Digital economic park operation management system based on big data analysis

Abstract

The invention discloses a digital economic park operation management system based on big data analysis, particularly relates to the technical field of smart park management, and comprises a data perception and fusion module, a dynamic knowledge graph engine module, a multi-objective optimization decision module and a smart application service module which work cooperatively in sequence. According to the method, a dynamic knowledge graph is constructed, a weight time sequence evolution mechanism is configured, multi-source data is integrated to eliminate information islands, industrial association degree indexes are quantified based on a graph topological structure, a multi-target optimization model is established, scientific configuration and prospective planning of park space and resources are realized, and the enterprise digital twinborn model is combined, so that the industrial association degree indexes are optimized. According to the system, active services such as accurate policy matching, intelligent demand recommendation and industrial coupling opportunity discovery are provided, the system realizes the transformation of park operation from decentralized management to global collaborative optimization and from passive response to active intelligent enabling, and the integrated management efficiency and industrial ecological aggregation activity of the park are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of smart park management technology, and more specifically, to a digital economy park operation and management system based on big data analysis. Background Technology

[0002] With the rapid development of the digital economy, various digital economy parks have become important carriers for industrial agglomeration and innovation incubation.

[0003] Traditional park operation and management systems often focus on basic property management, enterprise information registration, and simple data statistics, which have obvious limitations. First, the system's data source is singular, mostly isolated static data, which cannot integrate multi-source heterogeneous data such as IoT, government affairs, and enterprise dynamics, forming data silos. Second, the management approach is passive and lagging behind, lacking insight into the deep relationships within the park's industrial ecosystem, enterprise relationships, and resource efficiency, and thus unable to provide forward-looking decision support. Third, the service model is extensive and generalized, making it difficult to provide precise policy matching and resource docking services based on the personalized characteristics and real-time needs of enterprises.

[0004] While some existing solutions attempt to incorporate data analysis, they are mostly limited to single-dimensional report display or service push based on simple rules, failing to build a smart operation and management system capable of dynamic perception, intelligent analysis, collaborative optimization, and proactive empowerment.

[0005] Therefore, how to deeply integrate multi-source data, construct a dynamic model that reflects the complex relationships within the park, and on this basis achieve global optimization of resources and precise delivery of services has become a pressing technical problem to be solved in the intelligent upgrading of digital economy parks. In view of this, the present invention provides a digital economy park operation and management system based on big data analysis. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a digital economy park operation and management system based on big data analysis to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a digital economy park operation and management system based on big data analysis, comprising the following core modules that work collaboratively:

[0008] The data sensing and fusion module is used to collect and fuse multi-source heterogeneous data within the park. The multi-source heterogeneous data includes at least IoT sensing data, enterprise operation data, government-related data, and park resource data.

[0009] The dynamic knowledge graph engine module is connected to the data perception and fusion module. It is used to build and continuously update the dynamic knowledge graph of the park based on the fused data. The dynamic knowledge graph serves as a unified data model and relation index basis for subsequent optimization decisions. The entities in the knowledge graph include at least enterprise entities, resource entities, and policy entities. The relation weights between entities change dynamically according to the time of related events.

[0010] The multi-objective optimization decision module is connected to the dynamic knowledge graph engine module. It is used to establish a multi-objective optimization model based on the dynamic knowledge graph in real time, with the goal of improving the overall efficiency of the park. The multi-objectives include at least economic output objectives, industrial synergy objectives and resource efficiency objectives, and are equipped with corresponding quantifiable evaluation indicators.

[0011] The intelligent application service module connects to the dynamic knowledge graph engine module and the multi-objective optimization decision module. It is used to respond to external instructions or events, collaboratively call the calculation results of the dynamic knowledge graph and the multi-objective optimization model, and generate and output optimization solutions for resource allocation, enterprise services or industry guidance.

[0012] Preferably, the dynamic knowledge graph engine module includes:

[0013] The entity relationship extraction unit is used to extract entities, attributes, and relationships between entities from multi-source heterogeneous data.

