Farmland design file compliance evaluation method, system and device and storage medium
The compliance of farmland design documents is automatically assessed through natural language processing and knowledge graph technology, which solves the problem of low efficiency of traditional manual review and achieves efficient and accurate compliance assessment.
Patent Information
- Application Number
- CN202510879094.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-27
- Publication Date
- 2025-09-26
AI Technical Summary
The compliance assessment of traditional farmland design documents relies on manual review, which is inefficient, time-consuming, and susceptible to subjective factors. It is difficult to comprehensively and accurately identify potential conflicts between design parameters and environmental constraints.
Natural language processing technology is used to identify construction objects and their parameters in design documents, and pre-built knowledge graphs are used for intelligent compliance assessment, and graph neural networks are used to perform parameter compliance analysis.
Significantly shorten the review cycle, significantly improve assessment efficiency, reduce manual dependence, improve the accuracy and coverage of assessment results, and effectively identify potential conflicts between design parameters and environmental conditions.
Smart Images
Figure CN120705969A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of data processing technology, and specifically relates to a compliance assessment method, system, device and storage medium for farmland design documents. Background Art
[0002] In the field of agricultural engineering, the design of farmland construction projects (such as irrigation systems, terraces, roads, and shelterbelts) must strictly adhere to relevant technical specifications, environmental protection requirements, and land use policies. Traditionally, compliance assessments of farmland design documents (such as design specifications, drawings, and reports) have relied heavily on manual review. Reviewers manually extract construction objects (such as canals, pumping stations, and ridges) and their key parameters (such as size, materials, and layout) from the design documents. Reviewers then compare these with the complex standard clauses that may be scattered across various regulatory documents, taking into account the specific environmental parameters of the project site (such as soil type, slope, hydrological conditions, and proximity to protected areas). This process is inefficient, time-consuming, and susceptible to subjective factors. It also makes it difficult to comprehensively and accurately identify potential conflicts between design parameters and environmental constraints.
[0003] Although natural language processing and knowledge graph technologies have been applied in information extraction and intelligent decision-making in other fields, their deep integration for automated and intelligent evaluation of the compliance of construction object parameters in farmland design documents under specific plot environments remains a technical difficulty that needs to be addressed urgently. Summary of the Invention
[0004] In view of the above-mentioned deficiencies in the prior art, the present invention provides a compliance assessment method, system, device and storage medium for farmland design documents to solve the above-mentioned technical problems.
[0005] In a first aspect, the present invention provides a compliance assessment method for farmland design documents, comprising: Obtaining a farmland design file, and identifying a construction object and design parameters of the construction object from the farmland design file using natural language processing technology; Obtain the environmental parameters of the plot corresponding to the farmland design file; Based on the environmental parameters, a pre-built knowledge graph is used to perform a compliance assessment on the design parameters of the construction object.
[0006] In an optional embodiment, the farmland design file includes: Farmland design drawings and construction project documents.
[0007] In an optional embodiment, using natural language processing technology to identify construction objects and design parameters of the construction objects from the farmland design file includes: Extracting layer information from the farmland design drawing, wherein the layer information includes ditches, roads, and field boundaries; By constructing a spatial index, the geographical parameters of ditches, roads and fields are extracted, and the geographical parameters include coordinates, area and slope; Entities are extracted from the construction project file using a BERT model. The entities include construction objects and design parameters of the construction objects. The design parameters include specifications, dimensions, locations, construction materials, construction processes, and construction time.
[0008] In an optional embodiment, the method further comprises: Establish associations between construction objects based on their functional categories, locations, and specifications.
[0009] In an optional embodiment, based on the environmental parameters, a compliance assessment is performed on the design parameters of the construction object using a pre-built knowledge graph, including: Group related construction objects into the same group; Convert each construction object and corresponding design parameter in the group into a construction object coding vector to obtain multiple construction object coding vectors; Taking the multiple construction object encoding vectors and the environmental parameters as input parameters, a graph neural network is used to evaluate the compliance of the input parameters based on the knowledge graph.
[0010] In an optional embodiment, the knowledge graph includes: Construction objects include canals, pumping stations, drainage ditches, sedimentation tanks, and agricultural roads; Functional category entities, including irrigation facilities, drainage facilities, and transportation facilities; Environmental parameter entities, including soil type, slope class, precipitation zones, and field topology; Constraint rule entities include size constraints, spacing constraints, material rules, and process rules.
