An AI intelligent agent for automatic selection and assembly design of production line floor rail functional components

By using AI intelligent agents for the automatic selection and assembly design of ground rail drive systems, the problems of scattered calculation basis and assembly errors in the selection of functional components of ground rail drive systems are solved. This enables efficient and accurate generation of design schemes and ground rail assembly, and improves the reusability and traceability of the design.

CN123116501APending Publication Date: 2026-11-17SHANGHAI JIAOTONG UNIV +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611495718.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-23
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In the selection process of existing ground rail drive system functional components, designers need to convert between multiple tables and product samples. The calculation basis is scattered, making it difficult to form a complete design scheme. Moreover, errors and repetitive operations are prone to occur during the ground rail assembly process, making it difficult to meet multiple performance requirements and redundancy issues.

Method used

AI intelligent agents are used for automatic selection and assembly design of ground rail functional components. Through structured input and continuous calculation of transmission requirements, combined with a hierarchical collaborative selection method of reducer-helical gear-servo motor, intelligent selection of transmission system components and intelligent assembly of drag chain open ground rail model are realized. The user interface, physical calculation module, rule constraint reasoning module and other modules work together to generate traceable design schemes.

Benefits of technology

This enables efficient and accurate selection and assembly of the ground rail drive system, reduces errors from manual operation, improves the reusability and traceability of the design, ensures the performance matching of the drive system and the accuracy of ground rail assembly, and reduces repetitive work and uncertainty in manual selection.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123116501A_ABST
    Figure CN123116501A_ABST
Patent Text Reader

Abstract

This invention discloses an AI agent for the automatic selection and assembly design of functional components for production line floor rails, relating to the field of production line floor rail equipment. Based on a unified structured project configuration, it completes the selection and generation of functional component solutions for production line floor rails, the intelligent establishment of floor rail assembly, and the intelligent assembly verification of production line floor rails within a single design process. Input and result identifiers are shared within a unified project, and collaborative design outputs a complete production line floor rail design scheme. This invention utilizes deterministic mechanical models, a structured product database, a standard floor rail module knowledge base, and executable assembly rules to complete candidate scheme generation, verification, iterative iteration, and 3D assembly. The AI ​​agent system of this invention can be deployed in an enterprise's computer and modeling software environment, transforming design experience into repeatable, interpretable, and traceable digital design capabilities, possessing a clear software deployment foundation and promising application prospects in production line floor rail equipment design.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of production line floor rail equipment, and in particular to an AI intelligent agent for automatic selection and assembly design of functional components for production line floor rails. Background Technology

[0002] Automated production lines are crucial equipment for achieving efficient and flexible production in modern manufacturing. Ground rail systems, as key foundational equipment enabling the spatial movement of robots and supporting mechanisms within production line systems, are widely used in flexible production systems. The ground rail drive system uses a servo drive system to drive the supporting components along the track to achieve high-precision linear motion. Its drive system typically consists of servo motors, reducers, and functional components such as gears and racks. The proper selection of these functional components directly affects the equipment's load-bearing capacity, motion accuracy, and operational stability, and is a critical factor in ensuring that the ground rail equipment meets the requirements for specified loads, rapid traverse speeds, acceleration times, and strokes. In current engineering designs, the selection of functional components for the ground rail drive system is usually done by designers based on project inputs, completing calculations of speed and acceleration, frictional resistance and acceleration force, gear parameters, reducer output torque verification, iterative calculations of motor speed and torque, and load inertia conversion, before consulting different product catalogs to select a specific model. Because parameters and reference materials are scattered among mechanical design reference manuals, component product manuals and personal experience, unit conversions, formula selections and design constants are prone to differences due to different personnel. The calculation basis and component model screening and selection iteration process are also difficult to retain completely, which is not convenient for subsequent changes and traceability.

[0003] Furthermore, the connections and interactions between components in a ground rail drive system are not independent. For rack and pinion drives used in ground rail equipment, the module and number of teeth of the helical gear determine the pitch circle diameter and the required speed and torque at the reducer's output end. The discrete reduction ratio of the reducer also alters the required speed, rated torque, acceleration torque, and equivalent inertia on the motor side. The reducer's output shaft and gear bore diameter, as well as the motor shaft diameter and allowable dimensions at the reducer's input end, must be matched. Therefore, meeting the performance parameters of a single component does not equate to forming a usable combination. Designers often have to manually select components by searching through reducer, helical gear, and motor catalogs. If the selected component combination does not meet the requirements in terms of rated torque, peak torque, speed, inertia ratio, or installation dimensions, the component must be replaced and recalculated. This process involves numerous repetitive calculations, easily overlooking constraints, and making it difficult to determine the performance utilization of each component in a feasible combination, whether it meets the given requirements, and the redundancy situation. Existing general selection tables or single-component selection tools often focus on formula calculations or single product catalog searches, making it difficult to organize the motion requirements of ground rail functional components, the physical model of the transmission system, the component database, the coupling relationships between components, multiple rule constraints, and the candidate replacement process into a repeatable closed loop. Even if designers obtain several components that meet basic capabilities, it is difficult to determine whether the utilization of various performance aspects is balanced. If only the utilization rate of one aspect is pursued to be close to the upper limit, it may lead to other components being significantly oversized; if only familiar models are selected, it may result in unnecessary performance redundancy and increased costs.

[0004] For common floor rail production line designs, enterprises can adopt standardized methods for floor rail designs under different loads. Based on common design requirements, several standard floor rail models can be established to meet different load conditions. These models typically include head floor rail sections, intermediate floor rail sections, functional floor rail sections, and tail floor rail sections. Therefore, for the creation and assembly of floor rail 3D models, existing design methods rely on designers manually selecting head floor rail sections, intermediate standard sections, floor rail sections with cable chain openings, and tail section models from a model library after calculating the target length, and then inserting, positioning, and saving them segment by segment in CAD design software. Floor rail assembly is usually used in conjunction with RGVs or robots, requiring the configuration of cable chain structures. Using standardized floor rail models, the cable chain only has a limited number of actual opening positions. For different design requirements, the user-specified cable chain opening coordinates are usually continuous values. The user-specified coordinates and the standard opening positions often do not completely coincide. Designers need to first determine which standard segment the specified coordinates are located in, then select the closest model from multiple opening models, and redetermine the actual opening coordinates of the entire floor rail line. If this process relies solely on manual judgment, it is inefficient and prone to problems such as incorrect selection of opening sections, incorrect number of intermediate sections of the ground rail, failure to provide feedback on the actual opening position, and the drag chain installation position still being calculated based on the original specified coordinates.

[0005] Furthermore, the actual length of the ground rail required in production is not always directly assembled from several fixed-length standard segments. Designers need to calculate the tail segment length while retaining standard modules and choose between standard integer tail segments and parameterized tail segments. If the tail segment length, opening segment position, number of standard segments, and drag chain positioning coordinates are calculated separately, problems such as the total ground rail length not being able to close, overlapping or missing segments of adjacent ground rail segments, and inconsistencies between the drag chain installation position and the actual opening can easily occur. After completing the material calculation, designers also need to repeatedly perform operations such as creating new assemblies, opening or modifying parameterized templates, inserting components, setting absolute coordinates, saving models, and collecting dependency files in CAD software (such as NX, SolidWorks, etc.), making the entire process highly repetitive. For repetitive tasks, programming languages ​​(such as Python) are often used in conjunction with CAD software for secondary development to improve design efficiency. However, ordinary programming environments are suitable for parameter management, physical calculations, knowledge rule reasoning, and user interaction, while only the CAD software environment can directly call software interfaces to complete modeling and automated assembly. If there is no unified configuration and status feedback mechanism between the programming environment and the modeling software environment, problems such as updated operating parameters but the modeling software still reading the old parameters, the inability to provide feedback on assembly execution results, and the need for manual re-entry of calculation results into the CAD software can easily occur. Therefore, when carrying out secondary development for the ground rail assembly function, a stable data transfer and task collaboration relationship needs to be established between the design decision-making environment and the CAD software execution environment.

[0006] Therefore, it is necessary to establish an AI agent for the automatic selection of functional components and the efficient assembly design of 3D models for production line ground rail equipment. This agent integrates the design requirement input, transmission parameter calculation, component knowledge base query, component selection scheme generation and verification evaluation in the selection of transmission functional components with the standard ground rail section module mapping, ground rail section assembly calculation, tail section parameterization, local track section location positioning, and automatic assembly organization using CAD software in the ground rail assembly model building process as collaborative design tasks. This agent does not rely on manual operation but completes the design through physical models, structured knowledge, engineering rules, finite space search, interactive decision-making, and the invocation of external engineering software tools, ensuring that both functional component schemes and ground rail assembly schemes have clear, interpretable, and traceable formation bases.

[0007] Therefore, those skilled in the art are dedicated to developing an AI agent for the automatic selection and assembly design of production line floor rail functional components. Summary of the Invention

[0008] In view of the above-mentioned deficiencies of the prior art, the technical problem to be solved by the present invention is that the selection of ground rail transmission involves multiple types of inputs such as load, speed, acceleration time, friction, transmission efficiency and gear parameters. Manual calculation and copying are required between multiple tables and product samples, and the process from the original input to the formation of the required parameters is difficult to verify.

