Multi-source collaborative checking system and method for real quantity of natural resource assets

Abstract

The invention discloses a multi-source collaborative checking system and method for the real quantity of natural resource assets, and relates to the technical field of natural resource management. The system comprises a data acquisition module used for acquiring multi-source data and real-time data and preprocessing the multi-source data and the real-time data; the knowledge graph processing module is used for constructing a forest resource knowledge graph according to the multi-source data and the real-time data; and the check analysis module is used for constructing an intelligent check model based on the knowledge graph constructed by the knowledge graph processing module. By automatically collecting and preprocessing multi-source data, constructing and dynamically maintaining a knowledge graph, and intelligently checking and analyzing and visually presenting a result based on the graph, efficient data integration, accurate and intelligent checking, trend prediction and active early warning are realized, and a scientific decision basis is provided for urban forest resource management.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of natural resource management, in particular to a natural resource asset physical quantity multi-source collaborative verification system and method. BACKGROUND

[0002] In the field of natural resource management, with the development of digital technology, multi-source data integration and verification have become the key to accurately grasp the physical quantity of natural resource assets, especially in the context of urban forest resource management, the demand for multi-department collaborative management is increasingly prominent. However, the current forestry, urban planning, environmental protection and other departments operate independently information management system, data format, statistical standards, storage architecture are significantly different. Each department collects and statistics data based on its own business needs, such as forestry department focuses on the statistics of tree biological characteristics, urban planning department pays attention to the definition of forest range under land use planning, environmental protection department focuses on forest ecological function index, but there is a lack of unified data integration mechanism.

[0003] This leads to frequent data cross-overlapping and missing report problems, and the traditional data integration method relying on manual intervention is not only inefficient, but also prone to errors. The inaccuracy of data makes it difficult to accurately verify the physical quantity of urban forest resources, seriously affecting the scientificity of urban ecological planning and resource management decision-making, and cannot meet the requirements of urban sustainable development for fine management of natural resources, and the verification process lacks a dynamic optimization mechanism. When the data deviation exceeds the threshold, manual intervention is required to adjust the parameters, which has poor adaptability and affects the timeliness of decision-making. SUMMARY

[0004] The purpose of the present application is to provide a natural resource asset physical quantity multi-source collaborative verification system and method to solve the problems raised in the background art.

[0005] To achieve the above purpose, the present application provides the following technical scheme: a natural resource asset physical quantity multi-source collaborative verification system, comprising: A data acquisition module is used to acquire multi-source data of forestry department, urban planning department and environmental protection department, and real-time data collected through Internet of Things sensors, and to preprocess the multi-source data and real-time data through cleaning, format conversion and geographic coordinate system unification; A knowledge graph processing module is used to extract entities and relationships by using named entity recognition algorithm in natural language processing technology and conditional random field model in machine learning on the multi-source data and real-time data preprocessed by the data acquisition module, eliminate redundant and conflicting information by entity alignment technology and calculate cosine similarity between entities, construct a forest resource knowledge graph, store the knowledge graph in a graph database, and establish a real-time updating mechanism to dynamically update the knowledge graph; The verification analysis module is configured to construct an intelligent verification model based on the knowledge graph constructed by the knowledge graph processing module, and to identify cross-overlap and false reporting problems in the data by setting reasoning rules based on the relationships and rules in the knowledge graph for the forestry department to count the number of trees in a certain piece of forest land, for the urban planning department to calculate the theoretical number of trees based on land use, is a threshold value; and based on the knowledge graph, an anomaly detection rule is set for anomaly detection, wherein is the forest carbon sink, is the tree growth, is the forest area, is the theoretical carbon sink calculated according to and is a threshold value; and combined with geographic information system technology, the verification results are visualized and presented, and a verification report containing data problem conditions, cause analysis and rectification suggestions is generated; The verification analysis module is further configured to: based on time series analysis technology, the trend of the verification results of the continuous N periods is predicted, and a time series model is constructed wherein is the verification deviation rate of the tth period, is a random error term; When the predicted deviation rate exceeds the threshold range, early warning information is generated and the verification process is optimized.

[0006] Preferably, the data collection module includes a multi-source data acquisition unit and a sensor data collection unit: The multi-source data acquisition unit is configured to interface with external business databases through a data interface, and to extract, convert and load multi-source data using ETL tools; The sensor data collection unit is configured to deploy sensors in the resource area, and to collect and transmit data to the data collection module and multi-source data integration in real time through a wireless sensor network.

[0007] Preferably, the multi-source data acquisition unit converts the position data of different data sources into the WGS84 coordinate system when performing data conversion.

[0008] Preferably, the knowledge graph processing module includes a graph construction unit and a graph maintenance unit: ​The atlas construction unit is configured to implement natural language processing and machine learning algorithms by using a programming language and a corresponding tool kit, automatically complete entity and relationship extraction and knowledge fusion of input data of the data collection module, and construct a resource knowledge graph. The atlas construction unit defines matters, spatial ranges and measurement parameters related to resources as entities, defines the belonging relationship between matters and spaces, the spatial attribute change relationship, and the association relationship between measurement parameters and resources as relationships, and identifies the relationship between entities through semantic analysis. The atlas maintenance unit is configured to establish a knowledge graph quality evaluation mechanism, regularly check the accuracy and integrity of entity relationships in the knowledge graph, correct errors, and update entity attributes and relationships of the knowledge graph in real time according to newly entered resource data.

[0009] Preferably, the verification analysis module includes a verification model unit, an anomaly detection unit and a result presentation unit. The verification model unit is configured to construct an intelligent verification model based on the knowledge graph constructed by the knowledge graph processing module, input the resource data to be verified, use the relationships and rules in the knowledge graph, and judge whether the judgment is correct, if so, combine the related relationships and constraint conditions in the knowledge graph to reason about possible problems in the data, identify cross-overlapping and missing report problems in the data, and output verification results; The anomaly detection unit is configured to set anomaly detection rules based on the knowledge graph, judge whether the judgment is correct, and if so, issue an anomaly warning; The result presentation unit is configured to visualize the verification results of the verification model unit in the form of a map by combining geographic information system technology, mark the areas with data problems by different colors, and generate a verification report containing specific conditions of cross-overlapping and missing reports, cause analysis and rectification suggestions.

