GIS application development method and system based on low codes

Abstract

The invention relates to the technical field of GIS application development, and discloses a GIS application development method and system based on a low code, and the method comprises the steps: extracting a space entity, a space relation and a space behavior from a real scene, constructing a demand relation topology, inputting the demand relation topology into a network analysis unit, arranging a workflow at the same time, and analyzing the workflow. Generating a plurality of candidate processes in parallel by the arranged workflow, simulating and deducing a plurality of schemes, selecting an optimal scheme through priority ordering, if the plurality of schemes do not meet requirements, disassembling the arranged workflow into fingerprint high-order components, freely combining the fingerprint high-order components, abstracting the high-order components into topological embedding vectors, and performing topological embedding on the topological embedding vectors. The components with similar functions are adjacent to the topology embedded vector in a vector space, a virtual model is established to simulate a combined fingerprint high-order component, virtual random disturbance is injected, a high-order component operation result is output, and if the operation result fails, high-order component simplification is triggered, and high-efficiency GIS application development is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of GIS application development, and discloses a low-code-based GIS application development method and system. Background Art

[0002] In current fields such as geographic information systems and intelligent decision-making systems, there are many challenges in the spatial information processing and process optimization of real-world scenarios. When extracting spatial entities, spatial relationships, and spatial behaviors in the prior art, there are often problems of incomplete and inaccurate information extraction, resulting in defects in the subsequent constructed requirement relationship topology. At the same time, in the process of workflow orchestration and scenario deduction, it is difficult to efficiently generate candidate processes in parallel, the deduced scenarios are limited and lack a scientific priority ranking mechanism, resulting in the inability to quickly find the optimal solution. In addition, when the solution does not meet the requirements, the prior art lacks flexible and effective means for component disassembly, combination, and optimization, and cannot adapt to complex and changing business requirements, seriously restricting the decision-making efficiency and accuracy of the system. Summary of the Invention

[0003] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this part, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this part, the abstract, and the title. However, such simplifications or omissions shall not be used to limit the scope of the present invention.

[0004] To solve the above technical problems, the main purpose of the present invention is to provide a low-code-based GIS application development method, including: Extracting spatial entities, spatial relationships, and spatial behaviors from a real-world scenario, constructing a requirement relationship topology, inputting the requirement relationship topology into a network analysis unit, and at the same time orchestrating a workflow; Generating multiple candidate processes in parallel by the orchestrated workflow, simulating and deducing multiple scenarios, selecting the optimal scenario through priority ranking, and if the multiple scenarios do not meet the requirements, disassembling the orchestrated workflow into fingerprint high-order components; Freely combining the fingerprint high-order components, abstracting the high-order components into topological embedding vectors, and making components with similar functions adjacent to the topological embedding vectors in the vector space; Establishing a virtual model to simulate the combination of fingerprint high-order components and injecting virtual random perturbations, outputting the operation results of the high-order components, and if the operation results are invalid, triggering high-order component simplification and at the same time adjusting the constraints to repair the invalid results.

[0005] As a preferred solution of a low-code-based GIS application development method of the present invention, wherein: Extract multiple spatial entities from the real-world scenario through multimodal interaction, obtain the longitude and latitude coordinates of the multiple spatial entities, and at the same time access the spatial database to assign spatial fingerprints to the spatial entities, where the spatial fingerprints include type, location, and attributes; The spatial relationships include topological relationships, metric relationships, and directional relationships. Through the longitude and latitude coordinates of the spatial entities and the attributes of the spatial fingerprints, determine the topological relationships between multiple spatial entities, calculate the distances between multiple spatial entities from the longitude and latitude coordinates of the multiple spatial entities to determine the metric relationships, and determine the directional relationships between multiple spatial entities from the vector directions of the longitude and latitude coordinate sequences; The user operates the spatial entity to output an operation instruction, and by parsing the operation instruction, bind the spatial fingerprint to extract the spatial behavior; Construct a requirement relationship topology from the spatial entities, spatial relationships, and spatial behaviors.

[0006] As a preferred solution of a low-code based GIS application development method of the present invention, wherein: Import the requirement relationship topology into graph calculation, use spatial entities and spatial behaviors as heterogeneous nodes, and spatial relationships and logical dependencies as weighted edges to construct the requirement relationship topology; Convert the requirement relationship topology into a network model, serialize the node and edge information of the topology into a format compatible with the network analysis unit, input it into the network analysis unit, and construct a triple association matrix to convert the requirement relationship topology into computable features. The triple association matrix includes an adjacency matrix, a metric matrix, and a direction matrix; Parse the relationship topology through the network analysis unit. At the same time, the network analysis unit dynamically inserts mutex locks to eliminate spatial conflicts and resource conflicts, divide the behavior nodes without dependency relationships into the same level, and output the nodes at the same level by spatial position partition to generate a directed acyclic graph with hierarchical numbers; Convert each spatial behavior of the directed acyclic graph into an executable unit.

[0007] As a preferred solution of a low-code based GIS application development method of the present invention, wherein: The generation of the multiple candidate processes includes: based on the feature representation of the requirement relationship topology, retrieve historical process templates through a case matching engine, and directly reuse them when there are templates that meet the similarity requirements; if there is no matching template, then start a generative adversarial mechanism to create a new candidate workflow and verify its logical completeness; Simulate and verify the candidate workflow in a digital sandbox environment. By injecting multimodal perturbations such as simulated device failures and data anomalies, detect whether the scheme output deviates from the expectation; when the scheme is marked as invalid, trigger a dynamic degradation mechanism, disassemble the corresponding high-order components into basic operation units, and collaborate with the federated knowledge base to refine the degradation processing rules to update the knowledge base.

