CN114841402A - Underground water level height prediction method and system based on multi-feature map network - Google Patents
Underground water level height prediction method and system based on multi-feature map network Download PDFInfo
- Publication number
- CN114841402A CN114841402A CN202210300649.XA CN202210300649A CN114841402A CN 114841402 A CN114841402 A CN 114841402A CN 202210300649 A CN202210300649 A CN 202210300649A CN 114841402 A CN114841402 A CN 114841402A
- Authority
- CN
- China
- Prior art keywords
- nodes
- underground water
- water level
- graph
- node
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 120
- 238000000034 method Methods 0.000 title claims abstract description 77
- 239000013598 vector Substances 0.000 claims abstract description 37
- 238000012549 training Methods 0.000 claims abstract description 34
- 238000000547 structure data Methods 0.000 claims abstract description 32
- 238000012544 monitoring process Methods 0.000 claims abstract description 12
- 238000005303 weighing Methods 0.000 claims abstract description 9
- 239000011159 matrix material Substances 0.000 claims description 36
- 239000003673 groundwater Substances 0.000 claims description 30
- 238000003973 irrigation Methods 0.000 claims description 24
- 230000002262 irrigation Effects 0.000 claims description 24
- 239000002352 surface water Substances 0.000 claims description 23
- 230000035699 permeability Effects 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 13
- 238000009933 burial Methods 0.000 claims description 10
- 238000011156 evaluation Methods 0.000 claims description 10
- 238000010606 normalization Methods 0.000 claims description 9
- 238000010276 construction Methods 0.000 claims description 7
- 238000005273 aeration Methods 0.000 claims description 6
- 230000002776 aggregation Effects 0.000 claims description 6
- 238000004220 aggregation Methods 0.000 claims description 6
- 238000001556 precipitation Methods 0.000 claims description 6
- 239000007787 solid Substances 0.000 claims description 6
- 230000006872 improvement Effects 0.000 claims description 5
- 238000005192 partition Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 2
- 239000010410 layer Substances 0.000 description 27
- 230000008569 process Effects 0.000 description 22
- 238000010586 diagram Methods 0.000 description 14
- 230000015654 memory Effects 0.000 description 11
- 238000004364 calculation method Methods 0.000 description 9
- 238000004590 computer program Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 8
- 238000012360 testing method Methods 0.000 description 8
- 239000002689 soil Substances 0.000 description 7
- 238000003860 storage Methods 0.000 description 7
- 230000007246 mechanism Effects 0.000 description 5
- 238000013507 mapping Methods 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 238000012795 verification Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 238000001704 evaporation Methods 0.000 description 3
- 230000008020 evaporation Effects 0.000 description 3
- 230000004931 aggregating effect Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 239000004927 clay Substances 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000013589 supplement Substances 0.000 description 2
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 0.000 description 1
- 235000017784 Mespilus germanica Nutrition 0.000 description 1
- 244000182216 Mimusops elengi Species 0.000 description 1
- 235000000560 Mimusops elengi Nutrition 0.000 description 1
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 241000209140 Triticum Species 0.000 description 1
- 235000021307 Triticum Nutrition 0.000 description 1
- 235000007837 Vangueria infausta Nutrition 0.000 description 1
- 235000009754 Vitis X bourquina Nutrition 0.000 description 1
- 235000012333 Vitis X labruscana Nutrition 0.000 description 1
- 240000006365 Vitis vinifera Species 0.000 description 1
- 235000014787 Vitis vinifera Nutrition 0.000 description 1
- 240000008042 Zea mays Species 0.000 description 1
- 235000005824 Zea mays ssp. parviglumis Nutrition 0.000 description 1
- 235000002017 Zea mays subsp mays Nutrition 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 235000005822 corn Nutrition 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000008595 infiltration Effects 0.000 description 1
- 238000001764 infiltration Methods 0.000 description 1
- 239000011229 interlayer Substances 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 229920006395 saturated elastomer Polymers 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
- 235000013311 vegetables Nutrition 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Economics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Probability & Statistics with Applications (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention discloses a method and a system for predicting the height of underground water level based on a multi-feature map network, wherein the method comprises the following steps: selecting multiple underground water level monitoring points at different positions as space nodes for constructing a graph structure data set, acquiring characteristic information of each node, and constructing a characteristic vector of each node; weighing a correlation relation between two nodes based on the feature vector, constructing edge structure information for describing node connectivity, and taking the reciprocal of the geographic distance between the two connected nodes as the edge connection weight of the nodes; constructing a graph structure data set based on the information; improving the GCN to obtain an underground water level prediction model; training the underground water level prediction model based on the graph structure data set; and predicting the underground water level of the spatial position based on the trained underground water level prediction model. The method can realize efficient and accurate prediction of the underground water level of the unknown position.
Description
Technical Field
The invention relates to the technical field of hydrologic and geographic intelligent decision making, in particular to a method and a system for predicting the height of underground water level based on a multi-feature map network.
Background
Surface water and ground water resources are important components of complex ecological water circulation systems. The main water sources in the water resource system are atmospheric water, surface water and underground water, and water taken from outside the system. Various water sources are mutually converted under certain conditions. For example, rainfall infiltration and irrigation can supplement soil water, and the soil water is saturated and then continuously infiltrated to form underground water. For irrigation type agricultural areas, the underground water level is high due to long-term diversion irrigation of the opening of the river channel and unsmooth flood irrigation and drainage, and the surface evaporation of the irrigation type agricultural areas is usually strong, so that the soil salinization degree is deepened, and the ecological environment is gradually worsened. Along with the gradual scarcity of surface water resources, the runoff of the river channel is greatly reduced, the irrigation diversion amount is also reduced year by year, the current agricultural irrigation requirements cannot be met, and the agricultural development is greatly restricted.