[0014] The weight dynamic management unit is used to assign initial weights to the relationships and configure the weight change function. The weight change function is configured such that for the relationship edge representing the dynamic relationship between entities, its weight increases with the recent occurrence of the relationship event and decreases with the passage of time, so as to realize the timely representation of the ecological evolution of the park by the graph.

[0015] Preferably, the objective function constructed in the multi-objective optimization decision module is:

[0016]

[0017] in, , and This is a configurable balance coefficient;

[0018] The evaluation index corresponding to the industrial synergy target is the industrial relevance index. Its calculation is based on the connection strength and network density formed between enterprise entities in the dynamic knowledge graph through technology chains, supply chains or talent chains, and is quantitatively calculated based on the topological structure of the dynamic knowledge graph.

[0019] Preferably, when performing spatial resource allocation decisions, the multi-objective optimization decision module is specifically configured to: use the average correlation degree between candidate enterprise entities and enterprise clusters within a preset range surrounding the proposed allocation space in the knowledge graph as the industry correlation degree index in the objective function. The core computational factors are optimized to ensure that spatial allocation satisfies physical constraints while enhancing the aggregation of the industrial ecosystem.

[0020] Preferably, it further includes:

[0021] The enterprise digital twin construction module, connected to the data perception and fusion module, is used to build and maintain digital twin models for each enterprise in the park based on the dynamic data flow of individual enterprises. The digital twin model includes at least an innovation sub-model for evaluating innovation capabilities and a health sub-model for evaluating operational stability. The digital twin model forms the basis for conducting micro-profiles and macro-correlation analysis of enterprises.

[0022] The intelligent application service module is also connected to the enterprise digital twin construction module, which is used to generate service solutions by combining the output of the digital twin model.

[0023] Preferably, the smart application service module includes a precise service recommendation unit, which is configured as follows:

[0024] Analyze the potential needs of the target company based on its digital twin model;

[0025] Parallel queries of the dynamic knowledge graph match policy provisions, service resources, or other business entities with complementary capabilities that are relevant to the need.

[0026] Generate and output a personalized service list that includes matching results, reasons for recommendation, and expected benefit assessment.

[0027] Preferably, the smart application service module includes a policy simulation and matching unit, which is configured as follows:

[0028] Parse the policy text into structured conditional rules;

[0029] Based on the attribute data of enterprise entities and the prediction data of their digital twin models, the simulation calculation is performed to determine the policy implementation effect after the conditions are met.

[0030] For companies that do not fully meet the criteria, a reverse analysis of the gaps is conducted to generate a pre-assessment report containing quantitative indicators and improvement paths.

[0031] Preferably, the smart application service module includes an industry coupling discovery unit, which is configured as follows:

[0032] Regularly analyze dynamic knowledge graphs and, based on graph structure similarity measurement or semantic association analysis, identify enterprise entity pairs that have potential supply and demand or technological cooperation relationships but have not established direct connections.

[0033] After obtaining authorization, information on potential cooperation opportunities and the basis for correlation analysis will be proactively pushed to relevant business entities.

[0034] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, controls the processor to implement the functions of each module in the above-described system to perform the following operations:

[0035] Collect and integrate heterogeneous data from multiple sources within the park;

[0036] Build and update the park's dynamic knowledge graph, and manage the dynamic changes in relation weights;

[0037] A multi-objective optimization model was established and run based on the graph.

[0038] In response to operational needs, generate and output optimization solutions.

[0039] This invention also provides a method for the operation and management of digital economy parks based on big data analysis, implemented by the aforementioned system, the method comprising:

[0040] Data is collected and fused through the data sensing and fusion module;

[0041] The knowledge graph is built and updated through the dynamic knowledge graph engine module;

[0042] An optimization model is established through a multi-objective optimization decision-making module;

[0043] The intelligent application service module responds to requests and generates optimization solutions.

