A knowledge graph-based flood control scheduling rule intelligent recommendation method and device

Abstract

This invention discloses a method and apparatus for intelligent recommendation of flood control scheduling rules based on knowledge graphs. It extracts basic water conservancy object entities and attribute data, establishes relationships between water conservancy objects, and constructs a water conservancy object graph for the target basin. It designs a flood control scheduling rule ontology, integrates relevant flood control scheduling data by referencing the target basin scheduling report, and constructs a flood control scheduling graph for the target basin. Based on the business requirements for intelligent recommendation of flood control scheduling rules for the target basin, it instantiates a rule recommendation event graph using the Flowable workflow engine. It constructs a scheduling rule inference engine, driven by the rule recommendation event graph, and recommends flood control scheduling rules according to different water and rainfall conditions in the target basin. This invention solves the problems of multi-source heterogeneous data, complex storage departments, and diverse data formats in flood control scheduling auxiliary decision-making, making it difficult to query and access. It can effectively organize and represent flood control scheduling rules and comprehensively grasp flood control scheduling rules and related water conservancy object data.

Description

Technical Field

[0001] This invention relates to the fields of knowledge graph technology and flood control scheduling, specifically to a method and apparatus for intelligent recommendation of flood control scheduling rules based on knowledge graphs. Background Technology

[0002] Flood control scheduling is a measure to effectively manage floods using water conservancy projects such as reservoirs and dikes, aiming to mitigate the losses caused by floods to people's lives and property. It is a crucial component of the "four early warnings" (flood prevention, flood control, and flood preparedness). In conducting flood control scheduling operations, flood control departments coordinate the work of multiple water conservancy projects, requiring the acquisition and processing of large amounts of hydrological data, water engineering data, scheduling rules, expert experience, and other data and knowledge to rationally manage water conservancy projects and achieve precise decision-making. However, the required data and knowledge come from heterogeneous sources, have diverse structures, and suffer from poor sharing. The lack of an effective organizational structure and unified management system significantly slows down the operation of flood control scheduling processes, making it difficult for flood control departments to respond quickly to flood disasters. Therefore, it is necessary to establish a data and knowledge management system related to flood control scheduling operations to improve the decision-making level and response capabilities of flood control scheduling.

[0003] Knowledge graphs are a technology for structurally representing knowledge, capable of integrating and analyzing large amounts of data and knowledge, thereby improving the efficiency and accuracy of information retrieval. In the field of water conservancy, the application of knowledge graphs has already reached a considerable scale, but research on knowledge graphs specifically for flood control scheduling is relatively limited. Establishing a flood control scheduling graph can structure knowledge such as scheduling rules, organically integrating the relationships and functions between water conservancy projects such as reservoirs and dikes. It has advantages such as fast retrieval speed and high degree of knowledge association, meeting the application needs of knowledge retrieval, knowledge matching, knowledge reasoning, and rule recommendation, and improving the flood control department's ability to respond to and make decisions regarding floods.

[0004] Existing technical solutions typically use relational databases to store scheduling rules. However, the data stored in relational databases lacks semantic relationships, making it difficult to query scheduling rules and quickly and accurately push related knowledge and information under complex water and rainfall conditions. Furthermore, the lack of semantic relationships can also make the maintenance and updating of scheduling rules difficult. Therefore, existing flood control scheduling systems cannot achieve automated recommendation of flood control scheduling rules and other flood control scheduling knowledge, and cannot support the integrated forecasting and scheduling requirements of the business system. Summary of the Invention

[0005] Purpose of the invention: The purpose of this invention is to realize a method and device for intelligent recommendation of flood control scheduling rules based on knowledge graphs, so as to comprehensively grasp the flood control scheduling rules and related data of water conservancy objects, and assist in improving the decision-making ability of flood control scheduling.