[0010] Preferably, when constructing the intelligent verification model, the verification model unit trains historical verification data in the knowledge graph by using a machine learning algorithm to optimize the identification rules of the intelligent verification model.

[0011] Preferably, the system further includes a data sharing module configured to establish a data sharing platform, implement centralized storage and distributed management of data by using distributed database technology, provide data access services for the data collection module, the knowledge graph processing module and the verification analysis module through a data interface, and record and trace the uploading, accessing and using processes of data by using blockchain technology.

[0012] The application also provides a natural resource asset physical quantity multi-source collaborative verification method, which includes: Acquire multi-source data and real-time data collected by Internet of Things sensors, and preprocess the multi-source data and real-time data by cleaning, format conversion and unification of geographic coordinate systems; For the preprocessed multi-source data and real-time data, the named entity recognition algorithm in natural language processing technology and the conditional random field model in machine learning are used to extract entities and relationships, the entity alignment technology is used to eliminate redundant and conflicting information by calculating the cosine similarity between entities, a resource knowledge graph is constructed, the knowledge graph is stored in a graph database, and a real-time updating mechanism is established to dynamically update the knowledge graph; Based on the constructed knowledge graph, an intelligent verification model is constructed, for the input to be verified resource data, the relationship and rule in the knowledge graph are used to identify the cross-overlapping and missing report problems in the data by setting reasoning rules , wherein, the number of trees in a piece of forest land counted by the forestry department, the theoretical bearing tree quantity calculated based on land use by the urban planning department, is a threshold value; at the same time, abnormal detection rules are set based on the knowledge graph to perform abnormal detection, wherein, is the forest carbon sink, is the tree growth, is the forest area, is the theoretical carbon sink calculated according to and , and is a threshold value; and the verification result is visualized and presented by combining the geographic information system technology, and a verification report containing data problem conditions, cause analysis and rectification suggestions is generated; Based on time series analysis technology, the trend of the verification results of N consecutive periods is predicted, and a time series model is constructed, wherein, is the verification deviation rate of the t period, is a random error term; when the predicted deviation rate exceeds the threshold range, early warning information is generated and the verification process is optimized.

[0013] The application also provides an electronic device, which is an entity device, and the electronic device comprises: a processor, a memory, the memory being in communication connection with the processor; The memory is used to store executable instructions executed by the at least one processor, and the processor is used to execute the executable instructions to realize the natural resource asset physical quantity multi-source collaborative verification method as described above.

[0014] The application also provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the natural resource asset physical quantity multi-source collaborative verification method.

[0015] Compared with the prior art, the application has the following beneficial effects: By collecting multi-source data in real time and standardizing preprocessing, the data format and source barriers are broken, efficient data aggregation and standardization are realized, a solid foundation is laid for subsequent analysis, knowledge graph technology is used to deeply mine data correlation, a structured knowledge network is constructed and dynamically updated and maintained, data cross-overlapping and missing report problems are accurately identified, the accuracy and reliability of verification are greatly improved, deviation trends are predicted in advance through time series analysis, the effects of early detection of abnormalities and avoidance of systematic data deviation risks are achieved, an intelligent verification model is constructed based on the knowledge graph, reasoning analysis is performed in combination with set rules and algorithms, automation and intelligentization of the verification process are realized, data abnormalities are detected and warned in a timely manner, finally, complex verification results are visualized and professional reports are generated through a geographic information system, and scientific decision-making basis is intuitively and clearly provided for urban forest resource management, and the efficiency and level of resource management are comprehensively improved. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 A structural schematic diagram of a natural resource asset physical quantity multi-source collaborative verification system provided by the embodiment of the application is shown in the figure. Figure 2 A main flowchart of a natural resource asset physical quantity multi-source collaborative verification method provided by the embodiment of the application is shown in the figure. Figure 3 A structural schematic diagram of an electronic device provided by the embodiment of the application is shown in the figure. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the application.

[0018] The execution subject of the method in the embodiment is a terminal, which can be a mobile phone, a tablet computer, a palm computer PDA, a notebook computer or a desktop computer, and of course, can also be other devices with similar functions, which are not limited in the embodiment.

[0019] Please refer to Figure 1 The application provides a natural resource asset physical quantity multi-source collaborative verification system, which comprises: The data acquisition module 11 is configured to acquire multi-source data of forestry departments, urban planning departments and environmental protection departments, and real-time data collected by Internet of Things sensors, and to perform preprocessing on the multi-source data and the real-time data, including cleaning, format conversion and unification of geographic coordinate systems. The knowledge graph processing module 12 is configured to extract entities and relationships by using a named entity recognition algorithm in natural language processing technology and a conditional random field model in machine learning, eliminate redundant and conflicting information by using entity alignment technology and calculating cosine similarity between entities, construct a forest resource knowledge graph, store the knowledge graph in a graph database, and establish a real-time updating mechanism to dynamically update the knowledge graph based on the multi-source data and the real-time data preprocessed by the data acquisition module 11. The verification analysis module 13 is configured to construct an intelligent verification model based on the knowledge graph constructed by the knowledge graph processing module 12, use relationships and rules in the knowledge graph to identify cross-overlapping and missed reporting problems in input forest resource data to be verified by setting reasoning rules , wherein is the number of trees in a piece of forest land counted by the forestry department, is the theoretical number of trees calculated based on land use by the urban planning department, is a threshold value, and abnormal detection rules are set based on the knowledge graph to perform abnormal detection, wherein is the forest carbon sink, is the tree growth, is the forest area, is the theoretical carbon sink calculated according to and , and is a threshold value; and the verification result is visualized and presented by combining geographic information system technology, and a verification report containing data problem conditions, cause analysis and rectification suggestions is generated.

[0020] The data acquisition module 11 is a functional unit for acquiring multi-source data, which aims to provide a comprehensive and accurate data basis for subsequent data processing. Through this module, the problem of scattered data sources and inconsistent formats can be effectively solved.