[0008] As a preferred solution of a low-code based GIS application development method of the present invention, wherein: The choreographed workflow is a shunt engine, which generates multiple candidate processes in parallel through the shunt engine, and the shunt engine automatically selects and activates three types of generation strategies; The three types of generation strategies retrieve the historical process template library to reuse the workflow that meets the topological similarity requirements; apply the predefined business logic rule chain to infer and generate the standard process that meets the industry specifications; use the generative adversarial mechanism to dynamically innovate and generate and verify the new workflow with reasonable spatial logic.

[0009] As a preferred solution of a low-code based GIS application development method of the present invention, wherein: Perform multi-dimensional simulation and deduction on multiple candidate solutions generated in parallel in the digital sandbox environment, and the sandbox dynamically injects multi-level disturbances such as data anomalies, equipment failures, and business conflicts to simulate the complexity of the real environment; During the deduction process, the adaptive evaluation system real-time collects the robustness, efficiency, cost, and user adaptability of each solution to form a dynamic score. By screening out the invalid solutions that are significantly inferior to other solutions in all evaluation dimensions, retain the solution set that achieves the best balance in multiple dimensions. The system sorts the optimal solution set according to the preset priority rules and user-defined preferences, and outputs the optimal solution to be executed first; When all the candidate solutions optimized through deduction fail to meet the core requirements, the system triggers the workflow intelligent decomposition mechanism.

[0010] As a preferred solution of a low-code based GIS application development method of the present invention, wherein: Perform multi-dimensional feature fusion on the functional semantics, topological association, and industry attributes of the fingerprint high-order components through the feature extraction engine, and train the topological embedding model based on the contrast learning framework; For the topological embedding model, the model maps the components with similar functional semantics and topological roles to adjacent positions in the vector space.

[0011] As a preferred solution of a low-code based GIS application development method of the present invention, wherein: The virtual model loads the fingerprint high-order components and their topological embedding vectors through the adapter, and generates the initial connection logic according to the functional semantic similarity; The virtual model also predefines the data layer, device layer, and business layer through the perturbation rule library, and eliminates the interference data in the database, device layer, and business layer through multi-level perturbations; Automatically connect the adjacent fingerprint high-order components in the vector space through the topological semantic matching algorithm.

[0012] As a preferred solution of the low-code GIS application development system of the present invention, When the running results deviate from the safety tolerance or the process is interrupted, the data is traced back to the failed high-level component. The disassembler is called to deconstruct the failed high-level component into fine-grained basic units. At the same time, the disassembly depth is limited, the decomposition of key components is blocked, and a degradation fallback plan is generated. Retesting whether the topology embedding vector library is complete by dynamically adjusting constraints, including loosening boundaries, parameter calibration, federation rule evolution, and incremental verification; The relaxation bound is used for gradient extension failure constraint; The parameter calibration adjusts input parameters based on feedback; The federated rule evolution extracts similar failure strategies from a cross-domain knowledge base; The incremental verification is used to retest whether the topology embedding vector library is constructed. If the construction is successful, the topology embedding vector library is updated to form a closed-loop optimization.

[0013] As a preferred solution of the low-code GIS application development system of the present invention, Extract modules, arrange and disassemble modules, combine modules and virtual models; The extraction module extracts spatial entities, spatial relations and spatial behaviors from real scenes and constructs demand relationship topology, and the network analysis unit choreographs the workflow; The orchestration and disassembly module is used to generate multiple candidate processes in parallel from the orchestrated workflow, simulate and deduce multiple solutions, and disassemble the orchestration workflow into fingerprint high-level components; The combination module freely combines the fingerprint high-order components, abstracts the high-order components into topological embedding vectors, and makes components with similar functions adjacent to the topological embedding vectors in the vector space; The virtual model includes a component topology adapter, a perturbation rule base and dynamic constraints.

[0014] Beneficial effects of the present invention: The present invention accurately extracts spatial entities, spatial relationships and spatial behaviors from real-world scenarios, constructs a demand relationship topology, inputs the demand relationship topology into a network analysis unit and orchestrates the workflow. It can generate multiple candidate processes in parallel and simulate and deduce them. Combined with a priority sorting mechanism, it can quickly screen out the optimal solution, significantly improving decision-making efficiency.

[0015] In this application, when multiple solutions do not meet the requirements, the orchestration workflow is decomposed into fingerprint high-level components, which are freely combined and abstracted into topological embedding vectors, so that similar functional components are adjacent in the vector space, which is convenient for management and calling, and enhances the flexibility and scalability of the system.

[0016] By establishing a virtual model to simulate the high-order components of the combined fingerprint and injecting virtual random perturbations, this application can detect potential problems in advance. When the operation result fails, it triggers the simplification of the high-order components and adjusts the constraint to repair the result, effectively improving the stability and reliability of the system and providing an efficient and accurate solution for intelligent decision-making in complex scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them: Figure 1 is a flowchart of a method for developing a GIS application based on low-code according to the present invention; Figure 2 is a composition diagram of a system for developing a GIS application based on low-code according to the present invention; Figure 3 is an overall simulation topology diagram of a method for developing a GIS application based on low-code according to the present invention; Figure 4 is a simulation schematic diagram of a system for developing a GIS application based on low-code according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will provide a detailed description of the specific embodiments of the present invention with reference to the accompanying drawings of the specification.