[0006] Technical Solution: This invention provides an intelligent recommendation method for flood control scheduling rules based on knowledge graphs, comprising the following steps:

[0007] (1) Extract basic water conservancy object entities and water conservancy object attribute data from the target watershed data base, establish the relationship between water conservancy objects, and construct a water conservancy object map of the target watershed;

[0008] (2) Design the flood control dispatch rules ontology, refer to the target basin dispatch report, integrate flood control dispatch-related data, and construct the target basin flood control dispatch map;

[0009] (3) Intelligently recommend business requirements based on the flood control scheduling rules of the target watershed, and use the Flowable workflow engine to instantiate the rule recommendation event graph;

[0010] (4) Construct a scheduling rule inference engine. Based on the flood control scheduling map of the target watershed, the rule recommendation reasoning map drives the flood control scheduling rules to recommend flood control scheduling rules according to different water and rainfall conditions in the target watershed.

[0011] Furthermore, the flood control scheduling rule ontology described in step (2) adopts a top-down domain ontology design method to realize the structured representation of the scheduling rules.

[0012] Further, the flood control scheduling rule ontology described in step (2) is denoted as {key concepts, attributes, relationships}, where:

[0013] Key Concepts: Key concepts related to flood control scheduling knowledge include water conservancy objects, flood control initiation timing, scheme triggering timing, scheduling scheme, initiation object, triggering object, scheduling object, flood protection object, constraint object, control method, attributes and their value ranges; among them, water conservancy objects are specific things, facilities or systems existing within a watershed; scheduling schemes are the smallest unit of scheduling rules; flood control initiation timing is the condition for determining whether a water project should begin scheduling; scheme triggering timing is the condition for matching specific scheduling rules; initiation object is the water conservancy object related to the flood control initiation timing; triggering object is the water conservancy object related to the scheme triggering timing; scheduling object is the water project participating in the scheduling; control method is the specific action measures that the scheduling object should take; constraint object provides water level, flow rate, and other constraints for the action measures of the scheduling object; flood protection object is the area or water conservancy object protected by the scheduling scheme; attributes and their value ranges correspond to attributes such as water level and flow rate and their value ranges.

[0014] Attributes: Key concepts possess certain attribute characteristics, including attribute name, maximum value, minimum value, scheme type, and control method order;

[0015] Relationships: The relationships between key concepts are denoted as <Key Concept 1, Relationship, Concept 2>; where <Flood Control Response Timing, Relationship, Response Object>, <Response Object, is, Water Conservancy Object>, <Response Object, Response Conditions, Attributes and Value Range> indicate that the response object is a water conservancy object within the basin, and the water project starts scheduling when a certain attribute of the water conservancy object meets the value range constraint; <Scheme Trigger Timing, Relationship, Trigger Object>, <Trigger Object, is, Water Conservancy Object>, <Trigger Object, Response Conditions, Attributes and Value Range> indicate that the trigger object is a water conservancy object within the basin, and a specific scheduling rule is matched when a certain attribute of the water conservancy object meets the value range constraint; <Schedule Scheme, Scheduling, Scheduling Object>, < The following are definitions of the scheduling objects: "<Scheduling object, is a water conservancy object>", "<Scheduling object, adopts, control method>", "<Control method, constraints, attributes and value range>", indicating that the scheduling object is a water project within the basin, and the water project is scheduled using control methods such as pre-release, controlled release, and controlled storage. The control method requires that a certain attribute of the water project meets the constraint of the value range. "<Scheduling scheme, restriction, restriction object>", "<Restriction object, is a water conservancy object>", "<Restriction object, constraints, attributes and value range>", indicating that the restriction object is a water conservancy object within the basin, and the control method adopted by the scheduling object cannot cause a certain attribute of the restriction object to exceed the constraint of the value range. "<Scheduling scheme, protection, flood control protection object>", indicating that the purpose of the scheduling scheme is to protect a flood control protection object within the basin.

[0016] Furthermore, the process of constructing the target basin flood control scheduling map in step (2) is as follows:

[0017] Determine the naming convention for entity nodes in the flood control scheduling map of the target watershed: Based on the design of the flood control scheduling rule ontology, unify the naming convention for various entity nodes that have a mapping relationship with the ontology; the naming of instance nodes consists of a fixed naming part and a numbering part, which are combined with the name of the water conservancy object in the target watershed;

[0018] Collect scheduling data related to the target watershed and perform data preprocessing to remove invalid and redundant data;

[0019] Extract flood control scheduling rules and map them to the ontology: Following entity node naming conventions, extract entities and entity relationships of flood control scheduling rules from the collected data to represent them in a structured manner; for structured data, use the D2RQ automatic extraction technology to convert scheduling rule data in databases or tables into RDF files; for unstructured data, automatically label flood control scheduling rules and expert experience with high degree of terminology standardization contained in unstructured text, and export them as RDF files; import the RDF files into the graph database.