[0011] In an optional embodiment, the plurality of construction object encoding vectors and the environmental parameters are used as input parameters, and compliance of the input parameters is evaluated based on the knowledge graph using a graph neural network, including: Query related constraint rule entities from the knowledge graph based on input parameters; Embed logical reasoning rules as constraints into the pre-trained graph neural network to obtain a compliance classifier. Input parameters and constraint rule entities are input into the compliance classifier to obtain a compliance assessment result output by the compliance classifier.
[0012] In a second aspect, the present invention provides a compliance assessment system for farmland design documents, comprising: a file recognition module for obtaining a farmland design file and identifying a construction object and its design parameters from the farmland design file using natural language processing technology; Environmental detection module, used to obtain environmental parameters of the plot corresponding to the farmland design file; The compliance assessment module is used to perform compliance assessment on the design parameters of the construction object based on the environmental parameters using a pre-built knowledge graph.
[0013] According to a third aspect, a device is provided, comprising: a memory for storing the compliance assessment program of the farmland design document; A processor is used to implement the steps of the compliance assessment method for farmland design documents provided in the first aspect when executing the compliance assessment program for the farmland design documents.
[0014] In a fourth aspect, a computer-readable storage medium is provided, on which a compliance assessment program for a farmland design file is stored. When the compliance assessment program for a farmland design file is executed by a processor, the steps of the compliance assessment method for a farmland design file provided in the first aspect are implemented.
[0015] The beneficial effects of the present invention lie in the following: the compliance assessment method, system, equipment, and storage medium for farmland design documents provided herein automatically identify construction objects and their parameters in design documents through natural language processing technology, and utilize knowledge graphs for intelligent assessment, significantly shortening the review cycle, significantly improving efficiency, and reducing manual reliance. The knowledge graph integrates regulatory standards, environmental constraints, and professional experience, dynamically linking plot-specific environmental parameters with design parameters for comprehensive judgment, effectively avoiding manual omissions and improving the accuracy and coverage of assessment results. It can systematically identify potential compliance conflicts between design parameters and environmental conditions, providing a precise basis for design optimization and reducing engineering violations and ecological risks at the source. This invention provides a standardized assessment process that integrates advanced technologies (NLP and knowledge graphs), promoting the upgrade of farmland construction design review towards intelligent, data-driven approaches. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.
[0017] Figure 1 is a schematic flow chart of a method according to an embodiment of the present invention.
[0018] Figure 2 FIG. 4 is a schematic block diagram of a system according to an embodiment of the present invention.
[0019] Figure 3 A schematic structural diagram of a device provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0020] In order to enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.
[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention.
[0022] The compliance assessment method for farmland design files provided in the embodiment of the present invention is executed by a computer device. Accordingly, the compliance assessment system for farmland design files runs in the computer device.
[0023] Figure 1 is a schematic flow chart of a method according to an embodiment of the present invention. Figure 1 The execution entity can be a compliance assessment system for farmland design documents. According to different needs, the order of the steps in the flowchart can be changed, and some steps can be omitted.
[0024] like Figure 1 As shown, the method includes: S1. Obtain farmland design files and use natural language processing technology to identify construction objects and design parameters of construction objects from the farmland design files; S2. Obtain environmental parameters of the plot corresponding to the farmland design file; S3. Based on the environmental parameters, use a pre-built knowledge graph to perform a compliance assessment on the design parameters of the construction object.
[0025] In an embodiment of the present invention, based on step S1, a possible embodiment will be given below to illustrate its specific implementation scheme in a non-limiting manner.
[0026] Farmland design documents include: farmland design drawings and construction project documents.
[0027] S101. Extracting layer information from the farmland design drawing, the layer information including ditches, roads and field boundaries; extracting geographic parameters of the ditches, roads and fields by constructing a spatial index, the geographic parameters including coordinates, area and slope.
