Method for constructing correlation relationship and knowledge graph of water conservancy objects in joint dispatch of control water conservancy projects in river basin

Abstract

The application discloses a kind of watershed control nature water conservancy engineering joint scheduling water conservancy object correlation and knowledge graph construction method, first according to the actual demand of watershed water engineering multi-objective joint scheduling, constructs watershed water engineering multi-objective joint scheduling spatial database;Secondly based on GIS spatial analysis method, water conservancy object spatial correlation is automatically extracted from spatial data, and water conservancy object spatial correlation data information is formed;Thirdly, based on the spatial topological relation of river system, a linear reference system is constructed, a mapping relationship between water conservancy objects and attribute data including spatial attributes, boundary constraints of joint scheduling, model services, etc. and river system network is established using dynamic segmentation technology, supporting the dynamic and flexible changes of multi-objective and multi-scenario joint scheduling is realized;Finally, the knowledge graph of water conservancy object correlation of watershed water engineering joint scheduling is constructed, which provides support for building joint scheduling business computing process, information query and visualization display.

Description

Technical Field

[0001] This invention relates to the field of joint scheduling technology for watershed water projects, and in particular to a method for constructing the association relationship and knowledge graph of water conservancy objects in the joint scheduling of watershed control water conservancy projects. Background Technology

[0002] The water conservancy projects involved in the joint scheduling of water conservancy projects in the basin are diverse in type and large in number, and the scheduling and control are complex. The scheduling objectives have also expanded from flood control scheduling to multiple aspects such as water resources, water ecology, sediment, navigation, and emergency response, and have been upgraded to achieve joint intelligent scheduling of multiple objectives in the basin. Therefore, the relationship between water conservancy objects is of great significance for the construction of the business flow and calculation flow of joint scheduling of water conservancy projects.

[0003] Traditional watershed water conservancy project scheduling either involves the scheduling of a single key project or the joint scheduling of a few projects with a fixed combination. The business application requirements are relatively simple, and the application scenarios are relatively fixed. Therefore, scheduling decision support systems can often meet the requirements by adopting a customized development model. However, as the number of projects involved in scheduling increases, the scope of joint scheduling becomes wider, the number of related water conservancy objects increases, and the variety of business combinations and needs increases, more, higher, and newer demands are placed on aspects such as flexible adjustment of multiple scheduling objectives, model assembly and rapid adaptation, dynamic construction and flexible adjustment of business application scenarios, and visualization of calculation results. The traditional customized development model is obviously unable to meet the above requirements.

[0004] Therefore, it is necessary to design a method for constructing the association relationship and knowledge graph of water conservancy objects in the joint scheduling of water conservancy projects for watershed control to solve the above problems. Summary of the Invention

[0005] The technical problem to be solved by this invention is to provide a method for constructing the association relationship and knowledge graph of water conservancy objects for joint scheduling of water conservancy projects that control water conservancy projects in a basin. This solves the problem that the existing customized development model of water conservancy project scheduling system is relatively simple and cannot meet the requirements of joint scheduling under large-scale projects, multiple types of water conservancy objects, multiple business combinations and needs.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: Methods for constructing knowledge graphs and establishing relationships among water conservancy objects involved in the joint scheduling of watershed-controlling water conservancy projects, including: S1, Construct a spatial database for joint scheduling of watershed water projects: By utilizing watershed water conservancy data resources, spatial data is extracted through data aggregation, extraction, and processing to construct a spatial database for joint scheduling of watershed water projects. The database data includes basic geospatial data, remote sensing image data, water conservancy project spatial data, and economic and social data. S2, Extraction of spatial associations of water conservancy objects: GIS analysis methods are used to extract spatial relationships between water conservancy objects; GIS hydrological analysis methods are used to determine the relationship between the main stream and tributaries of river lines by classifying river lines, and adjacency analysis methods are used to determine the adjacency relationship between water conservancy point objects and river line objects; S3, Construct spatial relationships between water conservancy objects; S4, Construct the attribute mapping relationship for water conservancy objects: A linear reference system is constructed based on the river system network of the basin. Dynamic segmentation technology is used to realize the mapping between engineering scheduling-related attributes and river system water conservancy objects, including engineering design data mapping, monitoring data mapping, model service mapping and computational boundary condition mapping. S5, construct a knowledge graph of the relationship between water conservancy objects in the joint scheduling of water conservancy projects; A knowledge graph of water conservancy object associations is constructed based on the generated spatial associations and attribute mappings of water conservancy objects. The knowledge graph is constructed and stored using the RDF triple data model of the resource description framework. S6, building knowledge graph applications.

[0007] Preferably, in step S1, constructing the joint scheduling spatial database for watershed water projects includes: Utilize the water conservancy data resources of the basin data resource catalog, data resource center and big data center to extract spatial data required for the joint scheduling of water projects; For a single controlling water conservancy project, determine and extract its watershed range. Dispatch Scope Scope of Dispatch Impact The three polygons describing the above ranges are merged into a large polygon S, forming the total distribution range of all water conservancy objects related to the water project scheduling. Assuming the number of water projects participating in the joint scheduling of water conservancy projects in the basin is n, what is the distribution range of all water conservancy objects related to the m-th (m ≤ n) water project? and the overall distribution range of joint scheduling of all water projects Expressed as follows: ; ; in, , , These represent the basin area, scheduling area, and impact area of ​​the m-th controlling water conservancy project, respectively. The total distribution range of all water conservancy objects is obtained by merging them. Then, spatial data of all water conservancy objects required for joint scheduling of water projects are extracted from existing data resources using GIS spatial query technology. A spatial data model is established using the idea of ​​data classification and hierarchical organization. Water conservancy objects are classified into point, line, and surface objects according to their geometric types. They are also classified and organized into layers according to their attribute types, including hydrological control station layers, river channel measurement control section layers, river layers, reservoir layers, and hydropower station layers.