[0020] The knowledge fusion of water conservancy objects matches the same water conservancy object entity in the flood control scheduling map and the water conservancy object map, and updates the attributes and relationships of the entity in the water conservancy object map to the flood control scheduling map; based on spatial topology and other relationships, for water conservancy objects involved in the scheduling rules in the flood control scheduling map, the attributes and relationships in the water conservancy object map are integrated into the flood control scheduling map.

[0021] Furthermore, the instantiation rule recommendation event graph implementation process in step (3) is as follows:

[0022] S1: Constructing Situational Awareness Events: This event involves acquiring real-time and forecast data on water levels and flow rates of water conservancy objects such as cross-sections and reservoirs within the basin, in order to perceive the situation in the basin.

[0023] S2: Constructing the event for determining flood protection objects: This event is based on the results of watershed perception, which automatically generates a knowledge graph query statement to compare the water level, flow rate and characteristic water level and characteristic flow rate of water conservancy objects in the watershed, and to determine whether there are any water conservancy objects in the current watershed that are in a dangerous state, so as to determine the flood protection objects in the watershed that need to be protected by water projects.

[0024] S3: Construct a single-database or joint scheduling judgment event: This event automatically generates a knowledge graph query statement based on the determined flood control scheduling objectives. It is used to identify water projects that have hydraulic connections with the flood control protection objects and can affect them, and to determine whether the current scheduling scenario is single-database scheduling or joint scheduling. Single-database scheduling is a scheduling scenario in which water projects protect their own safety through scheduling, while joint scheduling is a scheduling scenario in which multiple water projects jointly participate in scheduling to protect the safety of the flood control protection objects. The multi-database scheduling also needs to consider the single-database scheduling of each water project.

[0025] S4: Construct Water Project Start-up Analysis Event: This event automatically generates a knowledge graph query statement, compares the water level of the water project with the flood season limit water level, and determines whether the water project participating in the scheduling meets the conditions for starting scheduling. When the water project does not meet the start-up conditions, it should prioritize water storage.

[0026] S5: Construction of Scheduling Scheme Trigger Event: This event automatically generates a knowledge graph query statement to query the flood control scheduling rules under the current water and rainfall conditions; in a single reservoir scheduling scenario, it triggers the scheduling rules for protecting the individual water project itself; in a multi-reservoir scheduling scenario, it triggers both the scheduling rules for protecting the water project itself and the scheduling rules for protecting the flood control of the water project.

[0027] S6: Construct a scheduling suggestion recommendation event: This event organizes flood control scheduling knowledge such as watershed status, relevant water conservancy object information, and scheduling rules into natural language and recommends it to the scheduler to assist the scheduler in decision-making;

[0028] S7: Use flowable to organize the relationships between the above events to form a rule recommendation event graph that can be executed automatically.

[0029] Furthermore, the implementation process of step (4) is as follows:

[0030] Using watershed data as input to the rule recommendation logic graph, the flood control scheduling knowledge in the RDF triple format in the flood control scheduling graph is queried. Using rule-based natural language generation technology, the RDF triple set is converted into scheduling suggestions in natural language form as output to assist the scheduler in decision-making and realize the positive reasoning of flood control scheduling rules.

[0031] Based on the dispatcher's needs, knowledge graph CRUD statements are automatically generated for CRUD requests related to dispatch rules, in order to manage dispatch rules and realize reverse reasoning of flood control dispatch rules.

[0032] The present invention also provides an apparatus comprising a memory and a processor, wherein:

[0033] Memory is used to store computer programs that can run on a processor;

[0034] The processor is configured to execute the steps of the above-described intelligent recommendation method for flood control scheduling rules based on knowledge graphs when running the computer program.

