An industrial park safe operation platform based on Beidou space-time big data technology

Abstract

The application discloses a kind of industrial park safe operation platform based on big data technology of Beidou space-time, the platform includes: industrial park safe operation platform and integrated multifunctional perusal display screen management system;The industrial park safe operation platform is based on Beidou grid data base and space-time grid framework, and is fused from space-time dimension and object dimension, including four intelligent modules of information acquisition, information calculation, risk early warning and safety command.The application breaks data barrier, realizes the global fusion of multi-source data, avoids the control blind area caused by data isolation, realizes fine, visual global control, promotes the change of control mode from passive disposal to active prevention, makes the park safety operation work more normative and scientific, further consolidates management responsibility, realizes whole-process closed-loop management, standardizes operation and maintenance management process, realizes whole-process closed-loop management;Adapt to the whole scene demand of park, strong expansibility.

Description

Technical Field

[0001] This invention relates to the field of industrial park safety management, and in particular to an industrial park safety operation platform based on BeiDou spatiotemporal big data technology. Background Technology

[0002] With the development of information technology, a large number of IoT devices and tools have been deployed and applied in all business links of the supply chain, and various information platforms and systems have been developed and launched, resulting in a large amount of basic data. Due to different management entities, this basic data cannot be discovered or exchanged at the overall supply chain level across professions, units, and platforms. Data cannot flow freely, and its value is greatly reduced. Industrial parks are home to high-risk enterprises and major hazard sources, with diverse, hidden, and easily spread risks. They are also densely populated with personnel and assets, making them prone to major accidents. Currently, enterprises suffer from severe data silos, lack high-precision spatiotemporal positioning capabilities, passive and lagging supervision, and difficulties in closing the responsibility loop. Traditional models cannot adapt to the refined safety requirements. The state mandates real-time monitoring of major hazard sources and closed-loop management of hidden dangers, and encourages technologies such as Beidou and big data to empower safety governance, making compliance an urgent need. The BeiDou system provides centimeter-level outdoor positioning and sub-meter-level indoor positioning, nanosecond-level timing, and short message communication without public network reliance. It is domestically developed and controllable, ensuring security and uninterrupted communication. Spatiotemporal big data technology enables spatial grid coding of industrial parks and the fusion of multi-source data (positioning, sensors, video, etc.). Combined with AI algorithms, it achieves risk identification, early warning, and emergency response, and with digital twins, it achieves full-domain visualization. With the steady growth of the BeiDou industry's output value and the widespread application of industrial-grade terminals, the "BeiDou + 5G + AI" integrated solution has been implemented in multiple industries and is ready for large-scale deployment. The security level of industrial parks needs to be improved, data barriers need to be broken down, and standardization, digitalization, and closed-loop management of security needs to be achieved. This will improve decision-making efficiency, enabling a shift from passive handling to proactive prevention, and from vague control to precise governance, reducing the accident rate. It also needs to meet national safety production and emergency management compliance requirements, reducing compliance risks. Furthermore, it needs to promote the deep integration of BeiDou with industrial security, assist in the digital transformation of industrial parks, and cultivate new business models. The technological accumulation is mature; only a breakthrough in research and development is needed.

[0003] CN118488384A discloses an industrial park monitoring and perception system based on the Beidou grid location code. This system relies on the Beidou grid coding to divide the park grid, constructs a dynamic and static attribute form for grid units, maps and binds the positions of monitoring and perception devices to the corresponding grids and configures acquisition rules; it collects audio-visual and environmental perception data in real time through various types of devices, which are uniformly analyzed and processed by the platform, dynamically updates the dynamic attributes of the grid and realizes the visualization display of全域 information. This solution breaks through the device data silos, realizes the linkage of monitoring and perception devices,弥补 the defects of the traditional system such as more manual operations, weak terrain combination, and insufficient data support, and improves the real-time and comprehensiveness of park monitoring. Although this invention proposes the linkage between grid coding and monitoring, it only stays at the basic monitoring application level, and can only perform data association and alarm through preset rules, unable to achieve the interconnection and sharing of data and emergency resources among enterprises in the park, unable to automatically discover the implicit spatio-temporal association relationships between devices, difficult to form a collaborative linkage ability, unable to achieve the synergy management effect of regional joint prevention and control, and even more unable to establish an evolution map of alarm events in the time dimension and space dimension, resulting in the problem that operation and maintenance personnel cannot quickly locate the root cause during an alarm storm.