[0021] In one possible implementation, the data acquisition module 11 acquires data from multiple external business databases through specific data interfaces, which covers various types of information related to natural resources; at the same time, real-time data is collected by Internet of Things sensors deployed in resource areas, including but not limited to resource environmental parameters, resource state changes and other data, and all collected data are preprocessed to ensure data availability and consistency, including cleaning, format conversion and unification of geographic coordinate systems.

[0022] Furthermore, the knowledge graph processing module 12 is the core component of the entire system for achieving deep data fusion and knowledge construction. Its main function is to transform the collected multi-source data into a structured and semantic knowledge graph. This module uses named entity recognition algorithms from natural language processing and conditional random field models from machine learning to extract entities and relationships. Entities refer to key information such as things related to resources, spatial ranges, and measurement parameters, while relationships include the affiliation between things and spaces, spatial attribute changes, and the association between measurement parameters and resources. Redundant and conflicting information is eliminated through entity alignment technology and the calculation of cosine similarity between entities, thereby constructing a resource knowledge graph, which is stored in a graph database. In addition, this module establishes a real-time update mechanism and a quality assessment mechanism to dynamically update and maintain the knowledge graph, ensuring that the knowledge graph accurately reflects the actual situation of natural resource assets.

[0023] In addition, the verification and analysis module 13 is a knowledge graph-based module constructed from the knowledge graph processing module 12, enabling the verification and analysis of the physical quantity of natural resource assets. It constructs an intelligent verification model, utilizing relationships and rules within the knowledge graph to analyze the input resource data to be verified, setting inference rules to identify overlaps and omissions in the data; simultaneously, it uses anomaly detection rules based on the knowledge graph to detect anomalies. Finally, combining geographic information system (GIS) technology, the verification results are visualized and a verification report is generated, including data issues, cause analysis, and rectification suggestions, providing decision support for natural resource asset management.

[0024] For example, in one specific implementation, for forest resource verification in a certain area, the data acquisition module 11 obtains data such as tree species and quantity from the forestry management system, and forest land planning data from the urban planning system. Simultaneously, it collects real-time data such as tree growth status and soil moisture through IoT sensors deployed in the forest, and performs preprocessing. The knowledge graph processing module 12 extracts entities and relationships from this data to construct a forest resource knowledge graph. For example, different types of trees and specific forest land plots are defined as entities, and the ownership relationship between trees and forest land is defined as a relationship. Based on the constructed knowledge graph, the verification and analysis module 13 uses set inference rules and anomaly detection rules to verify the input forest resource data, identify data overlaps, omissions, and anomalies, and presents the results in the form of a visual map, generating a detailed verification report to assist managers in making forest resource management decisions.

[0025] Among them, the function The specific calculation logic for calculating theoretical carbon sequestration is as follows: First, based on the forest ecosystem carbon sink model, the tree growth rate is determined. Biomass conversion coefficient The biomass conversion coefficient k is obtained by referring to the standard table of Forest Ecosystem Long-term Observation Method (LY / T1952-2011) according to the parameters such as tree species type (e.g. coniferous forest, broad-leaved forest), tree age, etc. Theoretical biomass Wherein, is the annual tree growth per unit area (tons / hectare / year); Carbon content conversion factor Taking 0.45 (i.e. the average proportion of carbon elements in forest biomass), the theoretical carbon sink amount is Wherein, is the forest area (hectare); Abnormality detection rule In the formula, is the threshold value set according to the regional carbon sink historical data (such as mean ± 2 times standard deviation), when the measured forest carbon sink amount deviates from the theoretical value by more than the threshold value, an abnormality early warning is triggered.

[0026] In an optional embodiment, the data collection module 11 comprises a multi-source data acquisition unit and a sensor data collection unit: The multi-source data acquisition unit is configured to connect with external business databases through a data interface, and to realize extraction, conversion and loading of multi-source data by using an ETL tool; The sensor data collection unit is configured to deploy sensors in a resource area, and to collect and transmit data to the data collection module 11 and the multi-source data integration in real time through a wireless sensor network.

[0027] The multi-source data acquisition unit is a functional unit responsible for establishing a data interaction channel with external business databases, and its core purpose is to break down data barriers and gather natural resource related data scattered in different systems.

[0028] In this embodiment, the multi-source data acquisition unit establishes a stable data connection link according to the communication protocol and data structure of each external business database through a customized data interface; and performs data extraction, conversion and loading operations by using an ETL tool. The operation can standardize the format of the original data that does not meet the system requirements, for example, unify the inconsistent data formats and numerical precision in different databases, so as to provide a data basis with unified format and clear structure for subsequent data processing, and effectively solve the problem of multi-source data format heterogeneity.

[0029] In addition, the sensor data collection unit is a key unit for real-time dynamic monitoring of natural resource areas, and its main role is to make up for the deficiencies of traditional database data in real-time and environmental parameter collection.

[0030] In the embodiment, the sensor data acquisition unit covers the resource area by deploying various types of sensors in the resource area, including but not limited to sensors for monitoring tree growth, humidity sensors for reflecting soil environment, and sensors for detecting air quality, etc. These sensors collect resource environment parameters, resource state changes, etc. in real time at a set frequency, and transmit the data to the data acquisition module 11 through the wireless sensor network in accordance with a specific communication protocol. The data is integrated with the business data obtained by the multi-source data acquisition unit, and the natural resource data from static information to dynamic changes is comprehensively collected, providing rich data dimensions for subsequent more accurate analysis and verification.

[0031] In an optional embodiment, the multi-source data acquisition unit converts the position data of different data sources into the WGS84 coordinate system uniformly when performing data conversion.

[0032] Specifically, the WGS84 coordinate system is an internationally recognized geographic coordinate system, which refers to a geocentric coordinate system with the center of the earth as the origin, and has the characteristics of high precision and global unity. In the embodiment, since different external business databases may use independent coordinate systems when storing natural resource related data, such as local coordinate systems, industry-specific coordinate systems, etc., these coordinate systems differ in coordinate origin, coordinate axis direction, length unit, etc., resulting in that the data cannot be directly fused and analyzed in spatial position expression.