[0035] The present invention also provides a storage medium storing a computer program, which, when executed by at least one processor, implements the steps of the above-described intelligent recommendation method for flood control scheduling rules based on knowledge graphs.

[0036] Beneficial effects: Compared with the prior art, the beneficial effects of the present invention are: The present invention has universality for scheduling operations in different river basins and different flood scenarios. By utilizing the semantic expression and association capabilities of knowledge graphs, it solves the problems of multi-source heterogeneous data sources, mixed storage departments, multiple data formats, and inconvenience in querying and accessing flood control scheduling auxiliary decision-making. It can effectively organize and represent flood control scheduling rules, comprehensively grasp flood control scheduling rules and related data of water conservancy objects, and help improve flood control scheduling decision-making capabilities. Attached Figure Description

[0037] Figure 1 This is a flowchart of the present invention;

[0038] Figure 2 This is a schematic diagram of the flood control dispatching rules designed in this embodiment of the invention;

[0039] Figure 3 This is a schematic diagram of the entity naming convention specified in the embodiments of the present invention;

[0040] Figure 4This is a schematic diagram illustrating the use of knowledge alignment technology in an embodiment of the present invention;

[0041] Figure 5 This is a schematic diagram of the principle atlas in an embodiment of the present invention;

[0042] Figure 6 This is a schematic diagram illustrating scheduling recommendations in an embodiment of the present invention;

[0043] Figure 7 This is a schematic diagram of scheduling rule management in an embodiment of the present invention. Detailed Implementation

[0044] The present invention will now be described in further detail with reference to the accompanying drawings.

[0045] This invention provides an intelligent recommendation method for flood control scheduling rules based on knowledge graphs, such as... Figure 1 As shown, it includes the following steps:

[0046] Step 1: Extract basic water conservancy object entities and water conservancy object attributes from the target watershed data base, establish the relationship between water conservancy objects, and construct a water conservancy object map of the target watershed.

[0047] Step 2: Design the flood control scheduling rule ontology, refer to the target basin scheduling report, integrate relevant flood control scheduling data, and construct the target basin flood control scheduling map.

[0048] The flood control scheduling rule ontology adopts a top-down domain ontology design method to realize the structured representation of scheduling rules.

[0049] Flood control dispatch rules ontology = {key concepts, attributes, relationships}, where:

[0050] Key Concepts: Key concepts related to flood control scheduling knowledge, including (hydraulic objects, flood control initiation timing, scheme triggering timing, scheduling scheme, initiation object, triggering object, scheduling object, flood protection object, constraint object, control method, attributes and their value ranges). Among these, a hydraulic object is a specific entity, facility, or system existing within a watershed; a scheduling scheme is the smallest unit of scheduling rules; the flood control initiation timing is the condition for determining whether a water project should begin scheduling; the scheme triggering timing is the condition for matching specific scheduling rules; the initiation object is the hydraulic object related to the flood control initiation timing; the triggering object is the hydraulic object related to the scheme triggering timing; the scheduling object is the water project participating in the scheduling; the control method is the specific action measures that the scheduling object should take; the constraint object provides constraints such as water level and flow rate for the actions of the scheduling object; the flood protection object is the area or hydraulic object protected by the scheduling scheme; and the attributes and their value ranges correspond to attributes such as water level and flow rate and their value ranges.

[0051] Attributes: Key concepts possess certain attribute characteristics, including attribute name, maximum value, minimum value, scheme type, and control method order.