[0028] When processing farmland design drawings, we first use vector graphics parsing techniques to extract layer information. For CAD-formatted drawings, we use the AutoCAD Graphics Development Interface (Object ARX) or open-source libraries such as Libre DWG to traverse the entity objects in the graphics database and accurately identify ditches, roads, and field boundaries based on layer names, color coding, and line types. For GIS-formatted drawings (such as GeoJSON and Shape files), we use the GDAL / OGR library to parse the vector feature collections within the files and select target layers based on the category fields in the attribute table.
[0029] After extracting ditches, roads, and field boundaries, a spatial index is constructed to improve the efficiency of geographic parameter extraction. Using a quadtree or R-tree spatial index structure, the minimum bounding rectangle (MBR) of a vector feature is used as an index node. Through recursive spatial partitioning or hierarchical organization of spatial regions, rapid feature retrieval and location are achieved. During the geographic parameter extraction phase, coordinate parameters are directly derived from the geometric coordinate pairs of the vector features. Area parameters are accurately calculated using Green's Theorem by calculating the vector cross product of the polygon boundary coordinate sequence. Slope parameters are based on digital elevation model (DEM) data. Using GIS software such as Arc GIS or QGIS, slope extraction tools are used to calculate the surface slope of the area corresponding to each field or linear feature (ditch, road). This algorithm is based on the first-order derivative approximation of the elevation data.
[0030] S102. Utilize the BERT model to extract entities from the construction project file, wherein the entities include construction objects and design parameters of the construction objects, and the design parameters include specifications, dimensions, location, construction materials, construction process, and construction time.
[0031] For construction project documents, data preprocessing begins with text cleaning (removing HTML tags, special characters, and stop words) and word segmentation (using the Jieba word segmentation tool or the Han LP natural language processing toolkit) to convert the document text into a format suitable for model input. A pretrained BERT Chinese model (such as RoBERTa-wwm-ext released by the Harbin Institute of Technology iFlytek Joint Laboratory) is used. Multi-layer bidirectional long short-term memory (BiLSTM) and conditional random field (CRF) layers are added to the model to construct an entity recognition model. The BiLSTM layer captures the contextual semantics of the text, while the CRF layer optimizes entity boundary recognition by considering dependencies between labels. During the model training phase, a corpus of construction project documents containing construction objects and design parameters was collected and annotated. The BIO (Begin, Inside, Outside) annotation system was used to label each character of these objects and design parameters. Multiple rounds of iterative training were performed on the annotated corpus using the Adam optimizer with a cross-entropy loss function as the optimization objective. During model inference, the preprocessed construction project document text was input into the trained model. The Softmax function calculated the probability of each character corresponding to a label. After decoding through the CRF layer, the final entity recognition results were output, enabling accurate extraction of construction objects (such as ditches, roads, and field ancillary facilities) and their design parameters (specifications, dimensions, location, construction materials, construction techniques, and construction time).
[0032] In an embodiment of the present invention, based on step S2, a possible embodiment will be given below to illustrate its specific implementation scheme in a non-limiting manner.
[0033] In the process of acquiring environmental parameters of farmland plots, multi-source data acquisition and fusion technology is used to achieve accurate acquisition of parameters such as humidity, soil type, temperature, and compaction degree. For soil moisture parameters, time domain reflectometry (TDR) or frequency domain reflectometry (FDR) is used for in-situ measurement. Within the target plot, random sampling or grid sampling is used to arrange the sampling points according to the size and shape of the field. Usually, one sampling point is set every 1-5 hectares, and each sampling point is measured at different soil depths such as 0-30cm, 30-60cm, and 60-90cm. The instrument calculates the volumetric water content of the soil based on the correlation between the soil dielectric constant and water content by emitting high-frequency electromagnetic pulses. At the same time, combined with remote wireless sensor networks (WSN), sensor nodes are deployed in farmland to collect data in real time and transmit it to the data center via LoRa or NB-IoT communication technology to achieve dynamic monitoring of moisture parameters. Soil type parameters are obtained using a combination of laboratory analysis and remote sensing interpretation. During field sampling, GPS tracking is used to record the precise coordinates of sampling points. Topsoil samples (0-20 cm) are collected and brought back to the laboratory for physical and chemical analysis, including particle size analysis (laser particle size analyzer), organic matter content determination (potassium dichromate oxidation method), and pH determination (potentiometric method). Soil types