation; performing forest ecological integrity heterogeneous graph construction according to the local voxel feature representation, to obtain a forest ecological integrity heterogeneous graph; and performing matrix construction according to the forest ecological integrity heterogeneous graph by using a meta-path mapping method, to obtain a forest ecological integrity heterogeneous graph matrix. According to the label vector of the several meta-paths in the forest ecological integrity heterogeneous graph matrix, a gated aggregated feature matrix is constructed, and a gated aggregated feature matrix is obtained; and the gated aggregated feature matrix is subjected to spatial three-dimensional compression to obtain a compressed feature matrix; According to the compressed feature matrix and forest ecological integrity heterogeneous graph meta-path data, a matrix is constructed to obtain a forest ecological integrity feature matrix. 3.The forest-ecological-integrity-based water quality management unit identification method of claim 1, wherein, The water quality data is subjected to water quality feature analysis to obtain water quality feature analysis data, including the steps of: The pH value, water temperature parameter and conductivity of each time point in the water quality data are subjected to nonlinear fusion and three-dimensional reorganization to obtain a three-dimensional hidden state feature matrix of each time point; According to the three-dimensional hidden state feature matrix of each time point and a preset rain intensity mask matrix, event correlation parameter calculation is performed to obtain an event correlation parameter set of each time point, wherein the event correlation parameter set includes correlation parameters of several events, and the event correlation parameters include acidic dissolution event correlation parameters, temperature and density stratification event correlation parameters, and salt resistance oxygen transmission event correlation parameters; According to the event correlation parameter set of each time point, a label vector conversion and matrix construction are performed to obtain an event label matrix of each time point; and according to the three-dimensional hidden state feature matrix of each time point, the event label matrix and a preset spatial feature algorithm, spatial weight distribution calculation is performed to obtain a spatial feature vector of each time point; The event label matrix of each time point is subjected to spatio-temporal convolution cooperative mapping to obtain a channel gating signal of each time point; and according to the three-dimensional hidden state feature matrix of each time point, the channel gating signal and a preset spatio-temporal feature aggregation algorithm, feature aggregation is performed to obtain a spatio-temporal aggregation feature of each time point; According to the spatial feature vector, the spatio-temporal aggregation feature, the event label vector and a preset early warning vector algorithm, an early warning vector of each time point is obtained; and according to the early warning vector of each time point, a sequence is constructed to obtain a precipitation event dissolved oxygen sudden drop mode analysis sequence. 4.The forest-ecological-integrity-based water quality management unit identification method of claim 1, wherein, The water quality data is subjected to water quality feature analysis to obtain water quality feature analysis data, including the steps of: The nitrogen form data and phosphorus form data of each time point in the water quality data are respectively encoded to obtain a nitrogen form encoding sequence and a phosphorus form encoding sequence; The nitrogen form encoding sequence and the phosphorus form encoding sequence are input into a bidirectional gated recurrent unit for effect state feature calculation to obtain an effect state feature sequence; and according to the effect state feature sequence and a preset pollutant effect classification algorithm, pollutant effect classification analysis is performed to obtain a pollutant effect category sequence; According to the pollutant effect category sequence, pollutant multi-time coupling analysis is performed to construct a water quality lag linkage response topology map. 5.The forest-ecological-integrity-based water quality management unit identification method of claim 1, wherein, The water quality data is subjected to water quality feature analysis to obtain water quality feature analysis data, including the steps of: The biochemical oxygen demand and initial permanganate index of each time point in the water quality data are respectively subjected to sequence construction to obtain an initial biochemical oxygen demand sequence and an initial permanganate index sequence; The initial biochemical oxygen demand sequence is subjected to composite noise modeling and signal separation to obtain a true biochemical oxygen demand sequence; The real biochemical oxygen demand sequence is subjected to asymmetric expansion convolution processing to obtain a reliable biochemical oxygen demand sequence; The initial permanganate index sequence is subjected to abnormal mask generation to obtain a corrected permanganate index sequence; a reliability index matrix is constructed according to the reliable biochemical oxygen demand sequence and the corrected permanganate index sequence to obtain a reliability index matrix; According to the reliability index matrix and a preset bidirectional attention gate algorithm, forward attention calculation and backward attention calculation are performed to obtain a forward attention vector and a backward attention vector at each time; deep fusion and tensor generation are performed according to the forward attention vector and the backward attention vector at each time to obtain a multi-modal water quality reliability tensor. 