[0004] CN118607738A discloses an emergency rescue solution combining remote sensing satellites and the Beidou grid location code. It divides the industrial park grid through the Beidou grid coding and builds a dynamic and static attribute data table, abstractly disassembles the elements of the emergency plan and realizes the grid visualization of the plan and emergency resources; when an incident occurs, it combines alarm information, remote sensing images and grid attribute data to judge the accident situation, matches and activates the corresponding emergency plan, plans the optimal passage route for the emergency team, and dynamically corrects the evacuation and refuge plan. However, this solution is specifically adapted to the military and police operation scenarios, with strong specificity, and unable to adapt to the actual management needs of industrial parks with a large number of people, high risks, dense property, insufficient emergency literacy and physical fitness of personnel, and easy to be disorderly and chaotic under emergencies.

[0005] It should be noted that the term "全域" in the original text is directly translated as "全域" here as it seems to be a specific term in the context. If there is a more appropriate English equivalent, it can be adjusted accordingly. Also, the "弥补" in the original text is translated as "弥补" first as it might be a specific term in the patent context. If a more accurate translation is required, it can be refined.CN103489326B discloses a vehicle positioning system based on spatiotemporal coding, including a passive vehicle-mounted terminal, an RFID tag reader, and a remote control center. The passive vehicle-mounted terminal is installed on the vehicle and contains a unique ID. It also includes a BeiDou module to obtain the current time and location's latitude and longitude in real time. The RFID tag reader is placed on the road. When a vehicle equipped with the passive vehicle-mounted terminal enters the effective reading range of the RFID reader, the reader reads the tag and its ID, along with the time and coordinates transmitted from the BeiDou satellite. The received time, longitude, and latitude are combined to form a spatiotemporal code. Every time interval T, the statistically obtained vehicle ID and spatiotemporal code are transmitted encrypted via a wireless network to the remote control center, which then decrypts and views the relevant information. This invention improves the accuracy of vehicle positioning, ensures communication quality through simple transmission content, reduces communication bandwidth usage, enhances data transmission security, and ensures communication quality. However, it also has limitations such as single application scenarios, shallow spatiotemporal applications, thin and isolated data dimensions, weak management and control capabilities, lack of collaborative sharing capabilities, poor architectural scalability, and inability to meet the needs of large-scale and integrated management and control in complex scenarios of industrial parks.

[0006] In summary, the current industrial park suffers from severe data silos among various enterprises and business systems, making it difficult to share data across units and platforms, resulting in low data integration and utilization rates. At the same time, the lack of high-precision spatiotemporal control capabilities leads to a passive and outdated safety supervision model, making it difficult to achieve closed-loop responsibility and refined governance.

[0007] Existing publicly available technical solutions have obvious limitations: some BeiDou grid-based systems only achieve basic monitoring and equipment linkage, and cannot explore the spatiotemporal correlation of multiple elements, making it difficult to support joint prevention and control in the park, and making it difficult to locate the root cause of alarm events; some emergency rescue solutions have poor scenario adaptability and are only applicable to military and police scenarios, which do not meet the actual management status of industrial parks with high risk, dense population and weak emergency foundation; other spatiotemporal positioning solutions are limited to single vehicle targets and cannot cover the security control needs of the entire park.

[0008] Currently, the national requirements for compliance with the monitoring of major hazard sources and the closed-loop management of potential hazards are becoming increasingly stringent. Leveraging BeiDou and spatiotemporal big data to empower industrial safety has become an inevitable trend. Therefore, it is urgent to overcome existing technological bottlenecks, break down data barriers, integrate spatiotemporal grid and knowledge graph capabilities, and construct a comprehensive, collaborative, proactive, and closed-loop industrial park safety management system. This will enable a shift in safety governance from passive response to proactive prevention and control, and from extensive management to precise control, meeting the needs of safety production compliance and the digital transformation of industrial parks. Summary of the Invention

[0009] This invention overcomes the shortcomings of existing technologies and provides an industrial park safety operation platform based on BeiDou spatiotemporal big data technology. It breaks through existing technological bottlenecks, eliminates data barriers, integrates spatiotemporal grid and knowledge graph capabilities, and constructs a park safety management and control system with full-domain collaboration, proactive early warning, and full-process closed loop. It realizes the transformation of safety governance from passive handling to proactive prevention and control, and from extensive management to precise control, meeting the needs of safety production compliance and park digital transformation.

[0010] To achieve the above objectives, the present invention adopts the following technical solution:

[0011] An industrial park security operation platform based on BeiDou spatiotemporal big data technology includes an industrial park security operation platform and an integrated multi-functional overview display screen management system; the industrial park security operation platform includes:

[0012] Spatiotemporal fusion coding module: used to divide the space within the park into grids and assign one code to each grid, assign one code to each device and person, and assign one code to each alarm event.

[0013] Data acquisition module: used to access alarm event data, video stream data, and device component data from the DCIM platform and video surveillance system;

[0014] Knowledge graph construction and reasoning module: used to associate the alarm event data with the one-bit-one-code, one-item-one-code and one-event-one-code, construct a dynamic semantic graph based on spatiotemporal constraints, and perform temporal evolution and spatial association reasoning on the dynamic semantic graph to generate a spatiotemporal discrete knowledge graph of alarm events.

