Search

CN-122021250-A - Knowledge decision supervision method for flood prevention in drainage basin

CN122021250ACN 122021250 ACN122021250 ACN 122021250ACN-122021250-A

Abstract

The invention discloses a knowledge decision supervision method for flood control of a river basin, which relates to the technical field of flood control of the river basin and comprises the steps of obtaining river basin flood control data and real-time meteorological data of a target area, carrying out knowledge extraction on the river basin flood control data to obtain space element information and non-space element information, encoding the space element information according to a preset identity encoding rule to obtain encoded space element information, constructing a space topological relation of the encoded space element information by utilizing a geographic information system, constructing a knowledge relation of a root primitive according to the constructed space topological relation to obtain the space element knowledge relation, fusing the non-space element information and the space element knowledge relation, converting a fused result into a knowledge map to obtain a fused knowledge map, inputting the meteorological data and the fused knowledge map into a preset knowledge auxiliary decision model, and outputting a decision result. The invention reduces and promotes the accuracy of decision, the response speed and the response efficiency of early warning.

Inventors

  • Qu yawei
  • HUANG XIAOYU
  • DING YINGXIA
  • Yao Dengguo

Assignees

  • 四川公用信息产业有限责任公司

Dates

Publication Date
20260512
Application Date
20251224

Claims (10)

