A base parameter intelligent design system and method for complex rule constraints

Abstract

The application discloses a base parameter intelligent design system and method for complex rule constraints, and relates to the technical field of computer-aided design.The system comprises an input module, a preprocessing module, a rule classification processing module, a multi-stage parameter generation module, a structured output and checking module, a parameter scheme preview and backwriting correction module, and a result output module.The system and method can realize stable generation, checking, correction and output of base parameter schemes under complex rule constraints, and improve the interpretability, reviewability and engineering deliverability of parameter design results.

Description

Technical Field

[0001] This invention relates to the field of computer-aided design technology, and in particular to an intelligent design system and method for base parameters oriented towards complex rule constraints. Background Technology

[0002] The design of base parameters is a crucial step in the formation and refinement of engineering schemes. Its design results not only affect the structural layout, hole location setting, and special configuration treatment, but also subsequent geometric modeling, verification analysis, and drawing or data writing. Therefore, the design of base parameters is not a simple data entry, but a comprehensive process that includes rule matching, parameter solving, conflict coordination, and result correction.

[0003] Existing base parameter design usually relies heavily on engineers' experience. When faced with a large number of rules, manual verification is required for each rule, which results in low efficiency, easy omissions, and design quality being greatly affected by individual abilities. When there is a strong coupling relationship between parameters, local modifications can easily trigger chain changes, and the management of solution versions and the work of writing back and correcting after multiple rounds of modifications are also more complicated.

[0004] On the other hand, as large language models are increasingly used in text understanding, rule-based text interpretation, and structured information extraction, existing large model-based auxiliary design methods, while flexible to a certain extent, can often only handle simple numerical relationships or a small number of rules in engineering scenarios. They are difficult to cover numerical calculation rules, text judgment rules, and geometric space rules simultaneously. Furthermore, when the number of rules increases and multi-parameter coupling is enhanced, "numerical illusion," unstable format, rule omissions, and parameter conflicts are prone to occur.

[0005] Furthermore, in industrial or classified application environments, locally deployed medium-sized models are often required. Compared to ultra-large cloud-based models, these models are more prone to numerical errors, format drift, and constraint violations when processing all rules, structural objects, and constraints at once. Therefore, single-agent end-to-end generation methods are difficult to directly meet engineering usability requirements.

[0006] Therefore, there is an urgent need for a base parameter intelligent design system and method that can adapt to complex rule constraints, multi-parameter coupling and local deployment environments, so as to improve the stability, controllability and engineering deliverability of parameter design. Summary of the Invention

[0007] The purpose of this invention is to provide an intelligent design system and method for base parameters oriented towards complex rule constraints, which solves the problems of high dependence on manual work, complex rule types, strong parameter coupling, low design efficiency, and the tendency of general large models to suffer from reasoning illusion, format drift, and difficulty in writing back results in complex constraint scenarios during the existing base parameter design process.

[0008] To achieve the above objectives, the present invention provides an intelligent design system for base parameters oriented towards complex rule constraints, comprising: The input module is used to perform the design requirements and basic information input steps, and to receive base design requirements, equipment information, cabin information, rule constraint information and additional user requirements; The preprocessing module is used to perform input parsing and structured preprocessing steps, parsing, splitting and structuring the information received by the input module; The rule parameter generation module is used to perform the rule parameter generation phase, generating rule parameter results based on the preprocessing results. The hole parameter generation module is used to perform the hole parameter generation stage, generating hole parameter results based on the rule parameter results and hole position-related constraints. The special parameter generation module is used to perform the special parameter generation stage, generating special parameter results based on the results of the previous stage and special constraints. The structured output and verification module is used to perform the structured output and verification stage, and to perform structured constraints, format verification and anomaly detection on the results of each stage. The parameter scheme preview and write-back module is used to perform the parameter scheme preview and write-back phase, and to summarize, revise, write back and confirm the items to be processed; The results output module is used to execute the steps of outputting the final parameter scheme, parameter change table, reference rule table, and final results.

