Geological disaster risk double-control data processing method and system based on data center

Abstract

The invention provides a geological disaster risk double-control data processing method and system based on a data center station, and the method comprises the steps: configuring a special rule for a geological disaster field, integrating a business system, Internet of Things equipment and external department heterogeneous data, unifying space coordinates, reconstructing a 15-bit hidden danger point coding system, employing a language large model to map a threat object to a standardized dictionary, and carrying out the construction of a standard dictionary. The method comprises the steps of constructing a data physical model containing association rules and constraints, splitting a core hidden danger library and a risk point library, dividing monitoring priorities, realizing three types of binding of hidden danger points, monitoring equipment, persons in charge and a disposal process based on administrative division, constructing a geological disaster type exclusive parameter library, and fusing monitoring data to calculate risk indexes. According to the method, emergency tasks are triggered and closed-loop feedback is carried out, hierarchical API service and three-order data quality detection are adopted, efficient treatment is realized, a whole-process closed loop of risk identification, management and control and feedback is formed, and the intelligent level of geological disaster prevention and control is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of geological data processing, in particular to a geological disaster risk double-control data processing method and system based on a data center. BACKGROUND

[0002] The current geological disaster prevention field has technical bottlenecks such as scattered data, non-uniform standards, and insufficient dynamic risk management and control. Traditional methods rely on manual patrols and single hidden danger point prevention and control, which are difficult to cope with the current situation of new hidden dangers occurring in risk areas rather than known points, and lack unified management capabilities for multi-source heterogeneous data.

[0003] The existing technology does not establish a standardized coding system and threat object classification dictionary, resulting in confusion in hidden danger point attribute fields, poor data correlation, and difficulty in supporting accurate risk identification and hierarchical response. At the same time, the dynamic binding mechanism of monitoring equipment and responsible persons is missing, the division of core hidden dangers and risk point libraries relies on static thresholds, and the priority cannot be adjusted in real time. In addition, the emergency task generation and disposal feedback lacks closed-loop management, and the data quality detection relies on manual verification, which can easily lead to distorted early warning.

[0004] To solve the above problems, the present method realizes heterogeneous data aggregation and standardized management through a data center, constructs an administrative division coding system and a dynamic risk model, combines three-level data quality detection and intelligent binding mechanism, forms a full-process closed loop of risk identification-control-feedback, and significantly improves the accuracy and response efficiency of geological disaster double control. SUMMARY

[0005] The purpose of the present application is to provide a geological disaster risk double-control data processing method and system based on a data center.

[0006] The problem to be solved by the present application is to solve the core problems of multi-source heterogeneous data integration difficulty, risk identification lag, dynamic response deficiency, and weak data management in geological disaster risk control.

[0007] The geological disaster risk double-control data processing method based on a data center adopts the following technical solutions: S1: Configure data extraction rules specific to the geological disaster field to access heterogeneous data scattered in business systems, Internet of Things devices, and external departments to the data center; S2: Convert spatial data uniformly to the national geodetic coordinate system, reconfigure the coding system for hidden danger point attribute fields according to the administrative division + hidden danger point + sequence number rule, map the threat object classification to the standardized dictionary, and build a geological disaster data physical model; S3: Based on the preset risk threshold, the original hidden danger point library is divided into core hidden danger and risk point library. The core hidden danger library corresponds to high monitoring priority, and the risk point library corresponds to regular patrol priority; S4: Based on the administrative division coding system, three types of bindings of hidden danger points-monitoring equipment, hidden danger points-responsible persons, risk event-disposal process are established. Hidden danger point-monitoring equipment binding is matched through grid location and equipment type, hidden danger point-responsible person binding is bound to group monitoring and group prevention personnel according to grid responsibility area, and risk event-disposal process binding is associated with disaster risk data and emergency tasks; S5: A geological disaster type exclusive parameter library is constructed, monitoring data and the parameter library are fused, a risk index is calculated, a high-risk event generates an emergency task, is pushed to the responsible person, and the disposal result is backfilled to the hidden danger point state field, forming a risk identification-control-feedback closed loop; S6: A hierarchical API service mechanism is adopted, internal systems are directly connected to database views, external departments obtain desensitized data based on API, and three-stage data quality detection of preliminary screening-error correction-backflow is adopted. Preliminary screening is a rule engine to detect coordinate out-of-bounds and missing fields, error correction is to match similar positions and complete attributes, and backflow is to update to the data center after manual confirmation.

[0008] Further, the data extraction rules specific to the geological disaster field in S1 are configured to access heterogeneous data scattered in business systems, Internet of Things devices, and external departments to the data center, including: Business system data includes geological disaster hidden danger point basic information including location, type, level, historical patrol record, and treatment engineering account, stored in an internal database; Internet of Things device data includes sensor monitoring data including displacement, underground water level, rainfall, and video monitoring stream data including landslide deformation video and unmanned aerial vehicle aerial image, transmitted through 4G and 5G network computing devices; External department data includes meteorological warning data including rainfall forecast, earthquake activity data including magnitude, focal depth, remote sensing image including satellite SAR data, and social and economic data including population density and land use type, accessed through API interface and file sharing; Special rules are designed for different data source types, SQL data query templates are configured through ETL tools to extract hidden danger point location and attribute fields, API interface call rules are designed based on RESTful protocol to send rainfall request to the meteorological bureau API and parse JSON response, CSV and XML format parsing rules are defined to parse unmanned aerial vehicle aerial image metadata including shooting time and coordinate range; The extraction logic is encapsulated as a reusable template, supporting parameterized configuration including input field name and output field name, and visual arrangement of rules is implemented based on the platform, and users define data extraction processes by dragging nodes.