[0033] The multi-source data acquisition unit maps the position data of different data sources to the WGS84 coordinate system through a coordinate conversion algorithm, so that the data from different channels can be superimposed, compared and analyzed on the same geographic spatial reference. For example, when analyzing forest resource distribution and urban planning land use, the accuracy and consistency of the two types of data in spatial position are ensured, effectively eliminating data errors and analysis deviations caused by coordinate system differences, and providing a reliable spatial data foundation for subsequent in-depth analysis and decision-making based on geographic information.

[0034] In an optional embodiment, the knowledge graph processing module 12 includes a graph construction unit and a graph maintenance unit: The graph construction unit is configured to implement natural language processing and machine learning algorithms using programming languages and corresponding toolkits, automatically extract entities and relationships from the input data of the data acquisition module 11, and fuse knowledge to construct a resource knowledge graph. The graph construction unit defines resources, spatial ranges, and measurement parameters as entities, defines the belonging relationship between things and spaces, the spatial attribute change relationship, and the association relationship between measurement parameters and resources as relationships, and identifies the relationship between entities through semantic analysis. The knowledge graph maintenance unit is used to establish a knowledge graph quality assessment mechanism, periodically check the accuracy and completeness of entity relationships in the knowledge graph, correct erroneous information, and update the entity attributes and relationships of the knowledge graph in real time based on newly entered resource data.

[0035] In this embodiment, the graph construction unit builds a natural language processing and machine learning algorithm execution environment based on a general programming language and corresponding toolkit.

[0036] Among them, things related to resources (such as specific natural resource objects, resource management agencies, etc.), spatial range (such as resource distribution areas, geographical boundaries, etc.), and measurement parameters (such as resource quantity, quality indicators, etc.) are defined as entities. These entities are the basic nodes that constitute the knowledge graph. Relationships are defined as the relationship between things and space, the relationship between spatial attribute changes, and the relationship between measurement parameters and resources. Relationships are used to describe the intrinsic connections between entities.

[0037] The graph construction unit uses named entity recognition algorithms and conditional random field models to perform semantic parsing on the data, automatically identifying and extracting entity and relation information. Then, using entity alignment technology, methods such as calculating the cosine similarity between entities are used to eliminate duplicate or conflicting entity information from different data sources. Finally, the processed information is integrated to construct a resource knowledge graph, which is stored in a graph database. This transforms the originally scattered and disordered data into a knowledge system with a clear semantic structure and logical connections, providing rich knowledge support for subsequent verification and analysis.

[0038] In this embodiment, the knowledge graph maintenance unit has established a complete knowledge graph quality assessment mechanism. It regularly checks the accuracy and completeness of entity relationships in the knowledge graph, uses automated algorithms to verify the data logic in the graph, such as checking whether the relationships between entities conform to business rules and whether the attribute values ​​are within a reasonable range, and combines manual review to verify complex or uncertain information, so as to promptly discover and correct errors.

[0039] Meanwhile, when new natural resource data is entered, the knowledge graph maintenance unit updates the entity attributes and relationships of the knowledge graph in real time according to the data update rules. For example, when the tree growth data in forest resources changes, the attribute information of the corresponding tree entity is updated in a timely manner, and the relationship with other related entities is adjusted to ensure that the knowledge graph always keeps in line with the actual situation, providing a reliable and accurate knowledge foundation for the verification and analysis module 13.

[0040] In an optional embodiment, the verification and analysis module 13 includes a verification model unit, an anomaly detection unit, and a result presentation unit: The verification model unit is used to construct an intelligent verification model based on the knowledge graph constructed by the knowledge graph processing module 12. It inputs the resource data to be verified and utilizes the relationships and rules in the knowledge graph to determine... If the condition is true, then combine the relevant relationships and constraints in the knowledge graph to infer the possible problems in the data, identify the cross-over and underreporting problems in the data, and output the verification results. The anomaly detection unit is used to set anomaly detection rules based on the knowledge graph and to determine... If the condition is met, an abnormal warning will be issued. The result presentation unit is used to visualize the verification results of the verification model unit in map form by combining geographic information system technology, marking areas with data problems with different colors, and generating a verification report that includes specific details of data overlap and omissions, cause analysis, and rectification suggestions.

[0041] In this embodiment, the verification model unit first designs and constructs an intelligent verification model based on the resource knowledge graph built by the knowledge graph construction unit, combined with natural resource verification business rules and historical verification experience. During operation, this model matches and analyzes the resource data to be verified with the relationships and rules in the knowledge graph, and sets inference rules accordingly. When the difference between the actual statistical quantity and the theoretical carrying capacity exceeds a threshold, the system combines the relationship between forest land and trees in the knowledge graph with relevant ecological indicator constraints to infer potential problems in the data, such as statistical errors, data duplication, or omissions, and outputs detailed verification results. This enables intelligent and precise verification of natural resource asset data, effectively improving verification efficiency and accuracy.

[0042] In this embodiment, the anomaly detection unit sets anomaly detection rules based on a knowledge graph. By continuously monitoring changes in resource-related indicators, growth, and regional data within the knowledge graph, an anomaly is identified and an early warning mechanism is triggered when the difference between actual and theoretically calculated indicators exceeds a set threshold. For example, if the trend of forest carbon sequestration does not match relevant data such as tree growth and forest area, the anomaly detection unit will promptly issue an anomaly warning, reminding relevant personnel to conduct further investigation and handling. This helps to promptly identify and resolve potential problems in natural resource management, ensuring the rational use of resources and the stability of the ecological environment.

[0043] Additionally, it should be noted that the results presentation unit is a functional unit that presents the verification and analysis results to users in an intuitive and easy-to-understand way. Its main purpose is to transform complex verification data and analysis conclusions into visualized information, providing clear and intuitive support for natural resource asset management decisions.

[0044] In this embodiment, the result presentation unit combines the geographic information system technology to visualize the checking results of the checking model unit and the early warning information of the anomaly detection unit in the form of a map. The areas with data problems are marked by different colors, symbols, and other visualization elements, such as red for areas with cross-overlapping data and yellow for areas that may have a false negative problem, so that the user can quickly and intuitively understand the spatial distribution of the problems.