[0044] The technical effects and advantages of this invention are as follows:

[0045] 1. This invention effectively solves the problems of data silos and static information in traditional park management systems by constructing a dynamic knowledge graph that integrates multi-source data and configuring a weight function that evolves over time for its relation edges. The system deeply integrates and maps multi-dimensional heterogeneous data such as IoT sensing, enterprise operations, government affairs, and park resources, transforming the data from fragmented static records into a reasonable and computable relational network that reflects the real-time relational strength and evolution trend between entities such as enterprises, resources, and policies within the park. This provides a unified, accurate, and semantically rich data foundation for subsequent intelligent analysis and decision-making.

[0046] 2. This invention introduces a multi-objective optimization decision-making model based on dynamic knowledge graphs to synergistically quantify and optimize abstract management objectives such as industrial relevance, economic output, and resource efficiency. This enables the park operation and management to leap from single-point, post-event management to global, forward-looking planning. The system can calculate the tightness of the industrial ecosystem in real time based on the topology of the knowledge graph, and synergistically optimize this indicator with economic and energy efficiency objectives in key decisions such as spatial allocation and resource allocation. This results in outputting a Pareto optimal solution that maximizes the overall efficiency of the park under multiple constraints, significantly improving the scientific nature of management decisions and the aggregation degree of the industrial ecosystem.

[0047] 3. By integrating enterprise digital twin models with intelligent application service units, this invention achieves accurate profiling and proactive services for enterprise needs, as well as intelligent discovery of industrial collaboration opportunities within the park. The system can dynamically diagnose the individual status and potential needs of enterprises based on the digital twin, and automatically match, recommend, or simulate the optimal policies, services, and cooperation resources to generate personalized service lists and pre-assessment reports. At the same time, the system can proactively explore potential enterprise cooperation pairings and promote connections, thereby greatly improving the accuracy, proactivity, and value creation capabilities of park services and effectively stimulating the park's endogenous innovation and cooperation vitality. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of the overall system architecture of the present invention.

[0049] Figure 2 This is a schematic diagram of the dynamic knowledge graph construction process of the present invention.

[0050] Figure 3 This is a schematic diagram of the spatial resource allocation optimization process of the present invention.

[0051] Figure 4 This is a schematic diagram of the precise service recommendation process of the present invention.

[0052] The attached diagram is labeled as follows: 1. Data perception and fusion module; 2. Dynamic knowledge graph engine module; 3. Multi-objective optimization decision-making module; 4. Smart application service module; 200. Entity relationship extraction unit; 201. Weight dynamic management unit; 5. Enterprise digital twin construction module; 400. Precision service recommendation unit; 401. Policy simulation and matching unit; 402. Industry coupling discovery unit. Detailed Implementation

[0053] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0054] As attached Figure 1As shown, this embodiment of the invention provides a digital economy park operation and management system based on big data analysis. The system adopts a layered collaborative architecture and mainly includes: a data perception and fusion module 1, a dynamic knowledge graph engine module 2, a multi-objective optimization decision module 3, a smart application service module 4, and an optional enterprise digital twin construction module 5. Each module exchanges data and calls instructions through predefined application programming interfaces or message middleware and works collaboratively.

[0055] Data perception and fusion module 1 provides a unified and standardized data foundation for the system. This module collects and fuses multi-source heterogeneous data within the park from multiple dimensions, mainly including:

[0056] IoT sensing data: Real-time physical environment and facility operation data collected by various sensor devices deployed in the park, including energy consumption, environmental parameters, security status and parking space occupancy information. The various sensors include smart meters, environmental monitors, security cameras and parking space detectors.

[0057] Enterprise operation data: Enterprise business registration information, recruitment dynamics, intellectual property information, project application data, and de-identified macroeconomic business trend data obtained through interfaces, provided that authorization is obtained and data security standards are met.

[0058] Government-related data: Legally disclosed information such as enterprise tax payment, social security payment, administrative compliance and qualification certification is obtained by connecting to the government data platform;

[0059] Park resource data includes spatial attribute information of all physical carriers in the park, reservation information of shared facilities, and textual content of various industry support policies. Among them, spatial attribute information includes location, area, rent and usage status, and shared facilities include meeting rooms and laboratories, etc.