[0009] Preferably, the preprocessing module includes: The basic information extraction unit is used to perform the steps of extracting basic information about the base, equipment, and ship cabin. The constraint information extraction unit is used to execute extraction rule constraints and user-required additional steps. The task context building unit is used to perform the steps of building a unified structured task context, organizing basic information, constraint information, and call templates into a unified data context. The unified structured task context includes at least user additional requirements, base triples, equipment triples, cabin triples, and structured data templates for inter-stage transfer. The structured data templates include at least one of the following: JSON format, XML format, key-value pair data structure, and database table structure.

[0010] Preferably, the rule parameter generation module completes rule clause matching and basic rule parameter solving based on the rule parameter generation rules and the structured task context, and outputs rule parameter results, geometric preview information, write-back prompts, parameter change table and reference rule table.

[0011] Preferably, the aperture parameter generation module generates interference aperture parameters or weight reduction aperture parameters based on the output results of the previous stage, aperture parameter generation rules, and structured task context, and outputs aperture parameter results, geometric preview information, write-back prompts, parameter change table, and reference rule table.

[0012] Preferably, the special parameter generation module processes the special hard point design based on the output results of the previous stage, the special parameter generation rules, and the structured task context, and outputs the special parameter results, geometric preview information, write-back prompts, parameter change table, and reference rule table.

[0013] Preferably, the structured output and verification module uses a JSON template to constrain the stage output results and performs null value detection, field missing detection, data type verification, numerical range check, format validity verification, and parameter naming consistency verification.

[0014] Preferably, the structured output and verification module includes: Anomaly detection unit is used to determine whether the anomaly detection output is abnormal; The anomaly rollback and regeneration unit is used to execute the trigger anomaly rollback and regeneration steps when the anomaly detection output result is abnormal, so as to return the regenerated result of the corresponding anomaly stage.

[0015] Preferably, the parameter scheme preview and write-back module includes: The summary unit is used to perform the steps of summarizing conflicting items, pending confirmation items, and direct replacement items; The user revision unit is used to perform user revision hit clauses, parameter formats, and geometric description steps; The write-back judgment unit is used to perform the step of judging whether the previous result needs to be written back; The result write-back unit is used to perform the step of writing back the corrected results to the rule parameter results, hole parameter results, and / or special parameter results when write-back is required.

[0016] A method for intelligent design system of base parameters oriented towards complex rule constraints includes the following steps: S100, Design Requirements and Basic Information Input: Receives base design requirements, equipment information, cabin information, rule constraint information, and additional user requirements; S200, Input Parsing and Structured Preprocessing: This involves performing structured parsing on the input content of S100, including: S210. Extract basic information about the base, equipment, and ship cabin; S220, Extraction rule constraints and additional user requirements; S230. Construct a unified structured task context; Among them, S230 uses JSON structured data as the data carrier for front-end and back-end parsing, inter-stage transmission, format validation, write-back correction and version management; S300, the rule parameter generation stage, generates rule parameter results based on the unified structured task context of S230; S400, Hole parameter generation stage: Hole parameter results are generated based on rule parameter results and unified structured task context; S500, Special Parameter Generation Stage: Special parameter results are generated based on the hole parameter results and the unified structured task context. In the S600 stage of structured output and verification, the outputs of S300, S400, and S500 are subjected to structured constraints, format verification, and anomaly detection, including: S610. Determine if an exception exists; if yes, proceed to S611 to trigger exception rollback and regeneration; if no, proceed to S700. S700, parameter scheme preview and write-back stage, including: S710, summarize conflict items, pending confirmation items, and direct replacement items; S720, User-revisioned hit clauses, parameter formats, and geometric descriptions; User revisions include at least revisions to hit clauses, parameter formats, and geometric descriptions; S730. Determine whether it is necessary to write back the previous results; if yes, proceed to S731 to write back the corrected results to the rule parameter results, hole parameter results, or special parameter results; if no, proceed to step S800. Among them, the write-back object of S731 includes at least one or more of the following: regular parameter results, hole parameter results, and special parameter results; S800 outputs the final parameter scheme, parameter change table, reference rule table, and final result.