[0045] At the same time, the result presentation unit also generates a detailed checking report containing the specific situation of data cross-overlapping and false negatives, cause analysis, and rectification suggestions. The report content covers data comparison analysis charts, problem detailed description, possible cause inference, and targeted solutions, etc., providing comprehensive and detailed basis for natural resource management departments to make scientific and reasonable decisions.

[0046] In an optional embodiment, the checking model unit adopts machine learning algorithms to train the historical checking data in the knowledge graph when constructing the intelligent checking model, and optimizes the identification rules of the intelligent checking model.

[0047] In this embodiment, historical checking data refers to the data set accumulated in the past natural resource asset physical quantity checking process, which contains correct checking results and related data characteristics. These data contain a large amount of checking experience and rules. The checking model unit selects appropriate machine learning algorithms, such as decision tree, random forest, neural network, etc., takes the historical checking data in the knowledge graph as training samples, and learns and analyzes the data through algorithms to mine the hidden features and patterns in the data, such as learning the distribution rules of different types of natural resource data under normal circumstances, the correlation between various types of data, and the factors that may cause data anomalies, etc. By continuously adjusting the model parameters, the identification rules of the intelligent checking model are optimized, so that the model can better adapt to the natural resource asset checking needs in different scenarios, improve the identification accuracy of data cross-overlapping and false negatives, and enhance the checking ability of the system in complex and variable data environment, providing more reliable technical support for the accurate management of natural resource assets.

[0048] In an optional embodiment, the checking analysis module 13 is also used for: Based on time series analysis technology, trend prediction is performed on the checking results of N consecutive periods to construct a time series model wherein, is the checking deviation rate of the t period, is a random error term; When the predicted deviation rate exceeds the threshold range, early warning information is generated and the checking process optimization is triggered.

[0049] It can be understood that in the natural resource asset verification process, the time dimension of the data contains rich change rules. The module is based on time series analysis technology to predict the trend of the verification results of N consecutive periods (N≥3). Through specific adoption of autoregressive model (AR), moving average model (MA) or combination model, a general time series model expression is constructed wherein, is the verification deviation rate of the t period, reflecting the deviation degree of the verification data of the period from the benchmark data or the expected value, is a random error term, representing random fluctuation factors that cannot be explained by the model. By introducing autocorrelation analysis of historical data, the long-term trend, seasonal change and periodic characteristics of the verification deviation can be effectively captured; In order to realize dynamic monitoring and risk prevention and control of the verification results, an intelligent early warning mechanism is set. When the deviation rate predicted by the model exceeds the pre-set threshold range, it means that there may be potential risks or systematic errors in the current verification process. At this time, the system will generate early warning information in advance and trigger the verification process optimization. The early warning information will be pushed to the relevant person in charge in the form of pop-up window, email, short message and other channels. At the same time, the system automatically starts the verification process optimization scheme, such as adjusting the verification sample size, optimizing the verification area distribution, introducing third-party data cross verification, etc., to reduce the verification error and improve the accuracy and reliability of the natural resource asset verification.

[0050] In an optional embodiment, the system further comprises a data sharing module for establishing a data sharing platform, using distributed database technology to realize centralized storage and distributed management of data, providing data access services for the data collection module 11, knowledge graph processing module 12 and verification analysis module 13 through data interface, and using blockchain technology to record and trace the uploading, accessing and using process of data.

[0051] Among them, the main function of the data sharing module is to establish an efficient and safe data sharing platform, break the data island phenomenon within the system, and at the same time guarantee the security and traceability of data in the sharing process.

[0052] In this embodiment, the data sharing module uses distributed database technology to build a data storage architecture, realizing centralized storage and distributed management of data. This storage method not only guarantees the unity of data, but also improves the efficiency of data storage and access. Through the data interface, the data sharing module provides standardized data access services for the data collection module 11, knowledge graph processing module 12 and verification analysis module 13. Each module can safely and conveniently obtain the required data through the interface according to its own needs, realizing efficient circulation and sharing of data within the system.

[0053] In addition, the data sharing module uses blockchain technology to record and trace the entire process of data uploading, access and use, ensures the integrity and non-tamperability of the data through the distributed ledger and encryption technology of the blockchain, and realizes the responsibility identification and audit of data operation. For example, when the data is abnormal, the source and operation history of the data can be traced through the blockchain record, ensuring the security and credibility of data sharing, and providing a solid guarantee for the stable operation and data collaboration and sharing of the natural resource asset verification system.

[0054] In the embodiment, by collecting multi-source data in real time and standardizing preprocessing, the data format and source barriers are broken, efficient convergence and standardization of data are realized, a solid foundation is laid for subsequent analysis, knowledge graph technology is used to deeply mine data correlation, a structured knowledge network is constructed and dynamically updated and maintained, data cross-overlap and missing report problems are accurately identified, the accuracy and reliability of verification are greatly improved, deviation trends are predicted in advance through time series analysis, the effect of early detection of abnormalities and avoidance of systematic data deviation risks is achieved, an intelligent verification model is constructed based on the knowledge graph, reasoning analysis is performed combined with setting rules and algorithms, automation and intelligentization of the verification process are realized, data abnormalities are detected and warned in time, finally, complex verification results are visualized and professional reports are generated through the geographic information system, and scientific decision-making basis is provided for urban forest resource management in an intuitive and clear manner, and the efficiency and level of resource management are comprehensively improved.