[0008] Furthermore, in step S1, the basic geospatial data includes natural geospatial data such as topography, rivers, lakes, roads, pipelines, and towns; the remote sensing image data includes up-to-date high-resolution remote sensing image data within the scope of the engineering scheduling impact, used to display the current status of the natural background within the basin; the water conservancy engineering spatial data includes the spatial layout of the control water conservancy projects participating in the joint scheduling, as well as various important water conservancy projects within the scope of impact, and related basic engineering design data, including reservoirs, hydropower stations, flood storage and detention areas, dikes, culverts, and pumping stations; and the economic and social data includes economic and social data on administrative divisions, land use, industrial layout, and population distribution.

[0009] Preferably, in step S2, the extraction of spatial association relationships of water conservancy objects includes: S201, Extraction of the relationship between main streams and tributaries of river systems, including: Using GIS hydrological analysis methods, watershed topographic DEM data was obtained, and depression filling preprocessing was performed on the DEM data. Perform traffic analysis on the extracted DEM data; Based on the flow analysis results, the river network of the watershed is extracted from the extracted DEM data; The Shreve river classification method was used to classify the extracted watershed river network. Vectorize the main streams and tributaries after river classification to achieve river network vectorization; By overlaying the water system data in the vectorized river network and water project joint scheduling spatial database, the spatial relationship between the main stream and tributaries of the river system is extracted. S202, The adjacency relationship between point objects and river systems is determined; S203, Determination of upstream and downstream relationships of point-like water conservancy objects.

[0010] Preferably, in step S201, the river classification of the extracted watershed river network using the Shreve river classification method includes: Assigning a class to each river in the river system network, represented by a corresponding numerical value, with higher classes and larger values ​​further downstream. This process involves processing all river lines in the river system network, specifically including: In S2011, a river line with one end not shared with any other river lines is called a river line without tributary connection, and the classification number of the river line without tributary connection is set to 1; the classification number of the currently processed river line is denoted as G, and initially G = 1; In S2012, process all river lines with a classification number of 1, find the river lines with a common end and a classification number of 1, set the classification number of the river lines connected to them to 2, and mark the processed river lines until all river lines with a classification number of 1 are processed, and update the value of G to 2; In S2013, process all river lines with a classification number of G, find the river lines connected to the classification number of G and not yet processed, set or modify their classification numbers, and the value of the classification number is the sum of the classification numbers of all river lines with set classification numbers connected to this river line, mark the processed river lines, and at the same time update the value of G, and set the value of G to the minimum value among the classification numbers newly assigned in this round; Among them, the unprocessed river lines include river lines with no set classification number or a classification number of 1; In S2014, check whether there are unprocessed river lines. If all river lines are processed, end; otherwise, repeat step S2013.

[0011] Preferably, in step S201, when extracting the main - tributary spatial relationship of the river system, determine the main - tributary relationship of the river based on the river classification number: Traverse each river line, obtain the classification number g_c of the current river line R_c, then query all the river lines connected to this river line, and respectively obtain their classification numbers, and take the maximum value g_max. If g_max > g_c, the river line R_max corresponding to the classification number g_max is the downstream section, and R_c is the upstream section, and the river line R_c flows into R_max; if g_max < g_c, the river line R_c is the most downstream section.

[0012] Preferably, in step S202, the determination of the adjacency relationship between the point object and the river system includes: Project the water conservancy point object P onto the surrounding river line objects, and select the river line L_min with the shortest projection distance, then it is determined that the point object P is adjacent to the river line L_min.

[0013] Preferably, in step S203, the determination of the upstream - downstream relationship of the point - type water conservancy object includes: The upstream - downstream relationship between point objects is judged according to the upstream - downstream relationship of the river lines they are adjacent to, and the upstream - downstream relationship of the point objects is the same as that of the river lines they are adjacent to; if two point objects are on the same river line, then it can be judged according to the elevation information of the point objects, and the one with a larger elevation value is upstream and the one with a smaller elevation value is downstream.

[0014] Preferably, in step S3, constructing the spatial association relationship of water conservancy objects includes constructing a linear reference system, specifically including: Water conservancy spatial data processing and quality inspection: For isal rivers or bilinear rivers, extract the river centerline and use the centerline to represent the river. Check the quality of the water conservancy spatial data, including: checking for breaks and self-intersections between lines; checking for hanging points and hanging lines; checking for topological connection errors; and checking for duplicate feature objects. Generate a reference model for a linear water system network: Based on natural river system data, a linear reference network of river systems is generated. Each natural river line is assigned a unique number as a river ID, and the starting point of the river line is set. Using GIS spatial data analysis tools, a measurement value M is generated for each point on the river line in the reference system. The value of M is the length along the river line to the starting point of the river line, which is the upstream or downstream endpoint of the river. The linear reference network of river systems is saved in the spatial database in the form of feature classes. Based on the linear reference network of the river system, river segments are divided, and the river lines in the reference system are divided, taking into account the meandering and tortuous nature of the river system and the natural morphological characteristics of the river gradient. When constructing the linear water system network reference model, a segmentation strategy is adopted, with the principle of keeping the segmented river segments straight, starting from the uppermost part of a river and gradually dividing it downwards.

[0015] Furthermore, in step S3, the water conservancy objects need to be classified. Based on the spatial characteristics of the water conservancy objects, they are abstracted into point objects, line objects, and surface objects. Point objects include water conservancy objects with point-like distribution characteristics, such as hydrological stations, control stations, river control monitoring sections, reservoirs, and hydropower stations. Line objects include water conservancy objects with linear characteristics, such as rivers, river section projects, and dikes. Surface objects include water conservancy objects with surface characteristics, such as flood storage and detention areas, lakes, watershed units, forecasting units, engineering impact areas, and important protection areas.