[0019] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar generalizations without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0020] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The "in one embodiment" that appears in different places in this specification does not all refer to the same embodiment, nor is it a separate or alternative embodiment that mutually excludes other embodiments.

[0021] Embodiment 1 As Figure 1 shown, a method for developing a GIS application based on low-code includes: Extracting spatial entities, spatial relationships, and spatial behaviors from the real-world scenario, constructing a demand relationship topology, inputting the demand relationship topology into the network analysis unit, and arranging the workflow at the same time.

[0022] Extract multiple spatial entities from the real scene through multimodal interaction, obtain the longitude and latitude coordinates of the multiple spatial entities, and at the same time access the spatial database to assign spatial fingerprints to the spatial entities, where the spatial fingerprints include type, location, and attributes: Capture geographical objects in the real scene through multimodal interaction. The user directly draws or selects on the map interface, or describes them in natural language. The system combines geocoding technology to convert semantics into longitude and latitude coordinates, and at the same time accesses the spatial database to automatically match preset entities, and identifies elements such as building outlines and water body boundaries. All entities are assigned spatial fingerprints including type, location, and attribute sets. For example, "Entity A: type = fire station, location = POINT(116.4, 39.9), attributes = {capacity: medium}".

[0023] The spatial relationships include topological relationships, metric relationships, and directional relationships. Through the longitude and latitude coordinates of the spatial entities and the attributes of the spatial fingerprints, determine the topological relationships between multiple spatial entities, calculate the distances between multiple spatial entities from the longitude and latitude coordinates of the multiple spatial entities to determine the metric relationships, and determine the directional relationships between multiple spatial entities from the vector directions of the longitude and latitude coordinate sequences; In this application, the directional relationships of multiple spatial entities include topological relationships, metric relationships, and directional relationships.

[0024] Specifically, the topological relationship is determined based on the intersection logic of the entity boundary coordinates. For example, through the overlay analysis of the administrative division polygon and the park polygon, confirm the spatial inclusion relationship.

[0025] The metric relationship calculates the true ground surface distance between entities, avoiding the map projection deformation error.

[0026] The directional relationship is determined through the vector direction analysis of the coordinate sequence. For example, the river flow direction depends on the derivation of the elevation attribute and the coordinate trend.

[0027] The spatial relationship extraction is based on the entity location, and the system uses spatial calculations to automatically deduce geometric relationships: Specifically, the topological relationship judges the boundary intersection. For example, check whether the park is included in the administrative region; The metric relationship calculates the true ground surface distance between locations.

[0028] The directional relationship depends on vector analysis to determine the azimuth angle.

[0029] For the semantic relationships described by the user, call the domain knowledge graph and map the influence to the downstream diffusion path in space.

[0030] The user operates the spatial entity to output an operation instruction, and by parsing the operation instruction, bind the spatial fingerprint to extract the spatial behavior.

[0031] Construct a topological relationship of requirements from the spatial entities, spatial relationships, and spatial behaviors.

[0032] Specifically, the topological relationship of requirements is constructed into a graph structure based on the extracted spatial entities, spatial relationships, and spatial behaviors, where entities are represented as nodes and relationships are represented as edges with attributes. For example, distance is used as the edge weight and direction is used as the edge label.

[0033] Furthermore, use a data conversion module to serialize the node and edge information of the topology into a format compatible with the network analysis unit, and input it into the real-time input unit through a predefined API interface or message queue.

[0034] Furthermore, the network analysis unit performs operations such as shortest path calculation, community detection, or connectivity analysis through the connectivity of nodes and edges, thereby inferring spatial interaction patterns, discretizing continuous space problems into computable networks, achieving efficient spatial optimization and decision support, and ensuring performance through memory optimization and parallel computing when dealing with large-scale relationships.

[0035] Import the topological relationship of requirements into graph computing. Spatial entities and spatial behaviors are used as heterogeneous nodes, and spatial relationships and logical dependencies are used as weighted edges to construct the topological relationship of requirements.

[0036] When the topology is input, the workflow orchestration system is activated to manage the end-to-end process. The orchestrator defines a task sequence, including topology input, network analysis execution, result output, and user feedback loop; for example, dynamically scheduling network analysis tasks through an event-driven mechanism and handling dependency relationships.

[0037] Furthermore, the workflow maintains a task state machine to ensure retry or rollback in case of analysis failure, integrates the input of the spatial database and user operation flow, and at the same time queue buffering and priority scheduling handle real-time interactions, ultimately achieving the robustness and low latency of scenario response to meet the requirements of dynamic real-world applications.

[0038] In this application, a specific implementation method of a preferred topological relationship of requirements includes: The topological relationship of requirements can be converted into a network model, serialize the node and edge information of the topology into a format compatible with the network analysis unit, and input it into the network analysis unit to construct a triple association matrix, which includes an adjacency matrix, a metric matrix, and a direction matrix.

[0039] The adjacency matrix is used to describe the existential association between entities, encoding the basic connectivity between entities through 0 / 1 binary values. For example, whether a road connects to a building. If a row / column is all 0, it represents an isolated entity to exclude invalid behaviors.

[0040] The metric matrix is used to quantify the spatial constraint strength, map the interaction cost between entities through the Euclidean distance calculation method, provide a conflict determination threshold for the conflict resolution mechanism, and provide a partitioning basis for the parallelism optimizer at the same time.

[0041] The direction matrix is used to capture the spatial behavior direction constraint and drive subsequent operations by encoding the direction dependence of the behavior with unit vectors.