[0009] To achieve the above objectives, this invention provides an AI agent for the automatic selection and assembly design of production line floor rail functional components. The AI ​​agent, based on a unified structured project configuration, completes the selection of production line floor rail functional component schemes and the intelligent establishment of floor rail assembly within a single design process. This includes intelligent selection of transmission system components and intelligent assembly of cable chain open floor rail models. The intelligent selection of transmission system components and the intelligent assembly of cable chain open floor rail models share input and result identifiers within a unified project, collaboratively designing and outputting a complete production line floor rail design scheme. The intelligent selection of transmission system components completes the selection of production line ground rail functional components and generates a solution. Taking the transmission chain composed of servo motor, reducer and drive gear as the design object, it completes motion requirement calculation, candidate combination generation, multi-constraint loop verification, performance redundancy evaluation and user interaction to filter and output the solution. The intelligent assembly of the cable chain opening ground rail model completes the intelligent assembly of the production line ground rail. Taking the standard ground rail section, the ground rail section with cable chain opening, the ground rail tail section and the cable chain as the design objects, it completes the specified cable chain opening coordinate mapping, ground rail section material calculation, tail section preparation, cable chain length, positioning calculation of each section and automatic assembly by modeling software.

[0010] Furthermore, the AI ​​agent includes a user interface, structured data configuration, a physical calculation module, a component knowledge base definition, a rule-constrained reasoning module, a scheme evaluation module, a standard ground rail module knowledge base, an assembly planning module, a modeling software task scheduling module, an execution module, and a result output module. Each module transmits data through structured configuration, candidate scheme objects, assembly planning objects, logs, and output files; among which, The user interface collects operating conditions and displays candidate solutions; The physical calculation module generates the calculation requirements of the transmission system based on the user input information. The rule-constrained reasoning module traverses and eliminates candidate combinations of functional component schemes; The assembly planning module converts the continuous opening coordinates of the cable chain into existing standard modules and the number of each segment. The modeling software task scheduling module starts the software's internal execution environment; The execution module calls relevant software functions to complete parametric modeling and component placement.

[0011] Furthermore, the intelligent selection of transmission system components, when completing the selection of production line floor rail functional components and generating a solution, includes the following steps: S101: Obtain the design parameters of the production line floor rail, convert the design parameters into a structured configuration, and verify the design parameters; S103: Calculate the kinematics and speed-related parameters of the transmission system; S105: Generate candidate solutions for functional components based on component parameters; S107: Perform multi-constraint verification on the candidate schemes for the functional components; S109: Evaluate and filter the candidate solutions for the verified functional components and output all feasible solutions; S111: Confirm the target solution and output traceable selection results.

[0012] Furthermore, in step S103, based on the structured input and transmission demand continuous calculation method for production line ground rail conditions, motion demand, transmission parameters and allowable utilization coefficient are unified as input, and driving force, gear size, output end demand, motor side demand and equivalent inertia are formed in sequence. By converting speed into linear velocity and calculating acceleration, acceleration force, friction force and total driving force are calculated. The pitch circle diameter and stroke per revolution are determined by gear module and number of teeth. Then, the requirements of reducer and motor are calculated step by step in combination with candidate reduction ratio and efficiency, so that intermediate quantities are transmitted through the same input chain.

[0013] Furthermore, in step S105, the hierarchical collaborative selection method of "reducer-helical gear-servo motor" based on the structured component database establishes performance relationships and two-level connection dimension relationships in the same combination, with the reducer as the outer layer, the helical gear as the middle layer, and the motor as the inner layer. By reading the reduction ratio, torque, speed, inertia, and end dimensions of the candidate reducer from the functional component data knowledge, the output shaft is first matched with the gear bore diameter and the output side requirements are calculated; then, the torque and speed requirements of the servo motor are calculated according to the reduction ratio, and the motor shaft diameter is matched with the allowable size of the reducer input end, forming a discrete combination with interrelated parameters.

[0014] Further, in step S107, a cyclic iterative verification and component replacement strategy oriented towards a finite candidate set is used to verify the combination of the reducer, helical gear, and motor under a unified hard constraint. If the current combination does not meet any constraint, it is eliminated, and the trial matching continues according to the nesting order of the motor, gear, and reducer. The process terminates after the finite candidate traversal is completed. Specifically, this includes: When the servo motor does not meet the usage requirements, the next model of motor will be automatically matched. If the current gear or motor candidate does not meet the requirements, select the matching gear; When there is no feasible combination of reducers, replace the reducer and recalculate the speed, torque and equivalent inertia according to the new discrete reduction ratio.

[0015] Furthermore, in step S109, based on the low-performance redundancy optimization method that prioritizes performance utilization constraints and then evaluates capacity utilization balance, when performance utilization evaluations are equal, the scheme with smaller rated torque and lower inertia ratio is prioritized. The speed and torque performance utilization scores of the reducer and motor are comprehensively considered, and secondary sorting is performed in combination with the rated torque and inertia ratio of the motor, while retaining the user's final selection. When calculating the performance utilization scores of the reducer and motor, the performance utilization rates of the reducer and motor are geometrically averaged according to their weights. The reducer speed includes the maximum input speed and rated input speed of the reducer, the reducer torque includes the maximum torque and rated torque of the reducer, the motor speed includes the acceleration speed and constant speed of the motor, and the motor torque includes the maximum torque and rated torque of the motor.

[0016] Furthermore, the intelligent assembly of the cable chain open track model is based on an automatic assembly method that utilizes the collaborative interaction between a general programming design decision-making environment and the built-in execution environment of the modeling software. The general automatic program environment is responsible for user information interaction, automatic parameter verification, knowledge base query, selection decision-making, and assembly configuration generation. Through structured configuration and task scripts, it calls the running engine of the modeling software. The built-in environment of the modeling software rereads the configuration, calls interfaces to modify the parameterized model, creates assemblies, inserts components according to absolute coordinates, saves the model, and generates a success marker. This includes the following steps: S201: Obtain the ground rail configuration parameters from the preprocessed production line ground rail design parameters, and load the standard ground rail module; S203: Decompose the total length configuration of the ground rail line, calculate the number and configuration of each ground rail segment, and determine the length of the tail segment: Determine the complete standard segment where the specified coordinates are located based on the standard ground rail segment length, and then calculate the target coordinates within the segment; construct a discrete set with the actual opening coordinates within each standard opening module, select the corresponding module according to the minimum absolute distance, and realize the function of accurately specifying the position of the drag chain opening; at the same time, the ground rail segment coordinates and deviation of the drag chain opening are fed back. S205: Jointly calculate the configuration of the head section and middle stage of the ground rail; S207: Perform modeling and assembly in the modeling software, wherein the modeling software is CAD software; S209: Collect the orbital dependency model and output the model scheme; S211: Output the functional component scheme for the production line floor rail and the assembly model of the production line floor rail.

[0017] Further, in step S205, based on the actual coordinates of the cable chain opening, coordinate position deviation, total number of track segments, and configuration, a joint calculation method is used to form a track line of target length by combining standard segments and parameterized tail segments, specifically including: Calculate the number of complete standard segments and the length of the tail segment based on the target total length and the standard segment length; Determine whether to set the head segment, the number of pre-mouth segments, and the number of post-mouth segments based on the opening segment number; If the tail segment has a standard integer length, the standard tail segment model is called; otherwise, the length expression of the parameterized tail segment template is modified so that the sum of the lengths of each segment equals the target length.

[0018] Further, in step S207, the cable chain length, number of links, and three-dimensional positioning coordinates are calculated based on the actual opening coordinates of the cable chain in the ground rail assembly, and the cable chain is automatically placed. The effective straight length is the distance from the actual opening to the far end of the ground rail. The required design length is obtained by superimposing the fixed end allowance, moving end allowance, and bending length allowance when placing the cable chain. The total cable chain length is calculated by rounding up according to the cable chain pitch. The three-dimensional installation coordinates of the cable chain are obtained by superimposing the actual opening coordinates with the installation offset corresponding to the ground rail specifications.

[0019] In a preferred embodiment of the present invention, compared with the prior art, the present invention has the following beneficial technical effects: 1. This invention utilizes a structured input and transmission demand continuous calculation method oriented towards ground rail working conditions. It unifies motion requirements, transmission parameters, and allowable utilization coefficients as inputs, sequentially forming driving force, gear size, output end requirements, motor side requirements, and converted inertia. This transforms scattered conversions into a unified calculation process, reducing deviations caused by omissions of units, formulas, and working conditions. Simultaneously, it retains inputs, intermediate quantities, and verification basis, facilitating review and reuse in different ground rail projects.

[0020] 2. The present invention is based on a structured component database and a hierarchical collaborative selection method of "reducer-helical gear-servo motor". The reducer is the outer layer, the helical gear is the middle layer, and the motor is the inner layer. Performance relationships and two-level connection dimension relationships are established in the same combination. This avoids using scattered components that meet individual indicators as assembly solutions, and allows the two main connection interfaces to be considered in conjunction with speed, torque and inertia requirements. Component parameters and combination rules can also be maintained as standard products are updated.