[0017] Preferably, in steps S300, S400, and S500, the rules are classified and processed. The rules include at least numerical calculation rules, text judgment rules, and geometric space rules. The numerical calculation rules are solved using fixed code or scripts, the text judgment rules are solved using logical text parsing and condition selection, and the geometric space rules are solved using two-dimensional parameter relationship or geometric relationship.

[0018] Therefore, the present invention employs the above-mentioned intelligent design system and method for base parameters oriented towards complex rule constraints, which has the following beneficial effects: (1) This invention reduces the complexity of single inference by dividing the complex base parameter design task into four stages: rule parameter generation, hole parameter generation, special parameter generation, and parameter scheme preview and write-back. It is suitable for completing complex parameter design tasks under local medium-sized model conditions.

[0019] (2) The present invention can simultaneously process numerical calculation rules, text judgment rules and geometric space rules, and adapt to multi-rule and multi-parameter coupling scenarios. It has a stronger ability to process complex rules than existing intelligent agents that can only process simple rules or a small number of rules.

[0020] (3) The present invention adopts JSON template, rule hit management, format verification and write-back correction mechanism to form a closed loop of "rule-generation-verification-write-back", which can improve the structural stability, traceability, reviewability and project deliverability of parameter output.

[0021] (4) This invention supports parameter import and transfer between multi-stage results, and supports returning to any previous stage input for adjustment in the final preview stage, which is more in line with the engineering review and manual intervention process.

[0022] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the overall architecture of an intelligent design system for base parameters oriented towards complex rule constraints according to the present invention; Figure 2 This is a flowchart of an intelligent design method for base parameters oriented towards complex rule constraints, according to the present invention. Detailed Implementation

[0024] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0025] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0026] Example Please see Figures 1-2 This invention provides an intelligent design system for base parameters oriented towards complex rule constraints. The system constructs an intelligent design process in the order of S100 to S800, and includes at least an input module, a preprocessing module, a rule parameter generation module, a hole parameter generation module, a special parameter generation module, a structured output and verification module, a parameter scheme preview and write-back module, and a result output module. The functions of each module are as follows: The input module is used for the S100 to perform the design requirements and basic information input steps, and receives base design requirements, equipment information, cabin information, rule constraint information and additional user requirements.

[0027] The preprocessing module is used to perform the input parsing and structured preprocessing steps of S200, which parses, splits and structures the information received by the input module.

[0028] The preprocessing module includes a basic information extraction unit, a constraint information extraction unit, and a task context construction unit. The basic information extraction unit extracts basic information about the base, equipment, and ship compartment. The constraint information extraction unit extracts rule constraints and user-defined additional requirements. The task context construction unit constructs a unified structured task context, organizing the basic information, constraint information, and call templates into a unified data context. This unified structured task context includes at least user-defined additional requirements, base triples, equipment triples, ship compartment triples, and structured data templates for inter-stage transfer. The structured data templates include at least one of the following formats: JSON, XML, key-value pairs, and database table structures.

[0029] The rule parameter generation module is used to perform the rule parameter generation phase of S300, generating rule parameter results based on the preprocessing results. According to the rule parameter generation rules and the structured task context, the rule parameter generation module completes rule clause matching, basic rule parameter solving, and outputs rule parameter results, geometry preview information, write-back prompts, parameter change tables, and referenced rule tables.

[0030] The hole parameter generation module is used to execute the hole parameter generation stage of S400. It generates hole parameter results based on the rule parameter results and hole position-related constraints. The hole parameter generation module generates interference hole parameters or weight reduction hole parameters according to the output results of the previous stage, the hole parameter generation rules, and the structured task context, and outputs the hole parameter results, geometric preview information, write-back prompts, parameter change table, and reference rule table.

[0031] The special parameter generation module is used to perform the special parameter generation phase of S500, generating special parameter results based on the results of the previous phase and special constraints. The special parameter generation module processes the special hard point design according to the output results of the previous phase, the special parameter generation rules, and the structured task context, and outputs the special parameter results, geometric preview information, write-back prompts, parameter change table, and reference rule table.