[0042] Multiple candidate processes are generated in parallel by the choreographed workflow, and multiple scenarios are simulated and deduced. The optimal scenario is selected through priority ranking. If multiple scenarios do not meet the requirements, the choreographed workflow is disassembled into fingerprint high-order components.

[0043] The generation of the multiple candidate processes includes: feature representation based on the requirement relationship topology, retrieving historical process templates through the case matching engine, and directly reusing them when there are templates that meet the similarity requirements; if there is no matching template, the generative adversarial mechanism is started to create a new candidate workflow and verify its logical completeness.

[0044] Specifically, the parallel process generation is based on the topological graph feature vector, retrieves historical templates through the case matching engine, and starts the generative adversarial network when there is no match: the generator outputs a candidate workflow, and the discriminator verifies the logical completeness.

[0045] In this application, a preferred method for establishing an adversarial mechanism specifically includes: When the network analysis fails, it automatically triggers a retry or rolls back to the state before the topology input. For the real-time geographical data stream and user operation event stream of the spatial database, a priority message queue is used to buffer high-concurrency requests, and tasks are dynamically scheduled according to the urgency of the instructions. During the implementation of the requirement relationship topology, the topology converter serializes the graph structure topology into a format compatible with the network analysis unit and constructs a triple association matrix.

[0046] The adjacency matrix encodes the connectivity between entities with 0 / 1 binary values. For example, the connection between a road and a building, and all-zero rows / columns identify isolated entities to exclude invalid behaviors.

[0047] The metric matrix quantifies the spatial constraint strength, such as logistics cost. When the distance is less than the safety threshold, an alarm is given, and the output value drives conflict resolution and parallel optimization. When the distance is greater than the parallel threshold, multi-node concurrency is allowed.

[0048] The direction matrix encodes the behavior direction constraint through unit vectors, defines the timing execution order, and provides a benchmark for direction mutation detection.

[0049] Based on the matrix feature representation, the parallel process generation is activated by intelligent shunting.

[0050] If there is no match in the confrontation, the generative adversarial mechanism is started, and by creating a candidate flow, the rationality of the plan is verified according to the spatial logic rule library.

[0051] When the digital sandbox test marking scheme fails, disassemble the high-order components and impose simplified restrictions to generate a degradation fallback plan. Finally, screen the multi-dimensional equilibrium solution set to achieve a closed-loop from topology construction to optimal execution.

[0052] In the digital sandbox environment, simulate and verify the candidate workflow. By injecting multi-modal perturbations such as simulated device failures and data anomalies, detect whether the scheme output deviates from the expectation.

[0053] When the scheme is marked as failed, trigger the dynamic degradation mechanism, disassemble the corresponding high-order components into basic operation units, and collaborate with the federated knowledge base to refine the degradation processing rules to update the knowledge base.

[0054] Specifically, the digital sandbox verification injects multi-modal perturbation parameters.

[0055] In the digital sandbox environment, the system dynamically generates multi-level perturbation parameters covering the data layer, device layer, and business layer through the perturbation rule base to holographically simulate the complexity of the real scenario.

[0056] For data layer perturbations, a probability model can be used to generate random coordinate offsets to simulate GPS signal interference and verify the fault tolerance of spatial analysis to positioning drift; randomly mask the key attributes of entities according to a preset ratio to test the adaptability of the scheme to data loss. Inject out-of-order and duplicate data streams reported by sensor anomalies to test the robustness of timing analysis. The device layer perturbations map the abnormal scenarios of hardware resources.

[0057] Specifically, the device layer perturbations mapping the abnormal scenarios of hardware resources are used to simulate memory overflow of edge nodes and CPU overload to trigger the computing degradation strategy, and periodically block the network link to test the offline decision-making ability.

[0058] For business layer perturbations, implant logic conflict events and construct mutually exclusive operation instructions, such as simultaneously requesting to demolish and protect the same building.

[0059] Verify that the conflict resolution mechanism is used to forge unauthorized operations to check the effectiveness of security interception. Verify the conflict resolution mechanism, for example, a low-privilege user triggers the control process.

[0060] The perturbation injection follows the principle of scene weight adaptability. When the sandbox detects that the output deviates from the expectation by more than the tolerance threshold, mark the scheme as failed, and immediately trigger the atomic degradation disassembly. Decompose the failed high-order components into basic operation units, and at the same time impose disassembly constraints.

[0061] Real-time collect degradation cases, establish a new policy perturbation rule base, and form a self-reinforcing mechanism for perturbation generation, failure detection, degradation processing, and knowledge evolution.

[0062] Verified in the rail transit dispatching scenario, after injecting disturbances such as positioning drift and communication interruption combinations, component degradation is completed, and the offline timetable strategy is enabled with the addition of signal loss, enhancing the resilience of the low-code GIS application development system.

[0063] The degradation case can extract the interpolation algorithm rules for sensor fault switching across projects.

[0064] The orchestrated workflow is a shunt engine that generates multiple candidate processes in parallel through the shunt engine. The shunt engine automatically selects and activates three types of generation strategies; Specifically, the multi-strategy parallel generation of candidate processes automatically selects and activates three types of generation strategies based on the structural characteristics of the demand topology diagram, such as node connection density and relationship type distribution.

[0065] The three types of generation strategies retrieve the historical process template library to reuse the workflow that meets the topological similarity requirements; apply the predefined business logic rule chain to infer and generate the standard process that complies with industry specifications; use the generative adversarial mechanism to dynamically innovate and generate and verify the new workflow with reasonable spatial logic.