[0052] Relationship: The connection between key concepts, denoted as <key concept 1, relationship, concept 2>, such as... Figure 2 As shown. Among them, <flood control initiation timing, association, initiation object>, <initiation object, is, water conservancy object>, <initiation object, initiation condition, attribute and value range> indicate that the initiation object is a water conservancy object within the basin. When a certain attribute of this water conservancy object meets the constraint of the value range, the water project begins scheduling; <scheme triggering timing, association, triggering object>, <triggering object, is, water conservancy object>, <triggering object, initiation condition, attribute and value range> indicate that the triggering object is a water conservancy object within the basin. When a certain attribute of this water conservancy object meets the constraint of the value range, a specific scheduling rule is matched; <scheme scheduling, scheduling, scheduling object>, <scheme object, is, water conservancy object>, <scheme... The `<Object, Control Method>` and `<Control Method, Constraints, Attributes and Value Range>` options indicate that the scheduling object is a water project within the basin, which is scheduled using control methods such as pre-release, controlled release, and controlled storage. The control method requires that a certain attribute of the water project meets the constraint of its value range. The `<Schedule, Restriction, Restricted Object>` and `<Restricted Object, is, Water Conservancy Object>` options, `<Restricted Object, Constraints, Attributes and Value Range>` options indicate that the restricted object is a water conservancy object within the basin, and the control method adopted by the scheduling object cannot cause a certain attribute of the restricted object to exceed the constraint of its value range. The `<Schedule, Protection, Flood Control Protection Object>` option indicates that the purpose of the scheduling scheme is to protect a flood control protection object within the basin.

[0053] The process of constructing the flood control scheduling map for the target watershed is as follows:

[0054] Determine the naming conventions for entity nodes in the flood control scheduling map of the target watershed: Based on the design of the flood control scheduling rule ontology, unify the naming conventions for various entity nodes that have a mapping relationship with the ontology; the naming of instance nodes consists of a fixed naming part and a numbering part, combined with the name of the water conservancy object in the target watershed, such as... Figure 3 As shown, the scheduling schemes for the Xia'an Reservoir can be named as Xia'an Reservoir Scheduling Scheme 1, Xia'an Reservoir Scheduling Scheme 2, and so on.

[0055] Collect scheduling data related to the target watershed and perform data preprocessing to remove invalid and redundant data; extract flood control scheduling rules and map them to the ontology: extract entities and entity relationships of flood control scheduling rules from the collected data according to entity node naming conventions to represent flood control scheduling rules in a structured manner; for structured data, use the automatic extraction technology D2RQ to convert scheduling rule data in databases or tables into RDF files; for unstructured data, automatically label flood control scheduling rules and expert experience with high degree of terminology standardization contained in unstructured text and export them as RDF files; import the RDF files into the graph database.

[0056] Knowledge fusion of water conservancy objects involves matching the same water conservancy object entity in the flood control scheduling map and the water conservancy object map, updating the attributes and relationships of the entity in the water conservancy object map to the flood control scheduling map; and using spatial topology and other relationships, for water conservancy objects involved in the scheduling rules in the flood control scheduling map, integrating the attributes and relationships in the water conservancy object map into the flood control scheduling map, such as... Figure 4 As shown, when the scheduling plan involves the Xia'an Reservoir, the flood control standard and other attributes of the Xia'an Reservoir in the water conservancy object map need to be aligned to the flood control scheduling map.

[0057] Step 3: Based on the flood control scheduling rules of the target watershed, intelligently recommend business requirements and use the Flowable workflow engine to instantiate the rule recommendation event graph.

[0058] S1: Construct a situational awareness event. This event involves acquiring real-time and forecast data on water levels and flow rates of water conservancy objects such as cross-sections and reservoirs within the watershed, automatically generating knowledge graph query statements to update the data in the water conservancy object graph.

[0059] S2: Construct the event for determining flood protection targets. This event is based on the results of watershed perception, which automatically generates a knowledge graph query statement to compare the water level, flow rate, and characteristic water level and flow rate of water conservancy objects within the watershed. It determines whether there are any water conservancy objects in the current watershed that are in a dangerous state, thereby identifying flood protection targets within the watershed that require the scheduling of water projects for protection.

[0060] S3: Construct Single-Reservoir / Joint Scheduling Judgment Events. This event automatically generates knowledge graph query statements based on the determined flood control scheduling objectives. It is used to identify water projects that have hydraulic connections with and can affect the flood control protected objects, such as reservoirs, sluices, and flood storage areas. It then determines whether the current scheduling scenario is single-reservoir scheduling or joint scheduling. Single-reservoir scheduling is a scenario where water projects protect their own safety through scheduling, while multi-reservoir scheduling is a scenario where multiple water projects jointly participate in scheduling to protect the safety of the flood control protected objects. Water projects need to ensure the safety of the flood control protected objects while ensuring their own safety; therefore, multi-reservoir scheduling also needs to consider the single-reservoir scheduling of each water project.