[0016] Furthermore, in step S3, the spatial relationships between water conservancy objects include: the topological connection relationship and the relationship between main streams and tributaries of the river system in the basin; the adjacency relationship between water conservancy point objects such as monitoring stations, reservoirs and control sections and river system line objects; and the upstream and downstream relationships between water conservancy point objects.

[0017] Preferably, in step S4, constructing the attribute mapping relationship of the water conservancy object includes: The river line reference model, which incorporates a reference system and measurement methods, is divided into networks using dynamic segmentation techniques. Specifically, this includes: An event table is constructed to store event information, including point events and line events. Point events and line events are associated with various constraints, scheduling procedures, and scheduling model services according to the actual needs of water project scheduling. The attribute field types and number of fields of point events and line events are different. The attribute values ​​of events are not directly stored in the point event and line event tables, but are stored in a separate attribute table and associated by event ID, event type, and attribute ID. A dynamic segmentation model is constructed to establish the transformation between the spatial location of water project scheduling events and the linear reference system, enabling information retrieval related to the spatial location of water project scheduling events, i.e., the transformation between one-dimensional and two-dimensional coordinates; the construction method is as follows: Based on the linear water system network reference model, a linear interpolation method based on adjacent reference points is used to obtain the coordinates of the interpolation points; the formula for interpolating the coordinates of the point event location points is as follows: ; ; ; In the formula, Pa is the location of the point event to be calculated, and P1 and P2 are the endpoints of the river segments closest to the point event Pa in the linear reference network model. The distance from Pa to the endpoints of river segments P1 and P2 is given by the linear reference method. The location coordinates of the point event can be interpolated using the above formula, which is used for querying and displaying point event information based on its spatial location. The coordinates of the start and end points of the line event are calculated using the point event location interpolation method.

[0018] Preferably, in step S6, constructing the knowledge graph application includes: Utilizing knowledge graphs to configure water project scheduling business processes: To address the diverse needs of joint scheduling of water conservancy projects in a river basin, a knowledge graph is used to organize the water conservancy objects participating in the scheduling, as well as the business calculation processes on the water conservancy objects. The models, boundary conditions, and model inputs required for business calculations are dynamically mapped to the water conservancy objects as event attributes. Using knowledge graphs to visualize and query water project scheduling systems: Based on knowledge graphs, the topological relationships of water conservancy objects are constructed, and the topological relationships and business connections between water conservancy objects participating in the joint scheduling of water projects are displayed in a visual way. According to the dynamic segmentation coordinate interpolation technology, the model calculation results are accurately located to the digital river system network, realizing the accurate mapping and visualization of model boundary and scheduling result attribute data to two-dimensional and three-dimensional dimensions.

[0019] The beneficial effects of this invention are as follows: 1. This invention organizes various spatial data required for the joint scheduling of watershed water projects and constructs a spatial database for joint scheduling of watershed water projects. It employs GIS spatial analysis methods to automatically extract and construct spatial relationships between water conservancy objects. Based on the river network within the watershed, a linear reference system is established, and dynamic segmentation technology is used to establish a dynamic mapping between various attribute data and water conservancy object entities. Building upon the aforementioned achievements, a knowledge graph of the relationships between water conservancy objects in the joint scheduling of watershed water conservancy projects is constructed. 2. The spatial database constructed by this invention provides basic data support for the visualization and spatial data services of joint scheduling of water conservancy projects in the basin; the use of GIS spatial analysis technology to automatically extract the spatial relationships of water conservancy objects greatly improves data processing efficiency and reduces the workload of manual processing; the use of dynamic segmentation technology to realize the dynamic mapping of attribute data provides technical support for the flexible adjustment of multi-objective and multi-scenario business in joint scheduling of water projects; the knowledge graph constructed by this invention provides support for data management, information query and visualization of scheduling results in joint scheduling of water projects. Attached Figure Description

[0020] The present invention will be further described below with reference to the accompanying drawings and embodiments: Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 This is a schematic diagram of the process for extracting the relationship between the main stream and tributaries of the river system in the present invention; Figure 3 This is a schematic diagram of river classification in this invention; Figure 4 This is a schematic diagram of the linear reference system model in this invention; Figure 5 This is a schematic diagram of the dynamic segmented data model in this invention; Figure 6 This is a schematic diagram of the construction of the knowledge graph of water conservancy objects in this invention. Detailed Implementation

[0021] Example 1: like Figure 1 As shown, the method for constructing the association relationship and knowledge graph of water conservancy objects in the joint scheduling of water conservancy projects controlling the watershed includes: S1, Construct a spatial database for joint scheduling of watershed water projects: By utilizing watershed water conservancy data resources, spatial data is extracted through data aggregation, extraction, and processing to construct a spatial database for joint scheduling of watershed water projects. The database data includes basic geospatial data, remote sensing image data, water conservancy project spatial data, and economic and social data. S2, Extraction of spatial associations of water conservancy objects: GIS analysis methods are used to extract spatial relationships between water conservancy objects; GIS hydrological analysis methods are used to determine the relationship between the main stream and tributaries of river lines by classifying river lines, and adjacency analysis methods are used to determine the adjacency relationship between water conservancy point objects and river line objects; S3, Construct spatial relationships between water conservancy objects; S4, Construct the attribute mapping relationship for water conservancy objects: A linear reference system is constructed based on the river system network of the basin. Dynamic segmentation technology is used to realize the mapping between engineering scheduling-related attributes and river system water conservancy objects, including engineering design data mapping, monitoring data mapping, model service mapping and computational boundary condition mapping. S5, construct a knowledge graph of the relationship between water conservancy objects in the joint scheduling of water conservancy projects; A knowledge graph of water conservancy object associations is constructed based on the generated spatial associations and attribute mappings of water conservancy objects. The knowledge graph is constructed and stored using the RDF triple data model of the resource description framework. S6, building knowledge graph applications.