The conversion relationship between surface water and underground water is very complex, and the change of underground water level is influenced by the surface irrigation quantity and is related to the landform and landform conditions. The research on the supplement and drainage relationship between surface water and underground water in different spatial and geographic positions can effectively relieve the problem of soil salinization caused by unreasonable utilization of water resources, and provides accurate guidance for agricultural production. The conventional research mainly adopts hydrogeology professional knowledge to research the evolution relation between surface water and underground water, and simulates the scheduling process of water resources in a system by establishing a traditional hydrogeology model and a mathematical model. Although the method has a strong theoretical basis, the precise evolution of water resources cannot be analyzed from the data perspective for a complex water resource system.
Therefore, an intelligent solution that comprehensively considers the data characteristics and the principle characteristics is needed.
Disclosure of Invention
The invention provides a multi-characteristic-diagram-network-based underground water level height prediction method and system, and aims to solve the technical problem that quantification and accuracy are lacked in the existing surface water and underground water dynamic evolution process.
In order to solve the technical problems, the invention provides the following technical scheme:
in one aspect, the present invention provides a method for predicting a height of a groundwater level based on a multi-feature map network, where the method for predicting a height of a groundwater level based on a multi-feature map network includes:
selecting multiple underground water level monitoring points at different positions as space nodes for constructing a graph structure data set, acquiring characteristic information of each node, and respectively constructing a characteristic vector of each node based on the characteristic information;
based on the characteristic vector, weighing the correlation relationship between the two nodes, constructing edge structure information for describing the connectivity of the nodes, and taking the reciprocal of the geographic distance between the two connected nodes as the edge connection weight of the nodes;
constructing a graph structure data set by taking the characteristic information and the edge structure information of the nodes as sample characteristics and taking the depth interval category to which the underground water level burial depth corresponding to the nodes belongs as a sample label;
improving a Graph Convolution Network (GCN) to obtain an underground water level prediction model; training the underground water level prediction model based on the graph structure data set;
and predicting the underground water level of the spatial position based on the trained underground water level prediction model.
Further, the characteristic information comprises surface water resource information and geospatial information;
the surface water resource information comprises irrigation quantity, water discharge quantity and total precipitation quantity in a preset period of the area where the node is located; the geospatial information comprises: the lithology of the aeration zone, the landform type, the permeability K1 partition, the name of the landform, the total solid solubility, the permeability coefficient KCP, and the distance between the node and the branch channel.
Further, respectively constructing a feature vector of each node based on the feature information, including:
vectorizing the characteristic information to obtain a characteristic vector of each node; the vectorization processing mode of the characteristic information is one-hot coding and normalization, wherein the one-hot coding and normalization processing are performed on unquantized information, and the normalization processing is performed on the quantized information directly.
Further, based on the characteristic vector weighing the correlation relationship between two nodes, constructing edge structure information for describing the connectivity of the nodes, including:
calculating cosine similarity, Pearson correlation coefficient and Euclidean distance among the feature vectors of all the nodes;
constructing a one-way connection relation graph of all nodes, reserving connection edges meeting preset conditions, and regarding nodes corresponding to the connection edges not meeting the preset conditions as non-connection relations; the preset conditions are that the cosine similarity between the characteristic vectors is more than 0.7, the Pearson correlation coefficient is more than 0.8, and the Euclidean distance is less than 1;
and counting the reserved connection relation after screening to obtain the side structure information for describing the node connectivity.
Further, the depth interval category includes 0-1m, 1-2m, 2-3m, 3-4m, 4-5m and 5m or more.
Further, improvements to the GCN are made, including:
adding a weight matrix in the convolution operation of the first layer and the last layer of the graph of the GCN network, and specifically realizing that: and the spatial node introduces edge connection weight as an aggregation coefficient in the aggregation neighbor node characteristic stage.
Further, the GCN is improved, and the method further comprises the following steps:
adding a graph attention convolution module after a first graph convolution layer of the GCN, wherein the input feature of the graph attention convolution module is the feature output after a neighbor node is aggregated by a first layer of graph convolution operation, and the output feature of the graph attention convolution module is the aggregated feature after the weights of the node and the neighbor node are automatically learned and updated; and after the feature output of the graph attention convolution module is obtained, the feature output is fused with the output feature of the first graph convolution layer, and the fused feature is input into the last graph convolution layer.
Further, after enabling the prediction of the groundwater level of the spatial location, the method further comprises:
and carrying out effectiveness evaluation on the prediction result of the underground water level prediction model.
Further, the evaluating the effectiveness of the prediction result of the groundwater level prediction model includes:
and comparing the types of the predicted result and the real result, and evaluating the effectiveness of the model through the prediction accuracy.
In another aspect, the present invention further provides a groundwater level height prediction system based on a multi-feature map network, where the groundwater level height prediction system based on the multi-feature map network includes:
a graph structure data set construction module to:
selecting multiple underground water level monitoring points at different positions as space nodes for constructing a graph structure data set, acquiring characteristic information of each node, and respectively constructing a characteristic vector of each node based on the characteristic information;
based on the characteristic vector, weighing the correlation relationship between the two nodes, constructing edge structure information for describing the connectivity of the nodes, and taking the reciprocal of the geographic distance between the two connected nodes as the edge connection weight of the nodes;
constructing a graph structure data set by taking the characteristic information and the edge structure information of the nodes as sample characteristics and taking the depth interval category to which the underground water level burial depth corresponding to the nodes belongs as a sample label;
the underground water height prediction model building and training module is used for improving a Graph Convolutional Network (GCN) to obtain an underground water level prediction model; training the underground water level prediction model based on the graph structure data set constructed by the graph structure data set construction module;
and the underground water height prediction module is used for constructing an underground water level prediction model trained by the training module based on the underground water height prediction model and realizing the prediction of the underground water level of the spatial position.