[0060] This module performs preprocessing on the collected raw data, including cleaning, noise reduction, format standardization, and key-value extraction. It then merges the data according to a unified spatiotemporal benchmark to form a standardized data pool for use by downstream modules.

[0061] Dynamic knowledge graph engine module 2, connected to data perception and fusion module 1, is the core data hub of the system, as shown in the attached diagram. Figure 2 As shown, its workflow includes:

[0062] Entity Relationship Extraction Unit 200: Utilizing natural language processing technology and predefined domain ontology rules, it automatically identifies and extracts the constituent elements of the knowledge graph from the fused data, including entities, entity attributes, and semantic relationships between entities. Among them, entities include enterprises, patented technologies, policy clauses, and spatial units, etc., and semantic relationships include supply chain relationships representing business connections, subordinate relationships representing organizational structures, and policy applicability relationships representing qualification conditions, etc.

[0063] The weighted dynamic management unit 201 is one of the key innovations of this invention. This unit assigns a weight value that changes dynamically over time to each relation edge in the knowledge graph. And configure the weight evolution function so that the knowledge graph can dynamically represent the evolution of the park's industrial ecosystem;

[0064] For example, the weights used to represent business partnerships can follow an exponential decay model:

[0065]

[0066] in, As the initial weights, This is a timestamp indicating the occurrence of the latest collaborative event. A positive attenuation coefficient occurs when a new event occurs. Updated, weight value The corresponding settings are reset and enhanced. For the relationship representing the policy's validity period, its weight can be linearly increased within a preset time period before the policy expires to indicate urgency.

[0067] The completed dynamic knowledge graph is stored in a graph database, such as Neo4j, providing efficient graph traversal, relationship chain query, and community discovery capabilities for upper-layer applications.

[0068] The multi-objective optimization decision module 3 is connected to the dynamic knowledge graph engine module 2. Based on the real-time relationship network provided by the dynamic knowledge graph, it constructs the following multi-objective optimization function. :

[0069]

[0070] in, , and A configurable balancing coefficient used to adjust the weights of different objectives;

[0071] Economic output indicators: These can be quantified as financial indicators such as total tax revenue of the park and output value per unit area;

[0072] The industry relevance index is another creative feature of this invention. This index is quantitatively calculated based on the topological structure of a dynamic knowledge graph. One feasible calculation method is to calculate the average clustering coefficient or network density of the subgraphs formed by enterprise entities in the graph. The higher the value, the closer the connection between enterprises through technology chains, supply chains or talent chains, and the better the synergy of the industrial ecosystem.

[0073] Resource efficiency indicators can be quantified as GDP per unit of energy consumption, average turnover rate of public facilities, etc.

[0074] This module embeds a multi-objective optimization algorithm library, such as the NSGA-II algorithm. When a decision-making scenario is triggered, such as recommending the optimal location for a newly established company, as shown in the attached diagram... Figure 3 As shown, this module is configured to: use the average correlation between candidate enterprise entities and existing enterprise clusters within a preset radius of each candidate space in the knowledge graph as... The core computational factors in this decision-making process are substituted into the aforementioned multi-objective function for solution. Ultimately, the algorithm outputs a Pareto optimal solution set, which is a series of site selection schemes that achieve the best balance among multiple objectives such as economic benefits, industrial synergy, and resource efficiency. This achieves the decision-making objective of strengthening the overall industrial ecosystem aggregation of the park while satisfying physical space constraints.

[0075] Optionally, the system of the present invention further includes an enterprise digital twin construction module 5, connected to the data perception and fusion module 1, used to construct and maintain digital twin models for each enterprise within the park. This model is updated by continuously ingesting dynamic data streams such as enterprise electricity consumption, recruitment, and patents, and includes at least...

[0076] Innovation Sub-model: Based on indicators such as enterprise R&D investment intensity, intellectual property output rate and quality, and demand for high-end technical positions, assess its innovation vitality and growth potential;

[0077] The health sub-model assesses a company's operational stability and risk profile based on information such as employment stability, operational energy consumption patterns, public opinion trends, and legal risks.