are then classified according to international soil texture classification standards (such as the Kachinsky system and the USDA system). Furthermore, multispectral remote sensing imagery (such as Landsat 8OLI and Sentinel-2MSI) is used for supervised classification (support vector machine (SVM) and random forest (RF) algorithms) or unsupervised classification (ISODATA algorithm), combined with ground sampling data for training and validation, to map the spatial distribution of soil types. Soil temperature parameters are measured using buried soil temperature sensors. DS18B20 digital temperature sensors, which offer high precision and strong anti-interference capabilities, are selected. The sensors are buried vertically in the soil at varying depths (e.g., 5cm, 10cm, and 15cm). Data is collected at 10-30 minute intervals, stored in a data logger, and uploaded to the cloud via a GPRS module. Surface temperature is inverted using thermal infrared remote sensing imagery. Combined with air temperature data from a meteorological station, soil temperature is spatially interpolated using an energy balance model to determine the temperature distribution across the entire plot. Soil compaction parameters are obtained by combining the knife-ring method with a soil compaction meter. The knife-ring method randomly selects measurement points within the plot and collects a certain volume of undisturbed soil samples. By measuring the wet and dry weights of the samples and the knife-ring volume, the soil bulk density is calculated, and soil compaction is subsequently estimated. A soil compaction meter (such as the Eijkelkamp cone index meter) presses a cone probe into the soil at a constant rate, measuring the resistance of the probe to the soil. Soil compaction data is obtained based on the empirical relationship between resistance and compaction. Furthermore, ground-penetrating radar (GPR) technology analyzes the propagation characteristics of radar waves in the soil to invert the soil's internal structure and compaction level, enabling rapid, non-invasive testing of soil compaction. Finally, the multi-source measurement data is imported into a geographic information system (GIS) such as ArcGIS, where kriging or inverse distance weighted (IDW) interpolation methods are used to generate spatial distribution maps of the plot's environmental parameters, providing comprehensive data support for farmland design and construction.
[0034] In an embodiment of the present invention, based on step S3, a possible embodiment will be given below to illustrate its specific implementation scheme in a non-limiting manner.
[0035] S301. Construction data organization.
[0036] Establish relationships between construction objects based on their functional categories, locations, and specifications. Group construction objects with these relationships into the same group. Each construction object and its corresponding design parameters within the group are converted into a construction object encoding vector, resulting in multiple construction object encoding vectors.
[0037] During the construction data collation phase, a structured data model based on an ontology is first constructed. Uniform Resource Identifiers (URIs) are used to uniquely identify construction objects. Semantic associations between objects are established by defining functional category, location, and size ontologies. Specifically, spatial topology analysis techniques within the Geographic Information System (GIS) are used to calculate spatial distances, adjacency, and inclusion relationships between construction objects. Based on the hierarchical structure of the functional category ontology (e.g., irrigation facilities → canals, pumping stations), semantic similarity algorithms (such as cosine similarity based on Word2Vec) are used to identify functional associations. During the encoding vector conversion process, the encoder module of the Transformer architecture is used. The design parameters of each building object (such as dimensions, location, and construction materials) are converted into a sequence of word vectors. A multi-head attention mechanism is used to capture long-range dependencies between these parameters. To enhance the representation of spatial information, the longitude and latitude coordinates of the building object are encoded into position embedding vectors in the form of sine and cosine functions and fused with the parameter word vectors. Finally, a multi-layer perceptron (MLP) is used to map this fused vector into a fixed-dimensional building object encoding vector. This vector not only preserves the semantic information of the design parameters but also incorporates the inter-object associations.
[0038] S302. Build a knowledge graph.
[0039] The entities of the knowledge graph include: Construction objects include canals, pumping stations, drainage ditches, sedimentation tanks, and agricultural roads; Functional category entities, including irrigation facilities, drainage facilities, and transportation facilities; Environmental parameter entities, including soil type, slope class, precipitation zones, and field topology; Constraint rule entities include size constraints, spacing constraints, material rules, and process rules.
[0040] The types of relationships between entities include: hasFunction: Constructs a relationship between an object and a functional category, for example, a canal has an irrigation function.
[0041] hasMaterial: Constructs a relationship between an object and its material.
[0042] locatedIn: The building object is located in a certain location (field).
[0043] hasEnvironment: The environmental parameters of the field.