[0032] The structured output and validation module is used to perform the structured output and validation phase of S600, applying structured constraints, format validation, and anomaly detection to the results of each phase. For the outputs of S300, S400, and S500, this module uses JSON templates to constrain the phase outputs and performs null value detection, field missing detection, data type validation, numerical range checks, format validity checks, and parameter naming consistency checks.

[0033] The structured output and verification module includes an anomaly detection unit and an anomaly rollback and regeneration unit. The anomaly detection unit determines whether the anomaly detection output is abnormal. The anomaly rollback and regeneration unit, when an anomaly is detected in the anomaly detection output, executes an anomaly rollback and regeneration step to return the regenerated result for the corresponding anomaly stage.

[0034] The parameter scheme preview and write-back module is used to perform the parameter scheme preview and write-back phase of S700, and to summarize, revise, write back and confirm the items to be processed.

[0035] The parameter scheme preview and write-back module includes a summary unit, a user revision unit, a write-back judgment unit, and a result write-back unit. The summary unit is used to summarize conflicting items, items awaiting confirmation, and direct replacement items. The user revision unit is used to revise the hit clauses, parameter formats, and geometric descriptions. The write-back judgment unit is used to determine whether to write back previous results. The result write-back unit, when write-back is required, is used to write back the corrected results to the rule parameter results, hole parameter results, and / or special parameter results.

[0036] The results output module is used to output the final parameter scheme, parameter change table, reference rule table and final result steps of S800 execution, for subsequent project review, data write-back and result delivery.

[0037] This system is deployed in a local large language model environment. By dividing the task into a multi-stage processing flow from S100 to S800, the complexity of the context of a single inference is reduced, thereby reducing numerical errors, format drift, rule omissions and parameter conflicts.

[0038] This system divides the base parameter design process into four stages: rule parameter generation, hole parameter generation, special parameter generation, and parameter scheme preview and write-back. Different solution mechanisms are used to process numerical calculation rules, text judgment rules, and geometric space rules. A closed-loop generation process is formed by combining the prompt construction mechanism based on example constraint templates, JSON template constraints, rule hit management, format verification, and write-back correction mechanism.

[0039] A method for intelligent design system of base parameters oriented towards complex rule constraints includes the following steps: S100, Design Requirements and Basic Information Input: Receives base design requirements, equipment information, cabin information, rule constraint information, and additional user requirements.

[0040] S200, Input Parsing and Structured Preprocessing: This involves performing structured parsing on the input content of S100, including: S210, Extract basic information about the base, equipment, and ship cabin.

[0041] S220, Extraction rule constraints and additional user requirements.

[0042] S230, Construct a unified structured task context.

[0043] S230 uses JSON structured data as the data carrier for front-end and back-end parsing, inter-stage transmission, format validation, write-back correction, and version management.

[0044] S300, the rule parameter generation stage, generates rule parameter results based on the unified structured task context of S230.

[0045] S400, Hole Parameter Generation Stage: Hole parameter results are generated based on rule parameter results and unified structured task context.

[0046] S500, Special Parameter Generation Stage: Special parameter results are generated based on the hole parameter results and the unified structured task context.

[0047] In the S600 stage of structured output and verification, the outputs of S300, S400, and S500 are subjected to structured constraints, format verification, and anomaly detection, including: S610. Determine if there is an anomaly; if yes, proceed to S611 to trigger an anomaly rollback and regeneration to improve the stability of the parameter results and the constraint satisfaction; if no, proceed to S700.

[0048] S700, parameter scheme preview and write-back stage, including: S710, summarize conflicting items, pending confirmation items, and direct replacement items.

[0049] S720, User-Revised Hit Terms, Parameter Formats, and Geometry Descriptions.

[0050] User revisions include at least revisions to hit clauses, parameter formats, and geometric descriptions.