[0078] This digital twin forms the basis for creating precise micro-level profiles of enterprises and connecting them with macro-level correlation analysis of the industrial park.

[0079] The intelligent application service module 4 is connected to the dynamic knowledge graph engine module 2, the multi-objective optimization decision-making module 3, and the enterprise digital twin construction module 5, respectively. It integrates multiple intelligent service units, including:

[0080] Precision Service Recommendation Unit 400: As attached Figure 4As shown, based on the potential needs analyzed by the enterprise digital twin model, the dynamic knowledge graph is queried in parallel to automatically match and recommend related policy provisions, park service resources or potential partners with complementary businesses, and generate a personalized service list with the reasons for the recommendation and the expected benefit assessment.

[0081] Policy Simulation and Matching Unit 401: Utilizes natural language processing technology to parse unstructured policy texts into structured, computable conditional rules. Subsequently, it combines the static attributes of enterprises with the prediction data of their digital twin models, such as the R&D expense forecast for the next quarter, to simulate and calculate the possible effects of meeting the policy conditions. For enterprises that do not fully meet the conditions, this unit can perform reverse gap analysis and generate a pre-assessment report containing specific quantitative improvement indicators.

[0082] Industry Coupling Discovery Unit 402: Regularly scans and analyzes the dynamic knowledge graph, and uses graph structure similarity measurement or semantic association analysis of entity description text to proactively discover enterprise entity pairs that have potential complementarity in technology or business but have not yet established a direct connection relationship. After obtaining authorization from the relevant enterprises, the system can proactively and targetedly push the potential cooperation opportunity and the basis for association analysis to both parties.

[0083] This invention also provides a computer-readable storage medium storing a computer program thereon. When the program is executed by one or more processors, the processors are able to implement all or part of the functions of the data perception and fusion module 1, dynamic knowledge graph engine module 2, multi-objective optimization decision module 3, smart application service module 4, and enterprise digital twin construction module 5 in the previous system, thereby executing the above-mentioned park operation and management method based on dynamic knowledge graph and multi-objective optimization.

[0084] The methods mainly include: collecting and fusing multi-source heterogeneous data, constructing and updating a park knowledge graph with dynamic weights, establishing and solving a multi-objective optimization model based on the graph, responding to operational scenarios, and collaboratively calling the graph and model to generate intelligent service solutions.

[0085] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A digital economy park operation and management system based on big data analysis, characterized in that: The core modules for collaborative work include the following: The data sensing and fusion module (1) is used to collect and fuse multi-source heterogeneous data in the park. The multi-source heterogeneous data includes at least IoT sensing data, enterprise operation data, government-related data and park resource data. The dynamic knowledge graph engine module (2) is connected to the data perception and fusion module (1) and is used to build and continuously update the dynamic knowledge graph of the park based on the fused data. The dynamic knowledge graph serves as the unified data model and relation index basis for subsequent optimization decisions. The entities in the knowledge graph include at least enterprise entities, resource entities and policy entities. The relation weights between entities change dynamically according to the time of the associated events. The multi-objective optimization decision module (3) is connected to the dynamic knowledge graph engine module (2) and is used to establish a multi-objective optimization model based on the dynamic knowledge graph in real time with the goal of improving the overall efficiency of the park. The multi-objective includes at least economic output target, industrial synergy target and resource efficiency target, and is equipped with corresponding quantifiable evaluation indicators. The intelligent application service module (4) is connected to the dynamic knowledge graph engine module (2) and the multi-objective optimization decision module (3) to respond to external instructions or events, coordinate the calculation results of the dynamic knowledge graph and the multi-objective optimization model, and generate and output optimization solutions for resource allocation, enterprise services or industry guidance.

2. The digital economy park operation and management system based on big data analysis according to claim 1, characterized in that: The dynamic knowledge graph engine module (2) includes: The entity relationship extraction unit (200) is used to extract entities, attributes, and relationships between entities from multi-source heterogeneous data; The weight dynamic management unit (201) is used to assign initial weights to the association and configure the weight change function, wherein the weight change function is configured such that for the relationship edge representing the dynamic association between entities, its weight increases with the recent occurrence of the association event and decreases with the passage of time, so as to realize the timeliness of the graph's representation of the ecological evolution of the park.