[0009] Further, the S2 reconfigures the coding system of the hazard point attribute field according to the administrative division + hazard point + sequence number rule, maps the threat object classification to the standardized dictionary, and builds a physical model of geological disaster data, including: A 15-digit hierarchical code of administrative division + hazard point + sequence number is adopted, the administrative division code is up to 9 digits, divided into four levels of province-city-county-town, the hazard point type code is 3 digits, including hazard point and type, and the sequence number is 3 digits, which is sequentially increased according to the newly added hazard point. When a new hazard point is added, a new sequence number is assigned and the database is updated; According to the geological disaster influence objects including residential areas, schools, roads and sensitivity including population density, economic value, the threat level is divided, and the dictionary structure is defined, including threat object name, classification code, sensitivity level and associated risk indicators; Based on the language large model, the semantic analysis of the original threat object description text is carried out, and the classification code in the dictionary is matched. For the text without explicit threat object field, the semantic keywords such as residential area and road are extracted and mapped to the classification. After manual verification, the dictionary is updated, and the corrected results are pushed to the data platform through API; The data physical model expression framework is constructed, including data storage structure, data association rule, data integrity constraint and data access optimization rule; The data storage structure includes field definition, primary key and foreign key, index and partition, defines field name, data type, length and constraint, data type includes VARCHAR, FLOAT, constraint includes non-empty, uniqueness, establishes association relationship through primary key and foreign key, primary key includes hazard point ID and administrative division code, foreign key includes monitoring point ID and organization ID, establishes index for high-frequency query field, selects regional and time partition to optimize performance, high-frequency query field includes hazard point ID and timestamp, and regional and time partition includes administrative division code and monitoring data time range; The data association rule includes 1:N, N:1 and many-to-many relationship, which is associated through foreign key, hazard point and monitoring point, hazard point and group monitoring and prevention personnel, and for many-to-many relationship including hazard point and responsible person, an intermediate table is designed to realize the association; The data integrity constraint includes primary key constraint, foreign key constraint, uniqueness constraint and check constraint, the primary key constraint is unique for each record, including hazard point ID and cutting slope point ID, the foreign key constraint ensures the consistency of associated data, including deleting the associated monitoring point when deleting the hazard point, the uniqueness constraint sets uniqueness limit for key fields including administrative division code and personnel ID, and the check constraint limits the field value range, including threat population ≥ 0 and rainfall ≤ threshold; The data access optimization rules include index optimization, cache strategy and parallel processing. The index optimization is performed on high-frequency query fields including administrative division code. The cache strategy is performed on core hidden danger library and monitoring data. The parallel processing is performed on large-scale data including historical monitoring data.

[0010] Further, the S3 splits the original hidden danger point library into core hidden danger and risk point library based on a preset risk threshold, including: A threat population threshold is set as a population risk evaluation parameter. A population of ≥ 50 people needs high priority monitoring, and a population of ≤ 10 people can be regularly patrolled. A stability index is calculated as a stability evaluation parameter based on historical monitoring data of hidden danger points. A threshold displacement rate of > 5 mm / month is set as a high risk. In combination with the administrative division + hidden danger point + sequence number rule in S2, a spatial relationship mapping is established between the hidden danger point and the population-dense area, the transportation hub sensitive area, the public facility and the economic key area. A sensitivity threshold is set as an environmental sensitivity evaluation parameter. The original hidden danger point data is preprocessed. The threat population, stability index and sensitivity threshold fields are interpolated to handle missing values. Based on the standardized dictionary in S2, the data consistency is checked. The threat object attribute is mapped to the resident household property classification system. According to the multi-dimensional evaluation parameters, a grading rule is established. For any two of the threat population threshold, the stability index and the sensitivity threshold, the core hidden danger library is classified into the core hidden danger library. The core hidden danger point is added with a high monitoring priority label. For any one of the threat population threshold, the stability index and the sensitivity threshold, the risk point library is classified into the risk point library. The risk point is set with a regular patrol cycle.

[0011] Further, the S4 establishes three types of binding of hidden danger point-monitoring device, hidden danger point-responsible person and risk event-disposal process based on the administrative division code system, including: The geographical distance between the hidden danger point and the monitoring device is calculated. According to the geographical distance, the monitoring device is divided into high-precision monitoring device and basic monitoring device. When the monitoring device installation position and the hidden danger point have a latitude and longitude distance of ≤ 500 meters, the high-precision monitoring device is associated. If the distance is > 500 meters, the basic monitoring device is associated. According to the hidden danger point type including landslide, collapse and threat object including residential area, highway, special monitoring device is matched. For landslide hidden danger point, displacement sensor and rain gauge are bound. For collapse hidden danger point, vibration sensor and crack meter are bound. When the monitoring device state and the hidden danger point risk level change, the device binding relationship is adjusted. The monitoring device state change includes fault and data anomaly. The hidden danger point risk level change is upgraded to the core hidden danger library in S3. Based on the administrative division + hidden point + sequence number coding rule in S2, the hidden point is attributed to a specific grid responsibility area. The responsible grid of the group monitoring and prevention personnel needs to completely cover the grid where the hidden point is located. The core hidden point is bound to the professional responsibility person of geological disaster prevention expert, and the risk point is bound to the part-time responsibility person of village group cadre. When the hidden point is migrated from the risk point library to the core hidden point library, the part-time responsibility person is upgraded to the full-time responsibility person. When the administrative division is adjusted, the responsibility person is matched again to avoid vacancy of responsibility. According to the disaster risk type including landslide and threat level, the corresponding emergency disposal process template, i.e. landslide emergency investigation process, is matched.

[0012] Further, the S5 constructs a geological disaster type exclusive parameter library and calculates a risk index, including: For different geological disaster types including landslide, collapse and debris flow, an exclusive parameter library is formulated. The parameters include rainfall, displacement rate, crack width, soil thickness index. The weight of each parameter is determined through an expert review meeting and recorded in the parameter library. Based on historical disaster risk data, the parameter threshold of a typical disaster event is extracted. The monitoring data is matched with the weight rules in the parameter library. The score of each parameter is calculated according to the weight and monitoring value. The final risk index is obtained by weighted summation of the scores of all parameters. The final risk index = Σ (monitoring value of parameter i / threshold of parameter i × weight of parameter i). According to the final risk index value, it is marked as different risk events. The low risk event is final risk index ≤ 0.5, the medium risk event is 0.5 < final risk index ≤ 0.8, and the high risk event is final risk index > 0.8.

[0013] Further, the S6 adopts three-stage data quality detection of preliminary screening, error correction and backflow, including: Based on the unified national geodetic coordinate system in S2, it is detected whether the hidden point coordinate exceeds the geographical boundary. The out-of-bound data is marked as to be manually corrected. The hidden point attribute fields including threat population and threat object are scanned. If the field missing rate exceeds 10%, an alarm is triggered and a log is recorded. For the hidden point with out-of-bound or missing coordinates, the administrative grid code in S4 is matched through GIS tools, combined with the administrative division + hidden point + sequence number rule in S2, the nearest correct position is associated, and the missing threat object field is inferred based on the geological disaster type exclusive parameter library in S5.