[0066] Retrieve the workflow of similar scenarios in the template library, control the reuse conditions through the topological similarity threshold, and apply the predefined business logic rules to generate the standard process that complies with industry specifications. For example, when monitoring pollution, a buffer zone needs to be delimited first.

[0067] Through the generative adversarial mechanism, the generator creates a new workflow, and the discriminator verifies its reasonable spatial logic.

[0068] In the digital sandbox environment, multi-dimensional simulation and deduction are carried out on multiple candidate solutions generated in parallel. The sandbox dynamically injects multi-level disturbances such as data anomalies, equipment failures, and business conflicts to simulate the complexity of the real environment.

[0069] During the deduction process, the adaptive evaluation system collects the robustness, efficiency, cost, and user adaptability of each solution in real time to form a dynamic score. By screening out the invalid solutions that are inferior to other solutions in all evaluation dimensions, the solution set that achieves the best balance in multiple dimensions is retained. The system sorts the optimal solution set according to the preset priority rules and user-defined preferences, and outputs the optimal solution to be executed first.

[0070] When all the candidate solutions optimized through deduction fail to meet the requirements, the system triggers the intelligent disassembly mechanism of the workflow. Specifically, the intelligent disassembly mechanism first accurately locates the core high-order functional modules that cause the systematic failure of the solution by analyzing the digital sandbox deduction logs and data provenance, disassembles the failed high-order components according to the predefined atomic rule library, and deconstructs them into a set of finer-grained fingerprint high-order components with clear input and output interfaces and independent functions. For example, "pollution diffusion prediction" is disassembled into basic steps such as "buffer generation", "flow direction simulation", "concentration calculation", and "heat map rendering".

[0071] Furthermore, during the disassembly process, the tight coupling dependencies of the original components are removed to ensure that each fingerprint high-order component represents a reusable business atomic operation unit with clear functional semantics, and a unique topological embedding vector is generated for it to identify its functional semantics and interface characteristics.

[0072] Furthermore, the digital sandbox environment construction can simulate digital twins with real-world complexity and dynamically inject multi-level perturbations: Specifically, the multi-level perturbations include data layer anomalies, device layer failures, and business layer conflicts.

[0073] Furthermore, data layer anomalies include real defects such as simulated location drift and data loss; device layer failures include hardware bottlenecks such as reproduced computing node delays and memory overflows.

[0074] Business layer conflicts include logical contradictions such as construction rule mutual exclusions and permission overstepping.

[0075] In this application, a preferred set of solutions that achieves the best balance in multiple dimensions. The multiple dimensions can include robustness, efficiency, cost, and user adaptability.

[0076] Specifically, robustness is used to measure the stability of the solution in a failure scenario; efficiency is used to evaluate the end-to-end processing delay and resource consumption; cost is used to quantify the computing resource and bandwidth overhead; user adaptability is used to analyze the fit between the operation path and the historical behavior pattern.

[0077] In this application, a preferred method for screening invalid solutions that are significantly inferior to other solutions in all evaluation dimensions can be to exclude the solutions that are comprehensively degraded and retain the optimal solution set with balanced multi-dimensional indicators as the final decision.

[0078] Freely combine the fingerprint high-order components, abstract the high-order components into topological embedding vectors, and make the components with similar functions adjacent to the topological embedding vectors in the vector space.

[0079] Through the feature extraction engine, multi-dimensional feature fusion is performed on the functional semantics, topological associations, and industry attributes of the fingerprint high-order components, and a topological embedding model is trained based on the contrast learning framework.

[0080] The topological embedding model maps components with similar functional semantics and topological roles to adjacent positions in the vector space.

[0081] For example, based on the spatial expansion of the distance between locations, perform a spatial buffer analysis function to generate the Euclidean distance of the topological embedding vector in the vector space. By calculating the Euclidean distance, determine whether the irrelevant function meets the user's needs. If the Euclidean distance is less than the distance of the irrelevant function component, output the topology that meets the user's needs. If the Euclidean distance is not less than the distance of the irrelevant function component, it does not meet the user's needs.

[0082] Establish a virtual model to simulate the combined fingerprint high-order components and inject virtual random perturbations, and output the operation results of the high-order components. If the operation results fail, trigger the simplification of the high-order components, and at the same time adjust the constraints to repair the failed results.

[0083] In this application, a preferred specific implementation method for simplifying high-order components includes.

[0084] When the virtual model detects that the operation results fail, the system triggers a cascading simplification process.

[0085] S1 Based on data lineage analysis technology, trace back the abnormal data flow, accurately mark the high-order component nodes that cause systematic deviations, and rely on sandbox logs to construct a fault propagation tree to quantify the failure contribution degree of sub-modules.

[0086] S2 Call the predefined atomic rule library to decouple the faulty components into a chain of fine-grained basic operation units, and at the same time impose triple simplification restrictions. The multiple simplification restrictions can include depth constraints, core protection, and interface inheritance.

[0087] S3 For failures caused by overly strict constraints, expand the tolerance range in gradients; based on historical sandbox data, the parameters can be re-calibrated through a PID controller; extract cross-domain repair cases from the federated knowledge base and generate adaptation strategies to inject new constraints.

[0088] S4 Only perform sandbox retesting on the simplified sub-chain, update the topological embedding vector to generate a vector for the new sub-unit and associate components with similar functions, shorten the Euclidean distance in the vector space to achieve semantic clustering, abstract the simplified path into a degradation rule template and map it to the federated knowledge base, forming a self-optimizing system for failure location, disassembly and repair, vector recombination, and rule evolution.