[0061] S4: Construct a water project initiation analysis event. This event automatically generates a knowledge graph query statement, compares the water level of the water project with the flood season limit water level, and determines whether the water project participating in the scheduling meets the conditions for initiating scheduling. When the water project does not meet the initiation conditions, it should be given priority to store water.

[0062] S5: Event triggering the construction of a scheduling scheme. This event automatically generates a knowledge graph query statement to retrieve the flood control scheduling rules under the current water and rainfall conditions. In a single-reservoir scheduling scenario, it triggers the scheduling rules for protecting the individual water project itself; in a multi-reservoir scheduling scenario, it triggers both the scheduling rules for protecting the individual water project and the scheduling rules for protecting flood control.

[0063] S6: Construct scheduling suggestion recommendation events. Organize flood control scheduling knowledge, including basin situation, relevant water conservancy object information, and scheduling rules, into natural language and recommend it to the dispatcher to assist in decision-making.

[0064] S7: Utilize flowable to organize the relationships between the above events, forming a rule recommendation event graph that can be automatically executed, such as... Figure 5 As shown.

[0065] Step 4: Construct a scheduling rule inference engine. Based on the flood control scheduling map of the target watershed, driven by the rule recommendation reasoning map, recommend flood control scheduling rules according to different water and rainfall conditions in the target watershed.

[0066] Driven by the rule recommendation logic graph, the process of recommending flood control scheduling rules based on different water and rainfall conditions in the target watershed is as follows:

[0067] Using watershed data as input to the rule recommendation logic graph and RDF triple-formatted flood control scheduling knowledge from the flood control scheduling graph as output, rule-based natural language generation technology is used to convert the RDF triple set into natural language scheduling suggestions to assist schedulers in decision-making and achieve positive reasoning for flood control scheduling rules. Figure 6 As shown, when the real-time water level of the downstream reservoir exceeds the warning value, the rule-based recommendation mechanism drives the query of the flood control scheduling mechanism graph, queries the downstream reservoir scheduling rules that meet the current water and rainfall conditions, and outputs natural language for the dispatcher's reference.

[0068] Based on the dispatcher's needs, the system automatically generates knowledge graph CRUD statements for CRUD requests related to dispatch rules, thereby managing dispatch rules and enabling reverse reasoning for flood control dispatch rules. Figure 7 As shown, you can query the scheduling rules of Xia'an Reservoir and perform operations such as adding, deleting, modifying and querying scheduling rules on the same interface.

[0069] The present invention also provides an apparatus comprising a memory and a processor, wherein: the memory is used to store a computer program capable of running on the processor; and the processor is used to execute the steps of the above-described intelligent recommendation method for flood control scheduling rules based on knowledge graphs when running the computer program.

[0070] The present invention also provides a storage medium storing a computer program, which, when executed by at least one processor, implements the steps of the above-described intelligent recommendation method for flood control scheduling rules based on knowledge graphs.

[0071] The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent transformations or modifications made in accordance with the spirit and essence of the present invention are covered within the scope of protection of the present invention.