[0014] Further, the geological disaster risk double-control data processing system based on the data middle platform is used for realizing the geological disaster risk double-control data processing method based on the data middle platform, and comprises a data acquisition module, a data aggregation module, a data conversion component module, a data warehousing module, a data storage module and a data updating module. The data acquisition module is configured with a multi-source heterogeneous data access interface, is used for collecting business system data, Internet of Things monitoring equipment data and external department shared data, integrates a rule engine to perform preliminary quality screening on the collected data, and performs data format standardization conversion; The data aggregation module integrates a heterogeneous data source adapter, is configured with a special data extraction rule library in the geological disaster field, and supports customizing data extraction, conversion and loading processes through a visual arrangement interface; The data conversion component module comprises an encoding system generation unit, a threat object mapping unit, a data relationship construction unit and a parameter library management unit, the encoding system generation unit constructs a hierarchical encoding based on an administrative division-hazard point-sequence number rule, the threat object mapping unit maps original threat objects to a standardized classification system, the data relationship construction unit is configured with 1:N, N:1 and many-to-many data correlation rules, and the parameter library management unit constructs special parameter libraries for different geological disaster types; The data warehousing module is configured with a data physical model definition unit and a hazard grading storage unit, the data physical model definition unit defines data storage structure, data correlation rules and data integrity constraints, and the hazard grading storage unit splits hazard point data into a core hazard library and a risk point library based on a risk threshold; The data storage module adopts a heterogeneous storage architecture, core hazard library data is stored in a high-performance memory database to support real-time access, risk point library data is stored in a distributed columnar database to optimize massive data query efficiency, and the system implements a hierarchical access control strategy; The data updating module comprises a binding relationship maintenance unit, an emergency task generation unit and a data quality guarantee unit, the binding relationship maintenance unit adjusts monitoring equipment and responsible person binding relationships according to risk level changes, the emergency task generation unit triggers an emergency disposal process based on a risk event, and the data quality guarantee unit implements a three-stage data quality control mechanism of preliminary screening-error correction-backflow.

[0015] The application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to realize steps in the geological disaster risk double-control data processing method based on the data middle platform.

[0016] The application further provides a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the geological disaster risk double-control data processing method based on a data middle platform when executing the computer program.

[0017] The application has the advantages that: based on multi-dimensional indexes such as threat population, stability index, and sensitivity threshold, the hidden danger points are divided into a core hidden danger library of high monitoring priority and a risk point library of regular patrol priority, the accuracy of resource allocation is realized, the hierarchical API service mechanism is adopted to meet the access requirements of different levels, and the data security is ensured; By constructing a geological disaster type exclusive parameter library, combining with monitoring data to calculate a risk index, realizing the upgrade from static threshold judgment to dynamic risk quantification, based on an administrative division coding system, realizing the intelligent matching of hidden danger points, monitoring equipment, responsible persons, and disposal processes, ensuring that the responsible persons can be quickly located and the standardized disposal process can be started when a risk event occurs, through the coding system, realizing the unique identification and quick search of hidden danger points, based on the physical model and data governance rules, ensuring the consistency, integrity, and efficient access of data. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 It is a flow chart of the geological disaster risk double-control data processing method based on a data middle platform; Figure 2 It is a module diagram of the geological disaster risk double-control data processing system based on a data middle platform. DETAILED DESCRIPTION

[0019] The application will be further and completely explained below, but the protection scope of the application is not limited to this.

[0020] Example 1 The technical scheme adopted by the geological disaster risk double-control data processing method based on a data middle platform is as follows: S1: configuring data extraction rules special for the geological disaster field, and connecting heterogeneous data scattered in business systems, Internet of Things equipment, and external departments to the data middle platform; S2: uniformly converting spatial data into the national geodetic coordinate system, reconfiguring the coding system according to the administrative division + hidden danger point + sequence number rule for the hidden danger point attribute field, mapping the threat objects to a standardized dictionary, and constructing a geological disaster data physical model; S3: based on a preset risk threshold, the original hidden danger point library is divided into a core hidden danger library and a risk point library, the core hidden danger library corresponds to a high monitoring priority, and the risk point library corresponds to a regular patrol priority; S4: Based on the administrative division coding system, three types of binding of hidden danger points-monitoring equipment, hidden danger points-responsible persons, risk event-disposal process are established. Hidden danger point-monitoring equipment binding is matched through grid location and equipment type, hidden danger point-responsible person binding is bound to group monitoring and group prevention personnel according to grid responsibility area, and risk event-disposal process binding is associated with disaster risk data and emergency tasks; S5: Construct a geological disaster type exclusive parameter library, integrate monitoring data and parameter library, calculate risk index, generate emergency tasks for high-risk events, push to the binding responsible person, backfill the disposal result to the hidden danger point state field, and form a risk identification-control-feedback closed loop; S6: Adopt hierarchical API service mechanism, internal system directly connects database view, external departments obtain desensitized data based on API, adopt three-stage data quality detection of preliminary screening-error correction-backflow, preliminary screening is rule engine detection of coordinate out-of-bound, field missing, error correction is matching similar position and attribute completion, and backflow is updated to data platform after manual confirmation.

[0021] Reference Figure 1 As shown, it is a geological disaster risk double-control data processing method flowchart based on data platform.

[0022] Further, the S1 configures data extraction rules specific to the geological disaster field, and connects heterogeneous data scattered in business systems, Internet of Things devices, and external departments to the data platform, including: Business system data includes geological disaster hidden danger point basic information including location, type, level, historical patrol record, and governance engineering account book, stored in internal database, mainly obtained from databases in provincial, municipal, and county geological disaster management systems. Historical patrol records contain unstructured data such as text descriptions and photos, which need to be extracted key information such as patrol time, patrol personnel, and hidden danger change through data cleaning tools; Internet of Things device data includes sensor monitoring data including displacement, underground water level, rainfall, and video monitoring stream data including landslide deformation video and unmanned aerial vehicle aerial image, transmitted through 4G and 5G network computing devices; sensor data is accessed in real time through MQTT protocol, transmitted in JSON format, video monitoring data is transmitted in streaming media protocol, and unmanned aerial vehicle aerial image is stored in GeoTIFF format; External department data includes meteorological warning data including rainfall forecast, earthquake activity data including magnitude, focal depth, remote sensing images including satellite SAR data, social and economic data including population density, land use type, which are accessed through API interface and file sharing; meteorological warning data request parameters include administrative division code, time range, and 24-hour rainfall forecast data in JSON format is returned; satellite SAR data file naming rules are "satellite type_ imaging time_ region code.tif"; social and economic data includes population density, land use type code and other fields.

[0023] Special rules are designed for different data source types, SQL data query templates are configured through ETL tools to extract hidden point positions and attribute fields; for hidden point spatial data, the SQL template is "SELECT ID, NAME, ST_TRANSFORM(geometry, 4490) AS geometry_cgcs2000 FROM landslide_points", where 4490 is the EPSG code of the coordinate system; API interface calling rules are designed based on RESTful protocol, rainfall request is sent to the meteorological bureau API and JSON response is parsed, CSV and XML format parsing rules are defined, metadata of unmanned aerial vehicle aerial image is parsed including shooting time, coordinate range, camera parameters, etc., CSV parsing rule specifies the separator as comma, the first line is field name, and the coordinate field format is "x, y"; Extracting logic is encapsulated as reusable templates such as hidden point coordinate conversion template, supporting parameterized configuration including input field name, output field name, visual arrangement of rules based on platform, users define data extraction process by dragging nodes; the platform provides node configuration panel, parameters of each node can be set; data source node → data parsing node → data conversion node → data storage node, nodes are connected through data flow, forming a complete data extraction link.