[0055] On the basis of the above-mentioned embodiment, the application further provides a use method of the natural resource asset physical quantity multi-source collaborative verification system, and the use method comprises the following steps: System deployment background: In view of the problems of cross-checking difficulty and abnormality finding lag of multi-source data (forestry statistical data, remote sensing image data and real-time sensor data) of a forestry department, a natural resources department and an ecological environment department, the system is deployed for quarterly verification; Collaborative process of each module: Data acquisition module: S11, connecting a forestry department database to obtain 2022-2023 provincial forest land subcompartment data (tree quantity, tree species type); S12, obtaining 2024Q1 high-resolution images (resolution 0.5m) through unmanned aerial vehicle remote sensing; S13, deploying Internet of Things sensors (collecting temperature and humidity, tree growth every hour) in 100 key monitoring areas, and converting the data into JSON format of WGS84 coordinates; Knowledge graph processing module: S21, extract entities (such as "Pinus massoniana forest - small class A" and "2024Q1 carbon sink") and relationships ("small class A - Pinus massoniana - carbon sink coefficient 0.42"), and construct a dynamic atlas containing 3000+ entities, with daily updates of real-time sensor data; Verification analysis module: S31, based on historical 3-phase (2023Q1-Q3) verification data, use ARIMA(2,1,1) model to predict 2023Q4 deviation rate, the result is 12.3% (threshold 17.5%), determine normal; S32, in 2024Q1 actual verification, through reasoning rules , found that the forestry statistical data of a certain area ( =800) and the remote sensing interpretation data ( =1050) deviate by 31.25%, triggering an abnormal mark; S33, time series model predicts that the deviation rate will reach 19.8% in 2024Q2 (exceeding the 3σ threshold of 17.5%), the system automatically pushes the warning to the relevant departments; S34, mark the above abnormal area (15km²) in red in ArcGIS, generate a verification report, including: Problem cause: the statistical lag is caused by the change of forest rights in this area; Rectification suggestion: supplement the 2024Q1 field sampling survey (sample size increased by 20%).

[0056] In this embodiment, after the system runs for 6 months, the cross-verification efficiency of multi-department data is improved by 60%, the average time to find abnormal problems is shortened from 45 days to 7 days, and the generation time of quarterly verification report is compressed from 5 working days to 8 hours.

[0057] Based on the above embodiments, as Figure 2 shown, the present application also provides a natural resource asset physical quantity multi-source collaborative verification method for supporting the natural resource asset physical quantity multi-source collaborative verification system of the above embodiments, the method comprising: Step 100, acquiring multi-source data and collecting real-time data through Internet of Things sensors, and preprocessing the multi-source data and real-time data by cleaning, format conversion and geographic coordinate system unification.

[0058] Specifically, step 100 includes: Step 110, through the data interface and external business database, use ETL tools to realize the extraction, conversion and loading of multi-source data; Step 220, deploy sensors in resource areas, collect and transmit data in real time through wireless sensor network and integrate multi-source data.

[0059] The multi-source data acquisition unit converts the position data of different data sources into a WGS84 coordinate system uniformly when performing data conversion.

[0060] In step 200, the pre-processed multi-source data and real-time data are used to extract entities and relationships by using a named entity recognition algorithm in natural language processing technology and a conditional random field model in machine learning. Redundant and conflicting information is eliminated by entity alignment technology and calculation of cosine similarity between entities. A resource knowledge graph is constructed. The knowledge graph is stored in a graph database, and a real-time updating mechanism is established to dynamically update the knowledge graph. The knowledge graph is maintained by a quality evaluation mechanism combining manual review and automatic algorithm.

[0061] Specifically, step 200 includes: In step 210, natural language processing and machine learning algorithms are implemented using programming languages and corresponding toolkits to automatically extract entities and relationships from data and integrate knowledge, and a resource knowledge graph is constructed. In step 210, things, spatial ranges, and measurement parameters related to resources are defined as entities, and the belonging relationships between things and spaces, spatial attribute change relationships, and the association relationships between measurement parameters and resources are defined as relationships. The relationships between entities are identified through semantic analysis. In step 220, a knowledge graph quality evaluation mechanism is established to regularly check the accuracy and completeness of the entity relationships in the knowledge graph, correct errors, and update the entity attributes and relationships of the knowledge graph in real time based on newly entered resource data.

[0062] In step 300, based on the constructed knowledge graph, an intelligent verification model is constructed to identify cross-overlapping and missed reporting problems in the input resource data to be verified by using the relationships and rules in the knowledge graph through setting reasoning rules , wherein, is the number of trees in a certain piece of forest land counted by the forestry department, is the theoretical number of trees calculated based on land use by the urban planning department, is a threshold value; and abnormal detection rules are set based on the knowledge graph to perform abnormal detection, wherein, is the forest carbon sink, is the tree growth, is the forest land area, is the theoretical carbon sink calculated based on and , and is a threshold value; and the verification results are visualized and presented by combining geographic information system technology, and a verification report containing data problem information, cause analysis, and rectification suggestions is generated.

[0063] wherein the function is used to calculate the theoretical carbon sink amount, and the specific calculation logic is as follows: First, based on the forest ecosystem carbon sink model, the tree growth amount is determined according to the relationship with the biomass conversion coefficient , the biomass conversion coefficient k is obtained according to the tree species type (such as coniferous forest, broad-leaved forest), tree age and other parameters through the standard lookup table of “Forest Ecosystem Long-term Positioning Observation Method” (LY / T1952-2011); the theoretical biomass , wherein is the annual growth amount of trees per unit area (tons / hm2·year); the carbon content conversion factor is 0.45 (i.e., the average proportion of carbon elements in forest biomass), and the theoretical carbon sink amount , wherein is the forest land area (hm2); the abnormality detection rule , wherein is a threshold value set according to regional carbon sink historical data (such as mean ± 2 times standard deviation), and when the measured forest carbon sink amount deviates from the theoretical value by more than the threshold value, an abnormality warning is triggered.

[0064] Specifically, the step 300 includes: Step 310, constructing an intelligent checking model based on the knowledge graph, inputting the to-be-checked resource data, utilizing the relationships and rules in the knowledge graph, judging whether is established, if established, combining the related relationships and constraint conditions in the knowledge graph to reason about possible problems in the data, identifying cross-overlapping and missing report problems in the data and outputting the checking result; Step 320, setting an abnormality detection rule based on the knowledge graph, judging whether is established, if established, issuing an abnormality warning; Step 330, combining geographic information system technology to visualize the checking result in the form of a map, marking the regions with data problems by different colors, and generating a checking report containing specific conditions of cross-overlapping and missing report, cause analysis and rectification suggestions.