Claims

1. A knowledge graph-based intelligent recommendation method for flood control scheduling rules, characterized in that, Includes the following steps: (1) Extract basic water conservancy object entities and water conservancy object attribute data from the target watershed data base, establish the relationship between water conservancy objects, and construct the target water conservancy object map; (2) Design the flood control dispatch rules ontology, refer to the target basin dispatch report, integrate relevant flood control dispatch data, and construct the target basin flood control dispatch map; (3) Based on the flood control scheduling rules of the target watershed, intelligently recommend business requirements and use the Flowable workflow engine to instantiate the rule recommendation event graph; (4) Construct a scheduling rule inference engine. Based on the flood control scheduling map of the target watershed, the rule recommendation reasoning map drives the flood control scheduling rules to recommend flood control scheduling rules according to different water and rainfall conditions in the target watershed. The instantiation rule recommendation event graph implementation process in step (3) is as follows: S1: Constructing Situational Awareness Events: This event involves acquiring real-time and forecast data on water levels and flow rates at cross-sections and reservoirs within the watershed to perceive the watershed situation. S2: Constructing the event for determining flood protection objects: This event is based on the results of watershed perception, which automatically generates a knowledge graph query statement to compare the water level, flow rate and characteristic water level and characteristic flow rate of water conservancy objects in the watershed, and to determine whether there are any water conservancy objects in the current watershed that are in a dangerous state, so as to determine the flood protection objects in the watershed that need to be protected by water projects. S3: Construct a single-database or joint scheduling judgment event: This event automatically generates a knowledge graph query statement based on the determined flood control scheduling target. It is used to identify water projects that have hydraulic connections with the flood control protection object and can affect the flood control protection object, and to determine whether the current scheduling scenario is single-database scheduling or joint scheduling. Single-database scheduling is a scheduling scenario in which water projects protect their own safety through scheduling, while joint scheduling is a scheduling scenario in which multiple water projects jointly participate in scheduling to protect the safety of the flood control protection object. S4: Construct Water Project Start-up Analysis Event: This event automatically generates a knowledge graph query statement, compares the water level of the water project with the flood season limit water level, and determines whether the water project participating in the scheduling meets the conditions for starting scheduling. When the water project does not meet the start-up conditions, it should prioritize water storage. S5: Triggering event for constructing a scheduling scheme: This event automatically generates a knowledge graph query statement to query the flood control scheduling rules under the current water and rainfall conditions; in a single-reservoir scheduling scenario, it triggers the scheduling rules for protecting the individual water project itself; in a multi-reservoir scheduling scenario, it triggers both the scheduling rules for protecting the water project itself and the scheduling rules for protecting the flood control objects of the water project. S6: Construct scheduling suggestion recommendation event: This event involves organizing flood control scheduling knowledge, including basin situation, relevant water conservancy object information, and scheduling rules, into natural language and recommending it to the scheduler to assist the scheduler in decision-making; S7: Use flowable to organize the relationships between the above events to form a rule recommendation event graph that can be executed automatically.

2. The intelligent recommendation method for flood control scheduling rules based on knowledge graphs according to claim 1, characterized in that, The flood control scheduling rule ontology described in step (2) adopts a top-down domain ontology design method to realize the structured representation of the scheduling rules.

3. The intelligent recommendation method for flood control scheduling rules based on knowledge graphs according to claim 1, characterized in that, The flood control scheduling rule ontology described in step (2) is denoted as {key concepts, attributes, relationships}, where: Key concepts: Key concepts related to flood control scheduling knowledge, including water conservancy objects, flood control initiation timing, scheme triggering timing, scheduling scheme, initiation object, triggering object, scheduling object, flood protection object, restriction object, control method, attributes and their value range; Attributes: Key concepts possess certain attribute characteristics, including attribute name, maximum value, minimum value, scheme type, and control method order; Relationships: The relationships between key concepts are denoted as <Key Concept 1, Relationship, Concept 2>; where <Flood Control Response Timing, Relationship, Response Object>, <Response Object, is, Water Conservancy Object>, <Response Object, Response Conditions, Attributes and Value Range> indicate that the response object is a water conservancy object within the basin, and the water project starts scheduling when a certain attribute of the water conservancy object meets the value range constraint; <Scheme Trigger Timing, Relationship, Trigger Object>, <Trigger Object, is, Water Conservancy Object>, <Trigger Object, Response Conditions, Attributes and Value Range> indicate that the trigger object is a water conservancy object within the basin, and a specific scheduling rule is matched when a certain attribute of the water conservancy object meets the value range constraint; <Schedule Scheme, Scheduling, Scheduling Object>, < The following are definitions of the scheduling objects: "<Scheduling object, is a water conservancy object>", "<Scheduling object, adopts, control method>", "<Control method, constraints, attributes and value range>", indicating that the scheduling object is a water project within the basin, and the water project is scheduled using pre-release, controlled release, or controlled storage control methods. The control method requires that a certain attribute of the water project meets the constraint of the value range. "<Scheduling scheme, restriction, restriction object>", "<Restriction object, is a water conservancy object>", "<Restriction object, constraints, attributes and value range>", indicating that the restriction object is a water conservancy object within the basin, and the control method adopted by the scheduling object cannot cause a certain attribute of the restriction object to exceed the constraint of the value range. "<Scheduling scheme, protection, flood control protection object>", indicating that the purpose of the scheduling scheme is to protect a flood control protection object within the basin.