[0015] Alarm analysis and prediction module: It is used to take the spatiotemporal discrete knowledge graph as input and predict the future state of entities through a graph neural network (GNN)-based model to achieve risk warning;

[0016] Safety Command Module: Based on the digital twin platform and combined with the results of the risk warning, it generates and manages maintenance work orders to achieve a closed loop of safety production management.

[0017] Preferably, the spatiotemporal grid coding system of the spatiotemporal fusion coding module divides the geographic space into multi-level grids, assigning a unique location code to each level of grid; for equipment objects, a spatial coding field is added to their production code or management code, so that their geographic location can be directly parsed, forming the one-item-one-code; for alarm events, the corresponding spatiotemporal code and object identity code are attached to form the one-event-one-code, and a digital raster map is constructed on this basis, and vectors are generated for all non-location code codes, forming a digital vector map attached to the digital raster map.

[0018] Preferably, a dynamic semantic graph based on spatiotemporal constraints is constructed, specifically including the changes in its nodes and relationships over time, and supporting the derivation of trajectory association based on the temporal continuity of adjacent grid overlays. By collecting and calculating information on people, equipment, flow, and fixed assets within the industrial park, and assigning attribute vector values ​​to vectors located on the digital vector map of the digital raster map, a vector digital elevation model is constructed. A primary location set code is generated using a certain number of adjacent location codes or location codes with the same attributes, serving as a primary overall adaptive microservice architecture route. Employing microservice architecture technology, each microservice can be deployed independently, and the microservices are loosely coupled. Similarly, several primary location set codes can generate intermediate location set codes, serving as an intermediate overall adaptive microservice architecture route, also employing microservice architecture technology. This layered architecture has good scalability and reusability, and application-level fault tolerance can be achieved through retry and circuit breaker mechanisms.

[0019] Preferably, space is divided using discrete grids, and time is segmented using discrete timestamps / time slices. The dynamic semantic graph under spatiotemporal constraints allows nodes / relationships to change with time and space, supporting temporal evolution and spatial association reasoning. Adjacent grids are combined with time to deduce trajectory associations. Based on spatiotemporal GNN, the future state of entities is predicted, and modeling is performed to establish a spatiotemporal discrete knowledge graph of alarm events.

[0020] Preferably, the alarm analysis and prediction module includes a layered structure of a data layer, a knowledge layer, a model layer, and an application layer, wherein:

[0021] The data layer acquires spatiotemporally discrete alarm data;

[0022] The knowledge layer sets up a spatiotemporally discrete alarm knowledge graph, including alarm events, devices, regions, buildings, time periods, fault types, spatiotemporal proximity, temporal sequence, causal relationships, topological subordination, concurrent alarms in the same region, and cascading triggers, transforming isolated alarms into a structured spatiotemporally related network.

[0023] The model layer includes an alarm analysis module and an alarm prediction module. The alarm analysis module performs similar clustering, correlation mining, causal tracing and storm noise reduction processing, while the alarm prediction module performs time series prediction and spatiotemporal prediction.

[0024] At the application layer, alarm noise reduction, root cause analysis, regional risk heat map, short-term alarm prediction, early fault warning, and operation and maintenance decision processing are performed.

[0025] Preferably, the twin system of the safety command module achieves virtual-real state synchronization through millimeter-level modeling and real-time data-driven operation, and carries the entire process of risk identification, early warning, source tracing, and emergency simulation. The enterprise safety production management twin system is a closed-loop platform based on digital twins, with a spatiotemporal knowledge graph as its central hub and AI algorithms as its core. It encompasses virtual-real mapping, real-time perception, risk prediction, intelligent handling, and final review and optimization. This platform completely solves the problems of passive response, information silos, reliance on experience, and difficulty in traceability. It accurately replicates all elements of the factory area, workshops, equipment, pipelines, personnel, and environment. Through millimeter-level modeling and real-time data-driven operation, the virtual-real state is completely synchronized. It integrates a spatiotemporal discrete knowledge graph and an alarm prediction model, carrying the entire process of risk identification, early warning, source tracing, emergency response, and training. AI drives risk simulation, trend prediction, and the generation of optimal handling solutions.