  1. 1. The knowledge decision supervision method for flood prevention in the river basin is characterized by comprising the following steps of: acquiring flood prevention data and real-time meteorological data of a river basin of a target area; Knowledge extraction is carried out on flood prevention data of the river basin to obtain space element information and non-space element information; Coding the space element information according to a preset identity coding rule to obtain coded space element information; Constructing a spatial topological relation of the encoded spatial element information by using a geographic information system, and constructing a knowledge relation of a root primitive according to the constructed spatial topological relation to obtain a spatial element knowledge relation; Fusing the non-space element information and the space element knowledge relationship, and converting the fusion result into a knowledge graph to obtain a fusion knowledge graph; The method comprises the steps of inputting meteorological data and a fusion knowledge graph into a preset knowledge auxiliary decision model, and outputting a decision result, wherein the knowledge auxiliary decision model comprises a calculation engine, a simulation engine and an early warning engine, the calculation engine is used for sequentially carrying out river basin runoff calculation, tributary confluence calculation, main flow interval confluence calculation and main flow confluence calculation according to the knowledge graph, the meteorological data and the hydrologic model, the simulation engine is used for carrying out main flow flood evolutionary simulation according to the calculation result of the calculation engine, the knowledge graph and the hydrologic coupling model, and the early warning engine is used for sequentially carrying out target section early warning, early warning notification, risk studying and judging, risk previewing and auxiliary decision emergency plan according to the simulation result of the knowledge graph and the simulation engine.
  2. 2. The basin-oriented knowledge decision supervision method according to claim 1, wherein obtaining basin flood control data of the target area comprises: The method comprises the steps of obtaining river basin flood control data of a target area, wherein the river basin flood control data comprise river basin information, sub-river basin partition information, river water system information, monitoring equipment information, risk area information, placement point information, flood control highland information, evacuation route information, historical scene information, emergency plan information, risk area house distribution information, population structure information and emergency organization system information; And carrying out data space processing on the flood control data of the river basin based on the geographic information system, and storing different knowledge elements in different thematic layers according to the management principle of classification and layering.
  3. 3. The knowledge decision supervision method for flood control in a river basin according to claim 2, wherein the knowledge extraction is performed on the flood control data, comprising: Constructing a database standard and a data dictionary of each class of flood control data according to industry specifications and business rules, wherein the database standard agrees with the data structure, the layer element code and the root primitive type of each class of flood control data; Extracting and cleaning flood prevention data of each type of river basin by adopting an ETL tool to obtain space element information and non-space element information, wherein the space element information comprises river basin river nets and related space element information of risk areas, and the non-space element information comprises early warning rules, historical scene data of the risk areas, emergency plans, emergency organization system information, material information and threat crowd information; And packaging the extracted and washed flood control data of each class of drainage basin into a preset standardized database structure according to the database standard and the data dictionary of the flood control data of each class of drainage basin to obtain a spatial database.
  4. 4. A knowledge decision supervision method for flood prevention in a watershed according to claim 3, wherein the encoding of the space element information according to a preset identity encoding rule comprises: Constructing a space element type coding rule, wherein the space element type coding rule comprises the steps of carrying out element geometric type identification by adopting initial letters of point elements, line elements, surface elements and three-dimensional models, and automatically assigning initial letters of corresponding space elements in the coding generation process according to the geometric type identification automatically recorded by a space database; constructing a position coding rule, wherein the position coding rule comprises position coding based on a Beidou grid position code grid subdivision mode; Constructing element coding rules, wherein the element coding rules comprise element classification and coding based on basic geographic information; Constructing a sequence code coding rule, wherein the sequence code coding rule comprises the steps of sequentially coding different element objects of an element layer, assigning numbers to elements according to the sequence codes from west to east and from north to south, sequencing according to the space coordinates of the elements, and distributing sequence numbers according to sequencing results; And coding the root graphic elements of each type of space element according to a space element type coding rule, a position coding rule, an element coding rule and a sequence code coding rule, and uniformly assigning values to other graphic elements according to a space inclusion relationship after the root graphic elements are coded when a plurality of space forms exist for the same type of space element.
  5. 5. The knowledge decision supervision method for flood prevention in a river basin according to claim 1, wherein the constructing the spatial topological relation of the encoded spatial element information by using the geographic information system, and constructing the knowledge relation of the root primitive according to the constructed spatial topological relation, to obtain the spatial element knowledge relation, comprises: constructing a space topological relation between every two space elements based on a geographic information system, wherein the space topological relation comprises a containing relation, a flowing relation, an upstream-downstream relation and an adjacent relation; And constructing a knowledge relationship between every two elements according to the space topological relationship to obtain a space element knowledge relationship, wherein the space element knowledge relationship is expressed by adopting five-tuple, and the five-tuple comprises space element identity codes, relationship types, states and extinction time corresponding to the two space elements, and the state field is used for identifying the space elements and the validity of the relationship.
  6. 6. The knowledge decision supervision method for flood prevention in a river basin according to claim 5, wherein the steps of fusing non-space element information and space element knowledge relations, converting the fusion result into a knowledge graph, and obtaining a fused knowledge graph comprise: In the knowledge extraction process, a cleaning tool of the ETL is utilized to assign values to fields of non-space element information, and a connection relation with space elements is established on a field value range; binding non-space elements with corresponding space elements based on code matching, and realizing association of business elements and space elements to obtain a fusion result; Warehousing the five-tuple knowledge relationship of the fusion result based on a Neo4J graph database; and constructing a fusion knowledge graph based on a visual graph technology according to the quintuple knowledge relationship.