[0022] Preferably, in step S1, constructing the joint scheduling spatial database for watershed water projects includes: Utilize the water conservancy data resources of the basin data resource catalog, data resource center and big data center to extract spatial data required for the joint scheduling of water projects; For a single controlling water conservancy project, determine and extract its watershed range. Dispatch Scope Scope of Dispatch Impact The three polygons describing the above ranges are merged into a large polygon S, forming the total distribution range of all water conservancy objects related to the water project scheduling. Assuming the number of water projects participating in the joint scheduling of water conservancy projects in the basin is n, what is the distribution range of all water conservancy objects related to the m-th (m ≤ n) water project? and the overall distribution range of joint scheduling of all water projects Expressed as follows: ; ; in, , , These represent the basin area, scheduling area, and impact area of ​​the m-th controlling water conservancy project, respectively. The total distribution range of all water conservancy objects is obtained by merging them. Then, spatial data of all water conservancy objects required for joint scheduling of water projects are extracted from existing data resources using GIS spatial query technology. A spatial data model is established using the idea of ​​data classification and hierarchical organization. Water conservancy objects are classified into point, line, and surface objects according to their geometric types. They are also classified and organized into layers according to their attribute types, including hydrological control station layers, river channel measurement control section layers, river layers, reservoir layers, and hydropower station layers.

[0023] Furthermore, in step S1, the basic geospatial data includes natural geospatial data such as topography, rivers, lakes, roads, pipelines, and towns; the remote sensing image data includes up-to-date high-resolution remote sensing image data within the scope of the engineering scheduling impact, used to display the current status of the natural background within the basin; the water conservancy engineering spatial data includes the spatial layout of the control water conservancy projects participating in the joint scheduling, as well as various important water conservancy projects within the scope of impact, and related basic engineering design data, including reservoirs, hydropower stations, flood storage and detention areas, dikes, culverts, and pumping stations; and the economic and social data includes economic and social data on administrative divisions, land use, industrial layout, and population distribution.

[0024] like Figure 2 As shown, preferably, in step S2, the extraction of spatial association relationships of water conservancy objects includes: S201, Extraction of the relationship between main streams and tributaries of river systems, including: Using GIS hydrological analysis methods, watershed topographic DEM data was obtained, and depression filling preprocessing was performed on the DEM data. Perform traffic analysis on the extracted DEM data; Based on the flow analysis results, the river network of the watershed is extracted from the extracted DEM data; The Shreve river classification method was used to classify the extracted watershed river network. Vectorize the main streams and tributaries after river classification to achieve river network vectorization; By overlaying the water system data in the vectorized river network and water project joint scheduling spatial database, the spatial relationship between the main stream and tributaries of the river system is extracted. S202, The adjacency relationship between point objects and river systems is determined; S203, Determination of upstream and downstream relationships of point-like water conservancy objects.

[0025] like Figure 3 As shown, preferably, in step S201, the river classification of the extracted watershed river network using the Shreve river classification method includes: Assigning a class to each river in the river system network, represented by a corresponding numerical value, with higher classes and larger values ​​further downstream. This process involves processing all river lines in the river system network, specifically including: In S2011, a river line with one end not shared with any other river lines is called a river line without tributary connection, and the classification number of the river line without tributary connection is set to 1; the classification number of the currently processed river line is denoted as G, and initially G = 1; In S2012, process all river lines with a classification number of 1, find the river lines with a common end and a classification number of 1, set the classification number of the river lines connected to them to 2, and mark the processed river lines until all river lines with a classification number of 1 are processed, and update the value of G to 2; In S2013, process all river lines with a classification number of G, find the river lines connected to the classification number of G and not yet processed, set or modify their classification numbers, and the value of the classification number is the sum of the classification numbers of all river lines with set classification numbers connected to this river line, mark the processed river lines, and at the same time update the value of G, and set the value of G to the minimum value among the classification numbers newly assigned in this round; Among them, the unprocessed river lines include river lines with no set classification number or a classification number of 1; In S2014, check whether there are unprocessed river lines. If all river lines are processed, end; otherwise, repeat step S2013.

[0026] Preferably, in step S201, when extracting the main - tributary spatial relationship of the river system, determine the main - tributary relationship of the river based on the river classification number: Traverse each river line, obtain the classification number g_c of the current river line R_c, then query all the river lines connected to this river line, and respectively obtain their classification numbers, and take the maximum value g_max. If g_max > g_c, the river line R_max corresponding to the classification number g_max is the downstream section, and R_c is the upstream section, and the river line R_c flows into R_max; if g_max < g_c, the river line R_c is the most downstream section.

[0027] Preferably, in step S202, the determination of the adjacency relationship between the point object and the river system includes: Project the water conservancy point object P onto the surrounding river line objects, and select the river line L_min with the shortest projection distance, then it is determined that the point object P is adjacent to the river line L_min.

[0028] Preferably, in step S203, the determination of the upstream - downstream relationship of the point - type water conservancy object includes: The upstream - downstream relationship between point objects is judged according to the upstream - downstream relationship of the river lines they are adjacent to, and the upstream - downstream relationship of point objects is the same as that of the river lines they are adjacent to; if two point objects are on the same river line, then it can be judged according to the elevation information of the point objects, and the one with a larger elevation value is upstream and the one with a smaller elevation value is downstream.