[0021] The following will further explain the concept, specific structure, and technical effects of the present invention in conjunction with the accompanying drawings, so as to fully understand the purpose, features, and effects of the present invention. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating the implementation of an AI agent according to a preferred embodiment of the present invention. Figure 2 This is a schematic diagram of the layout of a ground rail drive system according to a preferred embodiment of the present invention; Figure 3 This is a functional component selection process according to a preferred embodiment of the present invention; Figure 4 This is a schematic diagram of the input parameters and selection scheme results of a preferred embodiment of the present invention; Figure 5 This is a schematic diagram of the transmission system parameter calculation results according to a preferred embodiment of the present invention; Figure 6 This is a schematic diagram of a preferred embodiment of the ground track model of the present invention; Figure 7 This is a schematic diagram of the final intelligent assembly track model of a preferred embodiment of the present invention. Detailed Implementation

[0023] The following description, with reference to the accompanying drawings, illustrates several preferred embodiments of the present invention to make its technical content clearer and easier to understand. The present invention can be embodied in many different forms, and the scope of protection of the present invention is not limited to the embodiments mentioned herein.

[0024] In the accompanying drawings, components with the same structure are indicated by the same numerical designation, and components with similar structures or functions are indicated by similar numerical designations. The dimensions and thicknesses of each component shown in the drawings are arbitrary, and the present invention does not limit the dimensions and thicknesses of each component. To make the illustrations clearer, the thickness of some components has been appropriately exaggerated in the drawings.

[0025] like Figure 1 As shown, this invention addresses the problems existing in the automatic selection of functional components and the automatic and efficient assembly design of 3D models for existing production line floor rail equipment. It provides an AI agent for the automatic selection and assembly design of functional components for production line floor rails. The AI ​​agent uses a unified structured project configuration as its data foundation and completes the selection of functional component schemes and the intelligent establishment of the floor rail assembly within a single design process. This includes intelligent selection of transmission system components and intelligent assembly of the cable chain open floor rail model. The intelligent selection of transmission system components and the intelligent assembly of the cable chain open floor rail model share input and result identifiers within a unified project, collaboratively designing and outputting a complete production line floor rail design scheme. 1) Intelligent selection of transmission system components: Complete the selection of functional components of the production line floor rail and generate a solution. Taking the transmission chain composed of servo motor, reducer and drive gear as the design object, complete motion requirement calculation, candidate combination generation, multi-constraint loop verification, performance redundancy evaluation and user interaction to filter and output the solution. 2) Intelligent assembly of cable chain opening ground rail model: The intelligent assembly of the production line ground rail is completed. Taking the standard ground rail section, ground rail section with cable chain opening, ground rail tail section and cable chain as design objects, the software completes the specified cable chain opening coordinate mapping, ground rail section material calculation, tail section preparation, cable chain length, positioning calculation of each section and automatic assembly of the modeling software.

[0026] In this embodiment, the AI ​​agent includes a user interface, structured data configuration, a physical calculation module, a component knowledge base definition, a rule-constrained reasoning module, a scheme evaluation module, a standard ground rail module knowledge base, an assembly planning module, a modeling software task scheduling module, an execution module, and a result output module. Each module transmits data through structured configuration, candidate scheme objects, assembly planning objects, logs, and output files. 1) The user interface collects operating conditions and displays candidate solutions; 2) The physical calculation module generates the calculation requirements of the transmission system based on the user's input information; 3) The rule-constrained reasoning module traverses and eliminates candidate combinations of functional component schemes; 4) The assembly planning module converts the continuous opening coordinates of the cable chain into existing standard modules and the quantity of each segment; 5) The modeling software task scheduling module starts the software's internal execution environment; 6) The execution module calls the relevant software functions to complete parametric modeling and component placement.

[0027] In this embodiment, the intelligent selection of transmission system components includes the following steps when completing the selection of production line floor rail functional components and generating a solution: S101: Obtain the design parameters of the production line floor rails, convert the design parameters into structured configurations, and verify the design parameters; S103: Calculate the kinematics and speed-related parameters of the transmission system; S105: Generate candidate solutions for functional components based on component parameters; S107: Perform multi-constraint verification on candidate solutions for functional components; S109: Evaluate and screen the candidate solutions for the verified functional components and output all feasible solutions; S111: Confirm the target solution and output traceable selection results.

[0028] In this embodiment, in step S103, based on the structured input and transmission demand continuous calculation method for production line ground rail conditions, motion demand, transmission parameters and allowable utilization coefficient are unified as input, and driving force, gear size, output end demand, motor side demand and equivalent inertia are formed in sequence. By converting speed into linear velocity and calculating acceleration, acceleration force, friction force and total driving force are calculated. The pitch circle diameter and stroke per revolution are determined by gear module and number of teeth. Then, the requirements of reducer and motor are calculated step by step in combination with candidate reduction ratio and efficiency, so that intermediate quantities are transmitted through the same input chain.

[0029] In this embodiment, in step S105, the hierarchical collaborative selection method of "reducer-helical gear-servo motor" based on the structured component database is used. The reducer is the outer layer, the helical gear is the middle layer, and the motor is the inner layer. Performance relationships and two-level connection dimension relationships are established in the same combination. By reading the reduction ratio, torque, speed, inertia and end dimensions of the candidate reducer from the functional component data knowledge, the output shaft is first matched with the gear bore diameter and the output side requirements are calculated. Then, the torque and speed requirements of the servo motor are calculated according to the reduction ratio, and the motor shaft diameter is matched with the allowable size of the reducer input end to form a discrete combination with interrelated parameters.

[0030] In step S107, this embodiment utilizes a cyclic iterative verification and component replacement strategy for a finite candidate set. The combination of the reducer, helical gear, and motor is verified under a unified hard constraint. If the current combination does not meet any constraint, it is eliminated. The process continues with trial fitting according to the nesting order of the motor, gear, and reducer. The process terminates after the finite candidate set has been traversed. Specifically, this includes: 1) When the servo motor does not meet the usage requirements, the next model of motor will be automatically matched; 2) Select a matching gear if the current gear or motor candidate does not meet the requirements; 3) When there is no feasible combination of reducers, replace the reducer and recalculate the speed, torque and equivalent inertia according to the new discrete reduction ratio.

[0031] In step S109, this embodiment uses a low-performance redundancy optimization method based on performance utilization constraints screening first and capacity utilization balance evaluation later. When the performance utilization rate evaluations are equal, the scheme with smaller rated torque and lower inertia ratio is selected first. The speed and torque performance utilization rate scores of the reducer and motor are combined, and secondary sorting is performed by combining the rated torque and inertia ratio of the motor, while retaining the user's final selection. When calculating the performance utilization rate scores of the reducer and motor, the performance utilization rates of the reducer and motor are geometrically averaged according to their weights. The reducer speed includes the maximum input speed and rated input speed of the reducer, the reducer torque includes the maximum torque and rated torque of the reducer, the motor speed includes the acceleration speed and constant speed of the motor, and the motor torque includes the maximum torque and rated torque of the motor.

[0032] In this embodiment, the intelligent assembly of the cable chain open ground rail model is based on an automatic assembly method that uses the collaborative interaction between the ordinary programming design decision environment and the built-in execution environment of the modeling software. The ordinary automatic program environment is responsible for user information interaction, automatic parameter verification, knowledge base query, selection decision and assembly configuration generation. The modeling software's running engine is called through structured configuration and task scripts. The built-in environment of the modeling software rereads the configuration, calls the interface to modify the parameterized model, creates the assembly, inserts components according to absolute coordinates, saves the model and generates a success mark.

[0033] In this embodiment, the intelligent assembly of the cable chain open ground rail model includes the following steps: S201: Obtain the floor rail configuration parameters from the pre-processed production line floor rail design parameters and load the standard floor rail module; S203: Decompose the total length configuration of the ground rail line, calculate the number and configuration of each ground rail segment, and determine the length of the tail segment: Determine the complete standard segment where the specified coordinates are located based on the standard ground rail segment length, and then calculate the target coordinates within the segment; construct a discrete set with the actual opening coordinates within each standard opening module, select the corresponding module according to the minimum absolute distance, and realize the function of accurately specifying the position of the drag chain opening; at the same time, the ground rail segment coordinates and deviation of the drag chain opening are fed back. S205: Jointly calculate the configuration of the head section and middle stage of the ground rail; S207: Perform modeling and assembly in modeling software, which uses CAD software; S209: Collect the orbital dependency model and output the model scheme; S211: Output the functional component scheme for the production line floor rail and the assembly model of the production line floor rail.

[0034] In step S205, this embodiment uses a joint calculation method based on the actual coordinates of the cable chain opening, coordinate position deviation, total number of track segments, and configuration to form a track line of target length by combining standard segments and parameterized tail segments. Specifically, this includes: 1) Calculate the number of complete standard segments and the length of the tail segment based on the target total length and the standard segment length; 2) Determine whether to set the head segment, the number of pre-mouth segments, and the number of post-mouth segments based on the opening segment number; 3) If the tail segment is a standard integer length, call the standard tail segment model; otherwise, modify the length expression of the parameterized tail segment template so that the sum of the lengths of each segment equals the target length.