[0044] hasRule: The compliance rule associated with a functional category or material. For example, a canal for an irrigation function must meet certain size rules.
[0045] ruleCondition: The conditions of the rule, such as environmental conditions (such as slope) and design parameters (such as depth) associated with the rule.
[0046] Triple example: (drain, hasFunction, irrigation), (drain, hasMaterial, concrete), (irrigation, hasRule, minimum depth rule), (minimum depth rule, ruleCondition, "depth>=0.5m"), (minimum depth rule, affectedBy, slope) # Indicates that the rule is affected by the slope, and the depth requirement may change with the slope.
[0047] The knowledge graph was constructed using a hybrid model, combining top-down ontology design with bottom-up data extraction. The Protégé tool was used to construct a domain ontology, clearly defining the attributes and relationships of four entity types: construction objects, functional categories, environmental parameters, and constraint rules. Regarding entity extraction, construction object and functional category entities were directly extracted from the collated construction data. Environmental parameter entities were converted from discrete measurement point data into field-level attribute data using spatial interpolation algorithms (such as Kriging interpolation). Constraint rule entities were extracted from construction specification documents using natural language processing techniques, with key rule statements extracted and structured using rule template matching and semantic role labeling. Relationship extraction utilizes a semi-supervised learning approach based on a graph convolutional network (GCN). Labeled triples are used as seed data, and pseudo-labels are generated using a graph diffusion algorithm. The GCN model is then trained to identify semantic relationships between entities. To handle complex rule relationships (such as ruleCondition and affectedBy), a rule engine (such as Drools) is introduced to parse rule conditions and convert them into directed edges in the knowledge graph. Finally, the knowledge graph is stored in the Neo4j graph database, and efficient entity retrieval and relationship traversal are achieved using the Cypher query language.
[0048] S303. Take the multiple construction object encoding vectors and the environmental parameters as input parameters, and use the graph neural network to evaluate the compliance of the input parameters based on the knowledge graph.
[0049] (1) Query relevant constraint rule entities from the knowledge graph based on input parameters.
[0050] A query optimization strategy based on semantic similarity was designed. After converting input parameters into vector representations, knowledge graph embedding technologies (such as TransE and ComplEx) were used to calculate the similarity between the parameter vector and the rule entity vector, prioritizing the search for constraint rules with high similarity. To improve query efficiency, an index structure was established within the graph database, creating composite indexes based on key attributes of rule entities (such as rule type and applicable context) to enable rapid filtering and location.
[0051] (2) The logical reasoning rules are embedded as constraints into the pre-trained graph neural network to obtain a compliance classifier.
[0052] The model uses a Graph Attention Network (GAT) as its underlying architecture, combined with a Markov Logic Network (MLN) to embed logical reasoning rules into the model. Specifically, constraint rules are converted into first-order logic formulas, and weights are assigned to each rule. The logical rules and graph neural network parameters are jointly optimized by minimizing the weighted soft clause loss function. During training, an adversarial training strategy is employed, introducing an adversarial network to learn characteristic patterns of rule violations, enhancing the model's ability to identify complex compliance scenarios.
[0053] The compliance classifier uses a fusion architecture of the Graph Attention Network (GAT) and the Markov Logic Network (MLN). The system architecture is divided into three core components: Graph feature extraction layer: The GAT network captures the complex relationships between entities in the knowledge graph through a multi-head attention mechanism and automatically learns the importance weights of nodes. Logical rule embedding layer: converts constraint rules into differentiable logical expressions, and implements joint training of logical rules and neural networks through weighted soft clause loss function Adversarial training module: Introducing the Generative Adversarial Network (GAN) framework, the generator attempts to generate samples that violate the rules, while the discriminator learns to distinguish between compliant and non-compliant samples. Through adversarial game, the model's generalization ability for complex compliance scenarios is enhanced. The training process uses a multi-task learning strategy to simultaneously optimize classification accuracy and logical consistency. Dynamic adjustment of the learning rate and batch normalization techniques ensure the stability and convergence speed of model training.
[0054] (3) Inputting both the input parameters and the constraint rule entity into the compliance classifier to obtain the compliance assessment result output by the compliance classifier.