[0029] like Figure 4 As shown, preferably, in step S3, constructing the spatial association relationship of water conservancy objects includes constructing a linear reference system, specifically including: Water conservancy spatial data processing and quality inspection: For isal rivers or bilinear rivers, extract the river centerline and use the centerline to represent the river. Check the quality of the water conservancy spatial data, including: checking for breaks and self-intersections between lines; checking for hanging points and hanging lines; checking for topological connection errors; and checking for duplicate feature objects. Generate a reference model for a linear water system network: Based on natural river system data, a linear reference network of river systems is generated. Each natural river line is assigned a unique number as a river ID, and the starting point of the river line is set. Using GIS spatial data analysis tools, a measurement value M is generated for each point on the river line in the reference system. The value of M is the length along the river line to the starting point of the river line, which is the upstream or downstream endpoint of the river. The linear reference network of river systems is saved in the spatial database in the form of feature classes. Based on the linear reference network of the river system, river segments are divided, and the river lines in the reference system are divided, taking into account the meandering and tortuous nature of the river system and the natural morphological characteristics of the river gradient. When constructing the linear water system network reference model, a segmentation strategy is adopted, with the principle of keeping the segmented river segments straight, starting from the uppermost part of a river and gradually dividing it downwards.

[0030] Furthermore, in step S3, the water conservancy objects need to be classified. Based on the spatial characteristics of the water conservancy objects, they are abstracted into point objects, line objects, and surface objects. Point objects include water conservancy objects with point-like distribution characteristics, such as hydrological stations, control stations, river control monitoring sections, reservoirs, and hydropower stations. Line objects include water conservancy objects with linear characteristics, such as rivers, river section projects, and dikes. Surface objects include water conservancy objects with surface characteristics, such as flood storage and detention areas, lakes, watershed units, forecasting units, engineering impact areas, and important protection areas.

[0031] Furthermore, in step S3, the spatial relationships between water conservancy objects include: the topological connection relationship and the relationship between main streams and tributaries of the river system in the basin; the adjacency relationship between water conservancy point objects such as monitoring stations, reservoirs and control sections and river system line objects; and the upstream and downstream relationships between water conservancy point objects.

[0032] like Figure 5 As shown, preferably, in step S4, constructing the attribute mapping relationship of the water conservancy object includes: The river line reference model, which incorporates a reference system and measurement methods, is divided into networks using dynamic segmentation techniques. Specifically, this includes: An event table is constructed to store event information, including point events and line events. Point events and line events are associated with various constraints, scheduling procedures, and scheduling model services according to the actual needs of water project scheduling. The attribute field types and number of fields of point events and line events are different. The attribute values ​​of events are not directly stored in the point event and line event tables, but are stored in a separate attribute table and associated by event ID, event type, and attribute ID. A dynamic segmentation model is constructed to establish the transformation between the spatial location of water project scheduling events and the linear reference system, enabling information retrieval related to the spatial location of water project scheduling events, i.e., the transformation between one-dimensional and two-dimensional coordinates; the construction method is as follows: Based on the linear water system network reference model, a linear interpolation method based on adjacent reference points is used to obtain the coordinates of the interpolation points; the formula for interpolating the coordinates of the point event location points is as follows: ; ; ; In the formula, Pa is the location of the point event to be calculated, and P1 and P2 are the endpoints of the river segments closest to the point event Pa in the linear reference network model. The distance from Pa to the endpoints of river segments P1 and P2 is given by the linear reference method. The location coordinates of the point event can be interpolated using the above formula, which is used for querying and displaying point event information based on its spatial location. The coordinates of the start and end points of the line event are calculated using the point event location interpolation method.

[0033] The above steps established spatial and attribute relationship data for water conservancy objects. Based on this data, a knowledge graph of water conservancy object relationships was constructed. The Resource Description Framework (RDF) triplet data model was used to construct and store the knowledge graph. Taking the spatial topology of water conservancy objects as an example, the construction of the water conservancy object knowledge graph is illustrated below. Figure 6 As shown.

[0034] Preferably, in step S6, constructing the knowledge graph application includes: Utilizing knowledge graphs to configure water project scheduling business processes: To address the diverse needs of joint scheduling of water conservancy projects in a river basin, a knowledge graph is used to organize the water conservancy objects participating in the scheduling, as well as the business calculation processes on the water conservancy objects. The models, boundary conditions, and model inputs required for business calculations are dynamically mapped to the water conservancy objects as event attributes. Using knowledge graphs to visualize and query water project scheduling systems: Based on knowledge graphs, the topological relationships of water conservancy objects are constructed, and the topological relationships and business connections between water conservancy objects participating in the joint scheduling of water projects are displayed in a visual way. According to the dynamic segmentation coordinate interpolation technology, the model calculation results are accurately located to the digital river system network, realizing the accurate mapping and visualization of model boundary and scheduling result attribute data to two-dimensional and three-dimensional dimensions.

[0035] Example 2: The implementation steps of this invention are as follows: Step 1: Construct a spatial database for joint scheduling of watershed water projects. Design a spatial database for the joint scheduling of watershed water projects, and complete the data collection, organization, updating and maintenance work, mainly including basic geospatial data, remote sensing image data, water project spatial data, economic and social data, etc.

[0036] Step 2: Extraction of spatial relationships of water conservancy objects: 1) Extraction of the relationship between main streams and tributaries in river systems: Using GIS hydrological analysis methods, the river network of the watershed is extracted from the watershed topographic data (DEM data), and the relationship between main streams and tributaries is automatically generated. This data is then overlaid with the water system data in the joint water project scheduling spatial database to extract the spatial relationship between main streams and tributaries (river inflows). The process for extracting the relationship between main streams and tributaries (river inflows) based on hydrological analysis methods is described below. Figure 2 As shown.