[0051] S730. Determine whether it is necessary to write back the previous results; if yes, proceed to S731 to write back the corrected results to the rule parameter results, hole parameter results, or special parameter results, thereby forming a parameter revision closed loop; if no, proceed to step S800.

[0052] The write-back object of S731 includes at least one or more of the following: regular parameter results, hole parameter results, and special parameter results.

[0053] S800 outputs the final parameter scheme, parameter change table, reference rule table, and final result.

[0054] In S300, S400, and S500, rules are classified and processed. The rules include at least numerical calculation rules, text judgment rules, and geometric space rules. Numerical calculation rules are solved using fixed codes or scripts, text judgment rules are solved using logical text parsing and condition selection, and geometric space rules are solved using two-dimensional parametric relationships or geometric relationships.

[0055] Example 1 This embodiment 1 is a combination Figure 1 The overall system architecture implementation method.

[0056] like Figure 1 As shown, this invention provides an intelligent design system for base parameters oriented towards complex rule constraints. The system is built around the base parameter design task, corresponding to the processing chain from steps S100 to S800, including an input module, a preprocessing module, a rule parameter generation module, a hole parameter generation module, a special parameter generation module, a structured output and verification module, a parameter scheme preview and write-back module, and a result output module. The system can be further divided into an input layer, a temporary processing layer, and a knowledge support layer. The input layer organizes the original design input, the temporary processing layer completes the phased intelligent generation and intermediate result transmission, and the knowledge support layer provides rule knowledge, triplet knowledge, and parameter change knowledge support for each stage.

[0057] In this embodiment, the input module is used to perform step S100, which involves inputting design requirements and basic information. The user must input at least the base template triplet, equipment triplet, rule constraint information, and any additional user requirements. In some implementations, special constraint triplets, pipeline / cable triplets, or compartment triplets can also be input to support parameter generation tasks in subsequent stages. The input content can be manually entered or derived from data files or structured results imported from the previous design stage.

[0058] The preprocessing module is used to perform input parsing and structured preprocessing in step S200. In this step, the system first performs step S210 to extract basic information about the base, equipment, and cabin, parsing the structural parameters, relational parameters, and necessary identification information in the base, equipment, and cabin objects. Then, step S220 extracts rule constraints and additional user requirements, separating rule clauses, constraints, supplementary restrictions, and personalized user requirements from the original input. Finally, step S230 constructs a unified structured task context, organizing the above information into structured context data that can be called by subsequent intelligent agents. The structured context is preferably stored in JSON format to support multi-stage transmission, front-end and back-end parsing, format validation, and version management.

[0059] Example 2 This embodiment 2 is a combination Figure 2 The multi-stage method and process implementation method.

[0060] like Figure 2 As shown, this embodiment provides an intelligent design method for base parameters oriented towards complex rule constraints, which mainly includes three parts: input preprocessing, multi-stage parameter generation, and verification write-back output.

[0061] First, the input preprocessing steps S100 to S230 are performed. The system receives base design requirements, equipment information, cabin information, rule constraint information, and additional user requirements, and parses and structures this information. Specifically, in step S100, design requirements and basic information are received; in step S200, the input information is parsed and preprocessed in a structured manner; in step S210, basic parameters of the base, equipment, and cabin are extracted; in step S220, rule constraints and additional user requirements are identified; and in step S230, a unified structured task context is constructed. The unified structured task context is used for subsequent parameter generation, inter-stage data transfer, structured verification, and result write-back.

[0062] Secondly, the system performs multi-stage parameter generation steps S300 to S500. Based on a unified structured task context, the system sequentially generates rule parameter results, hole parameter results, and special parameter results. Specifically, in step S300, rule parameter results are generated by matching applicable rules and processing numerical calculation rules, text-based rules, and geometric space rules. In step S400, hole parameter results are generated based on the rule parameter results and hole position-related constraints; these results include interference hole parameters and weight-reduction hole parameters. In step S500, special parameter results are generated based on the results of the preceding stages and special constraints; these results include parameters for special hard points, local reinforcement, special avoidance, or interface adaptation. The results generated at each stage can include parameter results, geometric preview information, write-back prompts, parameter change tables, and reference rule tables.