[0055] A hierarchical reasoning mechanism was designed. First, a graph neural network was used to fuse the features of input parameters and constraint rules, calculating the association scores between entities. A rule engine was then used to logically verify the association scores and generate compliance assessment results based on the rule conditions. To enhance the interpretability of the assessment, a visualization technique based on attention weights was employed to display the association paths between input parameters and rule entities, quantifying the contribution of each factor to the assessment results. Finally, model parameters were optimized through multiple rounds of cross-validation to ensure the accuracy and reliability of the compliance assessment results.
[0056] Design a hierarchical reasoning mechanism to achieve efficient compliance assessment: Feature fusion layer: Calculates the correlation strength between input parameters and constraint rules through graph neural networks to generate a correlation score matrix; Rule execution layer: uses a rule engine to perform logical verification and converts correlation scores into interpretable rule execution paths; Result generation layer: Based on attention weight visualization technology, it generates an evaluation report containing the parameter-rule association path and quantifies the contribution of each factor to the evaluation results.
[0057] To improve the interpretability of the evaluation, the system provides two visualization methods: Association path diagram: shows the reasoning path from input parameters to constraint rules; Contribution heat map: intuitively presents the influence weight of each parameter on the final evaluation results.
[0058] A five-fold cross-validation strategy was adopted in the model evaluation phase, and indicators such as precision, recall, and F1 score were used to comprehensively evaluate the model performance.
[0059] In some embodiments, the farmland design document compliance assessment system may include multiple functional modules composed of computer program segments. The computer program of each program segment in the farmland design document compliance assessment system may be stored in a memory of a computer device and executed by at least one processor to perform (see Figure 1 Description) Function of compliance assessment of farmland design documents.
[0060] In this embodiment, the compliance assessment system for farmland design documents can be divided into multiple functional modules according to the functions it performs, such as Figure 2 As shown. The module referred to in the present invention refers to a series of computer program segments that can be executed by at least one processor and can perform fixed functions, which are stored in a memory. In this embodiment, the functions of each module will be described in detail in subsequent embodiments.
[0061] a file recognition module for obtaining a farmland design file and identifying a construction object and its design parameters from the farmland design file using natural language processing technology; Environmental detection module, used to obtain environmental parameters of the plot corresponding to the farmland design file; The compliance assessment module is used to perform compliance assessment on the design parameters of the construction object based on the environmental parameters using a pre-built knowledge graph.
[0062] Figure 3 The compliance assessment method of the farmland design document provided for the embodiment of the present application can be applied to equipment. Those skilled in the art will understand that the device structure involved in the embodiment of the present invention does not constitute a limitation on the device, and the device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. In the embodiment of the present invention, the device includes but is not limited to a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of the present application described and / or required herein.
[0063] The device 300 may include a processor 310, a memory 320, and a communication unit 330. These components communicate via one or more buses. Those skilled in the art will appreciate that the server structure shown in the figure does not limit the present invention. The server structure may be a bus structure or a star structure, and may include more or fewer components than shown, or combine certain components, or arrange the components differently.
[0064] The memory 320 can be used to store execution instructions of the processor 310. The memory 320 can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk. When the execution instructions in the memory 320 are executed by the processor 310, the device 300 can perform some or all of the steps in the above-described method embodiments.
[0065] The processor 310 is the control center of the storage device, which uses various interfaces and lines to connect various parts of the entire electronic device. It executes various functions of the electronic device and / or processes data by running or executing software programs and / or modules stored in the memory 320, and calling data stored in the memory. The processor can be composed of an integrated circuit (IC), for example, it can be composed of a single packaged IC, or it can be composed of multiple packaged ICs with the same or different functions. For example, the processor 310 can only include a central processing unit (CPU). In an embodiment of the present invention, the CPU can be a single computing core or multiple computing cores.
[0066] The communication unit 330 is configured to establish a communication channel so that the storage device can communicate with other devices, receive user data sent by other devices, or send user data to other devices.
[0067] The present invention also provides a computer storage medium, wherein the computer storage medium may store a program that, when executed, may include some or all of the steps of each embodiment provided by the present invention. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).
[0068] Those skilled in the art will clearly understand that the techniques in the embodiments of the present invention can be implemented using software and a necessary general-purpose hardware platform. Based on this understanding, the technical solutions in the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, among other media capable of storing program code, and includes instructions for causing a computer device (which can be a personal computer, a server, or a second device, a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention.