In yet another aspect, the present invention also provides an electronic device comprising a processor and a memory; wherein the memory has stored therein at least one instruction that is loaded and executed by the processor to implement the above-described method.
In yet another aspect, the present invention also provides a computer-readable storage medium having at least one instruction stored therein, the instruction being loaded and executed by a processor to implement the above method.
The technical scheme provided by the invention has the beneficial effects that at least:
the invention provides a method for predicting the height of underground water level, which is a method for predicting the height of underground water level through surface water resources and geographic space attributes. A graph structure data set is constructed by taking spatial position nodes as units, distance weights among the nodes are added, and an attention mechanism and a feature fusion mechanism are introduced, so that the method can obtain better feature expression of the nodes, and meanwhile, the method can train a model in a semi-supervised learning mode under the condition that the underground water level of part of the nodes is unknown, and therefore accurate prediction of the underground water level height of the spatial position is achieved. The invention researches the evolution relation of surface water and underground water through a data-oriented artificial intelligence technology, and solves the problems that the existing surface water and underground water evolution method can not quantify and has insufficient accuracy under complex conditions.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic diagram illustrating an implementation flow of a groundwater level height prediction method based on a multi-feature map network according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a spatial node feature descriptor provided by an embodiment of the present invention;
FIG. 3 is a flow chart of attention factor calculation provided by an embodiment of the present invention;
figure 4 is a schematic diagram of an improved GCN network architecture provided by an embodiment of the present invention;
FIG. 5 is a schematic diagram of a confusion matrix of predicted results for nodes in a test set space according to an embodiment of the present invention;
FIG. 6 is a schematic structural diagram of a groundwater level height predicting system based on a multi-feature map network according to an embodiment of the present invention;
fig. 7 is a block diagram of an electronic device to which the method of the invention is applied.
Detailed Description
To make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
First embodiment
The embodiment provides a groundwater level height prediction method based on a multi-feature map network, which can be realized by electronic equipment, and the electronic equipment can be a terminal or a server. The execution flow of the underground water level height prediction method based on the multi-feature map network is shown in fig. 1, and the method comprises the following steps:
s1, selecting multiple underground water level monitoring points at different positions as space nodes for constructing a graph structure data set, acquiring characteristic information of each node, and constructing a characteristic vector of each node based on the characteristic information;
s2, weighing the correlation relation between two nodes based on the characteristic vector, constructing edge structure information for describing the connectivity of the nodes, and taking the reciprocal of the geographic distance between the two connected nodes as the edge connection weight of the nodes;
s3, constructing a graph structure data set by taking the characteristic information and the edge structure information of the nodes as sample characteristics and taking the depth interval type to which the underground water level burial depth corresponding to the nodes belongs as a sample label;
s4, improving a Graph Convolution Network (GCN) to obtain an underground water level prediction model; training the underground water level prediction model based on the graph structure data set;
and S5, based on the trained underground water level prediction model, realizing the prediction of the underground water level of the spatial position.
Wherein the characteristic information in S1 includes surface water resource information and geospatial information; the surface water resource information comprises irrigation quantity, water discharge quantity and total precipitation quantity in a preset period of the area where the node is located; the geospatial information includes: the lithology of the aeration zone, the landform type, the permeability K1 partition, the name of the landform, the total solid solubility, the permeability coefficient KCP, and the distance between the node and the branch channel. Based on this, the implementation process of S1 is as follows:
acquiring earth surface irrigation and drainage, meteorological data and geographic attribute information of a plurality of positions in a space, and constructing a feature vector for describing earth surface information of nodes at the positions in the space based on the acquired information; the earth surface irrigation and drainage data is flow statistics, the meteorological data is height statistics, and the geographic attribute data is character information; generating characteristics of spatial nodes by establishing a mapping of groundwater monitoring points and surface spatial location nodes, comprising: and carrying out surface water quantity standardized calculation on irrigation and drainage data and meteorological data, distributing the surface water quantity to a grid unit where the node is located, carrying out feature coding on geographic attribute information, and fusing to obtain a node feature vector. The specific implementation process comprises the following steps:
dividing an irrigation area into a plurality of sub-drainage basins according to an irrigation canal system, wherein each irrigation sub-drainage basin has a clear water diversion source and water diversion data statistics, the water drainage source considers the whole agricultural irrigation area as a whole, and water drainage data distribution is carried out according to the corresponding sub-drainage basin irrigation area weight.
Dividing the sub-watershed into grid units of 100m by 100m, performing spatial superposition on the divided grid units and the drainage sub-watershed, establishing a topological relation between the grid units and the drainage units, and determining the drainage area and the drainage area where each grid unit is located, wherein the calculation mode of the water resource characteristics of the grid units is as follows:
daily irrigation discharge capacity is equal to daily average irrigation discharge capacity multiplied by 24 multiplied by 60
Daily precipitation evaporation capacity is equal to daily precipitation evaporation height multiplied by sub-basin area multiplied by 666.7 multiplied by 10 -4
And establishing a mapping relation between the underground water monitoring points and the surface grid units to form space nodes. The method for counting the geographic spatial attributes of the space nodes according to the coordinate positions of the space nodes comprises the following steps: the lithology of the aeration zone, the landform type, the permeability K1 subarea, the name of the land type, the total soluble solid, the permeability coefficient KCP and the distance between the node and the branch channel. Wherein the values of different permeability of the lithologic character of the aeration zone comprise sand, clay soil, an interlayer water distribution area, sandy soil below the clay soil at the upper part, and the like; the landform types comprise aeolian dunes, river and lake plains, fluvial alluvial plains, modern alluvial plains, hill terrains and the like; the partition value range of the permeability K1 is 1-3, 3-5, 5-10 and 10-20 corresponding to different permeabilities; the land names include corn, wheat, grape, medlar, rice, village, ground vegetable, city, abandoned land, etc. The geospatial attributes are collected through an ArcGis spatial connection function, a surface in which the nodes fall is selected for a surface type graph layer, and points or lines which are closest to the nodes are selected for a point type and line type graph layer. After the collection is completed, the geospatial attributes and the water resource characteristics are fused to construct a node characteristic descriptor, and the spatial node characteristic descriptor is shown in fig. 2.