[0063] Finally, the verification and write-back output of steps S600 to S800 are executed. The system performs structured output and verification of the rule parameter results, hole parameter results, and special parameter results. In step S600, field integrity, data type, numerical range, format legality, and parameter naming consistency are verified; in step S610, it is determined whether there are any anomalies; if there are any anomalies, the process proceeds to step S611, where the process reverts to the corresponding parameter generation stage based on the source of the anomaly, and regenerates the abnormal or conflicting fields; if there are no anomalies, the process proceeds to step S700, where the parameter scheme preview results are summarized, and direct replacement items, pending confirmation items, and conflicting items are generated. In step S720, the system receives the user's revision information on the parameter scheme preview results; in step S730, it determines whether it is necessary to write back the previous results; if it is necessary to write back, it proceeds to step S731 to write back the revised results to the corresponding rule parameter results, hole parameter results, or special parameter results, and can execute step S600 again; if it is not necessary to write back, it proceeds to step S800 to output the final parameter scheme, parameter change table, reference rule table, and final result file.

[0064] Through the above methods, this embodiment will Figure 2 The intelligent design process for base parameters shown is divided into three consecutive parts: input preprocessing (steps S100 to S230), multi-stage parameter generation (steps S300 to S500), and verification and write-back output (steps S600 to S800). It can generate, verify, revise, and output parameter results under complex rule constraints, thereby improving the stability and traceability of base parameter design results.

[0065] Therefore, the present invention employs the above-mentioned intelligent design system and method for base parameters under complex rule constraints, which can realize the stable generation, verification, correction and output of base parameter schemes under complex rule constraints, and improve the interpretability, reviewability and engineering deliverability of parameter design results.

[0066] Finally, 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 preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A smart design system for base parameters oriented towards complex rule constraints, characterized in that, include: The input module is used to perform the design requirements and basic information input steps, and to receive base design requirements, equipment information, cabin information, rule constraint information and additional user requirements; The preprocessing module is used to perform input parsing and structured preprocessing steps, parsing, splitting and structuring the information received by the input module; The rule parameter generation module is used to perform the rule parameter generation phase, generating rule parameter results based on the preprocessing results. The hole parameter generation module is used to perform the hole parameter generation stage, generating hole parameter results based on the rule parameter results and hole position-related constraints. The special parameter generation module is used to perform the special parameter generation stage, generating special parameter results based on the results of the previous stage and special constraints. The structured output and verification module is used to perform the structured output and verification stage, and to perform structured constraints, format verification and anomaly detection on the results of each stage. The parameter scheme preview and write-back module is used to perform the parameter scheme preview and write-back phase, and to summarize, revise, write back and confirm the items to be processed; The results output module is used to execute the steps of outputting the final parameter scheme, parameter change table, reference rule table, and final results.

2. The intelligent design system for base parameters oriented towards complex rule constraints according to claim 1, characterized in that, The preprocessing module includes: The basic information extraction unit is used to perform the steps of extracting basic information about the base, equipment, and ship cabin. The constraint information extraction unit is used to execute extraction rule constraints and user-required additional steps. The task context building unit is used to perform the steps of building a unified structured task context, organizing basic information, constraint information, and call templates into a unified data context. The unified structured task context includes at least user additional requirements, base triples, equipment triples, cabin triples, and structured data templates for inter-stage transfer. The structured data templates include at least one of the following: JSON format, XML format, key-value pair data structure, and database table structure.

3. The intelligent design system for base parameters oriented towards complex rule constraints according to claim 2, characterized in that: The rule parameter generation module, based on the rule parameter generation rules and the structured task context, completes the rule clause matching, basic rule parameter solving, and outputs the rule parameter results, geometric preview information, write-back prompts, parameter change table, and reference rule table.