[0089] The virtual model loads the fingerprint high-order components and their topological embedding vectors through an adapter, and generates the initial connection logic based on functional semantic similarity.

[0090] The virtual model also predefines the data layer, device layer, and business layer through a perturbation rule library, and eliminates the interference data in the database, device layer, and business layer through multi-level perturbations.

[0091] A topological semantic matching algorithm automatically concatenates adjacent fingerprint high-order components in the vector space. For example, buffer analysis is mapped to heat map rendering, forming a complete workflow and synchronously activating multi-level perturbation injection.

[0092] The specific multi-level layers include data layer, device layer and business layer; the data layer injects simulated sensor errors such as coordinate drift and attribute missing; the device layer triggers virtual hardware failures to reproduce edge bottlenecks; the business layer implants rule-based mutually exclusive events to test logical fault tolerance, which is used to simulate real-world complexity.

[0093] When the running results deviate from the safety tolerance or the process is interrupted, the data is traced back to the failed high-order component, and the disassembler is called to deconstruct the failed high-order component into fine-grained basic units. At the same time, the disassembly depth is limited, the decomposition of key components is shielded, and a degradation fallback plan is generated.

[0094] The topology embedding vector library is retested to see if it is constructed by dynamically adjusting constraints, including relaxing boundaries, parameter calibration, federation rule evolution, and incremental verification.

[0095] The relaxation bound is used for gradient extension failure constraint.

[0096] The parameter calibration adjusts input parameters based on feedback.

[0097] The federated rule evolution extracts similar failure strategies from a cross-domain knowledge base.

[0098] The incremental verification is used to retest whether the topology embedding vector library is constructed. If the construction is successful, the topology embedding vector library is updated to form a closed-loop optimization.

[0099] In this application, a preferred technical solution includes: The initial connection logic principle of the virtual model establishes component connections through a two-dimensional matching mechanism: functional semantic matching compares the relative positions of component topological embedding vectors in vector space. When the core functional goals of two components are strongly semantically related, as manifested by a close distance in vector space, the system determines that they are composable.

[0100] Furthermore, a two-dimensional matching mechanism establishes component connections: interface contract compatibility verifies whether the input / output data types and structures of candidate components meet the upstream and downstream transmission requirements, and connection edges are only allowed to be established when the data contracts are consistent.

[0101] The disassembly operation relies on a predefined library of atomic operations in the industry. According to domain knowledge, complex functions are disassembled into indivisible standard operation units. The sub-units inherit the input / output conventions of the original components to ensure that they can still interact with upstream and downstream components after degradation. Disassembly is prohibited for modules involving safety or core logic to maintain their atomicity and ensure that the key capabilities of the system do not degrade.

[0102] The database can learn cross-domain migration by establishing federated rules.

[0103] Cross-domain policy migration is achieved through a three-layer adaptation mechanism.

[0104] Furthermore, scene similarity matching is based on the distance of topological embedding vectors. It retrieves historical cases with similar functional semantics to the current failure scene, extracts the applicable boundaries of historical policies, verifies their constraint compatibility with the current scene, and automatically inserts lightweight adaptation modules to address technical differences.

[0105] Incremental verification and vector library update only perform sandbox testing on the simplified sub-component chain, significantly reducing verification overhead. When a new sub-unit passes verification, the system dynamically adjusts the position of its topological embedding vector in space to make it closer to the group of components with similar functions. The successful simplified paths are abstracted into reusable degradation rules and stored in the federated knowledge base for global invocation.

[0106] As Figure 3 shown, the starting node on the left side of the S1 requirements relationship topology construction diagram is labeled S1. Spatial entities, relationships, and behaviors are extracted from the real-world scene to construct the requirements relationship topology, and the steps of extracting spatial entities and constructing the topology are carried out.

[0107] L1 is the topology input workflow orchestration code. The connection line pointing to the right from S1 is labeled L1, which inputs the topology into the network analysis unit and orchestrates the workflow to input the topology into the network analysis unit and, at the same time, orchestrates the workflow.

[0108] S2 generates candidate processes in parallel. In the middle area, it is labeled S2 and branches such as the first candidate solution. Multiple candidate processes are generated in parallel by the orchestrated workflow.

[0109] L2 prioritizes candidate solutions. The connection line between S2 and S3 is labeled L2, which performs multi-dimensional scoring and ranking on the candidate solutions to screen the optimal solution set.

[0110] S3 injects perturbations into the virtual model. The virtual simulation model node in the figure is labeled S3, which injects perturbations such as data anomalies and equipment failures into the digital sandbox to simulate the real environment.

[0111] L3 has a two-way branch for simulation results. The two branch lines output from S3 are labeled L3 success and L4 failure. If the simulation is successful, the solution is output; if it fails, component disassembly is triggered.

[0112] S4 high - order component disassembly and repair. On the right side of the figure, the high - order component simplifies the node, marked as S4. The failed workflow is disassembled into fingerprint high - order components, and the constraints are adjusted to repair the failure.

[0113] Re - measurement of the optimized L4 solution. The loop line where S4 points to S3 is marked as L4. The repaired solution re - enters the virtual simulation to form a closed - loop optimization.

[0114] Embodiment 2 As Figure 2 shown, a low - code - based GIS application development system.

[0115] An extraction module, an orchestration disassembly module, a combination module, and a virtual model.