[0035] In step S207, the cable chain length, number of links, and three-dimensional positioning coordinates are calculated based on the actual opening coordinates of the cable chain in the ground rail assembly, and the cable chain is automatically placed. The effective straight length is the distance from the actual opening to the far end of the ground rail. The required design length is obtained by superimposing the fixed end allowance, moving end allowance, and bending length allowance when placing the cable chain. The total cable chain length is calculated by rounding up according to the cable chain pitch. The three-dimensional installation coordinates of the cable chain are obtained by superimposing the actual opening coordinates with the installation offset corresponding to the ground rail specifications.

[0036] Compared with existing technologies, the AI ​​intelligent agent for automatic selection and assembly design of production line floor rail functional components provided by this invention has the following beneficial effects: 1. Addressing the challenges of selecting ground rail drive systems involving multiple inputs such as load, speed, acceleration time, friction, transmission efficiency, and gear parameters, which require manual conversion and transcription across various tables and product samples, and inconsistent handling of units, operating conditions, and design margins by different personnel, as well as the difficulty in verifying the process from initial input to required parameters, this invention proposes a structured input and continuous calculation method for transmission requirements oriented towards ground rail operating conditions. This method unifies motion requirements, transmission parameters, and allowable utilization coefficients as inputs, sequentially forming driving force, gear dimensions, output requirements, motor-side requirements, and converted inertia. By converting speed to linear velocity and calculating acceleration, the acceleration force, friction force, and total driving force are calculated. The pitch circle diameter and stroke per revolution are determined from the gear module and number of teeth. Then, combining candidate reduction ratios and efficiencies, the requirements for the reducer and motor are calculated step-by-step, ensuring intermediate quantities are transmitted through the same input chain. This transforms fragmented conversions into a unified calculation process, reducing deviations caused by omissions in units, formulas, and operating conditions. Simultaneously, inputs, intermediate quantities, and verification bases are retained for easy verification and reuse in different ground rail projects.

[0037] 2. Given the coupling between the performance and installation relationships of functional components in the transmission system, individual component selection may fail to form a usable transmission chain; gear dimensions affect output load, discrete reduction ratios alter motor speed, torque, and inertia, and the two connection points are constrained by shaft diameter and bore diameter respectively, segmented selection can easily lead to local satisfaction but overall mismatch. This invention proposes a hierarchical collaborative selection method based on a structured component database, using a "reducer-helical gear-servo motor" layered collaborative selection approach. This method establishes performance relationships and two-level connections within the same combination, with the reducer as the outer layer, the helical gear as the middle layer, and the motor as the inner layer. The dimensional relationships are determined by reading the reduction ratio, torque, speed, inertia, and end dimensions of the candidate reducer from the functional component data knowledge. First, the output shaft is matched with the gear bore diameter, and the output side requirements are calculated. Then, the torque and speed requirements of the servo motor are calculated according to the reduction, and the motor shaft diameter is matched with the allowable size of the reducer input end, forming a discrete combination with interrelated parameters. This effectively avoids using scattered components that meet single indicators as an assemblable solution, and ensures that the two main connection interfaces and speed, torque, and inertia requirements are considered in a coordinated manner. The component parameters and combination rules can also be maintained with the updates of standard products.

[0038] 3. When the selected combination of functional components does not meet the requirements of rated torque, peak torque, speed, inertia ratio, or installation dimensions, traditional methods require returning to the product sample for reselection and recalculation. With a large number of candidates, manual matching is prone to repetitive trials and omissions of feasible combinations, and it is also difficult to determine when the verification of all functional components has been completed. This invention proposes a cyclical iterative verification and component replacement strategy for a finite candidate set. Each combination of reducer, helical gear, and motor is judged under a unified hard constraint. If the current combination does not meet any constraint, it is eliminated, and the matching continues according to the nesting order of motor, gear, and reducer. When the servo motor does not meet the usage requirements, the next model of motor is automatically matched. If the current gear or motor candidate does not meet the requirements, a matching gear is selected. If no feasible combination is found for a certain reducer, the reducer is replaced, and the speed, torque, and converted inertia are recalculated according to the new discrete reduction ratio. The process terminates once the finite number of candidates has been traversed, thus solidifying manual table lookup, back substitution, and type replacement into a repeatable closed loop. The system covers the candidate set and automatically excludes combinations that do not meet the conditions, reducing missed detections and redundant work. Clear candidate boundaries and termination conditions make the iteration process controllable and the verification results traceable.

[0039] 4. When multiple combinations need to meet usage constraints, manual selection often relies on familiarity or individual margins, easily leading to the selection of components with high performance redundancy. If only a certain indicator is pursued close to its upper limit, the matching balance of other operating conditions may be overlooked. This invention proposes a low-performance redundancy optimization method that prioritizes performance and usage constraints, followed by capacity utilization balance evaluation. It comprehensively scores the speed and torque performance utilization rates of the reducer and motor, and performs secondary ranking based on the motor's rated torque and inertia ratio, while retaining the user's final selection. By geometrically averaging the utilization rates of the reducer's maximum and rated torque, maximum and rated input speed, and the motor's maximum and rated torque, acceleration and constant speed, etc., according to weights; when performance utilization rate assessments are equal, priority is given to smaller rated torque and lower inertia ratio, achieving fuller utilization of component capacity and lower performance redundancy without compromising usage constraint safety. Multi-indicator evaluation reflects the overall matching degree of the transmission chain, and the ranking criteria are transparent, providing designers with a quantitative reference for comparing feasible solutions.

[0040] 5. Actual design requirements for cable carrier installation positions are typically defined by continuous coordinates. However, enterprise standard cable carrier model libraries usually only contain a limited number of opening structures. Manual selection makes it difficult to ensure consistency between the cable carrier segment model, actual opening coordinates, and calculated positional deviations. This invention establishes several standardized cable carrier opening cable carrier segment models and proposes a mapping method from the continuous coordinates of the cable carrier opening specified in actual requirements to several standard opening modules. First, the complete standard segment containing the specified coordinates is determined based on the length of the standard cable carrier segment. Then, the target coordinates within the segment are calculated. A discrete set is constructed using the actual opening coordinates within each standard opening module segment. The corresponding module is selected based on the minimum absolute distance, achieving accurate specification of the cable carrier opening position. Simultaneously, the coordinates and deviations of the cable carrier opening cable carrier segment are fed back, thus converting continuous design requirements into discrete structures that can be actually assembled in the enterprise standard module library, avoiding incorrect manual selection of the cable carrier segment model. Furthermore, the actual coordinates and deviations are fed back to the user, ensuring consistency between subsequent cable carrier positioning and on-site design based on the actual model.

[0041] 6. After the location of the cable chain opening in the ground rail production line is determined, the combination of the head section, middle section, and tail section of the ground rail line must simultaneously meet the target total length. If each section of the ground rail is calculated independently, problems such as missing sections, overlaps, unsatisfactory cable chain opening locations, or the inability to form a closed loop in the total length are likely to occur. This invention proposes a joint calculation method for the actual coordinates of the cable chain opening, coordinate position deviations, the total number of ground rail sections, and their configuration. It uses a combination of standard sections and parameterized tail sections to form a ground rail line of the target length. The number of complete standard sections and the length of the tail section are calculated based on the target total length and the length of the standard sections. The number of head sections, the number of sections before the opening, and the number of sections after the opening are determined based on the opening section sequence number. When the tail section is a standard integer length, the standard tail section model is called; otherwise, the length expression of the parameterized tail section template is modified so that the sum of the lengths of each section equals the target length. This allows the selection of the opening module, the calculation of ground rail sections, and the closure of the length to be completed in the same calculation result, reducing errors in manual calculation. While reusing the standard model, it also supports non-standard target lengths, improving the adaptability of the ground rail solution.

[0042] 7. Traditional ground rail design requires manual placement of cable chain installation positions and manual calculation of the number of cable chain links and total cable chain length. The assembly position may misalign with the actual opening after discrete mapping. If the purchased cable chain length is not rounded up according to the link pitch, insufficient length or inability to purchase standard links may occur. This invention proposes a method for automatically placing the cable chain in the overall ground rail assembly by calculating the cable chain length, number of links, and three-dimensional positioning coordinates based on the actual opening coordinates. The effective straight-line length is the distance from the actual opening to the far end of the ground rail. The required design length is obtained by adding the fixed end allowance, moving end allowance, and bending length allowance during cable chain placement. The total cable chain length is calculated by rounding up according to the cable chain pitch. The three-dimensional installation coordinates of the cable chain are obtained by adding the installation offset corresponding to the ground rail specifications to the actual opening coordinates, thus ensuring that the purchased cable chain length is the complete number of links and covers the farthest movement position. The cable chain model is aligned with the actual standard opening, reducing problems such as insufficient actual cable chain length, installation misalignment, and manual secondary adjustments.