4. The intelligent design system for base parameters oriented towards complex rule constraints according to claim 3, characterized in that: The aperture parameter generation module generates interference aperture parameters or weight reduction aperture parameters based on the output results of the previous stage, aperture parameter generation rules, and structured task context, and outputs aperture parameter results, geometric preview information, write-back prompts, parameter change table, and reference rule table.

5. The intelligent design system for base parameters oriented towards complex rule constraints according to claim 4, characterized in that: The special parameter generation module processes the special hard point design based on the output results of the previous stage, the special parameter generation rules, and the structured task context, and outputs the special parameter results, geometric preview information, write-back prompts, parameter change table, and reference rule table.

6. The intelligent design system for base parameters oriented towards complex rule constraints according to claim 5, characterized in that: The structured output and validation module uses JSON templates to constrain the stage output results and performs null value detection, field missing detection, data type validation, numerical range check, format validity validation, and parameter naming consistency validation.

7. The intelligent design system for base parameters oriented towards complex rule constraints according to claim 6, characterized in that, The structured output and verification module includes: Anomaly detection unit is used to determine whether the anomaly detection output is abnormal; The anomaly rollback and regeneration unit is used to execute the trigger anomaly rollback and regeneration steps when the anomaly detection output result is abnormal, so as to return the regenerated result of the corresponding anomaly stage.

8. The intelligent design system for base parameters oriented towards complex rule constraints according to claim 7, characterized in that, The parameter scheme preview and write-back module includes: The summary unit is used to perform the steps of summarizing conflicting items, pending confirmation items, and direct replacement items; The user revision unit is used to perform user revision hit clauses, parameter formats, and geometric description steps; The write-back judgment unit is used to perform the step of judging whether the previous result needs to be written back; The result write-back unit is used to perform the step of writing back the corrected results to the rule parameter results, hole parameter results, and / or special parameter results when write-back is required.

9. The method for intelligent design system of base parameters oriented towards complex rule constraints as described in claim 8, characterized in that, Includes the following steps: S100, Design Requirements and Basic Information Input: Receives base design requirements, equipment information, cabin information, rule constraint information, and additional user requirements; S200, Input Parsing and Structured Preprocessing: This involves performing structured parsing on the input content of S100, including: S210. Extract basic information about the base, equipment, and ship cabin; S220, Extraction rule constraints and additional user requirements; S230. Construct a unified structured task context; Among them, S230 uses JSON structured data as the data carrier for front-end and back-end parsing, inter-stage transmission, format validation, write-back correction and version management; S300, the rule parameter generation stage, generates rule parameter results based on the unified structured task context of S230; S400, Hole parameter generation stage: Hole parameter results are generated based on rule parameter results and unified structured task context; S500, Special Parameter Generation Stage: Special parameter results are generated based on the hole parameter results and the unified structured task context. In the S600 stage of structured output and verification, the outputs of S300, S400, and S500 are subjected to structured constraints, format verification, and anomaly detection, including: S610. Determine if an exception exists; if yes, proceed to S611 to trigger exception rollback and regeneration; if no, proceed to S700. S700, parameter scheme preview and write-back stage, including: S710, summarize conflict items, pending confirmation items, and direct replacement items; S720, User-revisioned hit clauses, parameter formats, and geometric descriptions; User revisions include at least revisions to hit clauses, parameter formats, and geometric descriptions; S730. Determine whether it is necessary to write back the previous results; if yes, proceed to S731 to write back the corrected results to the rule parameter results, hole parameter results, or special parameter results; if no, proceed to step S800. Among them, the write-back object of S731 includes at least one or more of the following: regular parameter results, hole parameter results, and special parameter results; S800 outputs the final parameter scheme, parameter change table, reference rule table, and final result.

10. The method for an intelligent design system of base parameters oriented towards complex rule constraints according to claim 9, characterized in that: In steps S300, S400, and S500, the rules are classified and processed. The rules include at least numerical calculation rules, text judgment rules, and geometric space rules. Numerical calculation rules are solved using fixed code or scripts, text judgment rules are solved using logical text parsing and condition selection, and geometric space rules are solved using two-dimensional parameter relationships or geometric relationships.

Images