[0069] In this specification, the same or similar parts between the various embodiments can be referred to each other. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method embodiment.
[0070] In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of the modules is merely a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of systems or modules, and can be electrical, mechanical or other forms.
[0071] The modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules may be selected to achieve the purpose of the present embodiment according to actual needs.
[0072] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.
[0073] Although the present invention has been described in detail with reference to the accompanying drawings and in conjunction with preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, persons of ordinary skill in the art may make various equivalent modifications or substitutions to the embodiments of the present invention, and such modifications or substitutions shall be within the scope of the present invention. Any changes or substitutions that can be easily conceived by persons skilled in the art within the technical scope disclosed in the present invention shall be within the scope of protection of the present invention.
Claims
1. A compliance assessment method for farmland design documents, characterized in that: include: Obtaining a farmland design file, and identifying a construction object and design parameters of the construction object from the farmland design file using natural language processing technology; Obtain the environmental parameters of the plot corresponding to the farmland design file; Based on the environmental parameters, a pre-built knowledge graph is used to perform a compliance assessment on the design parameters of the construction object.
2. The method according to claim 1, characterized in that The farmland design documents include: Farmland design drawings and construction project documents.
3. The method according to claim 2, characterized in that Using natural language processing technology to identify construction objects and design parameters of the construction objects from the farmland design file, including: Extracting layer information from the farmland design drawing, wherein the layer information includes ditches, roads, and field boundaries; By constructing a spatial index, the geographical parameters of ditches, roads and fields are extracted, and the geographical parameters include coordinates, area and slope; Entities are extracted from the construction project file using a BERT model. The entities include construction objects and design parameters of the construction objects. The design parameters include specifications, dimensions, locations, construction materials, construction processes, and construction time.
4. The method according to claim 3, characterized in that The method further comprises: Establish associations between construction objects based on their functional categories, locations, and specifications.
5. The method according to claim 4, characterized in that Based on the environmental parameters, a pre-built knowledge graph is used to perform compliance assessment on the design parameters of the construction object, including: Group related construction objects into the same group; Convert each construction object and corresponding design parameter in the group into a construction object coding vector to obtain multiple construction object coding vectors; Taking the multiple construction object encoding vectors and the environmental parameters as input parameters, a graph neural network is used to evaluate the compliance of the input parameters based on the knowledge graph.
6. The method according to claim 5, characterized in that The knowledge graph includes: Construction objects include canals, pumping stations, drainage ditches, sedimentation tanks, and agricultural roads; Functional category entities, including irrigation facilities, drainage facilities, and transportation facilities; Environmental parameter entities, including soil type, slope class, precipitation zones, and field topology; Constraint rule entities include size constraints, spacing constraints, material rules, and process rules.
7. The method according to claim 5, characterized in that Taking the plurality of construction object encoding vectors and the environmental parameters as input parameters, and evaluating the compliance of the input parameters based on the knowledge graph using a graph neural network, including: Query related constraint rule entities from the knowledge graph based on input parameters; Embed logical reasoning rules as constraints into the pre-trained graph neural network to obtain a compliance classifier. Input parameters and constraint rule entities are input into the compliance classifier to obtain a compliance assessment result output by the compliance classifier.
8. A compliance assessment system for farmland design documents, characterized in that: include: a file recognition module for obtaining a farmland design file and identifying a construction object and its design parameters from the farmland design file using natural language processing technology; Environmental detection module, used to obtain environmental parameters of the plot corresponding to the farmland design file; The compliance assessment module is used to perform compliance assessment on the design parameters of the construction object based on the environmental parameters using a pre-built knowledge graph.
9. A compliance assessment device for farmland design documents, characterized in that: include: a memory for storing the compliance assessment program of the farmland design document; A processor, configured to implement the steps of the method for compliance assessment of a farmland design document as described in any one of claims 1 to 7 when executing the compliance assessment program of the farmland design document.
10. A computer-readable storage medium storing a computer program, characterized in that: The readable storage medium stores a compliance assessment program for a farmland design file. When the compliance assessment program for a farmland design file is executed by a processor, the steps of the compliance assessment method for a farmland design file as described in any one of claims 1 to 7 are implemented.