[0043] 8. While conventional automated operating procedures can perform track segment calculations and rule reasoning, they cannot directly execute assembly operations from CAD software. If manual intervention is required to transcribe the track segment assembly materials and calculation results into the modeling software for component insertion, positioning, and saving, the process is repetitive and prone to model selection or coordinate errors. This invention proposes an automated assembly method that integrates a conventional programming design decision-making environment with the built-in execution environment of the modeling software. The conventional automated program environment handles user information interaction, automatic parameter verification, knowledge base queries, selection decisions, and assembly configuration generation, calling the CAD modeling software's execution engine through structured configuration and task scripts. The CAD software's built-in environment re-reads the configuration, calls relevant interfaces to modify the parametric model, create assemblies, insert components using absolute coordinates, save the model, and generate a success marker. This decouples design decisions from CAD software execution and automatically connects them, reducing repetitive modeling operations. The assembly strategy and the underlying operation interface of the CAD software can be extended separately, improving system maintainability and generating traceable assembly logs and model outputs.

[0044] The present invention will now be described in detail with reference to preferred embodiments.

[0045] This embodiment provides an AI agent for the automatic selection and efficient assembly design of functional components for production line floor rails. This agent uses a unified structured project configuration as its data foundation and sets up two collaborative design tasks: intelligent selection of transmission system components and intelligent assembly of floor rail models with cable chain openings. The functional component selection task takes the transmission chain composed of servo motors, reducers, and drive gears as the design object, completing motion requirement calculations, candidate combination generation, multi-constraint loop verification, performance redundancy evaluation, and user-interactive filtering of output solutions. The floor rail intelligent assembly task takes standard floor rail sections, floor rail sections with cable chain openings, floor rail tail sections, and cable chains as design objects, completing the designer's specified cable chain opening coordinate mapping, floor rail section material calculation, tail section preparation, cable chain length, segment positioning calculations, and automatic assembly using modeling software. The collaborative design tasks share basic information such as load, movement speed, acceleration time, and total floor rail length, and each reads its own relevant selection and assembly parameters.

[0046] The AI ​​agent in this embodiment includes a user interface, structured data configuration, a physical calculation module, a component knowledge base definition, rule-based constraint reasoning, scheme evaluation, a standard ground rail module knowledge base, an assembly planning module, a CAD modeling software task scheduling module, an execution module, and a result output module. The user interface collects operating conditions and displays candidate schemes; the physical calculation module generates the transmission system calculation requirements based on user input; the constraint reasoning module iterates through and eliminates candidate combinations of functional component schemes; the assembly planning module converts the continuous opening coordinates of the cable chain into existing standard modules and the number of each segment; the modeling software task scheduling module starts the software's internal execution environment; and the CAD execution module calls relevant software functions to complete parametric modeling and component placement. Each module transmits data through structured configuration, candidate scheme objects, assembly planning objects, logs, and output files.

[0047] In this invention, the two design tasks—functional component selection and intelligent assembly—share input and result identifiers within a unified project, but maintain appropriate decoupling in technical calculations. The selection task outputs candidate transmission components that meet the operating conditions; the assembly task generates a ground rail line model based on the total length of the ground rail and assembly requirements within the same project. This approach allows a single design process to simultaneously complete the selection of ground rail functional component schemes and the intelligent establishment of the ground rail line assembly, while enabling the two types of results to together constitute a complete ground rail design scheme.

[0048] The overall flowchart of the embodiments of the present invention is as follows: Figure 1 As shown, the ground rail equipment transmission system described in the functional component selection is as follows: Figure 2 As shown. Simultaneously, the application verification of this intelligent agent system is conducted based on the actual design requirements of the designers, and calculation results and selection schemes are obtained based on the design inputs. The functional component selection process established by this method is as follows: Figure 3 As shown. The selection and generation of functional components and the intelligent assembly of the ground rail are two relatively independent functions in the intelligent agent, but they can be performed sequentially. For the convenience of the description of the subsequent embodiments, the selection function of the functional components of the ground rail drive system is described in steps 1-5; and the related functions of intelligent assembly of the ground rail assembly are described in steps 6-13.

[0049] Step 1: Obtain the design parameters of the ground rail production line.

[0050] Complete parameters should include: mass of the moved load. , fast movement speed Acceleration time Load movement stroke Total length of the target track Available length of site space coefficient of friction Transmission efficiency Functional component performance utilization rate Peak capability coefficient Maximum permissible inertia ratio of motor Rail specifications, installation and application configuration, whether to use integer standard tail sections, and specified drag chain opening coordinates. , Number of chain links remaining at the fixed end of the cable chain Mobile terminal remaining chain number And the arrangement of cable chains.

[0051] Loaded movement stroke Used to describe the motion requirements of the load-bearing components and calculate the total number of motor rotations for one round trip; target total length of the ground track. Used for assembling and calculating materials for each section and the end section of the ground rail. The remaining parameters are intermediate parameters for calculating the ground rail drive system and assembling the ground rail.

[0052] In this embodiment, both the load stroke and the target total length of the ground track are set to... Input load quality ( ), fast movement speed ( ), acceleration time ( ),journey( ), site length limit ( ), coefficient of friction ( ), gear and rack transmission efficiency ( ), reducer transmission efficiency ( ), functional component performance allowable utilization factor ( ) and maximum permissible inertia ratio ( The interface input is converted into a structured configuration and then passed to the unified selection process; where rapid traverse speed is expressed in meters per minute, acceleration time in seconds, mass in kilograms, and the gear module read subsequently in millimeters. The gear width and the shaft hole dimensions involved in the calculation are expressed in meters or millimeters according to the formula and installation verification requirements, respectively. The agent verifies the following for the input: mass, speed, acceleration time, and target length are greater than zero; the rated capacity utilization coefficient is within the preset range; the maximum inertia ratio is not less than zero; the specified opening coordinates are not less than zero and do not exceed the target total length; when selecting a standardized integer length tail segment, the target length and the calculated tail segment should be able to match the standard tail segment model library.

[0053] Step 2: Preprocess the design parameters of the ground rail production line

[0054] First, convert the kinetic speed from meters per minute to meters per second, and then calculate the acceleration:

[0055] Let the acceleration due to gravity be... Frictional resistance, acceleration force, total driving force under acceleration conditions, and uniform resistance are respectively:

[0056] Read the module of candidate gears from the drive gear database. Number of teeth Tooth width Mounting bore diameter and order code. Gear pitch circle diameter, linear stroke per revolution, and speed are:

[0057] The load torque, gear inertia, and angular acceleration under acceleration conditions are as follows:

[0058] Therefore, the required total acceleration torque and constant speed rated torque at the output of the reducer are as follows:

[0059] In this embodiment, based on the input information from step 1, the linear velocity is calculated as follows: acceleration is Frictional resistance and total driving force The candidate gear model in this example is F03L32B40 gear, with a module of [missing information]. Number of teeth Tooth width is The pitch circle diameter is obtained. Each transfer is approximately Rotation speed approximately The moment of inertia of the gear is approximately The total torque required at the output end is approximately and the required rated torque at constant speed .

[0060] Step 3: Generate candidate solutions for speed reducers, gears, and servo motors.

[0061] Retrieves performance data of speed reducers, gears, and servo motors from the component database. Performance parameters for the speed reducer include model and reduction ratio. Rated output torque, maximum output torque, rated input speed, maximum input speed, and the reducer's own inertia. Output shaft dimensions and permissible input dimensions; gear data including module, number of teeth, tooth width, bore diameter, and order code; motor performance data including at least rated torque, maximum torque, rated operating speed, maximum speed, and rotor inertia. Shaft diameter, supplier, and order number.

[0062] The reducer is considered the outer candidate, the gear the middle candidate, and the motor the inner candidate. First, determine if the gear bore diameter matches the reducer's output shaft, then calculate the motor's requirements based on the reduction ratio.

[0063] Then, it is determined whether the motor shaft diameter is not greater than the allowable size of the reducer input end. This traversal is pruned in advance according to the calculation dependency: if the shaft hole does not match, the performance is no longer calculated; if the reducer capacity is insufficient, the motor under it is no longer traversed; if the motor shaft diameter does not meet the requirements, it is not included in the inertia and scoring calculation, thereby reducing invalid combinations.

[0064] In this embodiment, a reduction ratio of [value] is selected. The AB142-8 reducer was tested with the F03L32B40 gear. The rated output torque of the reducer is... Rated input speed is The maximum input speed is Its own inertia is The calculated required motor speed is approximately... Acceleration load torque approximately and the rated torque at constant speed The rated torque of the αiF 30-4000-B motor is... Maximum torque is Rotor inertia is Its total peak acceleration torque is approximately .

[0065] Step 4: Perform multi-performance constraint verification on candidate combination schemes of reducer-gear-servo motor components, and perform candidate scheme screening and component replacement when the constraints are not met.