4. The intelligent recommendation method for flood control scheduling rules based on knowledge graphs according to claim 1, characterized in that, The process of constructing the flood control scheduling map of the target watershed in step (2) is as follows: Determine the naming convention for entity nodes in the flood control scheduling map of the target watershed: Based on the design of the flood control scheduling rule ontology, unify the naming convention for various entity nodes that have a mapping relationship with the ontology; the naming of instance nodes consists of a fixed naming part and a numbering part, which are combined with the name of the water conservancy object in the target watershed; Collect scheduling data related to the target watershed and perform data preprocessing to remove invalid and redundant data; Extract flood control scheduling rules and map them to the ontology: Following entity node naming conventions, extract entities and entity relationships of flood control scheduling rules from the collected data to represent them in a structured manner; for structured data, use the D2RQ automatic extraction technology to convert scheduling rule data in databases or tables into RDF files; for unstructured data, automatically label flood control scheduling rules and expert experience with high degree of terminology standardization contained in unstructured text, and export them as RDF files; import the RDF files into the graph database. The knowledge fusion of water conservancy objects matches the same water conservancy object entity in the flood control scheduling map and the water conservancy object map, and updates the attributes and relationships of the entity in the water conservancy object map to the flood control scheduling map; Using spatial topology, for the water conservancy objects involved in the scheduling rules in the flood control scheduling map, the attributes and relationships in the water conservancy object map are integrated into the flood control scheduling map.

5. The intelligent recommendation method for flood control scheduling rules based on knowledge graphs according to claim 1, characterized in that, The implementation process of step (4) is as follows: Using watershed data as input to the rule recommendation logic graph, the flood control scheduling knowledge in the RDF triple format in the flood control scheduling graph is queried. Using rule-based natural language generation technology, the RDF triple set is converted into scheduling suggestions in natural language form as output to assist the scheduler in decision-making and realize the positive reasoning of flood control scheduling rules. Based on the dispatcher's needs, knowledge graph CRUD statements are automatically generated for CRUD requests related to dispatch rules, in order to manage dispatch rules and realize reverse reasoning of flood control dispatch rules.

6. The intelligent recommendation method for flood control scheduling rules based on knowledge graphs according to claim 3, characterized in that, The water conservancy object refers to a specific thing, facility, or system existing within the watershed; the scheduling scheme is the smallest unit of the scheduling rules; the flood control initiation timing is the condition for determining whether a water project should begin scheduling; the scheme triggering timing is the condition for matching specific scheduling rules; the initiation object is the water conservancy object related to the flood control initiation timing; the triggering object is the water conservancy object related to the scheme triggering timing; the scheduling object is the water project participating in the scheduling; the control method is the specific behavioral measures that the scheduling object should take; the restriction object provides water level and flow constraints for the behavioral measures of the scheduling object; the flood protection object is the area or water conservancy object protected by the scheduling scheme; the attributes and their value ranges correspond to the water level and flow attributes and their value ranges.

7. The intelligent recommendation method for flood control scheduling rules based on knowledge graphs according to claim 1, characterized in that, The multi-reservoir scheduling needs to take into account the individual reservoir scheduling of each water project.

8. A device, characterized in that, Includes memory and processor, wherein: Memory is used to store computer programs that can run on a processor; A processor, configured to, while running the computer program, perform the steps of the knowledge graph-based intelligent recommendation method for flood control scheduling rules as described in any one of claims 1 to 7.

9. A storage medium, characterized in that, The storage medium stores a computer program, which, when executed by at least one processor, implements the steps of the knowledge graph-based intelligent recommendation method for flood control scheduling rules as described in any one of claims 1 to 7.

Images