[0037] Extracting the main stream-tributary (inflow) relationships from a river system network involves two steps: river classification and main stream-tributary relationship extraction. River classification: This involves assigning a classification level to each river in a river system network, represented by a numerical value. The further downstream, the higher the river's classification level and the larger the numerical value. See the diagram for a river classification system. Figure 3 As shown. Each river line has two endpoints. If two rivers are connected, they share a single endpoint, called the river confluence point. The specific steps for processing all river lines in the river network are as follows: Step 1: River lines with one endpoint that is not shared with any other river lines are called river lines without tributary connections. The classification number of these river lines is set to 1. The classification number of the currently processed river line is denoted as G, initially G=1. Step 2: Process all river lines with a grade number of 1, find river lines with a common endpoint and a grade number of 1, set the grade number of the river lines connected to them to 2, and mark the processed river lines until all river lines with a grade number of 1 have been processed, and update the value of G to 2.

[0038] Step 3: Process all river lines with a classification level of G. Find the river lines that are connected to the river lines with a classification level of G and have not been processed (river lines with no set classification level or a classification level of 1). Set or modify their classification levels. The value of the classification level is the sum of the classification levels of all the river lines with set classification levels that are connected to this river line. Mark the processed river lines. At the same time, update the value of G. Set the value of G to the minimum value among the classification levels newly assigned in this round.

[0039] Step 4: Check if there are any unprocessed river lines. If all river lines have been processed, end. Otherwise, repeat Step 3. Determination of the main - tributary relationship of rivers: Based on the river classification levels, determine the main - tributary (inflow) relationship of the rivers. The specific method is as follows: Traverse each river line, obtain the classification level g_c of the current river line R_c, then query all the river lines connected to this river line and obtain their classification levels respectively, and take the maximum value g_max. If g_max > g_c, the river line R_max corresponding to the classification level g_max is the downstream section, and R_c is the upstream section, and the river (section) R_c flows into R_max; if g_max < g_c, the river line R_c is the most downstream section. 2) Determination of the adjacency relationship between point - like objects and the river system. Project the water conservancy point object P onto the surrounding river line objects, and select the river line L_min with the shortest projection distance, then it is determined that the point object P is adjacent to the river line L_min. 3) Determination of the upstream - downstream relationship between point - like water conservancy objects. The upstream - downstream relationship between point objects can be judged according to the upstream - downstream relationship of the river lines they are adjacent to. The upstream - downstream relationship of point objects is the same as that of the river lines they are adjacent to. If two point objects are on the same river line, then it can be judged according to the elevation information of the point objects. The one with a larger elevation value is upstream, and the one with a smaller elevation value is downstream.

[0040] Step Three: Construct the attribute association relationship of water conservancy objects: The water system network of a basin is relatively fixed, but the business applications of basin water project scheduling are changing. It is necessary to select scheduling objects and set corresponding scheduling constraint conditions according to different application scenarios and scheduling objectives. Therefore, relative to the fixed water system network, the scheduling attribute mapping is flexibly changing. To solve this problem, a linear reference system is proposed based on the water system network, and dynamic segmentation technology is used to dynamically map the attributes related to water project scheduling. The specific operations are as follows: 1) Construct a linear reference system: When processing linear elements based on a linear reference system, the position information of unknown linear elements can be represented by the position information of known linear elements and their relative position relationships. The specific construction of the linear reference system is as follows: Water conservancy spatial data processing and quality inspection: For isal rivers or bilinear rivers, extract the river centerline and use the centerline to represent the river. Quality inspection of water conservancy spatial data mainly includes: checking for breaks and self-intersections between lines; checking for hanging points and hanging lines; checking for topological connection errors; and checking for duplicate feature objects. Generating a River System Network Reference Model: First, based on natural river system data, a linear reference network of rivers is generated. Each natural river line is assigned a unique number as a river ID. A starting point for each river line is set. Using GIS spatial data analysis tools, a measurement value M is generated for each point on the river line in the reference system. This M value represents the length along the river line to the starting point. The starting point can be either the upstream or downstream endpoint of the river; this patent uses the downstream endpoint as the starting point of the river line in the reference system. The linear reference river system network is saved in the spatial database in the form of feature classes. The path data attribute table is shown in Table 1.

[0041] Table 1: Riverline Attributes in the Linear Reference Model

[0042] Secondly, based on the linear reference network of the river system, river segments are divided, and the river lines in the reference system are divided. Considering the meandering and tortuous features of the river system and the natural morphological features of the river such as the river gradient, a segmentation strategy is adopted when constructing the linear reference water system network model. That is, starting from the uppermost part of a river, segments are gradually divided downwards, and the segmented river segments are kept as straight as possible. The river segment attribute table is shown in Table 2.

[0043] Table 2: River segment attributes in the linear reference model

[0044] 2) Dynamic segmentation technology enables dynamic mapping of scheduling attributes: Dynamic segmentation technology divides the network into parts based on a riverline reference model with a reference system and measurement methods, rather than performing physical segmentation. See the dynamic segmentation data model for details. Figure 5 As shown.

[0045] Construct an event table, which is used to store event information, including point events and line events. The attribute tables for point events and line events are shown in Tables 3 and 4, respectively.

[0046] Table 3: Point Event Attribute Table

[0047] Table 4: Line Event Attribute Table

[0048] Point events and line events can be associated with various constraints, scheduling procedures, scheduling model services, etc., according to the actual needs of water project scheduling. The attribute field types and number of fields are different. In order to facilitate flexible processing and expansion, the attribute values ​​of the events are not directly stored in the point event and line event tables, but are stored in a separate attribute table, which is associated by event ID, event type and attribute ID.

[0049] The dynamic segmentation model is constructed to establish the transformation between the spatial location of water project scheduling events and the linear reference system, and to realize the information query of water project scheduling events and spatial location, that is, the transformation between one-dimensional and two-dimensional coordinates.

[0050] Based on the linear reference network model, the coordinates of the interpolation points are obtained by using a linear interpolation method based on adjacent reference points.