[0066] The verification items for functional components include: the rated output torque of the reducer should cover the rated requirements at constant speed, the allowable peak capacity should cover the output torque under acceleration conditions, and the rated and maximum input speeds should cover the corresponding required speeds; the rated torque, maximum torque, rated speed, and maximum speed of the motor should cover the requirements of the corresponding operating conditions; the inertia matching ratio should not exceed the allowable upper limit; and the allowable dimensions of the reducer output shaft and helical gear bore diameter, and the motor shaft diameter and reducer input end should meet the connection requirements.

[0067] The equivalent inertia of a linear load is: , The converted load and gear inertia are: , Superimposed speed reducer inertia ( After that, we get: , The inertia matching ratio is: .

[0068] When traversing the candidate functional components, the reducer is selected first, then the gear, and finally the motor to form the overall selection scheme. If the current combination scheme does not meet any constraint, the next candidate is tested. After replacing the reducer, the motor side requirements are recalculated according to the new discrete reduction ratio until the database of all functional components is traversed.

[0069] This embodiment verifies the candidate combination consisting of an AB142-8 reducer, an F03L32B40 helical gear, and an αiF30-4000-B servo motor, based on the selection results. The reducer output shaft diameter is ( ), and the diameter of the helical gear bore ( Matching; the input end of the reducer allows the motor shaft diameter to be no greater than ( The candidate motor shaft diameter is ( The servo motor-reducer connection dimensions are met. The rated torque, peak torque, and speed capability of both the reducer and motor cover the requirements obtained in steps 2 and 3. The calculated equivalent inertia of the linear load is approximately ( The converted load and gear inertia are approximately ( ), after adding the inertia of the reducer, it becomes ( The ratio of the moment of inertia of the motor rotor to the moment of inertia of the motor rotor is approximately (9.563), which does not exceed the allowable upper limit (10). Therefore, this combination is included in the set of feasible solutions. Other combinations that do not meet any of the constraints are eliminated, and the model replacement and matching are continued in the nesting order described above.

[0070] Step 5: Evaluate the redundancy of component performance for functional component combinations that have passed all performance and component coordination constraints, screen them, and output all feasible solutions.

[0071] The evaluation indicators include the performance utilization rate of the reducer's maximum torque, rated torque, maximum input speed, and rated input speed, as well as the utilization rate of the motor's maximum torque, rated torque, acceleration speed, and constant speed speed, totaling eight evaluation indicators. The utilization rates are weighted geometrically by preset weights to obtain a balanced score reflecting the overall capacity utilization of the transmission chain. The intelligent agent ranks components according to their performance utilization scores from high to low and selects the component combination with the higher performance utilization rate. The user provides multiple feasible combinations along with their reducer, helical gear, and motor parameters, while simultaneously recording the original design inputs, major intermediate calculations, rated and peak capacity verification, speed verification, inertia matching ratio, installation dimension relationships, and capacity utilization rate. The user can determine the final solution from the feasible options by considering spatial layout, product supply, and enterprise standards. If no feasible combination is formed after traversing the finite candidate set, a result indicating no feasible combination is output, allowing designers to adjust motion requirements, constraint boundaries, or candidate component data for re-selection.

[0072] The eight weights used in this embodiment are as follows: ( The utilization rates of the reducer's maximum torque, rated torque, maximum input speed, and rated input speed, calculated based on the kinematic performance of the above feasible combinations, are respectively (). ), ( ), ( )and( The utilization rates of the motor's maximum torque, rated torque, acceleration speed, and constant speed speed are approximately ( ). ), ( ), ( )and( The final combination scheme selected in this embodiment is as follows: AB142-8 independent reducer, F03L32B40 helical gear and αiF 30-4000-B servo motor. The selection scheme is as follows: Figure 4 As shown. This solution can be selected by the user from multiple feasible solutions and is not limited to a single optimal solution. The output result stores the corresponding ( )load,( ) speed of movement and ( The calculation results in this embodiment are as follows: (The original inputs include acceleration time, total driving force, output torque, motor-side torque, required speed, converted inertia, inertia ratio, two-stage shaft hole dimensions, and eight capacity utilization rates.) Figure 5 As shown. This proves that this embodiment can form a reducer-helical gear-servo motor combination that meets the torque, speed, inertia, and main connection size conditions under given ground rail operating conditions, and retains the complete basis from input, calculation, candidate replacement to final selection result for reference by subsequent designers and engineers. The selection of functional components of the ground rail transmission system in this embodiment has completed the implementation operation. The following is the implementation flow of the intelligent establishment process of the ground rail assembly.

[0073] Step 6: Read the ground rail assembly parameters and load standard module knowledge.

[0074] This embodiment takes the standardized assembly process of ground rail as an example. The intelligent agent initiates the assembly task of ground rail with a single drag chain opening, determines the ground rail specifications according to the ground rail type and load level, and loads the standard section length. The models include: head rail model, middle rail model, middle rail model with standard cable chain opening positions, integer length tail rail model, parameterized tail rail model, cable chain model with specific model number, and cable chain pitch. Length of the curved section in different cable chain arrangements And the installation offset of the cable chain relative to the actual opening. A model of the head section, the middle section with the cable chain opening, and the tail section of a certain type of ground rail are shown below. Figure 6 As shown, there are 6 types of models with cable chain openings in the middle section. On a section of ground rail, the coordinates of the continuously specified cable chain openings need to be selected to install the nearest cable chain opening position.

[0075] This embodiment adopts The length of the ground rail line includes a cable chain opening, and the standard section of the ground rail is [length missing]. The specified cable chain opening coordinates are:

[0076] The cable chain has a allowance of three links each at both the fixed and moving ends, and the cable chain pitch is... The arrangement is a sliding type, and the corresponding bending section length is... The actual opening coordinates within a standard opening module are mapped to a set over a 5m length as follows:

[0077] Each coordinate in the set is associated with an actual standard opening model. During subsequent assembly, these coordinates are linked to the set. The position mapping in the code enables the selection of the open track section for the drag chain.

[0078] Step 7: Decompose the total length configuration of the ground rail line, calculate the number and configuration of each section of ground rail, and determine the length of the tail section.

[0079] Let the total length of the target track be... The standard section length is Then the length of the track tail section and the number of complete standard sections are:

[0080] The starting coordinates of the tail section model of the ground track are:

[0081] In this embodiment, the following case data is used:

[0082] That is, the entire ground track has three complete sections before the end. Standard section, end section is Its starting point is This embodiment uses a standard integer tail segment. If the tail segment is not a standard integer length, a parameterized tail segment template is used to convert the target length into millimeters and write it into the length expression. Then, CAD modeling is performed, updated, and saved.

[0083] The tail segment calculation needs to ensure that its length falls within the allowable range, and makes This establishes a closed loop for the length of the ground track. The calculations in this embodiment yield... The segment is still identified as a tail segment in the planning object, and is not confused with the standard segment after the ordinary port. The structural differences, end interfaces and file naming rules of the ground rail end are retained, and it is convenient to replace the same role with a parameterized tail segment under other lengths.

[0084] Step 8: Map the user-specified cable chain opening position to the middle ground rail section of the standard cable chain opening.

[0085] After obtaining the configuration of the ground rail line and the location of the cable chain opening of the ground rail base equipment, first determine Whether it falls into the final stage. When If the specified opening enters the tail segment that currently does not have an opening structure, the system stops planning and prompts for adjustment. Otherwise, it determines the standard segment it belongs to. Not located at the boundary of a non-zero standard segment:

[0086] like If it lies precisely at the boundary of a non-zero standard segment, it is assigned to the preceding complete standard segment to eliminate ambiguity caused by adjacent segments interpreting the same coordinate. Let the standard opening coordinates be... ,choose:

[0087] The actual opening and positional deviation of the entire line are:

[0088] In this embodiment , The closest one is the Type 4 standard opening. ,therefore:

[0089] This invention can save the specified cable car opening coordinates, actual coordinates, selected opening type, and signed deviation of the user-constructed ground track. If the two standard coordinates are equal to the target distance, a unique result can be determined according to the preset module priority. Subsequent cable car length and 3D positioning are all based on this. No longer in use The cable chain opening mapping result includes the start and end points of the standard segment where the opening is located, the specified coordinates within the segment, the actual coordinates within the segment, and the boundary assignment status. This allows for the distinction between "the coordinate positions required by the designers" and "the positions that the standard model of the ground track segment can achieve."

[0090] Positional deviation is indicated by a positive or negative sign: a positive value indicates that the actual opening is after the specified position, and a negative value indicates that the actual opening is before the specified position. This invention can request user confirmation or halt assembly and prompt for expansion of the standard opening module library if the deviation exceeds the company's allowable range.

[0091] For nonzero The specified coordinates of the boundary are assigned to the previous segment, and an opening structure near the end of that segment is selected to avoid interpreting the same boundary as both the end of the previous segment and the start of the next segment. For specified coordinates falling into the tail segment, the current embodiment does not use a tail segment without an opening to forcibly assemble the agent, but instead explicitly returns the reason why assembly is not possible, thereby ensuring that the planning results all correspond to a real and usable model.

[0092] Step 9: Jointly calculate the quantity of the head section and middle section of the ground rail, and calculate the procurement length and installation coordinates of the drag chain to achieve complete material calculation for the ground rail section and drag chain installation positioning.