[0065] Wherein, when constructing the intelligent checking model, the historical checking data in the knowledge graph is trained by using a machine learning algorithm to optimize the identification rules of the intelligent checking model.

[0066] Further, the step 330 further includes: Step 331, the verification result data (such as coordinate points, surface area) is matched with the basic geographic data (administrative division, remote sensing image) in the geographic information system (such as ArcGIS, QGIS) in the spatial coordinate system (using WGS84 coordinate system), and the data and map are accurately matched through spatial overlay analysis; Step 332, set up a three-level early warning color system-green (deviation ≤10%), yellow (10%<deviation ≤30%), red (deviation >30%), and automatically fill the corresponding area color according to the calculation results of ; Step 333, based on the visualization labeling results and abnormal detection data, automatically generate verification reports according to the modular process.

[0067] Among them, step 333 also includes: Data problem extraction: automatically extract data cross-over area coordinates, missing data list and abnormal detection results from the knowledge graph; Reason analysis generation: based on the entity relationship in the knowledge graph (such as "tree species-growth amount-carbon sink amount" correlation), generate problem reasons (such as "the number of trees in a certain area is missing due to misjudgment of land use type") through rule reasoning; Rectification suggestion generation: according to the problem type, match the preset rectification strategy library (such as data supplement, field verification, model parameter optimization), and generate specific rectification measures; Report template output: use DOCX template engine to automatically layout PDF report by generating map visualization results, problem list, reason analysis and rectification suggestion.

[0068] Further, step 300 also includes: Step 340, based on time series analysis technology, trend prediction is made on the verification results of continuous N periods, and time series model is constructed, wherein, is the verification deviation rate of the t period, and is the random error term; Step 350, when the predicted deviation rate exceeds the threshold range, early warning information is generated and the verification process is optimized.

[0069] Among them, N is set according to the periodicity of data, such as N=4 for quarterly verification.

[0070] Step 400, establish a data sharing platform, use distributed database technology to realize centralized storage and distributed management of data, provide data access services through data interface, and record and trace the uploading, access and use process of data by using blockchain technology.

[0071] In the embodiment, by collecting multi-source data in real time and standardizing preprocessing, the data format and source barriers are broken, efficient data aggregation and standardization are realized, a solid foundation is laid for subsequent analysis, knowledge graph technology is used to deeply mine data correlation, a structured knowledge network is constructed and dynamically updated and maintained, data cross-overlapping and missing report problems are accurately identified, the accuracy and reliability of the verification are greatly improved, the deviation trend is predicted in advance through time series analysis, the effect of early discovery of abnormalities and avoidance of systematic data deviation risk is achieved, an intelligent verification model is constructed based on the knowledge graph, reasoning analysis is carried out combined with setting rules and algorithms, automation and intelligentization of the verification process are realized, data abnormalities are detected and warned in time, finally, the complex verification result is visualized and presented through the geographic information system and a professional report is generated, and scientific decision-making basis is intuitively and clearly provided for urban forest resource management, and the efficiency and level of resource management are comprehensively improved.

[0072] On the basis of the above-mentioned embodiments, as shown in Figure 3 The electronic device comprises: At least one processor 22, at least one memory 21, a communication interface 23 and a communication bus 24, the processor 22 is in communication connection with the memory 21; In the embodiment, the memory 21 can be implemented in any appropriate manner, for example: the memory 21 can be a read-only memory, a mechanical hard disk, a solid state disk or a U disk, etc.; the memory 21 is used to store executable instructions executed by the processor; In the embodiment, the processor 22 can be implemented in any appropriate manner, for example, the processor 22 can take the form of, for example, a microprocessor or a processor and a computer readable medium storing computer readable program code (such as software or firmware) executable by the (micro) processor, logic gates, switches, application specific integrated circuits (ASIC), programmable logic controllers and embedded microcontrollers, etc.; the processor is used to execute the executable instructions to realize the natural resource asset physical quantity multi-source collaborative verification method as described above.

[0073] On the basis of the above-mentioned embodiments, the present application further provides a computer readable storage medium, the computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the natural resource asset physical quantity multi-source collaborative verification method as described above.

[0074] Those skilled in the art can clearly understand that the modules and method steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0075] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working processes of the above-described system, device and module can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

[0076] In several embodiments provided in the present application, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the above-described system embodiments are merely illustrative, for example, the division of the modules is only a logical function division, and actual implementation can have another division manner, for example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed each other can be through some interface, indirect coupling or communication connection between the system or device, which can be electrical, mechanical or other forms.

[0077] The modules described as separate components can or can not be physically separated, and the components displayed as modules can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0078] In addition, the functional modules in each embodiment of the present application can be integrated into a processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0079] If the functions are realized in the form of software function modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the part of the present application that essentially contributes to the prior art or the part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory server, a random access memory server, a magnetic disk or an optical disk, and various program instruction storage media.

[0080] In addition, it should be noted that the combination of the technical features in the present case is not limited to the combination of the claims or the combination of the embodiments. All technical features disclosed in the present case can be freely combined or combined, unless they are contradictory to each other.

[0081] It should be noted that the above only lists specific embodiments of the present application. Obviously, the present application is not limited to the above embodiments, and there are many similar changes. All modifications directly derived or inferred from the disclosure of the present application by those skilled in the art shall fall within the scope of the present application.