[0051] Step 4: Establish a knowledge graph of the relationships between water conservancy objects in water project scheduling: The above steps established spatial and attribute relationship data for water conservancy objects. Based on this data, a knowledge graph of water conservancy object relationships was constructed. The Resource Description Framework (RDF) triple data model was used to construct and store the knowledge graph. Taking the spatial topological relationships of water conservancy objects as an example, the construction of the water conservancy object knowledge graph is illustrated below. Figure 6 As shown.

[0052] Step 5: Application of Knowledge Graphs 1) Utilizing knowledge graphs to configure water project scheduling business processes. Addressing the diverse needs of joint scheduling of water projects in a river basin, knowledge graphs are used to organize the water conservancy objects participating in the scheduling, as well as the various business calculation processes on these objects. The models, boundary conditions, and model inputs required for various business calculations are dynamically mapped to the water conservancy objects as event attributes, effectively solving the need for rapid construction and flexible adjustment of business river basins. 2) Utilize knowledge graphs to achieve visualization and information retrieval of water project scheduling systems. Based on knowledge graphs, construct the topological relationships of water conservancy objects, and visualize the topological associations and business connections between water conservancy objects participating in the joint scheduling of water projects. According to the dynamic segmentation coordinate interpolation technology, the model calculation results can be accurately located to the digital river system network, realizing the accurate mapping and visualization of attribute data such as model boundaries and scheduling results to two-dimensional and three-dimensional dimensions.

[0053] The above embodiments are merely preferred technical solutions of the present invention and should not be considered as limitations on the present invention. The scope of protection of the present invention should be limited to the technical solutions described in the claims, including equivalent substitutions of the technical features described in the claims. That is, equivalent substitutions and improvements within this scope are also within the scope of protection of the present invention.

Claims

1. A method for constructing a correlation relationship and a knowledge graph of a water conservancy object of a joint dispatch of a watershed control water conservancy project, characterized in that, include: S1, Construct a spatial database for joint scheduling of watershed water projects: By utilizing watershed water conservancy data resources, spatial data is extracted through data aggregation, extraction, and processing to construct a spatial database for joint scheduling of watershed water projects. The database data includes basic geospatial data, remote sensing image data, water conservancy project spatial data, and economic and social data. S2, Extraction of spatial associations of water conservancy objects: GIS analysis methods are used to extract spatial relationships between water conservancy objects; GIS hydrological analysis methods are used to determine the relationship between the main stream and tributaries of river lines by classifying river lines, and adjacency analysis methods are used to determine the adjacency relationship between water conservancy point objects and river line objects; S3, Constructing spatial relationships among water conservancy objects, specifically including: Water conservancy spatial data processing and quality inspection: For isal rivers or bilinear rivers, extract the river centerline and use the centerline to represent the river. Check the quality of the water conservancy spatial data, including: checking for breaks and self-intersections between lines; checking for hanging points and hanging lines; checking for topological connection errors; and checking for duplicate feature objects. Generate a reference model for a linear water system network: Based on natural river system data, a linear reference network of river systems is generated. Each natural river line is assigned a unique number as a river ID, and the starting point of the river line is set. Using GIS spatial data analysis tools, a measurement value M is generated for each point on the river line in the reference system. The value of M is the length along the river line to the starting point of the river line, which is the upstream or downstream endpoint of the river. The linear reference network of river systems is saved in the spatial database in the form of feature classes. Based on the linear reference network of the river system, river segments are divided, and the river lines in the reference system are divided, taking into account the meandering and tortuous nature of the river system and the natural morphological characteristics of the river gradient. When constructing the linear water system network reference model, a segmentation strategy is adopted to keep the segmented river segments straight, starting from the uppermost part of a river and gradually dividing it downwards. S4, Construct the attribute mapping relationship for water conservancy objects: A linear reference system is constructed based on the river system network of the basin. Dynamic segmentation technology is used to realize the mapping between engineering scheduling-related attributes and river system water conservancy objects, including engineering design data mapping, monitoring data mapping, model service mapping and computational boundary condition mapping. S5, construct a knowledge graph of the relationship between water conservancy objects in the joint scheduling of water conservancy projects; A knowledge graph of water conservancy object associations is constructed based on the generated spatial associations and attribute mappings of water conservancy objects. The knowledge graph is constructed and stored using the RDF triple data model of the resource description framework. S6, building knowledge graph applications.

2. The method according to claim 1, wherein the method is characterized in that, In step S1, constructing the spatial database for joint scheduling of watershed water projects includes: Utilize the water conservancy data resources of the basin data resource catalog, data resource center and big data center to extract spatial data required for the joint scheduling of water projects; For a single controlling water conservancy project, determine and extract its watershed range. Dispatch Scope Scope of Dispatch Impact The three polygons describing the above ranges are merged into a large polygon S, forming the total distribution range of all water conservancy objects related to the water project scheduling. Assuming the number of water projects participating in the joint scheduling of water conservancy projects in the basin is n, what is the distribution range of all water conservancy objects related to the m-th (m ≤ n) water project? and the overall distribution range of joint scheduling of all water projects Expressed as follows: ； ； wherein, , , respectively represent the basin range, the dispatching range and the influence range of the mth control water conservancy project. By merging, the total distribution range of all water conservancy objects is obtained After that, all water conservancy object spatial data required by joint regulation of water projects is extracted from existing data resources by using spatial query technology of GIS. A spatial data model is established using the idea of ​​data classification and hierarchical organization. Water conservancy objects are classified into point, line, and surface objects according to their geometric types. They are also classified and organized into layers according to their attribute types, including hydrological control station layers, river channel measurement control section layers, river layers, reservoir layers, and hydropower station layers.