[0093] First, check which section of the ground rail the cable chain opening is located on. It is located in the first complete standard section. When the segment is open, it is directly used as the starting segment; when A ground rail head section needs to be installed. The intermediate ground rail sections, depending on their structure and characteristics, can be divided into sections before the cable chain opening and sections after the cable chain opening. The number of these two sections are as follows:

[0094] No. The millimeter-based coordinates of a complete standard segment in CAD modeling software are:

[0095] In actual model assembly, the present invention synchronously installs the cable chain structure. The cable chain model is directly related to the ground rail specifications and is not considered in the combination selection of this embodiment. The effective straight-line length is the distance from the actual opening to the far end of the ground rail:

[0096] The allowance for fixed and mobile terminals and the original required length are as follows:

[0097] Meanwhile, the total number of chain links and the purchased length of the installed cable chains are as follows:

[0098] In this embodiment, the intermediate segment parameter is: , The number of intermediate sections without a separate ground rail head section is:

[0099] The ground rail section is configured as follows: one type 4 ground rail drag chain open section, with reference coordinates... The number of intermediate segments is 2, and their reference coordinates are respectively and ; The length of the tail section of the ground track, with reference coordinates as follows: The total length of the track is:

[0100] Based on the calculated track section, the corresponding cable chain configuration is configured. In this embodiment:

[0101] The cable chain length must be rounded up to ensure the purchased length is not less than the calculated requirement, and it must consist of complete links. The intelligent agent of this invention simultaneously outputs the original length of the cable chain, the total number of links, and the purchased length. The intelligent agent generates a component list according to the rail direction, with each item containing the logical functional role of the rail segment, model identifier, instance name, quantity, reference coordinates, and direction matrix, enabling the calculation results to be directly converted into actual CAD modeling and assembly instructions. This ensures that the cable chain model maintains a preset installation relationship relative to the actual standard opening, rather than the originally specified opening, meeting the actual installation and usage requirements of the cable chain.

[0102] Step 10: Perform the track section modeling and automatic assembly task in the CAD modeling software.

[0103] In a standard external programming environment, generate the assembly configuration based on the planning results, including at least the ground rail specifications, total length, actual cable chain opening position, position coordinate deviation, number of intermediate sections, tail section length and pattern, cable chain specifications, length configuration, and installation coordinates. Create a project output directory within the external programming environment, determining the overall assembly name and save path.

[0104] This embodiment uses NX2506 as the CAD modeling software and Python as the external programming language. Within the NX modeling software, Python obtains the NX session and loads the rail specifications. For the parametric tail section, the tail section template is opened before creating the final assembly, the length expression is found, the target tail section length is converted to millimeters and written into the expression, the model is updated, saved, and closed. For this embodiment... For integer tail segments, the corresponding 5m standard tail segment model is directly called. In the NX modeling software, a new assembly is created based on a millimeter template. An interface is inserted into the underlying components to establish a unit direction matrix, converting the planned coordinates into NX 3D base points, and the components are inserted through the NXOpen assembly interface. The business strategy is only responsible for model selection, quantity, and coordinates, while the underlying interface is only responsible for execution of insertion and saving, thus decoupling the assembly planning rules from NX operations. This invention separates the assembly strategy from the specification configuration. The assembly strategy describes how to calculate the number of segments, how to map openings, and the assembly order. The specification configuration describes the actual model, standard length, cable chain parameters, and offsets corresponding to each logical role. When adding similar ground rail specifications, the assembly strategy for the intermediate ground rail segment with a specified opening can be reused, only requiring supplementation of the corresponding model mapping and physical parameters.

[0105] The NX assembly sequence in the modeling software in this embodiment is as follows: (1) In The intermediate ground rail section with a drag chain opening is of type 4. (2) In Place the first intermediate standard track segment model; (3) In Place the second intermediate standard track section model; (4) In place Model of the tail section of the ground track; (5) Place a fixed model cable chain at the cable chain opening position specified by the designer.

[0106] During component placement, the Python environment within the NX modeling software in this invention generates business-meaning names for each instance to distinguish different track opening segments, intermediate track segments with different sequence numbers, track tail segments, and cable chain models. The component orientation matrix, reference points, and model paths are passed to the NXOpen component assembly method via a unified underlying interface. All standard track segments are aligned along the same assembly direction... The step size arrangement is used, while the cable chain uses independent coordinates obtained by adding specification offset to the actual opening. After the components are placed, a global save is performed to ensure that the overall assembly and the modified parametric model are properly placed on the disk. The final completed ground rail assembly model is as follows. Figure 7 As shown.

[0107] Step 11: Collect the models required for this design and output the design results.

[0108] This invention, after completing the selection and intelligent assembly of the ground rail functional components, forms a design scheme. The functional component selection results include all feasible combinations, user selection criteria, target scheme, and complete calculation basis; the intelligent assembly results of the ground rail model record the number of standard sections, tail sections, and cable chains actually used in this project. After removing duplicates, the corresponding model folders and dependent parts are copied to form a project model package. It ensures that the overall assembly, dependent models, planning parameters, and logs are saved using the same project identifier, so that subsequent design and archiving can maintain complete references.

[0109] Through the above steps, this embodiment demonstrates... Under the unified operating conditions of ground rail, candidate servo motor-reducer-gear solutions are developed to meet the requirements of torque, speed, inertia, and two-stage interface, and all feasible combinations are provided for user selection; at the same time, The continuous opening requirement is mapped to a Type 4 standard opening module to obtain the actual opening. and deviation It consists of an open section, two standard sections after the opening, and... Tail section composition The ground rail equipment places the cable chain in the installation position and obtains... Each link and Procurement length. This enables integrated, interpretable, and traceable design, from unified working condition input, selection reasoning, user interaction to NX 3D assembly and result output.

[0110] Compared with existing technologies, the AI ​​intelligent agent for automatic selection and assembly design of production line floor rail functional components provided by this invention has the following characteristics: This invention addresses the selection of transmission components and the modular automated assembly design of ground rail equipment in automated production lines. It can receive engineering inputs such as load weight, rapid traverse speed, acceleration time, stroke, friction conditions, transmission efficiency, specific coordinates of the cable chain opening, and cable chain layout parameters. The intelligent system utilizes a deterministic mechanical model, a structured product database, a standard ground rail module knowledge base, and executable assembly rules to generate, verify, iteratively select, and sort candidate solutions for servo motors, reducers, and gear racks with low redundancy. Based on the ground rail model library data, it completes the mapping of standard modules with specified openings, ground rail segment material calculation, tail segment parameterization, cable chain length calculation, and 3D assembly using CAD software.

[0111] The intelligent agent employs a collaborative architecture driven by physical models, rule-based knowledge, and engineering modeling software tools, ensuring rapid and stable output of design solutions without the need for pre-collecting large-scale training samples. The intelligent agent of this invention continuously maintains existing product samples, component performance utilization coefficients, peak capacity coefficients, inertia ratio requirements, mechanical interface dimensions, standard rail section models, coordinate positions of standard cable chain opening modules, and assembly experience as structured knowledge. When candidate components or standard models change, system functionality can be expanded by updating the database and mapping tables without altering the entire design process, making it suitable for rapid reuse under different design loads, operating speeds, rail lengths, and cable chain opening position requirements.

[0112] Regarding functional component selection, the intelligent agent can automatically traverse the enterprise's candidate database and generate multiple sets of solutions that meet the requirements for torque, speed, inertia, and installation dimensions, avoiding designers repeatedly screening and iteratively verifying multiple product samples. The intelligent agent system can provide performance evaluations of the selected solutions and retain the original inputs, key intermediate data, performance constraints, and usage status of each functional component. This allows designers to understand why each solution meets the requirements and make final confirmation based on price, delivery time, spatial layout, and enterprise standards. This method improves selection efficiency while preserving the final decision-making authority of engineering designers. In terms of ground rail assembly, the intelligent agent converts the user-provided continuous drag chain opening coordinates into ground rail segment modules that can be directly called from the standard model library, simultaneously providing the actual opening coordinates and positional deviations. Through the combination of complete standard segments and parameterized tail segments, it can cover standard integer length and non-standard length ground rail lines. The cable chain length can be calculated based on the actual cable chain opening position, the allowance at the fixed and moving ends of the cable chain, the bending length, and the link pitch. The cable chain installation coordinates are based on the actual cable chain opening position, ensuring that the calculation results are consistent with the final 3D model. This method can significantly reduce manual material calculation, model selection, assembly positioning, and repetitive operations, making it suitable for the rapid design of serialized floor rail products.

[0113] Simultaneously, the intelligent agent system in this invention can be deployed in the computer and CAD modeling software environment of the mechanical design department. The common programming language side handles the graphical interface, configuration management, computational reasoning, and report output, while the modeling software side utilizes functional interfaces to complete parametric modeling and assembly without altering the enterprise's existing 3D model format. This invention can be used for the design of ground rails in new production lines, the reuse of transmission system solutions under different load and speed conditions, and the generation of serialized models under different ground rail target lengths and cable chain opening requirements. Its knowledge base, computational rules, and assembly strategies can be continuously maintained within the enterprise, transforming design experience into repeatable, interpretable, and traceable digital design capabilities, possessing a clear software deployment foundation and promising application prospects in production line ground rail equipment design.