[0026] Preferably, the integrated multi-functional overview display screen management system is used for managing equipment status, park system overview, maintenance work orders, maintenance history, and spatiotemporal grid knowledge graph. Equipment status management includes real-time equipment operation status monitoring, equipment anomaly and alarm management, equipment health status assessment, equipment lifecycle ledger management, maintenance management, fault diagnosis and root cause analysis, equipment operation statistics and analysis, and safety compliance and risk control. The park system overview management includes a comprehensive situational overview, unified management of equipment and facilities, environmental and safety monitoring management, comprehensive personnel and vehicle control, alarm and emergency management, video fusion and three-dimensional prevention and control, inspection and hazard closed-loop management, energy consumption and resource intensive management, data statistics and decision analysis, and integrated system operation and maintenance management; using a 3D park map... Using real-view maps as a medium, the system replicates the core areas of the park, including zones, factories, warehouses, hazardous chemical storage areas, major hazard sources, fire exits, and security points, at a 1:1 scale. It supports panoramic roaming and zoned viewing, and presents core safety indicators in real time. It categorizes current alarm counts, online status of major hazard sources, hazard rectification rate, inspection completion rate, emergency resource availability, and personnel on-duty status by level, intuitively reflecting the overall safety level of the park. It also colors-codes each area, equipment, and point in the park according to safety level, highlighting high-risk areas and abnormal points, and achieving control over major hazard sources and high-risk areas, as well as control over all types of alarms and anomalies.

[0027] Preferred maintenance work order management: For equipment and facility failures, anomalies, maintenance and repair requests, standardized work orders are generated as the core carrier for maintenance handling and resource allocation, realizing closed-loop fault handling, clear responsibility, and traceable process, ensuring efficient and standardized handling of problems;

[0028] Maintain historical management: Collect and store historical data on completed work orders, maintenance, and fault rectification to form a full lifecycle operation and maintenance ledger, accumulate static data and operation and maintenance experience, and support risk control, cost accounting, and optimization and upgrades;

[0029] The two work together to build a full-process operation and maintenance management system that includes prevention, control, and post-event review.

[0030] The maintenance work order management system is a standardized business document created when equipment, systems, sites, or facilities experience malfunctions, anomalies, expired maintenance, or require repair. It serves as the core carrier for maintenance services, problem handling, and resource allocation, enabling closed-loop management of faults, assigning responsibility to individuals, and ensuring traceable processes. It is also a dynamic execution carrier, ensuring rapid handling of current issues and standardized processes. The maintenance history management system is a unified collection, storage, organization, and query management of all completed and closed maintenance work orders, maintenance records, repair operations, fault rectification, and anomaly handling data, forming a full lifecycle maintenance ledger. This provides data support for equipment operation and maintenance, risk control, cost accounting, and optimization and upgrades. The maintenance history is a static data accumulation, accumulating operation and maintenance experience and tracing historical issues. The two-way linkage enables full-process operation and maintenance management, including pre-event prevention, in-event control, and post-event review.

[0031] The preferred approach is to integrate spatiotemporal grid knowledge graph management with spatiotemporal grids and knowledge graphs: using geographic grids as a base, overlaying time, hierarchy, and spatial scope to achieve refined grid-based control of regions in multiple dimensions; relying on knowledge graphs to integrate people, objects, and potential risks, realizing the relationships between element entities, attributes, and associations, and conducting integrated management through system, entity, and relationship modeling, data fusion, graph visualization, rule semantics, and version maintenance, breaking down data barriers and achieving full-dimensional integration and linkage of spatial, temporal, and business elements. The spatiotemporal grid knowledge graph management includes a spatiotemporal grid and a knowledge graph. The spatiotemporal grid is based on a geospatial grid, integrating time dimension, regional hierarchy, and spatial range to achieve refined and gridded division and management of the entire region, carrying basic spatial information such as spatial location, boundary range, hierarchical affiliation, and temporal changes. The knowledge graph is a structured knowledge base that integrates entities, attributes, relationships, events, and rules, connecting all elements of people, places, things, events, organizations, facilities, and hidden dangers. It combines spatial grid, time dimension, business entities, and relationships to achieve integrated management of spatial demarcation, time traceability, element association, and relationship visualization. The management content includes spatiotemporal grid system management, knowledge graph entity management, relationship modeling management, spatiotemporal data fusion governance, graph visualization and retrieval management, rule and semantic management, historical version and operation and maintenance management, breaking down data silos and achieving full-dimensional integrated management of gridded space, time, and business elements.

[0032] Preferably, the DCIM platform is specifically designed for safety management in industrial parks. It integrates multi-dimensional capabilities such as power environment, assets, energy consumption, security, operation and maintenance work orders, spatiotemporal grids, and knowledge graphs. Based on the characteristics of industrial parks, including multi-business formats, high compliance, complex equipment, and cross-network deployment, it constructs a unified overview, integrated monitoring, closed-loop process, and intelligent early warning capabilities. This enables integrated management and control of safety, operation and maintenance, emergency response, energy consumption, and assets. The DCIM platform is a DCIM platform for safety management in industrial parks. Its core is to integrate power environment, assets, energy consumption, security, operation and maintenance work orders, spatiotemporal grids, and knowledge graphs to achieve a unified overview, integrated monitoring, closed-loop process, and intelligent early warning. Considering the characteristics of large spaces, multi-business factory warehouses, comprehensive buildings, hazardous chemical areas, strong safety compliance, diverse equipment types, and cross-regional networks in industrial parks, it strengthens the comprehensive management and control of safety, operation and maintenance, emergency response, energy consumption, and assets.