After obtaining the feature descriptors (feature information) of each spatial node, vectorizing the spatial node descriptors in a specific vectorization mode: and one-hot coding and normalization, coding a non-quantized column first and then normalizing a quantized column, and thus obtaining a feature vector of each space node.
Further, in this embodiment, the feature matrix of the spatial nodes is represented by a two-dimensional matrix form, where the row index represents the number of the spatial nodes and the index, and the column index represents the attribute features of the nodes, including the ground irrigation and drainage, the meteorological data, and the multi-dimensional topographic attribute data.
Further, in this embodiment, in the above S2, the edge structure information is constructed by calculating specific indexes of the feature vector, including three indexes of cosine similarity, pearson correlation coefficient, and euclidean distance. The cosine similarity evaluates the similarity of two vectors by calculating the cosine value of the included angle of the two vectors, and judges whether the two vectors approximately point to the same direction, and the closer the cosine value is to 1, the higher the similarity is. The calculation formula is as follows:
the Pearson correlation coefficient is used for measuring whether two vectors are on the same line or not and measuring the linear correlation relationship between distance variables, and is obtained by calculating the covariance and the standard deviation between two samples, and the closer the absolute value of the correlation coefficient is to 1, the stronger the linear correlation is. The calculation formula is as follows:
the euclidean distance is a commonly used distance definition, which refers to a real distance between two points in an m-dimensional space, and is used to calculate a distance between two vectors, and the value range is from 0 to positive infinity, and obviously, if the distance between two vector spaces is smaller, the vectors are certainly more similar, and the calculation formula is as follows:
specifically, the construction process of the spatial node connectivity edge structure is as follows:
and after the connectivity edge structure information is obtained, corresponding to a geographic space according to the index, acquiring geographic coordinates of the spatial nodes, respectively calculating the geographic distance of the connected nodes through the geographic coordinates, taking the reciprocal of the distance as the edge connection weight of the spatial nodes in the network model, and outputting the edge connection weight to an n multiplied by n sparse matrix, wherein the distance weight corresponding to the nodes without edge connection is 0. In the graph volume process, more information can be aggregated by the neighbor nodes with closer distances, and less characteristic information is contributed by the neighbor nodes with farther distances.
Further, in this embodiment, each correlation evaluation index and the weight of the connected edge are represented by a two-dimensional matrix, a row value and a column value represent a spatial node index, and a matrix element is a calculation result of the correlation evaluation index and the reciprocal distance between nodes; the information of the connected edge structure is represented by an n-x 2 matrix, each row of the matrix is indexed by two spatial nodes, and connectivity between the two nodes is represented.
Further, in this embodiment, the depth interval categories in S3 include 0-1m, 1-2m, 2-3m, 3-4m, 4-5m, and 5m or more, which correspond to different interval categories; the graph structure data set is divided into a training set, a verification set and a test set; the training set is used for training the underground water level height prediction model, the verification set is used for verifying the model accuracy in the training process, and the test set is used for evaluating the trained model. In another semi-supervised learning mode, a model is trained by nodes with known class labels to predict spatial nodes with unknown class labels. Specifically, the implementation process of S3 is as follows:
200 ground water level monitoring points are selected in an agricultural irrigation area as space nodes of a graph data set constructed by the method, and according to the mapping relation between the ground water level monitoring points and a ground surface space grid, feature descriptors of the space nodes are organized to construct feature vectors, wherein the attributes used for describing the features of the space nodes comprise surface water resource attributes and geographic space attributes. The surface water resource attributes comprise space grid moon irrigation, drainage and precipitation amount, and the geospatial attributes comprise aeration zone lithology, landform type, permeability K1 partition, land type name, total soluble solid, permeability coefficient KCP and node-to-canal distance. The underground water level burial depth of the underground water monitoring points is used as a category label of the space nodes, and each category of the underground water level burial depth labels comprises a certain underground water level burial depth range. Such as buried depths of 0-1m, 1-2m, 2-3m, 3-4m, 4-5m, and 5m or more, respectively, corresponding to different categories.
In this embodiment, the graph data set attribute data is represented by a two-dimensional matrix, the dimension is 200 × 12, the first column is an index of a space node, and the last column is a category label corresponding to the groundwater level burial depth of the space node. The middle 10 columns are the feature vectors for each spatial node. The graph data set edge connection data is represented as a two-dimensional matrix with dimensions of 2315 × 2, 200 spatial nodes are represented to have 2315 connected edges, and each row represents two node indexes which are connected with each other.
Further, in this embodiment, in the above S4, the GCN is modified to: the GCN is improved, and a distance weight and an attention mechanism are added, so that the GCN can better utilize the characteristics of surrounding neighbors to learn the characteristics of the nodes; the method comprises the following steps: adding a weight matrix in the convolution operation of the first layer and the last layer of the graph, and specifically realizing that the distance factor is introduced as an aggregation coefficient in the aggregation neighbor node characteristic stage by the space node. Adding a graph attention convolution module after the first graph convolution layer, wherein the input characteristics of the module are the characteristics output after the first layer graph convolution operation aggregates neighbor nodes, and the output characteristics of the module are the aggregated characteristics after automatically learning and updating the weights of the nodes and the neighbor nodes. After obtaining the feature output of the graph attention convolution module, the feature output is fused with the output feature of the first graph convolution layer, and the fused feature is input into the last graph convolution layer.