[0114] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. An AI intelligent agent for automatic selection and assembly design of production line floor rail functional components, characterized in that, The AI ​​agent uses a unified structured project configuration as its data foundation to intelligently select functional components for the production line floor rail and intelligently establish the floor rail assembly within a single design process. This includes intelligent selection of transmission system components and intelligent assembly of the cable chain open floor rail model. The intelligent selection of transmission system components and the intelligent assembly of the cable chain open floor rail model share input and result identifiers within a unified project, collaboratively designing and outputting a complete production line floor rail design scheme. The intelligent selection of transmission system components completes the selection of production line ground rail functional components and generates a solution. Taking the transmission chain composed of servo motor, reducer and drive gear as the design object, it completes motion requirement calculation, candidate combination generation, multi-constraint loop verification, performance redundancy evaluation and user interaction to filter and output the solution. The intelligent assembly of the cable chain opening ground rail model completes the intelligent assembly of the production line ground rail. Taking the standard ground rail section, the ground rail section with cable chain opening, the ground rail tail section and the cable chain as the design objects, it completes the specified cable chain opening coordinate mapping, ground rail section material calculation, tail section preparation, cable chain length, positioning calculation of each section and automatic assembly by modeling software.

2. The AI ​​intelligent agent for automatic selection and assembly design of production line floor rail functional components as described in claim 1, wherein the AI ​​intelligent agent includes a user interface, structured data configuration, a physical calculation module, a component knowledge base definition, a rule constraint reasoning module, a scheme evaluation module, a standard floor rail module knowledge base, an assembly planning module, a modeling software task scheduling module, an execution module, and a result output module. Each module transmits data through structured configuration, candidate scheme objects, assembly planning objects, logs, and output files; wherein, The user interface collects operating conditions and displays candidate solutions; The physical calculation module generates the calculation requirements of the transmission system based on the user input information. The rule-constrained reasoning module traverses and eliminates candidate combinations of functional component schemes; The assembly planning module converts the continuous opening coordinates of the cable chain into existing standard modules and the number of each segment. The modeling software task scheduling module starts the software's internal execution environment; The execution module calls relevant software functions to complete parametric modeling and component placement.

3. The AI ​​intelligent agent for automatic selection and assembly design of production line floor rail functional components as described in claim 2, wherein the intelligent selection of transmission system components, when completing the selection of production line floor rail functional components and generating a solution, includes the following steps: S101: Obtain the design parameters of the production line floor rail, convert the design parameters into a structured configuration, and verify the design parameters; S103: Calculate the kinematics and speed-related parameters of the transmission system; S105: Generate candidate schemes for functional components based on component parameters; S107: Perform multi-constraint verification on the candidate schemes for the functional components; S109: Evaluate and filter the candidate solutions for the verified functional components and output all feasible solutions; S111: Confirm the target solution and output traceable selection results.

4. The AI ​​intelligent body for automatic selection and assembly design of production line floor rail functional components as described in claim 3, in step S103, based on the structured input and transmission demand continuous calculation method for production line floor rail working conditions, motion demand, transmission parameters and allowable utilization coefficient are unified as input, and driving force, gear size, output end demand, motor side demand and equivalent inertia are formed in sequence. By converting speed into linear velocity and calculating acceleration, acceleration force, friction force and total driving force are calculated. The pitch circle diameter and stroke per revolution are determined by gear module and number of teeth. Then, the requirements of reducer and motor are calculated step by step in combination with candidate reduction ratio and efficiency, so that intermediate quantities are transmitted through the same input chain.

5. The AI ​​intelligent body for automatic selection and assembly design of production line floor rail functional components as described in claim 4, in step S105, based on the hierarchical collaborative selection method of "reducer-helical gear-servo motor" in the structured component database, with the reducer as the outer layer, the helical gear as the middle layer, and the motor as the inner layer, establishes performance relationships and two-level connection dimension relationships in the same combination. By reading the reduction ratio, torque, speed, inertia, and end dimensions of the candidate reducer from the functional component data knowledge, the output shaft is first matched with the gear bore diameter and the output side requirements are calculated; then, the torque and speed requirements of the servo motor are calculated according to the reduction ratio, and the motor shaft diameter is matched with the allowable size of the reducer input end, forming a discrete combination with interrelated parameters.

6. The AI ​​intelligent agent for automatic selection and assembly design of production line floor rail functional components as described in claim 5, in step S107, utilizes a cyclic iterative verification and component replacement strategy oriented towards a finite candidate set to verify the combination of reducer, helical gear, and motor under a unified hard constraint. If the current combination does not meet any constraint, it is eliminated, and trial fitting continues according to the nesting order of motor, gear, and reducer. The process terminates after the finite candidate traversal is completed. Specifically, it includes: When the servo motor does not meet the usage requirements, the next model of motor will be automatically matched. If the current gear or motor candidate does not meet the requirements, select the matching gear; When there is no feasible combination of reducers, replace the reducer and recalculate the speed, torque and equivalent inertia according to the new discrete reduction ratio.

7. The AI ​​intelligent agent for automatic selection and assembly design of production line floor rail functional components as described in claim 6, in step S109, based on the low-performance redundancy optimization method of prioritizing performance utilization constraints and then evaluating capacity utilization balance, when performance utilization rate assessments are equal, the scheme with smaller rated torque and lower inertia ratio is selected first. The speed and torque performance utilization rate scores of the reducer and motor are comprehensively considered, and secondary sorting is performed in combination with the rated torque and inertia ratio of the motor, while retaining the user's final selection; when calculating the performance utilization rate scores of the reducer and motor, the performance utilization rates of the reducer and motor are geometrically averaged according to their weights; the reducer speed includes the maximum input speed and rated input speed of the reducer, the reducer torque includes the maximum torque and rated torque of the reducer, the motor speed includes the acceleration speed and constant speed of the motor, and the motor torque includes the maximum torque and rated torque of the motor.

8. The AI ​​intelligent agent for automatic selection and assembly design of production line floor rail functional components as described in claim 7, wherein the intelligent assembly of the cable chain open floor rail model is based on an automatic assembly method that uses the collaborative interaction between a general programming design decision environment and the built-in execution environment of the modeling software. The general automatic program environment is responsible for user information interaction, automatic parameter verification, knowledge base query, selection decision, and assembly configuration generation. It calls the running engine of the modeling software through structured configuration and task scripts. The built-in environment of the modeling software rereads the configuration, calls the interface to modify the parameterized model, creates the assembly, inserts components according to absolute coordinates, saves the model, and generates a success marker, including the following steps: S201: Obtain the ground rail configuration parameters from the preprocessed production line ground rail design parameters, and load the standard ground rail module; S203: Decompose the total length configuration of the ground rail line, calculate the number and configuration of each ground rail segment, and determine the length of the tail segment: Determine the complete standard segment where the specified coordinates are located based on the standard ground rail segment length, and then calculate the target coordinates within the segment; construct a discrete set with the actual opening coordinates within each standard opening module, select the corresponding module according to the minimum absolute distance, and realize the function of accurately specifying the position of the drag chain opening; at the same time, the ground rail segment coordinates and deviation of the drag chain opening are fed back. S205: Jointly calculate the configuration of the head section and middle stage of the ground rail; S207: Perform modeling and assembly in the modeling software, wherein the modeling software is CAD software; S209: Collect the orbital dependency model and output the model scheme; S211: Output the functional component scheme and assembly model of the production line floor rail.

9. The AI ​​intelligent agent for automatic selection and assembly design of production line floor rail functional components as described in claim 8, in step S205, based on the joint calculation method of the actual coordinates of the cable chain opening, coordinate position deviation, total number of floor rail segments, and configuration, uses a combination of standard segments and parameterized tail segments to form a floor rail line of target length, specifically including: Calculate the number of complete standard segments and the length of the tail segment based on the target total length and the standard segment length; Determine whether to set the head segment, the number of pre-mouth segments, and the number of post-mouth segments based on the opening segment number; If the tail segment has a standard integer length, the standard tail segment model is called; otherwise, the length expression of the parameterized tail segment template is modified so that the sum of the lengths of each segment equals the target length.

10. The AI ​​intelligent body for automatic selection and assembly design of production line floor rail functional components as described in claim 9, in step S207, the cable chain length, number of links and three-dimensional positioning coordinates are calculated based on the actual opening coordinate position of the cable chain in the overall assembly of the floor rail, and the cable chain is automatically placed. The effective straight length is the distance from the actual opening to the far end of the floor rail. The required design length is obtained by superimposing the fixed end allowance, moving end allowance and bending length allowance when placing the cable chain. The total cable chain length is calculated by rounding up according to the cable chain pitch. The three-dimensional installation coordinates of the cable chain are obtained by superimposing the actual opening coordinates with the installation offset corresponding to the floor rail specifications.