[0033] Preferably, the integrated multi-functional panoramic display screen management system includes a data acquisition and coding subsystem, a grid data table subsystem, and a grid data engine subsystem;

[0034] The data acquisition and coding subsystem includes a basic data grid coding subdivision module, a basic data grid coding module, a coverage coding module, an administrative division coding module, and a multi-source data coding module;

[0035] The grid data large table subsystem includes a data source management module, an index configuration management module, an index management module, a coding task management module, a grid data query module, and a grid data statistics module;

[0036] The grid data engine subsystem includes an encoding registration module and an encoding parsing module.

[0037] Compared with the prior art, the beneficial effects of the present invention are:

[0038] 1. Break down data barriers and achieve full-domain integration of multi-source data to avoid blind spots in management caused by data silos.

[0039] 2. Achieve refined and visualized full-area management and control, and improve the level of refined management and control in the park.

[0040] 3. Promote the transformation of the management and control model from passive response to proactive prevention, make the park's safety operation and maintenance work more standardized and scientific, further strengthen management responsibilities, and achieve closed-loop management throughout the entire process.

[0041] 4. Standardize operation and maintenance management processes, improve the scientific nature of management, further solidify management responsibilities, and achieve closed-loop management throughout the entire process.

[0042] 5. It adapts to the needs of all scenarios in the park, has strong scalability, and can be compatible with the access of massive sensing devices, supporting the digital transformation of the park. Attached Figure Description

[0043] To more clearly describe the inventive objectives, technical solutions, and advantages of the specific embodiments of this invention, the solutions in the specific embodiments will be described in detail below with reference to the accompanying drawings. The specific technical solutions involved in the following embodiments are merely for the purpose of clearly and completely describing the innovative technical solutions of this invention. They are only a part of the specific implementation methods that this invention can adopt, not all embodiments, and should not be construed as limiting the innovative solutions of this invention. Any solution that adopts the same inventive concept as this invention should be included within the protection scope of this invention.

[0044] Figure 1 This is a diagram illustrating the data distribution and logical architecture of the present invention.

[0045] Figure 2 This is a marginal overall flow diagram of data processing in this invention;

[0046] Figure 3 This is a design diagram of the interface for filtering, querying, and resetting equipment data in the park operation security one-screen overview management system of the present invention;

[0047] Figure 4 This invention provides an interface design diagram for a single-screen overview of escape route planning for park operation safety.

[0048] Figure 5 This invention provides an interface design diagram for a one-screen overview of security incidents in the park, enabling reasoning, tracing, and correlation analysis.

[0049] Figure 6 This is the interface design diagram for the spatiotemporal grid knowledge graph of safety management in the park operation safety of this invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0051] The technical solutions in the embodiments of the present invention will now be clearly and completely described in conjunction with the accompanying drawings.

[0052] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of the components in a specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly. Furthermore, descriptions involving "preferred," "second-best," etc., in this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "preferred" or "second-best" may explicitly or implicitly include at least one of those features.

[0053] An industrial park safety operation platform based on BeiDou spatiotemporal big data technology includes an industrial park safety operation platform and an integrated multi-functional overview display management system. It unifies and integrates different types of data distributed across various systems through spatiotemporal fusion coding, assigning codes to equipment, measurement points, and alarm events in each system according to spatiotemporal fusion coding elements to establish a spatiotemporal data mapping relationship. Through knowledge graphs and analytical prediction models, it improves the enterprise's safe operation and production efficiency while establishing a digital twin system based on spatiotemporal fusion coding.

[0054] This case focuses on events, establishing a unified data standard and a unified spatiotemporal element 3D visualization system. On one hand, it achieves full-element spatiotemporal discretization based on events, establishing an internal security operation knowledge graph and building an analysis and prediction model; on the other hand, based on video grid partitioning and video fusion technology, it establishes rapid video retrieval and retrieval based on events; and further ensures operational safety by extracting events in reverse through video grid partitioning.

[0055] The project will focus on building standards, multi-level data aggregation capabilities, and grid-based big data analysis service capabilities to create a model for data sharing and exchange, and standardized data services and operations. The main construction content includes: a unified spatiotemporal data overview based on events; event extraction through video grid segmentation, and video retrieval based on events, inspections, emergency drills, and maintenance plans. Based on the analysis and prediction of integrated IoT data and video grid segmentation, results will be generated and integrated with training materials, security drills, or maintenance inspections.