  7. 7. The knowledge decision supervision method for flood prevention in a river basin according to claim 5, wherein the river basin runoff production calculation, the tributary confluence calculation, the main flow interval confluence calculation and the main flow confluence calculation are sequentially performed according to a knowledge graph, meteorological data and a hydrological model, and the knowledge decision supervision method comprises the following steps: Extracting sub-drainage basin data according to the inclusion relation of 'drainage basin-sub-drainage basin' in the knowledge graph, and calculating the rainfall capacity of each sub-drainage basin through spatial superposition analysis of the rainfall surface data and the sub-drainage basin vector data on the basis of acquiring drainage basin rainfall surface data shared by a meteorological department, wherein each sub-drainage basin adopts different yield calculation parameters according to the underlying surface attribute to obtain the yield of each sub-drainage basin; Calculating branch flow converging, namely calculating slope branch flow converging and river channel branch flow converging, wherein calculating slope branch flow converging comprises calculating converging time and flow attenuation process from slope to branch according to characteristic parameters of a sub-river basin by adopting a distributed hydrological model, and storing calculated data in association with corresponding branches, calculating river channel branch flow converging comprises calculating propagation time, flow change and attenuation process from branch starting point to converging section by adopting the hydrological model according to branch converging relation of branch-converging section in the knowledge map as flow basis; The main flow section confluence calculation comprises the steps of calculating slope main flow section confluence and river course main flow section confluence, wherein the calculating slope main flow section confluence comprises the steps of defining the confluence object of the river basin produced flow according to the inclusion relation of a main flow section subriver region and a main flow section river course in a knowledge graph, adopting a hydrological model to combine with a terrain parameter in the calculation process, calculating confluence time and flow data from the slope produced flow to the main flow, correlating the confluence time and the flow data to corresponding main flow sections, calculating river course main flow section confluence comprises the steps of calculating the propagation time and the attenuation process from the upstream to the outlet section of the main flow section according to the upstream and the downstream relation of the main flow section and the main flow section in the knowledge graph, and forming a main flow section flow process line; The main flow converging calculation comprises the steps of according to the upstream-downstream relation between converging port sections of every two branches in a knowledge graph, the 'branch-main flow' relation and the branch converging calculation result, according to the 'first-come-last-add' principle, translating each branch flow process according to converging time, and then overlapping the flow process line of a main flow section in the main flow section converging calculation result with each other in a time interval to finally generate flow process data of a main flow outlet section.
  8. 8. The watershed-oriented flood prevention knowledge decision supervision method of claim 7, wherein the performing of the dry-flow flood evolutionary simulation according to the calculation result of the calculation engine, the knowledge graph and the hydrographic coupling model comprises: According to the hierarchical relation of the 'main flow-outlet section-target section' interrupt surface in the knowledge graph as a flood evolution path, calculating a flood evolution process by adopting a hydrologic-hydraulic coupling model, wherein the calculation flow comprises the following steps: calculating key parameters such as arrival time, flow peak value, water level change and the like of flood from a main flow outlet section to each target section; According to the map association relation of river-section-monitoring points, real-time monitoring data are dynamically fed back to the model, and a forecast result is corrected through rolling iteration; and outputting high-precision forecast data of each target section for 1-12 hours in the future, wherein the high-precision forecast data comprises water level and flow time sequence curves.
  9. 9. The knowledge decision supervision method for flood prevention in a river basin according to claim 8, wherein the target section early warning, the early warning notification, the risk studying and judging, the risk predicting and the auxiliary decision emergency predicting are sequentially performed according to the knowledge graph and the simulation result of the simulation engine, and the knowledge decision supervision method comprises the following steps: The target section early warning judgment comprises the steps of calling forecast data or actual measurement data of the target section in real time according to the association relation of the target section, the monitoring points and the early warning rules in the knowledge graph, and automatically judging early warning grades and triggering early warning signals of the corresponding monitoring points by matching the early warning rules in the graph through a preset rule engine; The early warning notification comprises automatically generating early warning notification according to the association relation of 'monitoring points-affiliated risk areas-affiliated villages-emergency organization systems' in the knowledge graph and a preset rule engine, positioning the associated risk areas and villages according to the monitoring points triggering early warning, and pushing early warning information to the corresponding emergency organization systems, wherein the early warning information comprises early warning grades and data conditions; Risk research and judgment, namely extracting real-time forecast data of a target section after early warning triggering according to the map relation of a risk area-historical scene in a knowledge map, and screening historical cases with the similarity more than or equal to 85% through similarity algorithm analysis, wherein the real-time forecast data of the target section comprises water level, flow and duration; After research and judgment, simulating a flood flooding process based on a two-dimensional hydrodynamic model and a GIS space analysis model according to the map relations of 'risk area-house-masses', 'risk area-flood prevention plateau', 'risk area-evacuation route' and 'risk area-rescue power' in the knowledge map, and outputting a pre-modeling report, wherein the pre-modeling report comprises a flooding range boundary, the number and area of the submerged houses, the total number of the threatened resident population, special population distribution such as the old, the weak, the sick and the disabled, security assessment of the flood prevention plateau, risk assessment of the evacuation route and accessibility analysis of rescue power; the auxiliary decision-making emergency plan comprises a scheme with highest matching degree from an emergency plan library according to the association logic of a risk area-emergency plan, a risk area-monitoring point-early warning level, an early warning level-emergency plan and a risk area-history scene in a knowledge graph and combining a risk research and judgment result and a risk prediction result, wherein the scheme comprises an emergency response level, a personnel transfer scheme and a rescue force allocation scheme.
  10. 10. The knowledge decision supervision method for flood prevention in a river basin according to claim 9, wherein the constructing step of the preset rule engine comprises: The method comprises the steps of constructing an early warning judging engine, wherein the early warning judging engine is used for presetting a priority, directly covering a low-level judging result by a high level, independently setting a condition combination for each level, judging a corresponding level according to indexes one by one, screening the highest level, checking whether an AND/OR rule of the level is met, triggering the level early warning if the AND/OR rule is met, AND matching the next highest level downwards if the AND/OR rule is not met; The method comprises the steps of constructing an early warning notification engine, wherein the early warning notification engine is used for constructing a two-dimensional rule base, the construction of the two-dimensional rule base comprises a grade-personnel mapping and a flow link-personnel mapping, and the method further comprises the steps of automatically matching corresponding rules after early warning determination, calling a short message interface to send in batches, and synchronously recording sending time, a receiver and a sending state.