[0024] Further, the S2 reconstructs the coding system according to the administrative division + hidden point + sequence number rule for the attribute fields of the hidden point, maps the threat object classification to the standardized dictionary, and builds a physical model of geological disaster data, including: A 15-digit hierarchical code of administrative division + hidden point + sequence number is used, the administrative division code is up to 9 digits, divided into four levels of province-city-county-town, the hidden point type code is 3 digits, including hidden point and type, the sequence number is 3 digits, increasing in sequence according to the new hidden point, the code of a landslide hidden point in Xinjian District, Nanchang City, Jiangxi Province is: 360112101001, 36 = Jiangxi, 01 = Nanchang, 12 = Xinjian, 1 = hidden point, 01 = landslide, 001 = sequence number, when a new hidden point is added, a new sequence number is assigned and the database is updated; According to the geological disaster influence object including residential area, school, road and sensitivity including population density, economic value division threat level, high sensitivity such as residential area, hospital, school, medium sensitivity such as factory, farmland, low sensitivity such as wasteland, forest, define dictionary structure including threat object name, classification code, sensitivity level, associated risk index, reference table one is an example entry; Table One: Threat object Classification code Sensitivity level Associated risk indicator Residential area T001 High Population density, difficulty of evacuation School T002 High Number of students, building seismic resistance Based on the language big model, the semantic analysis is carried out on the original threat object description text, and the classification code in the dictionary is matched, and the training data is derived from the threat object description in the geological disaster investigation report of Jiangxi Province in previous years;The model identifies the middle school as T002 classification, and the text without explicit threat object field is extracted, such as residential area, road and mapping classification, and the dictionary is updated after artificial verification, and the corrected results are pushed to the data platform through API.

[0025] The data physical model expression framework is constructed, including data storage structure, data association rule, data integrity constraint and data access optimization rule; The data storage structure includes field definition, primary key and foreign key, index and partition;Define field name, data type, length and constraint, data type includes VARCHAR, FLOAT, constraint includes non-empty, uniqueness;The association relationship is established through the primary key and the foreign key, the primary key includes the hidden danger point ID, the administrative division code, the foreign key includes the monitoring point ID, the organization ID, when deleting the hidden danger point record, deleting the associated monitoring data, keeping the data consistency;Index is established for high-frequency query field, and index is established for high-frequency query field;The high-frequency query field includes hidden danger point ID, time stamp, B+ tree index is established for administrative division code, hidden danger point ID and other fields;The selection is carried out according to the region and the time partition optimization performance, and the high-frequency query field includes hidden danger point ID, time stamp, B+ tree index is established for administrative division code, hidden danger point ID and other fields; The data association rule includes 1:N, N:1 and many-to-many relationship, the association is established through the foreign key, the hidden danger point and the monitoring point, the hidden danger point and the group measurement and prevention worker, for the many-to-many relationship including the hidden danger point and the responsible person, the association is realized by designing the intermediate table;The hidden danger point and the monitoring point adopt 1:N relationship, one hidden danger point can be associated with multiple monitoring points;The hidden danger point and the group measurement and prevention worker adopt N:1 relationship, multiple hidden danger points can be responsible by the same group measurement and prevention worker;The hidden danger point and the responsible person adopt many-to-many relationship; The data integrity constraint includes primary key constraint, foreign key constraint, uniqueness constraint and check constraint;The primary key constraint is unique for each record, including hidden danger point ID, cutting slope point ID;The foreign key constraint is associated with the consistency of data, including deleting the associated monitoring point when deleting the hidden danger point;The uniqueness constraint sets the uniqueness limit for the key field, including administrative division code, personnel ID number;The check constraint limits the field value range, including threat population ≥ 0, rainfall ≤ threshold value; The data access optimization rules include index optimization, cache strategy and parallel processing; the index optimization is established on high-frequency query fields including administrative division code, the cache strategy is enabled on core hidden danger library and monitoring data, and the parallel processing is adopted on large-scale data including historical monitoring data.

[0026] For the hidden danger point physical model, the table structure includes fields such as hidden_danger_id (primary key), threat_population (threat population), risk_index (risk index), etc., which meet the field definition and constraints, and are associated with the JC_MONITORPOINTINFO table through hidden_danger_id in a 1:N manner, meeting the association rules.

[0027] For the weather risk warning physical model, the table structure includes fields such as weather_id (primary key), rainfall (rainfall), temperature (temperature), etc., which meet the field definition, are partitioned by time range, and have an index established on the rainfall field, meeting the access optimization rules.

[0028] Further, the S3 splits the original hidden danger point library into core hidden danger and risk point library based on a preset risk threshold, including: Set the threat population threshold as the population risk assessment parameter, ≥ 50 people need high priority monitoring, ≤ 10 people can be regularly patrolled, and calculate the stability index as the stability assessment parameter through the hidden danger point historical monitoring data such as displacement rate; set the threshold displacement rate > 5 mm / month as high risk; combine the administrative division + hidden danger point + sequence number rule in S2 to establish a spatial relationship mapping between hidden danger points and population dense areas, transportation hub sensitive areas, public facilities and economic key areas, set the sensitivity threshold as the environmental sensitivity assessment parameter, and refer to Table 2.