[0116] The extraction module extracts spatial entities, spatial relationships, and spatial behaviors from the real - world scenario, constructs a requirement relationship topology, and the network analysis unit orchestrates the workflow.

[0117] The orchestration disassembly module is used to generate multiple candidate processes in parallel from the orchestrated workflow, simulate and deduce multiple solutions, and disassemble the orchestrated workflow into fingerprint high - order components.

[0118] The combination module freely combines the fingerprint high - order components, abstracts the high - order components into topological embedding vectors, so that components with similar functions are adjacent to the topological embedding vectors in the vector space.

[0119] The virtual model includes a component topology adapter, a perturbation rule library, and dynamic constraints.

[0120] As Figure 4 shown, there are three types of initial solutions: the first candidate solution, the second candidate solution, and the third candidate solution, representing workflows generated from different strategic paths.

[0121] Each candidate solution first enters the un - sorted solution state. After being screened by the rules, it becomes a sorted solution, completing the preliminary distinction between good and bad.

[0122] All sorted solutions are aggregated into the solution evaluation link. Based on user requirements, a judgment is made, and there are two branches: If the requirements are met, the final result is directly output, and the process ends.

[0123] If the requirements are not met, it enters the disassembly and virtual verification link, disassembles the orchestrated workflow into fingerprint high - order components, and prepares for further optimization.

[0124] The disassembled fingerprint high - order components are input into the virtual simulation model, and virtual random perturbations are injected. The verification is also divided into two branches.

[0125] Successful virtual simulation indicates that the component solution can still meet the requirements under perturbations, and outputs a solution that meets user requirements.

[0126] The failure of virtual simulation triggers the simplification of high - order components and the invalidation of repair results. Adjust the component logic, parameters, etc., and try to make the solution meet the standards. After that, it may re - enter the simulation verification to form an optimization closed - loop.

[0127] It is important to note that the construction and arrangement of the present application shown in multiple different exemplary embodiments are only illustrative. Although only two embodiments are described in detail in this disclosure, those who refer to this disclosure should easily understand that, without substantially departing from the subject matter described in this application, changes in the dimensions, scales, structures, shapes and proportions of various elements, as well as parameter values (e.g., temperature, pressure, etc.), installation arrangements, use of materials, colors, orientations, etc. For example, an element shown as integrally formed may be composed of multiple parts or elements, the position of the element may be inverted or otherwise changed, and the nature, number or position of discrete elements may be changed or altered. Therefore, all such modifications are intended to be included within the scope of the present invention. The order or sequence of any process or method steps may be changed or reordered according to alternative embodiments. Any "means - plus - function" clause is intended to cover the structures that perform the functions described herein, and not only structurally equivalent but also equivalent structures. Other substitutions, modifications, changes and omissions may be made in the design, operating conditions and arrangement of the exemplary embodiments without departing from the scope of the present invention. Therefore, the present invention is not limited to specific embodiments, but extends to various modifications that still fall within the scope of the appended claims.

[0128] In addition, in order to provide a concise description of the exemplary embodiments, not all features of the actual embodiments may be described (i.e., those features that are not relevant to the currently considered best mode of implementing the present invention or those that are not relevant to the implementation of the present invention).

[0129] It should be understood that in the development of any actual implementation, such as in any engineering or design project, a large number of specific implementation decisions may be made. Such development efforts may be complex and time - consuming, but for those of ordinary skill in the art who benefit from this disclosure, without excessive experimentation, the development efforts will be a routine task of design, manufacturing and production.

[0130] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention may be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A low-code based GIS application development method, characterized in that, Including: Extract spatial entities, spatial relationships, and spatial behaviors from the real-world scenario, construct a requirements relationship topology, input the requirements relationship topology into the network analysis unit, and at the same time orchestrate the workflow; Parallelly generate multiple candidate processes from the orchestrated workflow, simulate and deduce multiple solutions, select the optimal solution through priority ranking. If the multiple solutions do not meet the requirements, disassemble the orchestrated workflow into fingerprint high-order components; Freely combine the fingerprint high-order components, abstract the high-order components into topological embedding vectors, so that components with similar functions are adjacent to the topological embedding vectors in the vector space; Establish a virtual model to simulate the combined fingerprint high-order components and inject virtual random perturbations, output the operation results of the high-order components. If the operation results are invalid, trigger the simplification of the high-order components, and at the same time adjust the constraints to repair the invalid results.

2. A low-code-based GIS application development method according to claim 1, characterized in that: Extract multiple spatial entities from the real-world scenario through multimodal interaction, obtain the longitude and latitude coordinates of the multiple spatial entities, and at the same time access the spatial database to assign spatial fingerprints to the spatial entities. The spatial fingerprints include type, location, and attributes; The spatial relationships include topological relationships, metric relationships, and directional relationships. Determine the topological relationships between multiple spatial entities through the longitude and latitude coordinates of the spatial entities and the attributes of the spatial fingerprints. Calculate the distances between multiple spatial entities from the longitude and latitude coordinates of the multiple spatial entities to determine the metric relationship, and determine the directional relationships between multiple spatial entities from the vector directions of the longitude and latitude coordinate sequences; The user operates the spatial entity to output an operation instruction, and by parsing the operation instruction, bind the spatial fingerprint to extract the spatial behavior; Construct a requirements relationship topology from the spatial entities, spatial relationships, and spatial behaviors.