[0056] This project establishes an event-based spatiotemporal data overview system. It builds analytical and predictive models through spatiotemporal fusion computing and integrates the results with training materials, security drills, and maintenance inspections to improve the efficiency of safety production management. The main technical objective is to provide visual representation.

[0057] A three-dimensional grid model is built using existing GIS data. The grid model displays the equipment, measurement point data, affected spatial range, and other equipment corresponding to the alarm event.

[0058] Spatiotemporal fusion encoding is performed on the equipment data and measurement point data of the video surveillance system, DCIM system, and other systems.

[0059] This case involves spatiotemporal fusion coding based on a 3D model base map to form a unique spatiotemporal fusion code (one code per bit), which includes the spatiotemporal distribution and attribute information of data such as park buildings, floors, and unit information.

[0060] A full-element 3D display based on a spatiotemporal fusion coding grid (one item, one code), including: spatiotemporal distribution and attribute information of power, electricity, and video surveillance equipment.

[0061] The 3D display based on spatiotemporal fusion event coding (one code per event) includes: the event level, event type, device name, spatial location, and measurement point details corresponding to the alarm event.

[0062] This project, based on a spatiotemporal grid framework, integrates encoding from both spatiotemporal and object dimensions. At the overall level, it achieves unified organization and efficient retrieval and computation of big data, while forming continuous spatiotemporal data chains at the individual object level. It serves as infrastructure to support big data analysis and utilization in various fields such as digital government, digital economy, and digital life. The main construction content includes the following: the center's construction will closely revolve around standardization, multi-level data aggregation capabilities, and grid-based big data analysis service capabilities, creating a model for data sharing and exchange, standardized data services, and operations. The core of the center is to achieve the aggregation and fusion of various types of data (including heat, electricity, and video) through spatiotemporal fusion data construction, realizing its infrastructure construction goals and fully leveraging the center's value in the region. The system construction content mainly includes: establishing a standardized spatiotemporal fusion encoding grid organizational framework based on a grid data foundation; service categories include: query services for grid-coded data, block data services, visualization services based on digital twin technology, and intelligent analysis services based on spatiotemporal discrete alarm events.

[0063] Basic data retrieval includes alarm event data, video stream data, equipment component data, and basic 3D model data from the DCIM platform and video surveillance system; it uses a spatiotemporal fusion coding standard to encode each bit with a unique code, each event with a unique code, and each item with a unique code; and it establishes a spatiotemporal discrete knowledge graph of alarm events through a spatiotemporal discrete approach to complete the analysis and prediction model of alarm events, thereby realizing a twin system for enterprise safety production management.

[0064] like Figure 1 As shown, the data architecture, from bottom to top, includes data aggregation and access, data resource system, and data application system.

[0065] The technical architecture in this case, the spatiotemporal fusion grid coding indexing system, adaptively marks the data coverage area with multi-scale spatial grids, assigns a unique code to each grid, and, based on the grid coding, realizes the grid coding association of multi-source geographic information data without overturning or rebuilding the existing system. Non-geographic information data can also be effectively associated as attribute data of geographic information data through grid coding.

[0066] In this case, a unique digit-time encoding (spatial-temporal encoding) is used to encode spatial locations (equivalent to a vacuum grid), ensuring the encoding's uniqueness. This can be done according to relevant standard rules, but requires transformation and extension based on application scenarios. Furthermore, standard city addresses can be added to the spatial encoding field, allowing for direct parsing of geographical locations from standard addresses and facilitating true address standardization through this procedural encoding process.

[0067] The one-item-one-code (object identity code) in this case can be simply divided into two types: For above-ground attachments (and real estate) such as houses, buildings, and urban components, a spatial code field is added to the existing departmental management code, which also allows the geographical location to be directly parsed, and the code is unique; For vehicles such as automobiles and goods in circulation, whose geographical location is constantly changing, only spatial and temporal codes need to be added to their production codes (similar to the rules for human ID card coding), and the code is unique, while the temporal and spatial changes in the process are recorded synchronously (human ID cards can be included in the scope of object identity coding).

[0068] In this case, the event type code is a unique identifier for a specific type of event, and it is unique across categories. To better uncover correlation patterns, it is necessary to attach spatiotemporal codes and object identity codes (if applicable) to the event codes.

[0069] The status-based coding (big data application) in this case, based on the principles of one code per person, one code per object, and one code per event, can determine and assign codes based on information such as time and space, object, and event. Taking the health code as an example, firstly, confirmed cases are assigned a red code based on their test report information; secondly, close contacts are assigned a yellow code based on their temporal and spatial intersection with confirmed cases (the judgment threshold can be adjusted and refined). It is evident that the health code is essentially a status code, belonging to big data applications based on time and space codes, object codes (people), and event codes (virus infection).