[0029] Table 2 Sensitive area type Sensitivity threshold range Population-dense area Threat population ≥ 50 people (high sensitivity); 10 people ≤ threat population < 50 people (medium sensitivity); threat population < 10 people (low sensitivity) Transportation hub Distance from railway / highway < 300 meters (high sensitivity); 300 meters ≤ distance < 1000 meters (medium sensitivity); distance ≥ 1000 meters (low sensitivity) Public facilities (hospitals, schools) Directly associated as high sensitivity (no additional threshold required) Economic key area Threat property value ≥ 10 million yuan (high sensitivity); 5 million yuan ≤ value < 10 million yuan (medium sensitivity); value < 5 million yuan (low sensitivity) The original hidden danger point data is preprocessed, and the missing values of the threat population, stability index and sensitivity threshold fields are interpolated; the missing values of the threat population are estimated by the regional population density x hidden danger influence area, the missing values of the stability index are filled with the historical average value of the same type of hidden danger point, and the missing values of the sensitivity threshold are taken within a range of 500 meters. The nearest hidden danger point sensitivity value is taken; based on the standardized dictionary in S2, the data consistency is checked, the threat object attribute is mapped to the resident household property classification system through regular expression matching combined with manual checking, and is classified and coded according to the nine categories standard (resident household property, education facilities, mine, industrial facilities, agriculture, highway transportation facilities, railway transportation facilities, social public facilities, water conservancy and hydropower facilities). According to the multi-dimensional evaluation parameter to establish grading rules: for any two conditions of meeting the threat population threshold, stability index and sensitivity threshold, it is classified into the core hidden danger library, and a high monitoring priority label is added to the core hidden danger point. For any one condition of meeting the threat population threshold, stability index and sensitivity threshold, it is classified into the risk point library, and a regular patrol cycle is set for the risk point. The core hidden danger library adopts a real-time monitoring mechanism, and the monitoring data is collected every 15 minutes. The risk point library adopts a regular patrol mechanism, and the high-risk point is patrolled once a week, the medium-risk point is patrolled twice a month, and the low-risk point is patrolled once a quarter.

[0030] Further, the S4 is based on the administrative division coding system to establish three types of binding of hidden danger points-monitoring devices, hidden danger points-responsible persons, and risk events-disposal processes, including: The geographic distance between the hidden danger point and the monitoring device is calculated, and according to the geographic distance, the monitoring device is divided into high-precision monitoring device and basic monitoring device. When the latitude and longitude distance between the installation position of the monitoring device and the hidden danger point is less than or equal to 500 meters, it is associated with the high-precision monitoring device. If the distance is greater than 500 meters, it is associated with the basic monitoring device. The 500-meter threshold is determined according to the geological disaster monitoring specification, which represents the effective sensing range of the monitoring device. According to the hidden danger point type including landslide, collapse and threat object including residential area, highway, special monitoring device is matched. For landslide hidden danger point, displacement sensor and rain gauge are bound. For collapse hidden danger point, vibration sensor and crack meter are bound. When the monitoring device state and hidden danger point risk level change, the device binding relationship is adjusted. The monitoring device state change includes fault, data anomaly, and the hidden danger point risk level change is upgraded to the core hidden danger library in S3.

[0031] The geographic distance between the hidden danger point and the monitoring device is calculated, and according to the geographic distance, the monitoring device is divided into high-precision monitoring device and basic monitoring device. When the latitude and longitude distance between the installation position of the monitoring device and the hidden danger point is less than or equal to 500 meters, it is associated with the high-precision monitoring device. If the distance is greater than 500 meters, it is associated with the basic monitoring device. The 500-meter threshold is determined according to the geological disaster monitoring specification, which represents the effective sensing range of the monitoring device. According to the hidden danger point type including landslide, collapse and threat object including residential area, highway, special monitoring device is matched. For landslide hidden danger point, displacement sensor and rain gauge are bound. For collapse hidden danger point, vibration sensor and crack meter are bound. When the monitoring device state and hidden danger point risk level change, the device binding relationship is adjusted. The monitoring device state change includes fault, data anomaly, and the hidden danger point risk level change is upgraded to the core hidden danger library in S3. According to the disaster risk type including landslide and threat level, the corresponding emergency disposal process template, i.e., landslide emergency investigation process, is matched.

[0032] Further, the S5 constructs a geological disaster type exclusive parameter library to calculate a risk index, including: A dedicated parameter library is developed for different types of geological disasters, including landslides, collapses, and debris flows. The parameters are selected based on the geological disaster monitoring specifications and the historical case library of geological disasters in Jiangxi Province. The landslide parameter library includes rainfall, displacement rate, crack width, and soil thickness indicators. The collapse parameter library includes vibration frequency, crack rate, and rock body inclination. The debris flow parameter library includes hourly rainfall intensity, channel blockage, and loose material reserves. Through three rounds of anonymous expert review, the weights of each parameter are determined and recorded in the parameter library. In landslide disasters, the weight of rainfall is 0.7, and the weight of displacement rate is 0.3. Based on historical disaster data, the parameter threshold of typical disaster events is extracted, and the landslide trigger threshold is 3 consecutive days of rainfall ≥ 100 mm. The threshold extraction uses the percentile method, taking the 95th percentile of historical disaster events as the trigger threshold; The monitoring data is matched with the weight rules in the parameter library. The rainfall in the landslide monitoring data is associated with the 0.7 weight. The score is calculated for each parameter based on the weight and monitoring value. The rainfall score = monitoring value / threshold x rainfall weight. The scores of all parameters are weighted and summed to obtain the final risk index. The final risk index = Σ (monitoring value of parameter i / threshold of parameter i x weight of parameter i). This formula ensures that the risk index reflects the relative risk level of each parameter, and when all parameters reach the threshold, the risk index is 1.0; The landslide hazard point monitoring value is rainfall = 90 mm (threshold = 100 mm), displacement rate = 4 mm / month (threshold = 5 mm / month), and risk index = (90 / 100 x 0.7) + (4 / 5 x 0.3) = 0.63 + 0.24 = 0.87. According to the final risk index value, different risk events are marked. Low-risk events have a final risk index ≤ 0.5, medium-risk events have a final risk index between 0.5 and 0.8, and high-risk events have a final risk index > 0.8.

[0033] Further, the S6 adopts a three-stage data quality detection of preliminary screening, error correction, and reflow, including: Based on the unified national geodetic coordinate system in S2, the coordinates of the hazard points are detected to see if they exceed the geographical boundary. Out-of-bound data is marked for manual correction. The attribute fields of the hazard points include threatened population and threatened objects. If the field missing rate exceeds 10%, an alarm is triggered and a log is recorded. For the hidden danger points with out-of-bound coordinates or missing, the administrative grid code described in S4 is matched through the GIS tool ArcGIS, the nearest correct position is associated combining the administrative division + hidden danger point + sequence number rule in S2, and the spatial correlation algorithm is executed: 1) a 500-meter buffer zone is created with the out-of-bound point as the center; 2) all correct grids in the buffer zone are extracted; 3) the Euclidean distance of the out-of-bound point to each grid center is calculated; 4) the grid with the highest terrain similarity (slope, slope difference) and the nearest distance is selected as the matching target; if the coordinates of a hidden danger point are missing, it is matched to the landslide hidden danger point in Nanchang Qingshanhu District through the grid code 360100101001, and the coordinates are completed. Based on the geological disaster type exclusive parameter library such as landslide parameter library in S5, the missing threat object field is inferred, and the resident family property is filled according to the hidden danger point type.