6.The forest-ecological-integrity-based water quality management unit identification method of claim 1, wherein, The feature fusion is performed by using the spatio-temporal attention fusion mechanism to obtain the dynamic coupling matrix of each grid cell, including the following steps: A terrain weighted mask is obtained; a Hadamard product operation is performed according to the forest ecological integrity feature matrix and the terrain weighted mask to construct a terrain enhanced forest ecological field; Hydrological response modeling is performed on the water quality feature matrix to obtain a precipitation time lag sequence; a time-varying water quality decay feature matrix is obtained according to the water quality feature matrix, the precipitation time lag sequence, and a preset time-varying water quality decay feature algorithm; According to the terrain enhanced forest ecological field, the time-varying water quality decay feature matrix, and a preset spatio-temporal attention fusion algorithm, feature fusion is performed to obtain a fusion feature matrix as the dynamic coupling matrix. 7.The forest-ecological-integrity-based water quality management unit identification method of claim 1, wherein: The influence evaluation module includes a feature coupling module and a fully connected module; The dynamic coupling matrix of each grid cell is input into the influence evaluation module to evaluate the dynamic influence of forest ecological integrity on water quality, and the dynamic influence evaluation value of each grid cell is obtained, including the following steps: The dynamic coupling matrix is input into the feature coupling module to extract parameters, and the scale parameter and the translation parameter of the dynamic coupling matrix are obtained; the feature decoupling matrix is obtained by decoupling the dynamic coupling matrix according to the scale parameter and the translation parameter; The feature decoupling matrix is compressed to obtain a compressed feature matrix; The compressed feature matrix is input into the fully connected module to perform nonlinear dimension reduction to obtain a nonlinear dimension reduction vector; the dynamic influence evaluation value is obtained according to the nonlinear dimension reduction vector and a preset dynamic influence evaluation algorithm of forest ecological integrity on water quality. 8.The forest-ecological-integrity-based water quality management unit identification method of claim 1, wherein, The water quality management unit is identified according to the dynamic influence evaluation value of each grid cell to construct a water quality management unit topology graph, including the following steps: A spatio-temporal clustering method is used to perform clustering analysis according to a preset spatial neighborhood radius, a time neighborhood, a minimum point number, and the dynamic influence evaluation value of each grid cell to obtain a plurality of clustering clusters, wherein the clustering cluster includes a plurality of grid cells; According to the dynamic influence evaluation values of the plurality of grid cells in each clustering cluster, a target grid cell is extracted from the plurality of grid cells as a water quality management unit to construct a water quality management unit topology graph.

9. A water quality treatment unit identification device based on forest ecological integrity, characterized in that, It includes: The data obtaining module is configured to obtain multi-source data of a forest region and a dynamic coupling model, wherein the multi-source data comprises forest ecological integrity data and water quality data of a plurality of grid cells; and the dynamic coupling model comprises a feature matrix construction module, a feature fusion module, and an influence evaluation module. The matrix construction module is configured to input the forest ecological integrity data and the water quality data of each grid cell into the feature matrix construction module to construct a feature matrix of forest ecological integrity of each grid cell. The water quality data is subjected to water quality feature analysis to obtain water quality feature analysis data, wherein the water quality feature analysis data comprises a dissolved oxygen sudden drop mode analysis sequence of a precipitation event, a water quality parameter lag linkage response topology graph, and a multi-modal water quality credibility tensor. A hydrological topology node matrix is obtained; spatial feature vectors are obtained by performing spatial feature calculation according to the hydrological topology node matrix and the dissolved oxygen sudden drop mode analysis sequence of the precipitation event; and a spatiotemporal dynamic coding matrix is obtained by performing spatiotemporal dynamic coding according to the water quality parameter lag linkage response topology graph and the multi-modal water quality credibility tensor. Feature fusion is performed according to the spatial feature vectors, the spatiotemporal dynamic coding matrix, and the dissolved oxygen sudden drop mode analysis sequence of the precipitation event to obtain a fused node feature matrix; and adjacent node feature aggregation is performed on the fused node feature matrix according to a preset graph attention algorithm to obtain an aggregated node feature matrix. A precipitation intensity matrix is obtained by performing precipitation intensity mapping on the aggregated node feature matrix using a hyperspherical projection method; and a water quality feature matrix of each grid cell is obtained by performing water quality feature matrix construction according to the precipitation intensity matrix, litter nitrogen and phosphorus content data in the forest ecological integrity data, and a preset litter nitrogen and phosphorus release kinetics model. The matrix fusion module is configured to input the forest ecological integrity feature matrix and the water quality feature matrix of each grid cell into the feature fusion module, perform feature fusion using a spatiotemporal attention fusion mechanism, and obtain a dynamic coupling matrix of each grid cell. The grid influence evaluation module is configured to input the dynamic coupling matrix of each grid cell into the influence evaluation module to perform dynamic influence evaluation of forest ecological integrity on water quality, and obtain a dynamic influence evaluation value of each grid cell. The water quality management unit identification module is configured to identify a water quality management unit according to the dynamic influence evaluation value of each grid cell, and construct a water quality management unit topology graph.

Images