3. The digital economy park operation and management system based on big data analysis according to claim 1, characterized in that: The objective function constructed in the multi-objective optimization decision module (3) is: in, , and This is a configurable balance coefficient; The evaluation index corresponding to the industrial synergy target is the industrial relevance index. Its calculation is based on the connection strength and network density formed between enterprise entities in the dynamic knowledge graph through technology chains, supply chains or talent chains, and is quantitatively calculated based on the topological structure of the dynamic knowledge graph.

4. The digital economy park operation and management system based on big data analysis according to claim 3, characterized in that: When performing spatial resource allocation decisions, the multi-objective optimization decision module (3) is specifically configured to: use the average correlation between candidate enterprise entities and enterprise clusters within a preset range around the proposed allocation space in the knowledge graph as the industry correlation index in the objective function. The core computational factors are optimized to ensure that spatial allocation satisfies physical constraints while enhancing the aggregation of the industrial ecosystem.

5. The digital economy park operation and management system based on big data analysis according to claim 1, characterized in that: Also includes: The enterprise digital twin construction module (5) is connected to the data perception and fusion module (1) and is used to build and maintain digital twin models for each enterprise in the park based on the dynamic data flow of individual enterprises. The digital twin model includes at least an innovation sub-model for evaluating innovation capabilities and a health sub-model for evaluating operational stability. The digital twin model constitutes the basis for conducting micro-profile and macro-correlation analysis of enterprises. The smart application service module (4) is also connected to the enterprise digital twin construction module (5) to generate service solutions by combining the output of the digital twin model.

6. The digital economy park operation and management system based on big data analysis according to claim 5, characterized in that: The smart application service module (4) includes a precise service recommendation unit (400), which is configured as follows: Analyze the potential needs of the target company based on its digital twin model; Parallel queries of the dynamic knowledge graph match policy provisions, service resources, or other business entities with complementary capabilities that are relevant to the need. Generate and output a personalized service list that includes matching results, reasons for recommendation, and expected benefit assessment.

7. The digital economy park operation and management system based on big data analysis according to claim 6, characterized in that: The smart application service module (4) includes a policy simulation and matching unit (401), which is configured as follows: Parse the policy text into structured conditional rules; Based on the attribute data of enterprise entities and the prediction data of their digital twin models, the simulation calculation is performed to determine the policy implementation effect after the conditions are met. For companies that do not fully meet the criteria, a reverse analysis of the gaps is conducted to generate a pre-assessment report containing quantitative indicators and improvement paths.

8. The digital economy park operation and management system based on big data analysis according to claim 1, characterized in that: The smart application service module (4) includes an industry coupling discovery unit (402), which is configured as follows: Regularly analyze dynamic knowledge graphs and, based on graph structure similarity measurement or semantic association analysis, identify enterprise entity pairs that have potential supply and demand or technological cooperation relationships but have not established direct connections. After obtaining authorization, information on potential cooperation opportunities and the basis for correlation analysis will be proactively pushed to relevant business entities.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by the processor, it controls the processor to implement the functions of each module in the system as described in any one of claims 1-8, to perform the following operations: Collect and integrate heterogeneous data from multiple sources within the park; Build and update the park's dynamic knowledge graph, and manage the dynamic changes in relation weights; A multi-objective optimization model was established and run based on the graph. In response to operational needs, generate and output optimization solutions.

10. A method for the operation and management of digital economy parks based on big data analysis, characterized in that: The system implemented by any one of claims 1-8, the method comprising: Data is collected and fused through the data sensing and fusion module (1); The knowledge graph is constructed and updated through the dynamic knowledge graph engine module (2); An optimization model is established through the multi-objective optimization decision module (3); The intelligent application service module (4) responds to requests and generates optimization solutions.

Images