Description

Knowledge decision supervision method for flood prevention in drainage basin Technical Field The invention relates to the technical field of flood control in a river basin, in particular to a knowledge decision supervision method for flood control in the river basin. Background In traditional flood control work, core scenes such as risk analysis and decision making face significant challenges. The staff needs to comprehensively reference multi-dimensional and high-dynamic information, including real-time hydrological monitoring data, historical flood case data of different orders, flood evolution data, mass distribution and transfer requirements of a power-assisted area, emergency material reserve, rescue team positions, scheduling capability and the like. However, these key factors are often present in the form of independent unstructured data in the existing mode, and the internal association (such as matching relationship between flood evolution and mass transfer time, and corresponding relationship between material reserve and the scale of the area in the hypochondrium) between them cannot be effectively linked and integrated, so as to form an information island. The fragmented information state causes that when working personnel face flood conditions, the association logic among all factors is difficult to establish quickly, feasibility and potential risks of different decision schemes cannot be estimated efficiently, and finally the flood control decision efficiency is low, the precision is insufficient, and the quick response requirement under complex flood conditions is difficult to meet. Therefore, a knowledge decision supervision method for flood prevention in the river basin is developed to solve the problems. Disclosure of Invention The invention provides a knowledge decision supervision method for flood control in a river basin, which aims to solve the problems of low efficiency and insufficient accuracy of the existing flood control decision. The invention discloses a knowledge decision supervision method for flood prevention in a river basin, which comprises the following steps: acquiring flood prevention data and real-time meteorological data of a river basin of a target area; Knowledge extraction is carried out on flood prevention data of the river basin to obtain space element information and non-space element information; Coding the space element information according to a preset identity coding rule to obtain coded space element information; Constructing a spatial topological relation of the encoded spatial element information by using a geographic information system, and constructing a knowledge relation of a root primitive according to the constructed spatial topological relation to obtain a spatial element knowledge relation; Fusing the non-space element information and the space element knowledge relationship, and converting the fusion result into a knowledge graph to obtain a fusion knowledge graph; The method comprises the steps of inputting meteorological data and a fusion knowledge graph into a preset knowledge auxiliary decision model, and outputting a decision result, wherein the knowledge auxiliary decision model comprises a calculation engine, a simulation engine and an early warning engine, the calculation engine is used for sequentially carrying out river basin runoff calculation, tributary confluence calculation, main flow interval confluence calculation and main flow confluence calculation according to the knowledge graph, the meteorological data and the hydrologic model, the simulation engine is used for carrying out main flow flood evolutionary simulation according to the calculation result of the calculation engine, the knowledge graph and the hydrologic coupling model, and the early warning engine is used for sequentially carrying out target section early warning, early warning notification, risk studying and judging, risk previewing and auxiliary decision emergency plan according to the simulation result of the knowledge graph and the simulation engine. Further, obtaining flood control data of the river basin of the target area includes: The method comprises the steps of obtaining river basin flood control data of a target area, wherein the river basin flood control data comprise river basin information, sub-river basin partition information, river water system information, monitoring equipment information, risk area information, placement point information, flood control highland information, evacuation route information, historical scene information, emergency plan information, risk area house distribution information, population structure information and emergency organization system information; And carrying out data space processing on the flood control data of the river basin based on the geographic information system, and storing different knowledge elements in different thematic layers according to the management principle of classification and layering. Furt