[0070] This case follows a hierarchical logic of one code per entity, one code per item, and one code per event, enabling the operation of a comprehensive security overview system that penetrates all aspects of enterprise security management. It forms a unified digital twin model through a unified big data framework. Unlike traditional visualization models, this integrated digital twin model for operational security, based on spatiotemporal fusion coding, achieves full-time, full-element, and full-process coverage of data; furthermore, the model supports data queryability, displayability, computationability, and serviceability.

[0071] This case utilizes a deep fusion approach combining spatiotemporal fusion coding and vectorized data. The integer coding design of spatiotemporal fusion coding significantly simplifies the complexity of identifying, representing, and calculating location information, offering unique advantages in information processing speed, information indexing efficiency, information exchange and integration, and the expansion and enhancement of navigation and positioning capabilities. More importantly, it not only effectively compensates for the shortcomings and limitations of traditional latitude and longitude technology systems but also achieves perfect compatibility with latitude and longitude, serving as an excellent complement and improvement to these systems. Furthermore, it effectively solves the problem of organizing massive, multi-source, and heterogeneous spatial information and can be easily converted to information systems under various existing technological systems. This provides an ideal solution to the long-standing challenge of cross-industry and cross-departmental information sharing and efficient services in information technology development.

[0072] This project employs a microservices architecture, where each microservice can be deployed independently and is loosely coupled, offering greater flexibility. Each microservice represents a small business function, focusing solely on completing that function and performing it well, resulting in excellent scalability and reusability. In the event of a system failure, only the problematic service requires code modifications and service restarts; other services can achieve application-level fault tolerance through mechanisms such as retries and circuit breakers.

[0073] This case is based on a flattened network architecture combined with SDN. The flattened network architecture is divided into a service control layer and a user access layer. The service control layer consists of core layer devices, providing network user access control and service isolation functions. The user access layer consists of aggregation and access layer devices; aggregation and access directly use VLAN transparent transmission, providing only basic user access functions and security isolation functions. This structure clarifies device functions and network layers, reduces the complexity of services and management / maintenance, and facilitates network management and maintenance. SDN is a new innovative network architecture and a method of network virtualization. Its core technology, OpenFlow, separates the control plane and data plane of network devices. This separation of hardware control and data forwarding facilitates centralized network control, enabling the control layer to obtain global information on network resources and perform global resource allocation and optimization based on service requirements.

[0074] like Figure 2 As shown, the data flow design in this case is as follows: Data access and processing flow: Data from the DCIM platform, video surveillance data, access control system data, 3D model data, industry data, social enterprise data, and other relevant data are accessed through the security boundary. Simultaneously, for the various data resources accessed, they are first processed according to the system's data specifications to form various basic databases; secondly, various basic data are fused using spatiotemporal information to form various thematic databases and special topic databases; finally, after spatiotemporal coding of various data resources through the fusion analysis service platform, a big data product supermarket is formed. Based on the data product supermarket, data services are provided to various users such as enterprises, government sectors, special industries, social and third-party institutions, and the general public.

[0075] In this case, the integrated multi-functional overview display screen management system has the following structure:

[0076]

[0077]

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084] The event early warning application scenarios in this case include the following:

[0085]

[0086]

[0087]

[0088]

[0089] The following description uses the design interface of the Park Operation Safety One-Screen Overview Management System in actual application as an example.

[0090] Figure 3This is a design diagram of the interface for filtering, querying, and resetting equipment data in a comprehensive park operation security management system for practical applications. The equipment data filtering / querying / resetting function aims to provide a flexible data retrieval mechanism, enabling users to filter and query equipment data based on various equipment attributes (such as building, floor, room, equipment type, and equipment ID).

[0091] Figure 4 This is a user interface design for a one-screen overview of escape route planning in a practical application for park operation safety. By inputting a device ID, the system uses intelligent recognition to generate the shortest escape route and displays the result as a path on a map.

[0092] Figure 5 This is a user interface design diagram for a one-screen overview of security incident reasoning, tracing, and correlation analysis in practical applications of park operation security. Users configure the rule flow in the rule configuration, the system processes the data, generates the corresponding learning graph, and displays it on the screen.

[0093] Figure 6 This is a user interface design for a comprehensive overview of a spatiotemporal grid knowledge graph for park operation safety in practical applications. The grid coverage is displayed by overlaying a grid model onto a 3D park map, and different grids can be switched by clicking on different buildings.