Specifically, in this embodiment, the implementation process of improving the GCN is as follows:
firstly, it should be noted that the GCN is proposed to solve the problem that irregular data cannot share a convolution kernel during convolution, and a local sensing region and a shared weight are adopted in a conventional convolution, so that spatial features of an image can be well extracted. The graph structure does not have the translation invariance of the picture, so the traditional convolution mode is not suitable for the graph structure. GCN is the first-order local approximation of spectrogram convolution, neighborhood information is aggregated through convolution layers, information transmission of multi-order neighborhoods can be achieved by superposing a plurality of convolution layers, each layer obtains summary of neighbor characteristics of each vertex through multiplication of an adjacent matrix A and a characteristic matrix H, and then multiplies a trainable parameter matrix W, the adjacent matrix is normalized through a transition matrix, and large differences of vertexes with high degrees and vertexes with low degrees in characteristic distribution are prevented.
Based on the above, the first improvement of this embodiment is to add edge connection weight in the graph convolution process, where the weight is represented by the reciprocal of the geospatial distance, and the role of this embodiment is that each node needs to consider not only the degree of a neighbor node but also the distance between the neighbor node and the node when aggregating neighbor node information, so that it can be realized that a neighbor node farther away has a smaller contribution to the node, and a neighbor node closer to the node has a larger contribution to the node. The GCN calculation after adding the distance weight is as follows, wherein H (l+1) Representing the characteristic representation of the central node after one convolution,
obtained by summing the node adjacency matrix and the identity matrix I,
express according to
Degree matrix obtained by summing rows, W (l) And representing a trainable parameter matrix, Dist is an introduced distance weight matrix, and since diagonal elements of Dist are all 0, the Dist needs to be summed with an identity matrix I to ensure that the characteristics of the central node are also taken into consideration in the process of aggregating the characteristics of the neighbor nodes.
The second improvement of the GCN of this embodiment is to add a graph attention module, in which the attention factor is calculated as shown in FIG. 3. The network input is the constructed Graph structure data set, Graph represents the constructed Graph structure data, X represents an adjacent matrix corresponding to the Graph nodes, solid points represent that the corresponding nodes have edge connection relations, and hollow points represent that the nodes have no edge connection relations; after the first graph convolution layer, an attention module is added, and then the output characteristics of the attention module and the output characteristics of the graph convolution layer are fused, so that the improvement can enable the spatial node to more accurately utilize the information of the surrounding neighbor nodes, the prediction accuracy is improved, and the overall structure of the improved GCN model network is shown in figure 4.
Specifically, in this embodiment, after the spatial node feature and the edge connection information thereof are input into the first graph convolution layer, the graph node structure does not change, each node generates a feature representation with a length of 32, and the dimension of the parameter matrix W in this process is 10 × 32, which corresponds to the input feature dimension and the output feature dimension, respectively. The characteristics output by the first graph convolution layer are processed by a ReLU activation function, so that model training can be accelerated, the problem of gradient disappearance is solved, and model fitting is prevented by dropout processing.
The node features of the first graph convolutional layer after dropout processing are input into a graph attention module, since the GCN cannot distribute different weights according to the importance of neighbor nodes, although the reciprocal of the spatial distance is added into the first convolutional layer as the weights of different nodes, the weight information cannot be learned and updated in the network training process, and the graph attention layer can learn and update the weight coefficients of the nodes autonomously through training, so that the dependency relationship between global features can be better learned. The attention factor is calculated as follows, wherein
Representing a feed-forward neural network, the parameters can be updated by training, W represents a parameter matrix,
a representation of the characteristics of the node is represented,
center of representationAnd (3) multiplying the total number of the neighbor nodes of the node by the characteristics, splicing, performing LeakyReLU nonlinearity, and performing Softmax normalization to obtain an attention coefficient.
After the attention factor is aggregated on each node of the graph attention layer to the surrounding neighbor node features, a node feature representation with the dimension of 32 is also generated, in this embodiment, the feature output by the graph attention module is fused with the output feature of the first graph convolution layer, the fused feature dimension is n × 64, n represents the number of spatial nodes, and the method can obtain the enhanced representation of the spatial node features. Inputting the fused enhancement features into a graph convolution layer, outputting a result, normalizing an exponential function through log _ softmax, mapping the output to a range of 0-1, calculating loss in the training process of the model through NLLLoss, and correcting parameters of the network model through a back propagation algorithm.
When the underground water level height prediction model is trained, the space node characteristic matrix and the edge connection matrix are read into a memory, the node type is subjected to one-hot coding, an adjacent matrix and a degree matrix of the edge are constructed, the adjacent matrix and a transposed matrix of the edge are summed, a directed graph is converted into an undirected graph, and normalization operation is performed on the characteristic matrix and the degree matrix. When a training set, a verification set and a test set are divided, randomly disordering the space nodes according to the following steps of 3: 1: 2, inputting the training set into the model for training, verifying the effectiveness of the model through a verification set in each training epoch, testing the prediction precision of the model after the test set is used for all the training epochs, calculating the loss of the training set, and updating the model parameters through a back propagation algorithm. The loss of the validation set and the test set is not used for model optimization. Specifically, the training process of the groundwater level height prediction model is as follows:
and repeating the steps to train the improved GCN until the loss is stable and is not reduced any more, and converging the model to obtain the optimal training parameters.