3. The method according to claim 1, wherein the method is characterized by, In step S2, the extraction of spatial associations of water conservancy objects includes: S201, Extraction of the relationship between main streams and tributaries of river systems, including: Using GIS hydrological analysis methods, watershed topographic DEM data was obtained, and depression filling preprocessing was performed on the DEM data. Perform traffic analysis on the extracted DEM data; Based on the flow analysis results, the river network of the watershed is extracted from the extracted DEM data; The Shreve river classification method was used to classify the extracted watershed river network. Vectorize the main streams and tributaries after river classification to achieve river network vectorization; By overlaying the water system data in the vectorized river network and water project joint scheduling spatial database, the spatial relationship between the main stream and tributaries of the river system is extracted. S202, The adjacency relationship between point objects and river systems is determined; S203, Determination of upstream and downstream relationships of point-like water conservancy objects.

4. The method according to claim 3, wherein the method is characterized in that, In step S201, the river classification of the extracted watershed river network using the Shreve river classification method includes: Assigning a class to each river in the river system network, represented by a corresponding numerical value, with higher classes and larger values ​​further downstream. This process involves processing all river lines in the river system network, specifically including: S2011, a river line with one endpoint that is not shared with any other river line is called a river line without tributary connection, and the number of grades of the river line without tributary connection is set to 1; the number of grades of the river line currently being processed is denoted as G, and G=1 initially; S2012, process all river lines with a grade number of 1, find river lines with a common endpoint and a grade number of 1, set the grade number of the river lines connected to them to 2, and mark the processed river lines until all river lines with a grade number of 1 have been processed, and update the value of G to 2. S2013, process all river lines with a grade number of G, find the river lines connected to the river line with a grade number of G that have not been processed, set or modify its grade number, the value of the grade number is the sum of the grade numbers of all river lines connected to the river line with a set grade number, mark the processed river lines, and update the value of G at the same time, setting the value of G to the minimum value of the newly assigned grade number in this round. Among them, unprocessed river lines include river lines without a set number of grades or with a grade of 1; S2014, check if there are any unprocessed river lines. If all river lines have been processed, end the process; otherwise, repeat step S2013.

5. The method according to claim 4, wherein the method is characterized in that, In step S201, when extracting the spatial relationship between the main stream and tributaries of a river system, the relationship between the main stream and tributaries is determined based on the river classification number: Traverse each river line, obtain the grading number \(g_c\) of the current river line \(R_c\), then query all the river lines connected to this river line, and respectively obtain their grading numbers, and take the maximum value \(g_{max}\). If \(g_{max}>g_c\), the river line \(R_{max}\) corresponding to the grading number \(g_{max}\) is the downstream river section, and \(R_c\) is the upstream river section, and the river line \(R_c\) flows into \(R_{max}\); if \(g_{max}<g_c\), the river line \(R_c\) is the most downstream river section.

6. The method according to claim 5, wherein the method is characterized in that, In step S202, the determination of the adjacency relationship between the point object and the river water system includes: Project the water conservancy point object \(P\) onto the surrounding river line objects, and select the river line \(L_{min}\) with the shortest projection distance, then it is determined that the point object \(P\) is adjacent to the river line \(L_{min}\).

7. The method according to claim 5, wherein the method is characterized by, In step S203, the determination of the upstream and downstream relationship of the point-shaped water conservancy object includes: The upstream and downstream relationship between point objects is judged according to the upstream and downstream relationship of the river lines they are adjacent to. The upstream and downstream relationship of point objects is the same as that of the river lines they are adjacent to; if two point objects are on the same river line, then it can be judged according to the elevation information of the point objects. The one with a larger elevation value is upstream, and the one with a smaller elevation value is downstream.

8. The method according to claim 1, wherein the method is characterized by, In step S4, constructing the attribute mapping relationship of water conservancy objects includes: Perform network division on the river line reference model with a reference system and a measurement method through dynamic segmentation technology, specifically including: Construct an event table to store event information, including point events and line events; point events and line events are associated with various constraint conditions, scheduling regulations, and scheduling model services according to the actual needs of water project scheduling; the attribute field types and field quantities of point events and line events are different. The attribute values of events are not directly stored in the point event and line event tables, but are stored in another attribute table, and are associated by event ID, event type, and attribute ID; Construct a dynamic segmentation model to establish the conversion between the spatial position of water project scheduling events and the linear reference system, and realize the query of information related to the spatial position of water project scheduling events, that is, the conversion between one-dimensional and two-dimensional coordinates; the construction method is: Based on the linear water system network reference model, use the linear interpolation method based on adjacent reference points to obtain the coordinates of the interpolation points; the coordinate interpolation calculation formula for the position points of point events is as follows: ； ； ； In the formula, \(P_a\) is the position point of the point event to be calculated, \(P_1\) and \(P_2\) are the endpoints of the river section closest to the point event \(P_a\) in the linear reference network model respectively, and the distance from \(P_a\) to the endpoints of the river section \(P_1P_2\) is given by the linear reference method; through the above formula, the position coordinates of the point event can be interpolated for information query and display of the point event based on its spatial position; The coordinates of the start and end points of the line event are calculated respectively according to the interpolation method of the position points of the point event.

9. The method according to claim 1, wherein the method is characterized by, In step S6, constructing the knowledge graph application includes: Use the knowledge graph to realize the configuration of the water project scheduling business process: In view of the practical problem of the diversity of the joint scheduling requirements of basin water projects, use the knowledge graph to organize the water conservancy objects participating in the scheduling, as well as the business calculation processes on the water conservancy objects, and dynamically map the models, boundary conditions required for business calculation, and model inputs as event attributes to the water conservancy objects; Using knowledge graphs to visualize and query water project scheduling systems: Based on knowledge graphs, the topological relationships of water conservancy objects are constructed, and the topological relationships and business connections between water conservancy objects participating in the joint scheduling of water projects are displayed in a visual way. According to the dynamic segmentation coordinate interpolation technology, the model calculation results are accurately located to the digital river system network, realizing the accurate mapping and visualization of model boundary and scheduling result attribute data to two-dimensional and three-dimensional dimensions.

Images