[0034] Further, the geological disaster risk double-control data processing system based on the data center is used to realize the geological disaster risk double-control data processing method based on the data center, and the geological disaster risk double-control data processing system based on the data center comprises a data collection module, a data aggregation module, a data conversion component module, a data storage module, a data update module: The data collection module is configured with a multi-source heterogeneous data access interface, which is used to collect business system data, Internet of Things monitoring equipment data and external department shared data, integrate a rule engine to perform preliminary quality screening on the collected data, and perform data format standardization conversion; The data aggregation module integrates a heterogeneous data source adapter and is configured with a special data extraction rule library in the field of geological disasters, and supports customizing data extraction, conversion and loading processes through a visual arrangement interface; The data conversion component module includes an encoding system generation unit, a threat object mapping unit, a data relationship construction unit and a parameter library management unit. The encoding system generation unit constructs a hierarchical encoding based on the administrative division-hidden danger point-sequence number rule. The threat object mapping unit maps the original threat object to a standardized classification system. The data relationship construction unit is configured with 1:N, N:1 and many-to-many data correlation rules. The parameter library management unit constructs exclusive parameter libraries for different geological disaster types; The data storage module is configured with a data physical model definition unit and a hidden danger grading storage unit. The data physical model definition unit defines data storage structure, data correlation rules and data integrity constraints. The hidden danger grading storage unit splits hidden danger point data into core hidden danger library and risk point library based on risk threshold; The data storage module adopts a heterogeneous storage architecture, the core hidden danger library data is stored in a high-performance memory database to support real-time access, the risk point library data is stored in a distributed columnar database to optimize the query efficiency of massive data, and the system implements a hierarchical access control strategy; The data updating module includes a binding relationship maintenance unit, an emergency task generation unit, and a data quality guarantee unit. The binding relationship maintenance unit adjusts the binding relationship between monitoring equipment and responsible persons according to the risk level change. The emergency task generation unit triggers the emergency disposal process based on the risk event. The data quality guarantee unit implements a three-stage data quality control mechanism of preliminary screening, error correction, and backflow.

[0035] Reference Figure 2 As shown in the figure, it is a geological disaster risk double-control data processing system module diagram based on a data platform.

[0036] Embodiment Two The hidden danger point basic information of Qingyuan District in Ji'an City, Jiangxi Province is extracted from the geological disaster business support platform, including location, type, and level. An ETL tool is used to configure a SQL template, such as SELECT * FROM hidden_danger WHERE region_code='360801'. The monitoring data of displacement sensors and the rainfall of rain gauges are accessed, transmitted through a 5G network, and parsed in JSON data stream format. The weather bureau API is called based on the RESTful protocol to obtain rainfall forecasts. The remote sensing image is imported in a file sharing manner to parse CSV metadata. A reusable template such as the Qingyuan hidden danger point extraction template is built. The input / output fields are parameterized and configured. The hidden danger point ID is input, and the standardized location is output. The interface is visually arranged. The user drags the nodes to define the process such as data source→filter→conversion→warehouse. Eighty hidden danger point data, 50 sensor data, and daily weather warning data are successfully accessed. The preliminary screening data missing rate is less than 5%.

[0037] The spatial data of Qingyuan District is unified to the national geodetic coordinate system CGCS2000, the coordinate offset is corrected, and 15-bit hierarchical code is used, i.e. administrative division code + hidden point type + serial number, for example, the code of a landslide hidden point in Qingyuan District is 360801101001 (36 = Jiangxi, 08 = Ji'an, 01 = Qingyuan District, 1 = hidden point, 01 = landslide, 001 = serial number), when a new hidden point is added, a new serial number is assigned and the database is updated, the threat object classification is defined, such as residential area T001, school T002, the hidden point description text is input, such as threatening a middle school in Qingyuan District, the model semantic analysis is matched to the T002 classification, and the correction result is pushed through API after manual verification, the field hidden_danger_id is defined as the primary key, and the threat_population is defined as the non-empty constraint, the hidden point table HIDDEN_DANGER and the monitoring point table JC_MONITORPOINTINFO are associated in 1:N by the foreign key monitor_id, and the index is established for the high-frequency field region_code, and the monitoring data is partitioned by time.

[0038] The core hidden threat population threshold is ≥ 50 people, the risk point threat population threshold is ≤ 10 people, the displacement rate is calculated to be > 5 mm / month through historical monitoring data, the school is associated, the high-sensitive population dense area is associated, the missing value of the threat population field is interpolated, the consistency is verified based on the standardized dictionary, and is mapped to the residential property, the core hidden library meets any two threshold values of threat population ≥ 50 people and displacement rate > 5 mm / month, a landslide point in Qingyuan District (threat population 60 people, displacement 6 mm / month) is included in the core library, a high monitoring priority label is added, the risk point library meets any threshold value of threat population = 30 people, a monthly regular patrol cycle is set, among the 120 hidden points, 40 are included in the core library for real-time monitoring, and 80 are included in the risk point library for regular patrol, and the resource allocation efficiency is improved by 40%.

[0039] The geographical distance is calculated to be ≤ 500 meters, the high-precision device is associated, the code of a landslide point in Qingyuan District is 360801101001, the distance is 300 meters, and the displacement sensor is bound, when the device fails or the risk is upgraded, the binding is adjusted, the core hidden point is bound to the full-time person in charge of the disaster expert, the risk point is bound to the part-time person in charge of the village group cadre, the hidden point belongs to the grid 360801, the full-time person in charge is bound, the grid is adjusted, the landslide is matched with the landslide emergency investigation process.

[0040] For the landslide type in Qingyuan area, set the exclusive parameter library as rainfall weight 0.7, displacement rate weight 0.3, and trigger landslide when continuous 3-day rainfall ≥100mm. The final risk index = Σ (monitoring value / threshold value × weight), rainfall in Qingyuan area = 90mm (threshold value 100mm), displacement rate = 4mm / month (threshold value 5mm / month), index = (90 / 100 × 0.7) + (4 / 5 × 0.3) = 0.87, which is high risk, index > 0.8 generates emergency task, pushes to the binding person in charge, and the disposal result "has completed reinforcement" is backfilled to the hidden danger point state field, forming a closed loop. In the 2024 rainy season, the system generates 10 high-risk tasks, and the response time is shortened to 30 minutes.

[0041] Example Three Business system: scenic spot hidden danger point account, store internal database, contain historical debris flow record; Internet of Things equipment: pore water pressure gauge, ground sound alarm; external data: satellite SAR image monitoring ground deformation, meteorological bureau short-time heavy rain warning, API interface call; create scenic spot debris flow data template, parameterize configuration satellite image analysis rule, XML format metadata extraction.