[0082] The above is only a preferred embodiment of the present application, and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A natural resource asset physical quantity multi-source collaborative verification system, characterized in that, The method comprises the following steps: a data acquisition module is used to acquire multi-source data of forestry departments, urban planning departments and environmental protection departments, and real-time data collected through Internet of Things sensors, and to preprocess the multi-source data and real-time data through cleaning, format conversion and geographic coordinate system unification; a knowledge graph processing module is used to extract entities and relationships through a named entity recognition algorithm in natural language processing technology and a conditional random field model in machine learning on the multi-source data and real-time data preprocessed by the data acquisition module, eliminate redundant and conflicting information through entity alignment technology and calculation of cosine similarity between entities, construct a forest resource knowledge graph, store the knowledge graph in a graph database, and establish a real-time updating mechanism to dynamically update the knowledge graph; The verification analysis module is configured to construct an intelligent verification model based on the knowledge graph constructed by the knowledge graph processing module. For inputted forest resource data to be verified, the intelligent verification model is configured to utilize the relationships and rules in the knowledge graph to identify cross-overlapping and missed-reporting problems in the data by setting reasoning rules , wherein is the number of trees in a piece of forest land counted by the forestry department, is the theoretical number of trees calculated by the urban planning department based on land use, is a threshold value, and abnormal detection rules are set based on the knowledge graph is the forest carbon sink, is the tree growth, is the forest land area, is the theoretical carbon sink calculated according to and , and is a threshold value; and the verification result is visualized and presented by combining the geographic information system technology, and a verification report containing data problem conditions, cause analysis and rectification suggestions is generated. wherein the verification analysis module is further used to: Based on time series analysis technology, the trend of the verification results of N continuous periods is predicted, and a time series model is constructed wherein, is the verification deviation rate of the tth period, is a random error term; when the prediction deviation rate exceeds the threshold range, generate early warning information and trigger the verification process optimization.

2. The natural resource asset stock multi-source collaborative verification system of claim 1, wherein, The data acquisition module comprises a multi-source data acquisition unit and a sensor data acquisition unit: The multi-source data acquisition unit is used to connect with external business databases through data interfaces, and to realize extraction, conversion and loading of multi-source data through ETL tools; The sensor data acquisition unit is used to deploy sensors in resource areas, and to collect and transmit data to the data acquisition module and multi-source data integration through a wireless sensor network in real time.

3. The natural resource asset stock multi-source collaborative verification system of claim 2, wherein, When the multi-source data acquisition unit performs data conversion, the position data of different data sources is uniformly converted into the WGS84 coordinate system.

4. The natural resource asset stock multi-source collaborative verification system of claim 1, wherein, The knowledge graph processing module comprises a graph construction unit and a graph maintenance unit: The graph construction unit is used to realize natural language processing and machine learning algorithms through programming languages and corresponding toolkits, automatically complete entity and relationship extraction and knowledge fusion of input data of the data acquisition module, and construct a resource knowledge graph; wherein the graph construction unit defines things, spatial ranges and measurement parameters related to resources as entities, defines the belonging relationship between things and spaces, the spatial attribute change relationship, and the association relationship between measurement parameters and resources as relationships, and identifies the relationship between entities through semantic analysis; The graph maintenance unit is used to establish a knowledge graph quality evaluation mechanism, regularly check the accuracy and integrity of entity relationships in the knowledge graph, correct error information, and update entity attributes and relationships of the knowledge graph in real time according to newly entered resource data.

5. The natural resource asset stock multi-source collaborative verification system of claim 1, wherein, The verification analysis module comprises a verification model unit, an anomaly detection unit and a result presentation unit: The verification model unit is configured to construct an intelligent verification model based on a knowledge graph constructed by the knowledge graph processing module, input resource data to be verified, and use relationships and rules in the knowledge graph to determine whether the relationships and rules are valid. If the relationships and rules are valid, the verification model unit combines related relationships and constraint conditions in the knowledge graph to reason about possible problems in the data, identifies cross-overlapping and missed-reporting problems in the data, and outputs a verification result. The anomaly detection unit is used to set anomaly detection rules based on the knowledge graph and to determine... If the condition is met, an abnormal warning will be issued. The result presentation unit is used to visualize the verification results of the verification model unit in the form of a map through geographic information system technology, mark the areas with data problems through different colors, and generate a verification report containing specific conditions of data cross-overlap and missed reports, cause analysis and rectification suggestions.

6. The natural resource asset stock multi-source collaborative verification system of claim 5, wherein, When constructing an intelligent verification model, the verification model unit trains historical verification data in the knowledge graph through a machine learning algorithm, and optimizes the identification rules of the intelligent verification model.

7. The natural resource asset stock multi-source collaborative verification system of claim 1, wherein, The system also comprises a data sharing module for establishing a data sharing platform, using distributed database technology to realize centralized storage and distributed management of data, providing data access services for the data collection module, knowledge graph processing module and verification analysis module through a data interface, and using blockchain technology to record and trace the uploading, access and use of data.

8. A natural resource asset physical quantity multi-source collaborative verification method, characterized in that, Comprise: Acquire multi-source data and real-time data collected through Internet of Things sensors, and preprocess the multi-source data and real-time data by cleaning, format conversion and geographic coordinate system unification; For the preprocessed multi-source data and real-time data, extract entities and relationships using the named entity recognition algorithm in natural language processing technology and the conditional random field model in machine learning, eliminate redundant and conflicting information by entity alignment technology and calculate the cosine similarity between entities, construct a resource knowledge graph, store the knowledge graph using a graph database, and establish a real-time updating mechanism to dynamically update the knowledge graph; Based on the constructed knowledge graph, an intelligent verification model is constructed. For the input to be verified resource data, the relationship and rules in the knowledge graph are used to set reasoning rules , identify cross-overlap and false reporting problems in the data, wherein, is the number of trees in a piece of forest land counted by the forestry department, is the theoretical bearing tree number calculated by the urban planning department based on land use, is the set threshold value; and based on the knowledge graph, an anomaly detection rule is set for anomaly detection, wherein, is the forest carbon sink, is the tree growth, is the forest area, is the theoretical carbon sink calculated according to and , and is the set threshold value; and the verification result is visualized and presented by combining the geographic information system technology, and a verification report containing data problem situation, cause analysis and rectification suggestion is generated. Based on time series analysis technology, the trend of the verification results of N continuous periods is predicted, and a time series model is constructed wherein, is the verification deviation rate of the tth period, is a random error term; when the predicted deviation rate exceeds the threshold range, early warning information is generated and the verification process is optimized.

9. An electronic device, comprising: The electronic device comprises: A processor, a memory, the memory being in communication connection with the processor; The memory is used to store executable instructions executed by the processor, and the processor is used to execute the executable instructions to realize the multi-source collaborative verification method of natural resource asset physical quantity according to claim 8.

10. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the multi-source collaborative verification method of natural resource asset physical quantity according to claim 8.

Images