Further, in this embodiment, the implementation process of S5 is as follows:
further, after enabling the prediction of the groundwater level of the spatial location, the method further comprises:
and carrying out effectiveness evaluation on the prediction result of the underground water level prediction model. Specifically, in this embodiment, the validity evaluation index of the groundwater level height prediction model is Accuracy, Accuracy (Accuracy) refers to a proportion of a spatial node type with correct prediction to the total number of nodes in the test set, and the Accuracy calculation method is as follows:
in this embodiment, two data sets with different surface water characteristics and underground water level burying in the rich water period and the dry water period are respectively constructed, and the underground water level height prediction model is evaluated according to the evaluation indexes, and the evaluation results are shown in table 1. In order to explore the performance of the model, different training batches and learning rate sizes are adopted to evaluate the effectiveness of the model.
TABLE 1 groundwater level height prediction model evaluation results
In table 1, Epoch is the number of iterations when the model is trained using the training set, and different training iterations have an influence on the prediction effectiveness of the model. The Learning rate represents the influence range of the error in the training process on the parameter change when the error reversely propagates, and the prediction effect of the model is also influenced by the Learning rate. As can be seen from the above table, under the conditions that the iteration number is 1500 and the learning rate is 0.005, the underground water level height prediction model has better performance in both the rich water period and the dry water period. The prediction confusion matrix for spatial nodes is shown in fig. 5.
In summary, the present embodiment provides a method for predicting the height of the groundwater level through the surface water resource and the geospatial attributes. By constructing a graph structure data set by taking spatial position nodes as a unit, adding inter-node distance weight and introducing an attention mechanism and a feature fusion mechanism, the method can obtain better feature expression of the nodes, and meanwhile, the method of the embodiment can train a model in a semi-supervised learning mode under the condition that the underground water level of part of the nodes is unknown, so that the accurate prediction of the underground water level height of the spatial position is realized, and the problems that the existing surface water and underground water evolution method cannot be quantified and is insufficient in accuracy are solved.
Second embodiment
The embodiment provides a groundwater level height prediction system based on a multi-feature map network, and the structure of the groundwater level height prediction system based on the multi-feature map network is shown in fig. 6, and the system comprises the following modules:
a graph structure data set construction module to:
selecting a plurality of underground water level monitoring points at different positions as space nodes for constructing a graph structure data set, acquiring characteristic information of each node, and respectively constructing a characteristic vector of each node based on the characteristic information;
based on the characteristic vector, weighing the correlation relationship between the two nodes, constructing edge structure information for describing the connectivity of the nodes, and taking the reciprocal of the geographic distance between the two connected nodes as the edge connection weight of the nodes;
constructing a graph structure data set by taking the characteristic information and the edge structure information of the nodes as sample characteristics and taking the depth interval category to which the underground water level burial depth corresponding to the nodes belongs as a sample label;
the underground water height prediction model building and training module is used for improving a Graph Convolutional Network (GCN) to obtain an underground water level prediction model; training the underground water level prediction model based on the graph structure data set constructed by the graph structure data set construction module;
the underground water height prediction module is used for constructing an underground water level prediction model trained by the training module based on the underground water height prediction model and realizing prediction of the underground water level of the spatial position;
and the model evaluation module is used for evaluating the effectiveness of the prediction result of the underground water level prediction model.
The underground water level height prediction system based on the multi-feature map network of the embodiment corresponds to the underground water level height prediction method based on the multi-feature map network of the first embodiment; the functions realized by each functional module in the underground water level height prediction system based on the multi-characteristic diagram network correspond to each flow step in the underground water level height prediction method based on the multi-characteristic diagram network one by one; therefore, it is not described herein.
Third embodiment
The present embodiment provides an electronic device, which includes a processor and a memory; wherein the memory has stored therein at least one instruction that is loaded and executed by the processor to implement the method of the first embodiment.
The electronic device may have a relatively large difference due to different configurations or performances, and may include one or more processors (CPUs) and one or more memories, where at least one instruction is stored in the memory, and the instruction is loaded by the processor and executes the method.
Specifically, as shown in fig. 7, the electronic apparatus may include a processor (CPU)701, which may perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM)702 or a program loaded from a storage section 708 into a Random Access Memory (RAM) 703. The processor may include a general purpose microprocessor (e.g., CPU), an instruction set processor and/or related chip sets and/or a special purpose microprocessor (e.g., an Application Specific Integrated Circuit (ASIC)), among others. The processor may also include on-board memory for caching purposes.
Further, the device may also include an input/output (I/O) interface 705, the input/output (I/O) interface 705 also being connected to the bus 704. And may also include one or more of the following components connected to the I/O interface 705: an input portion 706 including a keyboard, a mouse, and the like; an output section 707 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 708 including a hard disk and the like; and a communication section 709 including a network interface card such as a LAN card, a modem, or the like. The communication section 709 performs communication processing via a network such as the internet. A drive 710 is also connected to the I/O interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like, is mounted on the drive 710 as necessary, so that a computer program read out therefrom is mounted into the storage section 708 as necessary.
Fourth embodiment
The present embodiments provide a computer-readable storage medium having stored therein at least one instruction, which is loaded and executed by a processor, to implement the method of the first embodiment. The computer readable storage medium may be, among others, ROM, random access memory, CD-ROM, magnetic tape, floppy disk, optical data storage device, and the like. The instructions stored therein may be loaded by a processor in the terminal and perform the above-described method.
Furthermore, it should be noted that the present invention may be provided as a method, apparatus or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media having computer-usable program code embodied in the medium.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
It should also be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.