[0042] The code of a debris flow gully in Qingyuan area is 360801203002, 20 = debris flow, 3 = scenic spot type, the language large model matches the description "threatens the visitor center" to the dictionary classification, T003: tourism facility, and indexes the high-frequency query field visitor_density of scenic spot.

[0043] Sensitivity threshold: scenic spot tourist ≥500 people / day (high sensitivity); stability index: material source reserve ≥10 4 m³, and the hidden danger point meeting "tourist ≥500 people + material source reserve exceeding standard" is included in the core library.

[0044] The debris flow gully is bound to the ground sound alarm, the distance is ≤300 meters, the rainy season is associated with the rain gauge, the special monitoring equipment is matched, the core library is bound to the full-time scenic spot safety supervisor, the risk library is bound to the part-time tour guide, and the debris flow warning triggers the tourist evacuation process.

[0045] Parameter 1-hour rainfall, corresponding weight 0.6, threshold value (historical data) ≥50mm; parameter 1-hour pore water pressure, corresponding weight 0.3, threshold value (historical data) ≥20kPa; parameter 1-hour material source reserve, corresponding weight 0.1, threshold value (historical data) ≥10 4 m³; risk index = (rainfall monitoring value / 50 × 0.6) + (pore water pressure / 20 × 0.3) + (material source reserve / 10 4 × 0.1), rainfall = 60mm, pore pressure = 25kPa, material source reserve = 1.35 × 10 4→ Index = 1.23 > 0.8, high risk, push evacuation task, and backfill the treatment result to the hidden danger point state field.

[0046] Embodiment Four The embodiment provides a computer readable storage medium, and a computer program is stored on the computer readable storage medium, and the computer program is executed by a processor to realize steps in the geological disaster risk double-control data processing method based on a data center as described in the embodiment one.

[0047] Embodiment Five The embodiment provides a computer device, including a memory, a processor, and a computer program stored on the memory and capable of running on the processor, and the processor realizes steps in the geological disaster risk double-control data processing method based on a data center as described in the embodiment one when executing the computer program.

[0048] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including, but not limited to, disk storage and optical storage, etc.) containing computer usable program code.

[0049] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be realized by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The device that realizes the functions specified in one flow or multiple flows and / or blocks. Figure 1 The device that realizes the functions specified in one flow or multiple flows and / or blocks.

[0050] Finally, the above only describes the preferred embodiments of the present application and is not used to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

[0051] The present application provides a geological disaster risk double-control data processing method and system based on a data center, which configures special rules in the field of geological disasters, integrates business systems, Internet of Things devices and external department heterogeneous data, unifies spatial coordinates and reconstructs a 15-bit hidden danger point coding system, maps threat objects to a standardized dictionary using a language large model, constructs a data physical model containing association rules and constraints, splits the core hidden danger library and the risk point library, divides the monitoring priority, realizes the three types of binding of hidden danger points-monitoring equipment, responsible persons and disposal processes based on administrative divisions, constructs a geological disaster type exclusive parameter library, fuses monitoring data to calculate risk indexes, triggers emergency tasks and closes the loop for feedback, adopts hierarchical API services and three-order data quality detection, realizes efficient management, forms a full-process closed loop of risk identification-management-feedback, and improves the intelligent level of geological disaster prevention and control.

Claims

1. A data center-based geological disaster risk double-control data processing method, characterized in that, Comprise: S1: configure data extraction rules special for geological disaster field, access heterogeneous data in data center from business systems, Internet of Things devices and external departments, the extraction rules are designed specially for different data source types; S2: convert spatial data into national geodetic coordinate system, reconfigure coding system for attribute fields of hidden danger points according to predetermined coding rules, map threat object classification to standardized dictionary, and build geological disaster data physical model, the predetermined coding rules are administrative division + hidden danger point + sequence number; S3: based on preset risk threshold, split original hidden danger point library into core hidden danger and risk point library, the core hidden danger library corresponds to high monitoring priority, the risk point library corresponds to regular patrol priority, the preset risk threshold includes threat population threshold, stability index threshold and sensitivity threshold; S4: based on administrative division coding system, establish three types of binding of hidden danger point-monitoring device, hidden danger point-responsible person, and risk event-disposal process; S5: build geological disaster type exclusive parameter library, fuse monitoring data and parameter library, calculate risk index, and generate emergency tasks based on risk index, push to binding responsible person, backfill disposal result to hidden danger point state field, form a closed loop of risk identification, control and feedback; based on each parameter, calculate score by weight coefficient and monitoring value, aggregate and calculate each parameter score by weighted summation algorithm to obtain the risk index; S6: adopt hierarchical API service mechanism, internal system directly connects database view, external departments obtain desensitized data based on API, and implement multi-stage data quality detection process, including preliminary screening, error correction and backflow stage. 2.The data center-based geological disaster risk double-control data processing method according to claim 1, wherein, The S1 further comprises: configure data query template through ETL tool, call external data based on API interface protocol, and define parsing rules of multiple formats; encapsulate extraction logic into reusable template, support parameterized configuration, and define data extraction process through visual arrangement platform; Business system data includes geological disaster hidden danger point basic information, historical patrol record and management engineering account; Internet of Things device data includes sensor monitoring data, video monitoring stream data and unmanned aerial vehicle aerial image; external department data includes meteorological warning, remote sensing image and social and economic data. 3.The data center-based geological disaster risk double-control data processing method according to claim 1, wherein, In the S2, the attribute fields of hidden danger points are reconfigured according to the predetermined coding rules, the threat object classification is mapped to the standardized dictionary, and the geological disaster data physical model is built, which comprises: build hierarchical coding system of administrative division + hidden danger point + sequence number, wherein the hierarchical coding system adopts fixed number of digital combination, including administrative division level code representing administrative region, hidden danger point classification code representing hidden danger type, and sequence code representing new sequence, assign unique sequence code when hidden danger point is added and update database; build standardized dictionary of threat objects based on geological disaster influence objects and sensitivity, the dictionary defines threat object attribute structure, including object name, classification code, sensitivity level and associated risk indicators; The original threat object description text is extracted and matched by the language large model engine, mapped to the standardized dictionary classification system, the missing threat object field text is implemented keyword extraction and semantic reasoning, the threat object inference result is generated, the dictionary is updated after artificial verification and confirmation, and the correction result is synchronized to the data platform through the interface; A geological disaster data physical model is constructed, including data storage structure design, data correlation rule definition, data integrity constraint setting and data access optimization strategy, wherein the data storage structure design includes field attribute definition, primary and foreign key relationship configuration, index strategy and data partition scheme, the data correlation rule definition supports one-to-many, many-to-one and many-to-many correlation modes, the data integrity constraint setting includes primary key uniqueness constraint, foreign key reference integrity constraint, field uniqueness constraint and value range constraint, and the data access optimization strategy includes high-frequency query field index optimization, core data cache mechanism and large data volume parallel processing mechanism. 4.The data center-based geological disaster risk double-control data processing method according to claim 1, wherein, In the S3, based on the preset risk threshold, the original hidden danger point library is divided into core hidden danger and risk point library, including: Set the threat population threshold as the population risk assessment parameter, calculate the stability index threshold based on the hidden danger point historical monitoring data as the stability assessment parameter, combine the administrative division + hidden danger point + sequence number rule in S2 to establish the spatial relationship mapping of hidden danger points and population dense areas, traffic hub sensitive areas, public facilities and economic key areas, and set the sensitivity threshold as the environmental sensitivity assessment parameter; The original hidden danger point data is preprocessed, the interpolation algorithm is used to complete the missing values of the threat population, stability index and sensitivity threshold fields, and the threat object attributes are mapped to the resident family property classification system based on the standardized dictionary in S2. According to the multi-dimensional evaluation parameters, the grading rules are established: when the hidden danger point meets any two of the threat population threshold, stability index threshold and sensitivity threshold, it is classified into the core hidden danger library and assigned a high monitoring priority label, and when the hidden danger point meets any one of the above three threshold conditions, it is classified into the risk point library and configured with a regular patrol cycle. 5.The data center-based geological disaster risk double-control data processing method according to claim 1, wherein, In the S4, based on the administrative division coding system, three types of bindings of hidden danger point-monitoring equipment, hidden danger point-responsible person and risk event-disposal process are established, including: The hidden danger point-monitoring equipment binding is matched by grid location and equipment type, the hidden danger point-responsible person binding is the group measurement and group prevention staff in the grid responsibility area, and the risk event-disposal process binding is the association of disaster and risk data and emergency tasks; The geographical distance between the hidden danger point and the monitoring equipment is calculated, the monitoring equipment is classified based on the geographical distance and the hidden danger point type and threat object attribute, the equipment-hidden danger point matching rule is established, and when the monitoring equipment state is abnormal or the hidden danger point risk level changes, the equipment binding relationship is reconfigured, wherein the monitoring equipment state abnormality includes equipment failure and data anomaly, and the hidden danger point risk level change includes migration from the risk point library to the core hidden danger library; Based on the administrative division coding rules described in S2, the hidden danger points are mapped to the corresponding responsibility grid area, and a hierarchical binding mechanism of core hidden danger points and full-time responsible persons and risk points and part-time responsible persons is established. When the hidden danger points migrate between warehouses, the type of the bound responsible person is adjusted. When the administrative division is adjusted, the matching and binding of the responsible person are re-executed. According to the matching of the disaster risk type and the threat level, the corresponding emergency disposal process template is matched to realize the association of the risk event and the disposal process. 6.The data center-based geological disaster risk double-control data processing method according to claim 1, wherein, The S5 constructs a geological disaster type exclusive parameter library, and calculates a risk index, including: A parameter library exclusive to different geological disaster types is constructed, which contains monitoring index parameters related to the geological disaster type. The weight coefficients of each parameter are determined through an expert review meeting, and the parameter threshold benchmark of a typical disaster event is extracted based on historical disaster risk data; The monitoring data is associated with the weight rules in the parameter library, and the score of each parameter is calculated according to the weight coefficient and the monitoring value. The parameter scores are aggregated by a weighted summation algorithm to generate a final risk index, wherein the final risk index calculation formula is: final risk index = Σ (monitoring value of parameter i / threshold value of parameter i × weight of parameter i); According to the final risk index value, different risk event levels are marked.