Claims

1. An industrial park security operation platform based on BeiDou spatiotemporal big data technology, characterized in that: This includes an industrial park safety operation platform and an integrated multi-functional overview display screen management system; The industrial park's security operation platform includes: Spatiotemporal fusion coding module: used to divide the space within the park into grids and assign one code to each grid, assign one code to each device and person, and assign one code to each alarm event. Data acquisition module: used to access alarm event data, video stream data, and device component data from the DCIM platform and video surveillance system; Knowledge graph construction and reasoning module: used to associate the alarm event data with the one-bit-one-code, one-item-one-code and one-event-one-code, construct a dynamic semantic graph based on spatiotemporal constraints, and perform temporal evolution and spatial association reasoning on the dynamic semantic graph to generate a spatiotemporal discrete knowledge graph of alarm events. Alarm analysis and prediction module: It is used to take the spatiotemporal discrete knowledge graph as input and predict the future state of entities through a graph neural network (GNN)-based model to achieve risk warning; Safety Command Module: Based on the digital twin platform and combined with the results of the risk warning, it generates and manages maintenance work orders to achieve a closed loop of safety production management.

2. The industrial park security operation platform based on BeiDou spatiotemporal big data technology according to claim 1, characterized in that: The spatiotemporal grid coding system of the spatiotemporal fusion coding module divides the geographic space into multi-level grids, with each level of grid assigned a unique location code; For equipment objects, a spatial code field is added to their production code or management code so that their geographical location can be directly parsed, forming the one-item-one-code; for alarm events, the corresponding spatiotemporal code and object identity code are attached to form the one-event-one-code.

3. The industrial park security operation platform based on BeiDou spatiotemporal big data technology according to claim 1, characterized in that: The project constructs a dynamic semantic graph based on spatiotemporal constraints, specifically including the changes in its nodes and relationships over time, and supports the derivation of trajectory association based on the temporal continuity of adjacent grid overlays.

4. The industrial park security operation platform based on BeiDou spatiotemporal big data technology according to claim 1, characterized in that: The spatiotemporal discrete knowledge graph of alarm events is obtained by dividing space into discrete grids and cutting time into discrete timestamps / time slices. It is a dynamic semantic graph under spatiotemporal constraints, where nodes / relationships change with time and space, supporting temporal evolution and spatial association reasoning. Adjacent grids are combined with time to deduce trajectory associations. The model is based on spatiotemporal GNN to predict the future state of entities.

5. The industrial park security operation platform based on BeiDou spatiotemporal big data technology according to claim 1, characterized in that: The alarm analysis and prediction module includes a layered structure of data layer, knowledge layer, model layer and application layer.

6. The industrial park security operation platform based on BeiDou spatiotemporal big data technology according to claim 1, characterized in that: The safety command module includes a twin system, which achieves virtual-real state synchronization through millimeter-level modeling and real-time data-driven operation, and carries out the entire process of risk identification, early warning, source tracing, and emergency simulation.

7. The industrial park security operation platform based on BeiDou spatiotemporal big data technology according to claim 1, characterized in that: The integrated multi-functional overview display screen management system is used for managing equipment status, park system overview, maintenance work orders, maintenance history, and spatiotemporal grid knowledge graph.

8. The industrial park security operation platform based on BeiDou spatiotemporal big data technology according to claim 7, characterized in that: Maintenance work order management: Standardized work orders are generated for equipment and facility malfunctions, anomalies, maintenance, and repair requests. These work orders serve as the core carrier for maintenance handling and resource allocation, achieving closed-loop fault management, clear responsibility, and traceable processes. Maintenance history management: Summarize and store historical data on completed work orders, maintenance, and fault rectification to form a full lifecycle operation and maintenance ledger; The two work together to build a full-process operation and maintenance management system that includes prevention, control, and post-event review.

9. The industrial park security operation platform based on BeiDou spatiotemporal big data technology according to claim 7, characterized in that: Spatiotemporal grid knowledge graph management integrates spatiotemporal grids and knowledge graphs: based on geographic grids, it overlays time, hierarchy, and spatial scope to achieve refined grid-based control of regions in multiple dimensions; relying on knowledge graphs, it integrates people, objects, and potential risks to realize the relationships between element entities, attributes, and associations. Through system, entity, and relationship modeling, data fusion, graph visualization, rule semantics, and version maintenance, it achieves integrated management and realizes the full-dimensional integration and linkage of spatial, temporal, and business elements.

10. The industrial park security operation platform based on BeiDou spatiotemporal big data technology according to claim 1, characterized in that: The integrated multi-functional panoramic display screen management system includes a data acquisition and coding subsystem, a grid data table subsystem, and a grid data engine subsystem; The data acquisition and coding subsystem includes a basic data grid coding subdivision module, a basic data grid coding module, a coverage coding module, an administrative division coding module, and a multi-source data coding module; The grid data large table subsystem includes a data source management module, an index configuration management module, an index management module, a coding task management module, a grid data query module, and a grid data statistics module; The grid data engine subsystem includes an encoding registration module and an encoding parsing module.

Images