Finally, it should be noted that while the above describes a preferred embodiment of the invention, it will be appreciated by those skilled in the art that, once the basic inventive concepts have been learned, numerous changes and modifications may be made without departing from the principles of the invention, which shall be deemed to be within the scope of the invention. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Claims (10)
1. A groundwater level height prediction method based on a multi-feature map network is characterized by comprising the following steps:
selecting multiple underground water level monitoring points at different positions as space nodes for constructing a graph structure data set, acquiring characteristic information of each node, and respectively constructing a characteristic vector of each node based on the characteristic information;
based on the characteristic vector, weighing the correlation relationship between the two nodes, constructing edge structure information for describing the connectivity of the nodes, and taking the reciprocal of the geographic distance between the two connected nodes as the edge connection weight of the nodes;
constructing a graph structure data set by taking the characteristic information and the edge structure information of the nodes as sample characteristics and taking the depth interval category to which the underground water level burial depth corresponding to the nodes belongs as a sample label;
improving a Graph Convolution Network (GCN) to obtain an underground water level prediction model; training the underground water level prediction model based on the graph structure data set;
and predicting the underground water level of the spatial position based on the trained underground water level prediction model.
2. The method for groundwater level height prediction based on a multi-feature map network as claimed in claim 1, wherein the feature information comprises surface water resource information and geospatial information;
the surface water resource information comprises irrigation quantity, water discharge quantity and total precipitation quantity in a preset period of the area where the node is located; the geospatial information comprises: the lithology of the aeration zone, the landform type, the permeability K1 partition, the name of the landform, the total solid solubility, the permeability coefficient KCP, and the distance between the node and the branch channel.
3. The method for predicting the height of the groundwater level based on the multi-feature map network according to claim 1, wherein the step of respectively constructing the feature vector of each node based on the feature information comprises the following steps:
vectorizing the characteristic information to obtain a characteristic vector of each node; the vectorization processing mode of the characteristic information is one-hot coding and normalization, wherein the one-hot coding and normalization processing are performed on unquantized information, and the normalization processing is performed on the quantized information directly.
4. The method for predicting the height of the groundwater level based on the multi-feature map network according to claim 1, wherein the step of constructing the edge structure information for describing the connectivity of the nodes by weighing the correlation relationship between two nodes based on the feature vectors comprises the following steps:
calculating cosine similarity, Pearson correlation coefficient and Euclidean distance among the feature vectors of all the nodes;
constructing a one-way connection relation graph of all nodes, reserving connection edges meeting preset conditions, and regarding nodes corresponding to the connection edges not meeting the preset conditions as non-connection relations; the preset conditions are that the cosine similarity between the characteristic vectors is larger than 0.7, the Pearson correlation coefficient is larger than 0.8, and the Euclidean distance is smaller than 1;
and counting the reserved connection relation after screening to obtain the side structure information for describing the node connectivity.
5. The method according to claim 1, wherein the depth interval categories include 0-1m, 1-2m, 2-3m, 3-4m, 4-5m, and 5m or more.
6. The method for predicting the height of the groundwater level based on the multi-feature map Network according to claim 1, wherein the improvement of a Graph volume Network (GCN) comprises:
adding a weight matrix in the convolution operation of the first layer and the last layer of the graph of the GCN network, and specifically realizing that: and the spatial node introduces edge connection weight as an aggregation coefficient in the aggregation neighbor node characteristic stage.
7. The method for predicting the height of the groundwater level based on the multi-feature map Network according to claim 6, wherein a Graph volume Network (GCN) is improved, and the method further comprises:
adding a graph attention convolution module after a first graph convolution layer of the GCN, wherein the input feature of the graph attention convolution module is the feature output after a neighbor node is aggregated by a first layer of graph convolution operation, and the output feature of the graph attention convolution module is the aggregated feature after the weights of the node and the neighbor node are automatically learned and updated; and after the feature output of the graph attention convolution module is obtained, the feature output is fused with the output feature of the first graph convolution layer, and the fused feature is input into the last graph convolution layer.
8. The multiple feature map network-based groundwater level height prediction method according to claim 1, wherein after the prediction of the groundwater level of the spatial location is achieved, the method further comprises:
and carrying out effectiveness evaluation on the prediction result of the underground water level prediction model.
9. The method for predicting the height of the underground water level based on the multi-feature map network as claimed in claim 8, wherein the evaluating the effectiveness of the prediction result of the underground water level prediction model comprises:
and comparing the types of the predicted result and the real result, and evaluating the effectiveness of the model through the prediction accuracy.