3. A low-code-based GIS application development method according to claim 1, characterized in that: Import the requirements relationship topology into graph calculation. The spatial entities and spatial behaviors are used as heterogeneous nodes, and the spatial relationships and logical dependencies are used as weighted edges to construct the requirements relationship topology; Convert the requirements relationship topology into a network model, serialize the node and edge information of the topology into a format compatible with the network analysis unit, input it into the network analysis unit, and construct a triple association matrix to convert the requirements relationship topology into computable features. The triple association matrix includes an adjacency matrix, a metric matrix, and a direction matrix; Parse the relationship topology through the network analysis unit. At the same time, the network analysis unit dynamically inserts mutex locks to eliminate spatial conflicts and resource conflicts, divide the behavior nodes without dependency relationships into the same level, and output the nodes in the same level partitioned by spatial position to generate a directed acyclic graph with hierarchical numbers; Convert each spatial behavior of the directed acyclic graph into an executable unit.

4. A low-code-based GIS application development method according to claim 1, characterized in that: The generation of the multiple candidate processes includes: based on the feature representation of the requirements relationship topology, retrieve historical process templates through a case matching engine. When there is a template that meets the similarity requirements, directly reuse it; if there is no matching template, start a generative adversarial mechanism to create a new candidate workflow and verify its logical completeness; Simulate and verify the candidate workflow in a digital sandbox environment. By injecting multimodal disturbances such as simulated device failures and data anomalies, detect whether the output of the detection scheme deviates from the expectation; when the scheme is marked as invalid, trigger the dynamic degradation mechanism, disassemble the corresponding fingerprint high-order components into basic operation units, and collaborate with the federated knowledge base to refine the degradation processing rules to update the knowledge base.

5. A low-code based GIS application development method according to claim 1, characterized in that: The orchestrated workflow is a shunt engine, which generates multiple candidate processes in parallel through the shunt engine, and the shunt engine automatically selects and activates three types of generation strategies; The three types of generation strategies retrieve the historical process template library to reuse the workflow that meets the topological similarity requirements; apply the predefined business logic rule chain to infer and generate the standard process that conforms to the industry specification; use the generative adversarial mechanism to dynamically innovate and generate and verify the new workflow with reasonable spatial logic.

6. A low-code based GIS application development method according to claim 5, characterized in that: Conduct multi-dimensional simulation and deduction on multiple candidate solutions generated in parallel in a digital sandbox environment. The sandbox dynamically injects multi-level disturbances such as data anomalies, device failures, and business conflicts to simulate the complexity of the real environment; During the deduction process, the adaptive evaluation system real-time collects the robustness, efficiency, cost, and user adaptability of each solution to form a dynamic score. By screening out the invalid solutions that are significantly inferior to other solutions in all evaluation dimensions, retain the set of solutions that achieve the best balance in multiple dimensions. The system sorts the optimal solution set according to the preset priority rules and user-defined preferences, and outputs the optimal solution to be executed first; When all the candidate solutions that have been deduced and optimized fail to meet the core requirements, the system triggers the workflow intelligent disassembly mechanism.

7. A low-code based GIS application development method according to claim 1, characterized in that: Perform multi-dimensional feature fusion on the functional semantics, topological associations, and industry attributes of the fingerprint high-order components through the feature extraction engine, and train the topological embedding model based on the contrast learning framework; For the topological embedding model, the model maps the components with similar functional semantics and topological roles to adjacent positions in the vector space.

8. A low-code based GIS application development method according to claim 7, characterized in that: The virtual model loads the fingerprint high-order components and their topological embedding vectors through the adapter, and generates the initial connection logic according to the functional semantic similarity; The virtual model also predefines the data layer, device layer, and business layer through the perturbation rule library, and eliminates the interference data in the database, device layer, and business layer through multi-level perturbations; Automatically connect the adjacent fingerprint high-order components in the vector space through the topological semantic matching algorithm.

9. A low-code based GIS application development method according to claim 8, characterized in that: When the operation result deviates from the safety tolerance or the process is interrupted, trace the data source to the failed high-order component, call the disassembler to decompose the failed high-order component into fine-grained basic units, and at the same time limit the disassembly depth and shield the decomposition of key components to generate a degradation fallback plan; By dynamically adjusting whether the constrained retest topology embedding vector library is constructed, the dynamic adjustment constraints include relaxation boundary, parameter calibration, federated rule evolution, and incremental verification; The relaxation boundary is used for gradient expansion failure constraints; The parameter calibration adjusts the input parameters based on feedback; The federated rule evolution extracts similar failure strategies from the cross-domain knowledge base; The incremental verification is used to check whether the retest topology embedding vector library is constructed. If the construction is successful, the topology embedding vector library is updated to form a closed-loop optimization.

10. A low-code based GIS application development system, which is implemented based on the low-code based GIS application development method described in any one of claims 1-9, characterized in that, The system includes: an extraction module, an orchestration and disassembly module, a combination module, and a virtual model; The extraction module extracts spatial entities, spatial relationships, and spatial behaviors from the real-world scenario, constructs a requirement relationship topology, and the network analysis unit orchestrates the workflow; The orchestration and disassembly module is used to parallelly generate multiple candidate processes from the orchestrated workflow, simulate and deduce multiple solutions, and disassemble the orchestrated workflow into fingerprint high-order components; The combination module freely combines the fingerprint high-order components, abstracts the high-order components into topology embedding vectors, and makes components with similar functions adjacent to the topology embedding vectors in the vector space; The virtual model includes a component topology adapter, a perturbation rule library, and dynamic constraints.

Images