7. The data center-based geological disaster risk double-control data processing method as claimed in claim 1, wherein, The S6 implements a multi-stage data quality detection process, including screening, error correction and backflow stages, including: The screening detection stage uses a rule engine to detect coordinate out-of-bounds and field missing. In the error correction stage, similar positions and attributes are completed through matching. In the backflow stage, the corrected data is updated to the data platform after being confirmed by artificial confirmation; Based on the unified geographic coordinate system described in S2, the geographic validity of the hidden danger point coordinates is verified. The out-of-bounds data is marked as a to-be-verified state. Through field integrity analysis, the threat population and threat object attribute fields of the hidden danger points are scanned. When the field missing rate exceeds the preset threshold, an alarm mechanism is triggered and a log record is generated; The hidden danger points with out-of-bounds coordinates and missing attributes are processed. Through GIS tools and administrative grid coding, the nearest correct spatial position is located in combination with the unique identification coding rule. Based on the geological disaster type exclusive parameter library described in S5, the missing threat object field is completed through rule inference.

8. The geological disaster risk double-control data processing system based on a data middle platform, characterized in that, The geological disaster risk double-control data processing system based on the data platform includes a data acquisition module, a data aggregation module, a data conversion component module, a data storage module, a data update module: The data acquisition module is configured with a multi-source heterogeneous data access interface for acquiring business system data, Internet of Things monitoring equipment data and external department shared data. The rule engine is integrated to perform preliminary quality screening on the collected data and execute data format standardization conversion; The data aggregation module integrates heterogeneous data source adapters and is configured with a special data extraction rule library for the geological disaster field. It supports customizing data extraction, conversion and loading processes through a visual programming interface. The data conversion component module comprises an encoding system generation unit, a threat object mapping unit, a data relationship construction unit and a parameter library management unit, the encoding system generation unit constructs hierarchical coding based on the administrative division-hazard point-sequence number rule, the threat object mapping unit maps the original threat object to the standardized classification system, the data relationship construction unit configures 1:N, N:1 and many-to-many data association rules, and the parameter library management unit constructs a dedicated parameter library for different geological disaster types; The data storage module is configured with a data physical model definition unit and a hazard grading storage unit, the data physical model definition unit defines data storage structure, data association rules and data integrity constraints, and the hazard grading storage unit splits hazard point data based on risk threshold and stores it in the core hazard library and the risk point library; The data storage module adopts a heterogeneous storage architecture, the core hazard library data is stored in a high-performance memory database, the risk point library data is stored in a distributed columnar database, and the system implements a hierarchical access control strategy; The data update module comprises a binding relationship maintenance unit, an emergency task generation unit and a data quality guarantee unit, the binding relationship maintenance unit adjusts the binding relationship between monitoring equipment and responsible persons according to the risk level change, the emergency task generation unit triggers the emergency disposal process based on the risk event, and the data quality guarantee unit implements a three-stage data quality control mechanism of preliminary screening-error correction-backflow.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps in the geological disaster risk double-control data processing method based on the data middle station according to any one of claims 1-7.

10. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to realize the steps in the geological disaster risk double-control data processing method based on the data middle station according to any one of claims 1-7.

Images