10. A groundwater level height prediction system based on a multi-feature map network is characterized by comprising:
a graph structure data set construction module to:
selecting multiple underground water level monitoring points at different positions as space nodes for constructing a graph structure data set, acquiring characteristic information of each node, and respectively constructing a characteristic vector of each node based on the characteristic information;
based on the characteristic vector, weighing the correlation relationship between the two nodes, constructing edge structure information for describing the connectivity of the nodes, and taking the reciprocal of the geographic distance between the two connected nodes as the edge connection weight of the nodes;
taking the characteristic information and the edge structure information of the nodes as sample characteristics, and taking the depth interval category to which the underground water level burial depth corresponding to the nodes belongs as a sample label to construct a graph structure data set;
the underground water height prediction model building and training module is used for improving a Graph Convolution Network (GCN) to obtain an underground water level prediction model; training the underground water level prediction model based on the graph structure data set constructed by the graph structure data set construction module;
and the underground water height prediction module is used for constructing an underground water level prediction model trained by the training module based on the underground water height prediction model and realizing the prediction of the underground water level of the spatial position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210300649.XA CN114841402A (en) | 2022-03-25 | 2022-03-25 | Underground water level height prediction method and system based on multi-feature map network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210300649.XA CN114841402A (en) | 2022-03-25 | 2022-03-25 | Underground water level height prediction method and system based on multi-feature map network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114841402A true CN114841402A (en) | 2022-08-02 |
Family
ID=82562803
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210300649.XA Pending CN114841402A (en) | 2022-03-25 | 2022-03-25 | Underground water level height prediction method and system based on multi-feature map network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114841402A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115689368A (en) * | 2022-11-10 | 2023-02-03 | 华能西藏雅鲁藏布江水电开发投资有限公司 | Runoff forecasting model evaluation method based on life cycle |
| CN117576099A (en) * | 2024-01-16 | 2024-02-20 | 季华实验室 | Liquid bottle separation detection method and device and computer readable storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030078901A1 (en) * | 2001-10-22 | 2003-04-24 | Coppola Emery J. | Neural network based predication and optimization for groundwater / surface water system |
| KR101703972B1 (en) * | 2015-09-30 | 2017-02-22 | 한국지질자원연구원 | System and method for predicting groundwater potential area using spatial information |
| CN112182063A (en) * | 2020-09-16 | 2021-01-05 | 河海大学 | Method for constructing hydrological forecasting model based on space-time characteristics |
| CN112561133A (en) * | 2020-11-30 | 2021-03-26 | 西安科锐盛创新科技有限公司 | Water flow prediction system and method based on neural network |
| CN112561134A (en) * | 2020-11-30 | 2021-03-26 | 西安科锐盛创新科技有限公司 | Neural network-based water flow prediction method and device and electronic equipment |
| CN112949944A (en) * | 2021-04-13 | 2021-06-11 | 北京科技大学 | Underground water level intelligent prediction method and system based on space-time characteristics |
| CN114118410A (en) * | 2021-11-24 | 2022-03-01 | 阿里巴巴达摩院(杭州)科技有限公司 | Method, device and storage medium for extracting node feature of graph structure |
-
2022
- 2022-03-25 CN CN202210300649.XA patent/CN114841402A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030078901A1 (en) * | 2001-10-22 | 2003-04-24 | Coppola Emery J. | Neural network based predication and optimization for groundwater / surface water system |
| KR101703972B1 (en) * | 2015-09-30 | 2017-02-22 | 한국지질자원연구원 | System and method for predicting groundwater potential area using spatial information |
| CN112182063A (en) * | 2020-09-16 | 2021-01-05 | 河海大学 | Method for constructing hydrological forecasting model based on space-time characteristics |
| CN112561133A (en) * | 2020-11-30 | 2021-03-26 | 西安科锐盛创新科技有限公司 | Water flow prediction system and method based on neural network |
| CN112561134A (en) * | 2020-11-30 | 2021-03-26 | 西安科锐盛创新科技有限公司 | Neural network-based water flow prediction method and device and electronic equipment |
| CN112949944A (en) * | 2021-04-13 | 2021-06-11 | 北京科技大学 | Underground water level intelligent prediction method and system based on space-time characteristics |
| CN114118410A (en) * | 2021-11-24 | 2022-03-01 | 阿里巴巴达摩院(杭州)科技有限公司 | Method, device and storage medium for extracting node feature of graph structure |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115689368A (en) * | 2022-11-10 | 2023-02-03 | 华能西藏雅鲁藏布江水电开发投资有限公司 | Runoff forecasting model evaluation method based on life cycle |
| CN117576099A (en) * | 2024-01-16 | 2024-02-20 | 季华实验室 | Liquid bottle separation detection method and device and computer readable storage medium |
| CN117576099B (en) * | 2024-01-16 | 2024-05-10 | 季华实验室 | Liquid bottle separation detection method and device and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Shu et al. | Monthly streamflow forecasting using convolutional neural network | |
| Jasiewicz et al. | A new GRASS GIS toolkit for Hortonian analysis of drainage networks | |
| Fonstad | Cellular automata as analysis and synthesis engines at the geomorphology–ecology interface | |
| CN106780089B (en) | Permanent basic farmland planning method based on neural network cellular automaton model | |
| Abdollahi et al. | Daily mean streamflow prediction in perennial and non-perennial rivers using four data driven techniques | |
| CN110232471B (en) | Rainfall sensor network node layout optimization method and device | |
| CN114841402A (en) | Underground water level height prediction method and system based on multi-feature map network | |
| Barnes et al. | Computing water flow through complex landscapes–Part 2: Finding hierarchies in depressions and morphological segmentations | |
| CN103093114A (en) | Distributed-type river basin water deficit calculating method based on terrain and soil characteristics | |
| Ditthakit et al. | Comparative study of machine learning methods and GR2M model for monthly runoff prediction | |
| Barnes et al. | Computing water flow through complex landscapes, Part 3: Fill-Spill-Merge: Flow routing in depression hierarchies | |
| Haag et al. | Development of a data model to facilitate rapid watershed delineation | |
| CN114399103A (en) | CNN-based land-water integrated river water quality space-time continuous prediction method | |
| CN116562414A (en) | Land water reserve change prediction method based on multisource hydrographic meteorological data | |
| CN116955450A (en) | Hydrologic forecasting method and system for fusion of drainage basin production confluence space-time process | |
| Ahmadi et al. | Streamflow prediction using a hybrid methodology based on variational mode decomposition (VMD) and machine learning approaches | |
| Gao et al. | A framework for automatic calibration of SWMM considering input uncertainty | |
| Chen et al. | Urban inundation rapid prediction method based on multi-machine learning algorithm and rain pattern analysis | |
| CN113282572B (en) | Gridding division method and system for farmland and terminal equipment | |
| Dehghani et al. | Evaluation of statistical models and modern hybrid artificial intelligence in the simulation of precipitation runoff process | |
| CN118035777A (en) | Air pollutant concentration prediction method and device based on space-time diagram data | |
| Santra et al. | Delineation of hydrologically similar units in a watershed based on fuzzy classification of soil hydraulic properties | |
| Saghafian et al. | Backcasting long-term climate data: evaluation of hypothesis | |
| Jiang et al. | A modified HASM algorithm and its application in DEM construction | |
| Ma et al. | A high-precision hydrodynamic model coupled with the hydrological habitat suitability model to reveal estuarine vegetation distribution |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |