A method, system, device and storage medium for generating a tooling fixture solution

Abstract

The application provides a jig scheme generation method, system, device and storage medium, relates to the technical field of artificial intelligence, and fundamentally changes the condition that traditional design seriously depends on manual experience, a process is discrete, and consistency is poor. Through standardization semantic modeling of clamping requirements, the bottleneck that a design intention is not machine-readable is solved. Through deep integration of structure constraint rules and a large model, it is ensured that the generated scheme meets functional requirements and engineering feasibility, so that automation, intelligentization and standardization of the design process are realized, and design efficiency, scheme reliability and knowledge reuse level are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and more specifically, to a method, system, device, and storage medium for generating tooling fixture solutions. Background Technology

[0002] Tooling fixtures are key process equipment in machining and assembly production lines, and their design quality directly determines the machining accuracy, production efficiency, and process stability of parts. Currently, fixture design mainly relies on the professional experience of designers, using computer-aided design (CAD) software for manual modeling and verification. This traditional approach faces the following prominent bottlenecks: First, clamping requirements are expressed in a fragmented and unstructured manner. Design inputs typically include heterogeneous information from multiple sources, such as 3D models of parts, 2D drawings, process cards, and technical requirements described in natural language. This information lacks a unified and computable semantic expression. In particular, key clamping intentions, such as the part's positioning datum, clamping points, force characteristics, and machining avoidance areas, are difficult for the design system to automatically and accurately identify and understand. This leads to a high degree of reliance on manual interpretation in the design process, making knowledge difficult to reuse and prone to ambiguity.

[0003] Secondly, the design process has a low level of intelligence and relies heavily on trial and error. Existing automated design-aided methods are mostly based on parametric templates or finite rule bases. When faced with complex and ever-changing part geometry, process requirements, and machine tool environments, they have poor generalization ability and find it difficult to generate novel and optimized structures.

[0004] Therefore, existing technologies struggle to achieve machine-understandable expressions of clamping requirements and lack guiding mechanisms for deeply integrating domain knowledge with large-scale model generation capabilities. This results in limited automation levels in tooling and fixture design, making it difficult to balance design efficiency and solution reliability. There is an urgent need for a new method capable of intelligently parsing and semantically modeling clamping requirements, and using structured engineering constraints to guide artificial intelligence models in manufacturable design reasoning, in order to substantially advance fixture design towards intelligence and automation. Summary of the Invention

[0005] The problem solved by this invention is one or more of the aforementioned related technical problems.

[0006] To address the aforementioned problems, this invention provides a method, system, device, and storage medium for generating tooling fixture solutions.

[0007] In a first aspect, the present invention provides a method for generating a tooling fixture solution, comprising: Obtain the initial clamping requirements information corresponding to the tooling fixture to be designed; the initial clamping requirements information includes at least the three-dimensional model data of the part to be processed or assembled, the processing procedure information, the processing equipment information, the clamping posture requirements, and the processing accuracy and stability requirements; The initial clamping requirements information is parsed and processed to extract key feature information for fixture design; Based on the aforementioned key feature information, a structured semantic model of clamping requirements is constructed. A fixture structure constraint rule system is constructed, and based on the clamping requirement semantic model and the fixture structure constraint rule system, a large language reasoning model is used to generate reasoning to obtain fixture structure description data; the fixture structure description data includes fixture structure topology, functional unit configuration and key parameters; Based on the fixture structure description data, at least one set of fixture design schemes corresponding to the tooling fixture to be designed is generated; The fixture design scheme includes the overall structural layout of the fixture, the configuration of positioning elements, the form of the clamping mechanism, the setting of the support structure, and the assembly relationship between the various structural elements.

[0008] Optionally, the large language reasoning model includes a clamping function reasoning layer, a structural topology reasoning layer, and a parameter and layout reasoning layer; the step of generating reasoning based on the clamping requirement semantic model and the clamping structure constraint rule system through the large language reasoning model to obtain clamping structure description data includes: Based on the clamping function inference layer, the set of functional units is determined according to the clamping requirement semantic model; Based on the structural topology reasoning layer, the structural topology scheme is determined according to the set of functional units and the fixture structure constraint rule system; Based on the parameters and layout inference layer, the fixture structure description data is determined according to the structural topology scheme.

[0009] Optionally, the key feature information includes at least: positioning reference features identified from the three-dimensional model data, freedom constraint-related features and force-sensitive features analyzed from the processing procedure information, and prohibited clamping areas and processing avoidance areas determined based on the processing procedure information and the processing equipment information.

[0010] Optionally, the clamping requirement semantic model is represented by a unified semantic tag system and a parameterized description method; wherein, the semantic tag system includes positioning requirement semantic tags for describing the positioning reference type and constraint degrees of freedom, clamping requirement semantic tags for describing the clamping direction and force range, and support requirement semantic tags for describing the support position and load-bearing requirements.

[0011] Optionally, the fixture structure constraint rule system includes at least one of the following: positioning structure constraints for limiting the number and arrangement of positioning elements, clamping constraints for limiting the direction and range of clamping force, spatial interference constraints for preventing interference with parts, cutting tools or machine tools, and manufacturing and assembly constraints for satisfying manufacturability and assemblability.

[0012] Optionally, after generating at least one set of fixture design schemes corresponding to the tooling fixture to be designed based on the fixture structure description data, the method further includes: The fixture design scheme is constrained and verified to obtain the verification results; the constraint verification includes at least semantic compliance verification based on the clamping requirement semantic model and structural compliance verification based on the fixture structure constraint rule system. If the verification result is unsuccessful, the verification result is fed back to the large language reasoning model, which then adjusts the fixture structure description data and regenerates the fixture design scheme until a design scheme that meets the verification requirements is obtained.

[0013] Optionally, after generating at least one set of fixture design schemes corresponding to the tooling fixture to be designed based on the fixture structure description data, the method further includes: The optimal fixture design is obtained by screening multiple different fixture design schemes according to the preset optimization rules.

[0014] Secondly, the present invention provides a tooling fixture solution generation system, comprising: The acquisition unit is used to acquire the initial clamping requirement information corresponding to the tooling fixture to be designed; the initial clamping requirement information includes at least the three-dimensional model data of the part to be processed or assembled, the processing procedure information, the processing equipment information, the clamping posture requirements, and the processing accuracy and stability requirements; The extraction unit is used to parse and process the initial clamping requirement information and extract key feature information for fixture design. The construction unit is used to construct a structured clamping requirement semantic model based on the key feature information; construct a fixture structure constraint rule system; and generate and reason through a large language reasoning model according to the clamping requirement semantic model and the fixture structure constraint rule system to obtain fixture structure description data; the fixture structure description data includes fixture structure topology, functional unit configuration and key parameters. The processing unit is used to generate at least one set of fixture design schemes corresponding to the fixture to be designed based on the fixture structure description data; wherein, the fixture design scheme includes the overall structural layout of the fixture, the configuration of positioning elements, the form of clamping mechanism, the setting of support structure and the assembly relationship between various structural elements.

[0015] Thirdly, the present invention provides a tooling fixture solution generation device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the tooling fixture solution generation method described in the first aspect.

[0016] Fourthly, the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the tooling fixture solution generation method described in the first aspect.

[0017] The beneficial effects of the tooling fixture generation method, system, equipment, and storage medium of the present invention are: By constructing a transformation path from unstructured clamping requirements to a structured semantic model, and innovatively inputting semantic design goals and formalized engineering constraint rules into a large model for guided reasoning, a complete fixture design scheme that can directly guide production is generated. This fundamentally changes the situation where traditional design relies heavily on human experience, has discrete processes, and poor consistency. The standardized semantic modeling of clamping requirements solves the bottleneck of machine-inreadable design intent. The deep integration of structural constraint rules and the large model ensures that the generated scheme simultaneously meets functional requirements and engineering feasibility, thereby realizing the automation, intelligence, and standardization of the design process, and significantly improving design efficiency, scheme reliability, and knowledge reuse level. Attached Figure Description

[0018] Figure 1 This is one of the flowcharts illustrating a method for generating a tooling fixture according to an embodiment of the present invention; Figure 2 This is a second flowchart illustrating a method for generating a tooling fixture according to an embodiment of the present invention. Figure 3 This is a schematic diagram of the structure of a tooling fixture generation system according to an embodiment of the present invention. Detailed Implementation

[0019] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the accompanying drawings and embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of protection of the present invention.

[0020] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0021] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to"; the term "based on" means "at least partially based on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; and the term "optionally" means "optional embodiments". Definitions of other terms will be given in the following description. It should be noted that the concepts of "first," "second," etc., mentioned in this invention are used only to distinguish different devices, modules, or units, and are not intended to limit the order of functions performed by these devices, modules, or units or their interdependencies.

[0022] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0023] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0024] like Figure 1 As shown in the figure, an embodiment of the present invention provides a method for generating a tooling fixture solution, comprising: Step S100: Obtain the initial clamping requirements information corresponding to the tooling fixture to be designed; the initial clamping requirements information includes at least the three-dimensional model data of the part to be processed or assembled, processing process information, processing equipment information, clamping posture requirements, and processing accuracy and stability requirements.

[0025] Specifically, the core task of step S100 is to completely and accurately collect the initial conditions and environmental parameters necessary to define a fixture design task. This step is not simply uploading a part drawing, but rather systematically acquiring a structured design task package to ensure that subsequent automated processing has a basis.

[0026] 3D model data refers to the digital geometric model of the part to be processed or assembled, usually a CAD file in a standard format (such as STEP, IGES). It accurately describes all geometric features of the part, including its shape, dimensions, holes, and surfaces, and is the sole data source for the system to automatically identify and analyze geometric features. For example, a STEP file of an engine block contains all the details of the cylinder bores, bolt connection surfaces, locating pin holes, etc.

[0027] Machining process information: This refers to the specific machining content, sequence, and process parameters that the part needs to be machined in this clamping operation. For example: "Process 20: Finish milling the upper surface, depth of cut 0.5mm, using an 80mm face mill." This information clarifies the machining area that the fixture needs to guarantee, the tool path, and the direction and magnitude of the cutting force generated during machining. It is crucial for judging the stress conditions and determining the avoidance area.

[0028] Machining equipment information: This refers to the specifications and interface data of the machine tool to be used. For example: "Vertical machining center, table size 800mm×500mm, T-slot specification 18mm, maximum spindle travel Z=600mm". This information specifies the installation method of the fixture (how to connect it to the table), the external dimensional boundaries (it must not interfere with the machine tool safety door), and the workspace constraints.

[0029] Clamping posture requirements: These refer to the predetermined placement of the part relative to the machine tool coordinate system during machining. For example, "the bottom surface of the part is parallel to the machine tool table, and its major axis is along the X-axis of the machine tool." This directly determines the selection of the positioning datum and the direction of force applied by the clamping mechanism.

[0030] Machining accuracy and stability requirements refer to the quality indicators and process constraints of the machining results. For example, "the flatness of the machined surface ≤ 0.02mm, and the workpiece amplitude during machining < 5μm." This translates into quantitative design targets for the fixture's positioning accuracy, rigidity, and clamping reliability.

[0031] By mandating and structurally collecting initial clamping requirements information covering all dimensions of "parts-process-equipment-posture-quality", the traditional design task that relies on manual interpretation and implicit experience is transformed into a complete, clear, and computer-processable "digital design order". This fundamentally avoids design rework caused by missing or ambiguous information, and lays a precise and reliable data foundation for subsequent automated feature analysis, semantic modeling, and intelligent reasoning. It is the primary prerequisite for realizing the intelligence of the entire process.

[0032] Step S200: The initial clamping requirement information is parsed and processed to extract key feature information for fixture design.

[0033] Specifically, the raw, multimodal data collected by the S100 is automatically decoupled, analyzed, and fused to mechanically identify and extract the core elements that directly affect the fixture structure design. This process replaces the designer's experience-based work of "reading drawings" and "analyzing processes," transforming unstructured input into a structured list of design features.

[0034] The parsing process may include: Geometric Feature Analysis and Positioning Reference Identification: This function automatically identifies the geometric topology and features of 3D model data. It not only recognizes all geometric entities (such as planes, cylindrical surfaces, and holes), but also automatically determines and extracts the most suitable features as fixture positioning references based on geometric attributes (such as area and precision markings) and process logic (such as the machining surfaces specified in the machining process information). For example, from a box-shaped part, it might identify the large flat surface on its bottom (largest area, unmarked for machining) as the primary positioning reference surface, and two relatively large precision holes as auxiliary positioning references.

[0035] Process mechanics analysis and force characteristic extraction: Combining machining process information (such as "milling the upper surface"), the main direction, magnitude, and area of ​​action of the cutting force during machining can be simulated or calculated. This automatically identifies which areas on the part are stress-sensitive features (such as thin walls or cantilever structures), requiring special protection or reinforced support during clamping. Simultaneously, it can also deduce the optimal direction (usually pointing towards the main locating surface) of the clamping force to ensure workpiece stability.

[0036] Manufacturing environment analysis and spatial constraint extraction: By integrating machining equipment information (such as table size and spindle travel) and machining process information (such as tool size and machining path), the system calculates the motion envelope space required by the tool during machining. Any part areas that conflict with this space are marked as machining avoidance zones, which the fixture structure must not intrude upon. Simultaneously, the surfaces on the parts about to be machined are automatically marked as prohibited clamping areas to prevent clamping devices from damaging machined surfaces or interfering with the tool.

[0037] By automatically mapping the original requirements to design features through algorithms, the feature recognition work, which relies on human experience, is prone to omissions and inconsistencies, is transformed into an objective, repeatable, and fully quantifiable calculation process. This not only greatly improves the efficiency and accuracy of feature extraction and avoids human negligence, but more importantly, it produces a structured and semantic "list of design elements". This provides a unique and authoritative data foundation for building a machine-understandable and reasonable semantic model of clamping requirements in the next stage. It is a key transformation hub for realizing the lossless transmission of design intent and automated decision-making.

[0038] Step S300: Based on the key feature information, construct a structured semantic model of clamping requirements.

[0039] Specifically, the discrete "key feature information" extracted in step S200 is organized into a complete, self-consistent "design instruction set" that can be directly interpreted by an artificial intelligence model, using a unified machine language. This is a structured assembly requirement semantic model. This model is no longer a human-readable drawing or list, but a machine-executable "assembly task program".

[0040] Core build process: Semantic encapsulation and unified expression: Various key feature information is encapsulated into several fixed semantic modules. Each module uses predefined semantic tags (a type of standardized keyword) and parameterized descriptions to express specific requirements.

[0041] The positioning requirement semantic module is used to declare "where to fix the workpiece". It translates the positioning reference features recognized by the S200 (such as "large flat bottom surface" and "two holes") into explicit instructions.

[0042] For example: Positioning reference: {Type: Plane, Geometric ID: Plane A}; {Type: Hole, Geometric ID: Hole 1, Hole 2}; Constrained degrees of freedom: [Translation along Z, Rotation around X, Rotation around Y]. This is equivalent to telling the system: "Use this plane and these two holes to restrict the workpiece's movement in these directions." Clamping Requirement Semantic Module: Used to declare "how to apply force to stabilize the workpiece". Based on the force analysis results, it specifies the point of application, direction, and magnitude range of the clamping force.

[0043] For example: Clamping direction: vector (0, 0, -1) (i.e., vertically downward); Clamping position: {area: upper surface non-machined area}; Clamping force range: [500N, 800N].

[0044] Support Requirement Semantic Module: Used to declare "where additional support is needed to prevent deformation". It specifies the location and requirements of auxiliary supports for stress-sensitive features.

[0045] For example: Support location: {Point coordinates: P1, P2, P3, P4} (located at the four corners of the bottom surface of the part); Support type: Elastic auxiliary support.

[0046] The process constraint semantic module is used to declare "what must be avoided during design". It transforms prohibited and avoidable areas into negative constraints.

[0047] For example: No clamping area: {Geometric ID: Face B (side to be machined)}; Clearance space: {Description: Cylindrical envelope of Φ10 drill bit along the negative Z-axis with a stroke of 50mm}.

[0048] Taking "aluminum alloy casing" as an example, S300 transforms the "Manifestation" output by S200 into the following semantic model (illustrated): {Location requirements:{ Primary reference: {Type: Plane, Reference: Shell bottom} Auxiliary reference: {Type: Hole group, Reference: [Hole A, Hole B]} Constrained degrees of freedom: [Tx, Ty, Tz, Rx, Ry]} Clamping requirements: {Direction: (0, 0, -1), Area of ​​action: Top boss of the housing, Force value: [600N, 750N]} Process constraints: {Prohibited clamping surface: [Side drilling area], clearance space: Drill bit path envelope}}.

[0049] By creating a structured semantic model of clamping requirements, a revolutionary transformation of design intent from human language to machine language has been achieved. It transforms the originally vague and experience-dependent process requirements into a precise, unambiguous, computable, and reasonable "digital contract." This model not only provides a clear and stable objective function for subsequent artificial intelligence generation, but more importantly, it establishes a reliable communication protocol between engineering domain knowledge and general AI models. It solves the problem of "unclear input intent" in intelligent design systems and is a decisive link in ensuring that the output results of automated design are correct in principle and feasible within constraints.

[0050] Step S400: Construct a fixture structure constraint rule system, and generate and reason using a large language reasoning model based on the clamping requirement semantic model and the fixture structure constraint rule system to obtain fixture structure description data; the fixture structure description data includes fixture structure topology, functional unit configuration and key parameters.

[0051] Specifically, step S400 is the core step in achieving intelligent generation in this method, which includes two key actions: knowledge injection and constraint-guided generative reasoning. This step does not allow the large model to run freely, but rather equips it with engineering knowledge and sets constraints, enabling it to perform creative design within clear objectives and strict rule boundaries. Among them, the large language reasoning model adopts a large language model or a multimodal large model that has been adapted and trained with data from the fixture design domain and engineering constraint rules.

[0052] Constructing a system of constraint rules for fixture structures (knowledge injection): This system is a formalized collection of engineering design knowledge and experience rules, derived from domain principles (such as the six-point positioning principle) and engineering specifications (such as safety standards and manufacturing processes). It is not specific to any particular part, but rather a general specification applicable to a wide range of fixture designs. For example: Positioning structure constraint: "If planar positioning is used, at least three non-collinear support points must be set up."

[0053] Clamping direction constraint: "The main component of the clamping force should be perpendicular to the main positioning surface."

[0054] Spatial interference constraint: "No fixture element may intrude into the tool motion envelope space defined by the machining equipment information and machining process information."

[0055] Manufacturability constraint: "Prioritize the use of components such as cylinders and pressure plates from the standard parts library."

[0056] Constraint-guided large-scale model generative reasoning (intelligent solution): The specific "task specification" (clamping requirement semantic model) and general specifications (clamping structure constraint rule system) are simultaneously input into the large language reasoning model (a large-scale artificial intelligence model fine-tuned with domain knowledge and possessing structural understanding and generative capabilities). The model then performs reasoning under this dual guidance: Reasoning objective: To achieve all the clamping requirements defined by the semantic model while satisfying all the rules.

[0057] Inference Output: Generates fixture structure description data, a machine-readable blueprint that fully defines the core architecture of the fixture, including: Fixture structure topology: describes the spatial connection and cooperation relationship between various functional components (such as base plate, support block, pressure plate). For example: "The positioning support pin is installed on the upper surface of the base plate, the lateral positioning pin is installed on the side upright plate of the base plate, and the hydraulic cylinder is connected to the pressure plate through a hinge."

[0058] Functional Unit Configuration: Specifies the component types and basic specifications to be used at each location in the topology. For example: "Locking Points 1-3: Adjustable support pins; Lateral positioning: Fixed cylindrical pins; Clamping mechanism: Hinged pressure plate driven by a single piston rod hydraulic cylinder."

[0059] Key parameters: Determine the specific dimensions and mechanical values ​​that ensure functionality and performance. For example: "Support pin height: 45mm; Clamping cylinder output force: 700N; Pressure plate rotation radius: 120mm."

[0060] For the aforementioned aluminum alloy housing, a semantic model containing instructions such as "bottom surface and two-hole positioning," "top clamping," and "avoiding drilling areas," along with general rules such as "three-point surface positioning," "clamping force vertically downward," and "standard parts priority," are input into the large model. The model performs reasoning under these rule constraints and may output the following descriptive data (illustrated): {Topology: [Three-point support on the base plate, cylindrical pins on the side uprights, top hinge pressure plate], Configuration: [Support pin type: SW18, Pressure plate type: standard swing pressure plate], Parameters: [Clamping force: 700N, Pressure plate opening height: 50mm]}. This output directly defines the fixture's skeleton.

[0061] By introducing a formalized system of fixture structure constraints, the generated results of the large model are constrained, ensuring that the output is correct in principle and feasible within the constraints. The generated fixture structure description data serves as a complete and structured intermediate representation, enabling a deterministic leap from abstract requirements to specific solutions. This provides precise and programmable input for the automatic construction of the final solution, which is a key guarantee for the innovativeness and engineering practicality of this method.

[0062] Step S500: Generate at least one set of fixture design schemes corresponding to the fixture to be designed based on the fixture structure description data; wherein, the fixture design scheme includes the overall structural layout of the fixture, the configuration method of the positioning elements, the form of the clamping mechanism, the setting of the support structure, and the assembly relationship between the various structural elements.

[0063] Specifically, the "design blueprint" (fixture structure description data) generated in step S400 and stored within the machine is automatically instantiated, specified, and visualized, transforming it into a "design scheme report" that engineers can directly review, evaluate, or import into a CAD system for detailed design. This process essentially "compiles" structured data into engineering language.

[0064] Generation process: Receive fixture structure description data and perform comprehensive processing on its three main elements: Based on the fixture structure topology (the spatial relationship between each component), determine the installation position and orientation of the components.

[0065] Based on the functional unit configuration (component type and specifications), call the specific 3D model from the standard parts library or model library.

[0066] Based on key parameters (dimensions and mechanical values), drive the model to complete parametric deformation (such as adjusting the height of the support block) or selection (such as selecting a cylinder model with an output force of 700N).

[0067] Based on the above instantiation results, one or more complete fixture design schemes are automatically organized and generated. These schemes are comprehensive descriptions that can be directly used to guide subsequent engineering implementation, specifically including: Overall fixture structure layout: refers to the overall arrangement and orientation of all fixture components in space. For example: "The fixture adopts a base plate structure, with positioning elements concentrated on the left side of the base plate and clamping mechanism located on the upper right of the part, resulting in a compact overall structure."

[0068] Positioning element configuration: A detailed explanation of the positioning function. For example: "Using 'one-sided two-pin' positioning: Three adjustable support pins are arranged on the base plate to form the main positioning surface, and a cylindrical pin and a diamond pin are used on the side to restrict the rotational degree of freedom."

[0069] Clamping mechanism type: A detailed description of how the clamping function is implemented. For example: "Using a 'hydraulic cylinder-driven hinged pressure plate' type, the pressure plate can be quickly lifted for workpiece loading and unloading."

[0070] Support structure setup: Specific description of auxiliary supports. For example: "Two elastic auxiliary supports are set in the suspended area at the bottom of the part to enhance process rigidity."

[0071] Assembly relationships between structural elements: Describe how components are connected and fixed. For example: "The locating pin is pressed into the base plate by an interference fit; the pressure plate is connected to the base by hinge bolts; the hydraulic cylinder is mounted on the fixture bracket by a flange."

[0072] For example: Following the output regarding the aluminum alloy housing in step S400, upon receiving the description data, automatically execute the following: ① Retrieve a standard base plate model, three adjustable support pins, one cylindrical pin, one standard hydraulic cylinder, and a set of hinge pressure plate assemblies from the library; ② Automatically assemble the support pins and positioning pins to the specified coordinate positions on the base plate according to the topology data; ③ Set the support pin height to 45mm according to the parameter data and select a 700N model for the hydraulic cylinder; ④ Automatically generate a solution document, which includes a 3D assembly drawing (showing the overall structural layout), a bill of materials (clarifying the configuration method and form), and a brief assembly instruction (describing the assembly relationships).

[0073] By automatically and one-click transforming machine-readable "design blueprints" into "design solutions" that engineers can directly understand and use, a closed-loop intelligent design process is achieved, seamlessly integrating cutting-edge artificial intelligence reasoning capabilities into mature engineering practices. It not only generates theoretically sound and detailed feasible solutions with extremely high efficiency, eliminating manual conversion between concepts and concrete designs and significantly increasing design output speed, but more importantly, the structured and parametric solutions it generates provide direct and accurate input for subsequent automatic simulation analysis, cost accounting, and even CNC programming. This truly extends the value of intelligent design from "solution generation" to "manufacturing preparation," significantly improving the automation level and collaborative efficiency of the entire process equipment development chain.

[0074] In this embodiment, a transformation path from unstructured clamping requirements to a structured semantic model is constructed. Semantic design goals and formalized engineering constraint rules are input into a large model for guided reasoning, ultimately generating a complete fixture design scheme that can directly guide production. This fundamentally changes the situation where traditional design heavily relies on human experience, has discrete processes, and poor consistency. Standardized semantic modeling of clamping requirements solves the bottleneck of machine-inreadable design intent. Deep integration of structural constraint rules and the large model ensures that the generated scheme simultaneously meets functional requirements and engineering feasibility. This achieves automation, intelligence, and standardization of the design process, significantly improving design efficiency, scheme reliability, and knowledge reuse.

[0075] Optionally, such as Figure 2 As shown, the large language reasoning model includes a clamping function reasoning layer, a structural topology reasoning layer, and a parameter and layout reasoning layer; the clamping requirement semantic model and the clamping structure constraint rule system are used to generate reasoning through the large language reasoning model to obtain clamping structure description data, including: Based on the clamping function reasoning layer, the set of functional units is determined according to the clamping requirement semantic model; Based on the structural topology reasoning layer, the structural topology scheme is determined according to the set of functional units and the fixture structure constraint rule system; Based on the parameter and layout inference layer, the fixture structure description data is determined according to the structural topology scheme.

[0076] Specifically, the hierarchical reasoning described is an intelligent solution strategy that decomposes and refines complex fixture design tasks layer by layer. Through three functionally defined and sequentially progressive reasoning layers, it simulates the thought process of an experienced engineer "from functional planning to detailed design," ensuring that the generated results both conform to the high-level intent and satisfy the low-level constraints.

[0077] Clamping Function Inference Layer: From "Requirements" to "Function List" Input: Clamping requirement semantic model (i.e., a machine-readable "design task book").

[0078] Processing and Output: This layer serves as the solution planning layer, and its core task is to answer the question, "What types of clamping actions are needed to achieve all the requirements?" It analyzes various requirements such as positioning, clamping, support, and constraints in the semantic model, maps them, and instantiates them into specific, executable functional units.

[0079] For example, for a box-shaped part that requires "bottom positioning, side clamping, and bottom auxiliary support," the output functional unit set of this layer of reasoning might be: {Positioning unit: planar positioning assembly (1 set), Clamping unit: lateral spiral pressure plate (2 sets), Support unit: adjustable auxiliary support (4 units)}. It determines "what to use" to achieve the function, but has not yet decided "how many to use, where to place them, and what they should look like."

[0080] Structural Topological Reasoning Layer: From "Function List" to "Spatial Relationship Diagram" Input: The set of functional units in the previous layer + the system of constraint rules for fixture structure.

[0081] Processing and Output: This layer serves as the layout design layer, and its core task is to answer the question, "How should these functional units be connected and arranged to achieve their functions in space and comply with all rules?" Under the strict constraints of structural rules (such as no fewer than three positioning points, clamping force direction should point towards the positioning surface, and processing areas must be avoided), it deduces the spatial connections and relative positional relationships between each functional unit.

[0082] Continuing with the previous example, this layer of reasoning determines that: the three support points in the planar positioning assembly should be arranged in the triangular area projected onto the bottom surface of the housing; the two sets of lateral spiral pressure plates should be symmetrically arranged on both sides of the housing, and their force axes should point towards the positioning surface; the four auxiliary supports should be located near the four corners of the bottom of the housing. The output structural topology scheme is a spatial relationship network describing relationships such as "point A (support pin) is installed at position P1 on the base plate, in contact with the bottom surface of the part; component B (pressure plate) is hinged to the base plate at position P2 through connector C...", clarifying "how to place and how to connect".

[0083] Parameter and Layout Inference Layer: From "Spatial Relationship Diagram" to "Engineering Blueprint" Input: The structural topology scheme of the previous layer.

[0084] Processing and Output: This layer serves as the detailed design layer, and its core task is to answer the question, "What are the key dimensions, mechanical parameters, and precise locations of each specific unit?" It transforms the qualitative relationships in the topology scheme into precise quantitative descriptions.

[0085] Continuing with the previous example, this layer of reasoning ultimately determines: the protrusion height of each support pin is H=32.5mm; the required clamping force for the lateral pressure plate is F=650N, and based on this, the M12 screw specification is selected; the spring stiffness of the auxiliary support is K=50N / mm; and the precise coordinates of the mounting holes of all components on the base plate are (X1, Y1), (X2, Y2)... The output fixture structure description data is a complete, structured engineering dataset that includes the aforementioned structural topology (connection relationships), functional unit configuration (specific selection), and all these key parameters.

[0086] At the structural topology inference layer, when faced with the same set of functional units (such as needing two clamping points), the model may infer different topology schemes based on different "priority rules" (such as "cost priority" or "stiffness priority"): the cost priority scheme may choose the simplest direct clamping plate; while the stiffness priority scheme may infer a more complex linkage hook clamping plate structure to achieve better force balance.

[0087] This hierarchical reasoning mechanism decomposes the complex fixture structure generation problem into three manageable and interpretable progressive stages: "functional planning, spatial layout, and parameter refinement." Each stage introduces relevant domain knowledge (semantic models, constraint rules) for guidance, significantly reducing the complexity and uncertainty of generating the entire solution from a large model at once. This not only ensures that the results of each layer of reasoning are strictly aligned with engineering logic, greatly improving the rationality and feasibility of the generated solution, but also makes the entire generation process highly traceable and debuggable. When the results do not meet requirements, the mechanism can quickly locate the problem level and perform targeted optimization. This is the core architectural advantage for achieving stable, reliable, and high-quality intelligent design.

[0088] Optionally, the key feature information includes at least: positioning reference features identified from the three-dimensional model data, freedom constraint-related features and force-sensitive features analyzed from the processing procedure information, and prohibited clamping areas and processing avoidance areas determined based on the processing procedure information and the processing equipment information.

[0089] Specifically, the positioning reference features are identified from the 3D model data: automatic geometric topology analysis and feature recognition are performed on the 3D CAD model (3D model data) of the part. It not only identifies all geometric entities (faces, edges, holes, shafts), but also automatically evaluates and selects the most suitable features as the positioning reference for the fixture based on geometric attributes (such as area size, shape regularity, and precision marking), process attributes (combined with process information to determine whether it is a machined or unmachined surface) and clamping stability principles.

[0090] Detailed explanation and examples: Positioning reference surface: This is usually selected from the surface of the part that has a large area, high flatness, and is typically unmachined or machined. For example, the bottom surface of a gearbox housing or the end face of a mounting flange.

[0091] Locating holes / locating shafts: Select hole or shaft features with high precision (e.g., marked with H7 tolerance) and larger spacing to enhance stability. For example, two engine mounting holes on a housing.

[0092] Different priority strategies can be configured. For example, in "accuracy priority" mode, features with the highest dimensions and geometric tolerances will be selected as the reference; in "stability priority" mode, the plane with the largest projected area will be selected.

[0093] Analyze the characteristics of freedom constraints and stress sensitivity from machining process information: Transform abstract process requirements into specific spatial and mechanical constraints. Analyze the process content (such as "milling a plane" or "drilling a through hole"), combine the tool type and cutting parameters, and deduce the degrees of freedom that must be restricted on the workpiece and the main stress conditions during the machining process through physical rules or empirical models.

[0094] Detailed explanation and examples: Freedom constraint related features: If the operation is "milling the upper surface", then the translational degree of freedom perpendicular to the surface (Z-axis translation) and the rotational degree of freedom about axes parallel to the surface (rotation about the X and Y axes) must be fully constrained. The system will identify the part features directly related to constraining these degrees of freedom, that is, the parts that need to be contacted or aligned by the corresponding fixture elements (support pins, stop pins, etc.).

[0095] Stress-sensitive features: When analyzing the direction and magnitude of cutting forces (main cutting force, feed force), the system simultaneously analyzes the geometric model of the part and automatically identifies areas with insufficient rigidity, such as thin walls, deep cavities, and overhanging structures. For example, when milling a box-shaped side wall, the thin wall on the opposite side may be identified as a "stress-sensitive feature," suggesting that the fixture design may need to add auxiliary support in that area to prevent deformation or vibration.

[0096] For stress analysis, simplified rule-based reasoning (e.g., the main force during drilling is axial thrust) can be used, or more complex finite element analysis (FEA) pre-calculation modules can be integrated to obtain more accurate force flow paths and identification of deformation weak points.

[0097] Based on machining process information and machining equipment information, prohibited clamping areas and machining avoidance areas are determined: this is a key safety analysis to ensure the compatibility of fixtures and processes. The system integrates the machining path of the process with the actual physical boundaries of the machine tool, performs Boolean operations and spatial reasoning, and dynamically generates an "unintrusive" virtual space.

[0098] Detailed explanation and examples: No clamping area: All surfaces and their adjacent areas that are clearly marked as to be machined in the machining process information. For example, on the outer cylindrical surface of a journal that requires fine grinding, no clamping elements are allowed to come into contact with it within a certain range above and around it to prevent damage to the machined surface or interference with measurements.

[0099] Machining clearance zone: Based on the tool type defined in the process (e.g., a Φ20 diameter end mill), the machining trajectory (e.g., cavity contour), and the machine tool's travel limits in the machining equipment information, the system calculates the tool motion envelope (including the tool body, clamping parts, and necessary safety clearances). For example, on a vertical machining center, when a long-blade end mill is machining a deep cavity, its entire cylindrical motion space from the spindle end to the tool tip is the machining clearance zone. No part of the fixture can enter this space. The size of the safety clearance can be configured according to the machining type (roughing / finishing) and material (steel / aluminum alloy). For high-speed machining, a larger dynamic clearance zone may be needed to account for potential vibrations or tool runout.

[0100] By conducting collaborative analysis and intelligent reasoning on multi-source data, the traditional feature recognition work, which relies on human experience, is prone to oversights, and is highly subjective, is transformed into a fully automated, quantifiable, and repeatable accurate calculation process. This not only greatly improves the efficiency and consistency of feature extraction, but also fundamentally ensures that all subsequent design decisions are based on a complete and accurate understanding of engineering constraints. This lays an indispensable and solid data foundation for generating safe, reliable, and process-compatible fixture solutions, and is a key prerequisite for avoiding design rework and achieving "first-time correct" intelligent design.

[0101] Optionally, the clamping requirement semantic model is represented by a unified semantic tag system and a parameterized description method; wherein, the semantic tag system includes positioning requirement semantic tags for describing the positioning reference type and constraint degrees of freedom, clamping requirement semantic tags for describing the clamping direction and force range, and support requirement semantic tags for describing the support position and load-bearing requirements.

[0102] Specifically, the core step is to transform discrete and heterogeneous "key feature information" into a unified, structured, and machine-understandable set of design instructions. Its core innovation lies in defining a standardized engineering semantic description language that encodes design intent in an unambiguous way.

[0103] Constructing a unified semantic tagging system: This system defines a series of standard "keywords" (semantic tags) for classifying and labeling core concepts in the field of fixture design. Its purpose is to create a "vocabulary" that both machines and systems can accurately understand, ensuring consistency and accuracy in information transmission.

[0104] Detailed explanation and examples: Location requirement semantic tags: used to precisely describe "where it is fixed" and "which movements are restricted".

[0105] Loc_BaseType (Location Base Type): Its parameter value can be "plane", "internal hole", "external cylinder", etc. For example, Loc_BaseType: "plane".

[0106] Loc_DOF (Degrees of Freedom of Constraint): Its parameter value is a list that specifies the degrees of freedom of the specific constraint. It is usually represented by standard symbols such as ["TX", "TY", "RZ"] to indicate that the translation in the X and Y directions and the rotation about the Z axis are constrained.

[0107] Clamping requirement semantic tag: used to precisely describe "how force is applied".

[0108] Clamp_Direction: Represented by a unit vector, such as (0, 0, -1) which means vertically downward.

[0109] Clamp_ForceRange (clamping force range): represented by parameter pairs, such as [min: 500N, max: 800N].

[0110] Supporting semantic tags: used to precisely describe "how to assist in support".

[0111] Support_Type: Its value can be "rigid", "elastic", or "adjustable".

[0112] Support_LoadCapacity (load requirement): e.g., "≥200N".

[0113] The tagging system can be extended or customized according to specific industry or enterprise standards. For example, in the field of aerospace composite material processing, a Clamp_PressureLimit tag can be added to prevent workpiece damage. The tag syntax can also adopt different formal languages, such as JSON Schema, XML Schema, or ontology-based descriptions, to adapt to different system integration needs.

[0114] Parametric description: This is the process of populating semantic tags with specific, quantifiable numerical or geometric references. It binds qualitative tags with quantitative engineering data to form executable instructions.

[0115] Detailed explanation and examples: A complete description of "positioning requirements" might be: {Loc_BaseType: "plane", Geometry_ID: "Face_5", Loc_DOF: ["TZ", "RX", "RY"]}. This means "using the plane with ID 'Face_5' in the 3D model as a reference, constraining the movement of the part along the Z-axis and its rotation around the X and Y axes."

[0116] A complete description of "clamping requirements" might be: {Clamp_Location: {Region: "Top_Flange"}, Clamp_Direction: [0,0,-1], Clamp_ForceRange: [600, 750]}. This means "apply a clamping force of 600 to 750 Newtons in the vertically downward direction in the 'top flange' region of the part".

[0117] Parametric design can have different levels of precision. In the initial conceptual design phase, Clamp_Location can be a coarse description of a region (such as "top surface"); while in the detailed design phase, it can be associated with precise geometric surface IDs or coordinate ranges. The system supports this progressive refinement from fuzzy to precise.

[0118] The semantic tags with parameters mentioned above are organized into a complete, structured document or data object according to a certain logical structure (such as JSON, XML or knowledge graph), which is the final semantic model of the binding requirements.

[0119] Example: A semantic model might be organized in the following structure (simplified illustration): Json: {"FixturingRequirements": { "Locating": [{ "tag": "Loc_BaseType", "value": "Plane", "ref": "Bottom_A"},{ "tag": "Loc_DOF", "value": ["TZ", "RX", "RY"]}], "Clamping": [{ "tag": "Clamp_Direction", "value": [0, 0, -1]},{ "tag": "Clamp_ForceRange", "value": {"min": 500, "max": 800, "unit": "N"}}], "Constraints": [{ "tag": "NoClamp_Zone", "value": "Sidewall_Processed Surface_B"}]}}.

[0120] Among them, "FixturingRequirements": the key name is "FixturingRequirements", and the value is an object containing all requirement categories; "Locating": The key is "Locating" (location requirement), and the value is an array containing multiple semantic description objects related to location; "tag": "Loc_BaseType": The key is "Locating" (location requirement), and the value is an array containing multiple semantic description objects related to location; "tag": "Loc_BaseType": Semantic tag: Indicates the base type for location, with a value of "Loc_BaseType"; "value": "plane", which is the value of the label: indicating that the specific type of the positioning reference is "plane"; "ref": "bottom face_A", which is a reference identifier: associated with the geometric face named "bottom face_A" in the 3D model; "tag": "Loc_DOF", which is a semantic tag: representing the constrained degrees of freedom, with a value of "Loc_DOF"; "value": ["TZ", "RX", "RY"], which is the value of the label: an array of strings listing the constrained degrees of freedom (TZ: translation along the Z-axis, RX: rotation around the X-axis, RY: rotation around the Y-axis); "Clamping": The key is "Clamping" (clamping requirement), and the value is an array containing semantic description objects related to clamping; "tag": "Clamp_Direction", which is a semantic tag indicating the clamping direction, with a value of "Clamp_Direction"; "value": [0, 0, -1], which is the value of the label: a three-dimensional vector indicating that the clamping force is in the negative Z-axis direction (vertically downward); "tag": "Clamp_ForceRange", which is a semantic tag: indicating the clamping force range, with a value of "Clamp_ForceRange"; "value": {"min": 500, "max": 800, "unit": "N"}: The value of the tag is an object describing the specific range of clamping force. The lower limit of clamping force is 500 (Newtons), the upper limit of clamping force is 800 (Newtons), and the unit of force is N (Newtons). "Constraints": The key name is "Constraints" (process constraints), and the value is an array containing semantic description objects related to the constraints; "tag": "NoClamp_Zone": Semantic tag: Indicates a region where clamping is prohibited, with a value of "NoClamp_Zone"; "value": "Sidewall_MachinedSurface_B", that is, the value of the label: refers to the geometric surface named "Sidewall_MachinedSurface_B" in the 3D model, indicating that clamping is prohibited in this area.

[0121] By employing a unified semantic tagging system and parametric description to construct a semantic model of clamping requirements, this embodiment transforms the traditionally implicit, vague design intent, which relies on natural language or drawings, into a precise, structured, unambiguous, and machine-interpretable standardized data format. This not only provides clear and unambiguous design goals for artificial intelligence models, improving the accuracy and reliability of intelligently generated results, but also constitutes a reusable, exchangeable, and iterative digital design asset. This lays the core technological foundation for the accumulation and sharing of clamping design knowledge and the seamless integration between different design automation systems, representing a crucial step in propelling clamping design from individual experience and skill to the software-based transformation of industrial knowledge.

[0122] Optionally, the fixture structure constraint rule system includes at least one of the following: positioning structure constraints for limiting the number and arrangement of positioning elements, clamping constraints for limiting the direction and range of clamping force, spatial interference constraints for preventing interference with parts, cutting tools or machine tools, and manufacturing and assembly constraints for satisfying manufacturability and assemblability.

[0123] Specifically, the positioning structure constraint: This constraint originates from the six-point positioning principle, and its core is to ensure that the workpiece has a definite and unique position in space, while avoiding over-positioning (interference or deformation caused by repeatedly constraining the same degree of freedom) or under-positioning (workpiece movement caused by insufficient constraints). It specifies the basic conditions that the number of components and spatial layout must meet when implementing different positioning methods.

[0124] For example, for planar positioning, the rule might be that at least three non-collinear support points must be used to constrain the three translational degrees of freedom (TX, TY, TZ).

[0125] For V-block positioning of an outer cylinder: the rules may be that a long V-block constrains four degrees of freedom (TY, TZ, RY, RZ), and a short V-block constrains two degrees of freedom (TY, TZ).

[0126] It can also be configured using rules. For example, for thin-walled parts with poor rigidity, the system can enable "deformation-resistant positioning constraints" to optimize "at least three support points" into "multiple evenly distributed support points to reduce local pressure".

[0127] Clamping constraint: This constraint ensures that the clamping force is effective, reliable, and safe. It is based on the principles of mechanics and tribology, and specifies the direction, point of application, magnitude, and sequence of the clamping force.

[0128] For example, directional constraints: "The direction of the main clamping force component should be as perpendicular as possible to the main positioning surface and point towards the positioning support point" to prevent the workpiece from being lifted or slipping under the action of cutting force.

[0129] Force range constraint: Calculate and limit the "upper limit of clamping force (to prevent workpiece deformation or crushing)" and "lower limit of clamping force (to prevent workpiece slippage)" based on the workpiece material, clamping contact area and required friction coefficient. For example, "the clamping force must be between 400N and 1200N".

[0130] For different clamping mechanisms (such as screw, eccentric, hydraulic), the rule base can include corresponding self-locking constraints (such as the eccentric wheel helix angle must be less than the friction angle) or force amplification factor constraints.

[0131] Spatial interference constraints are a safety guarantee, ensuring the harmonious coexistence of the fixture and the entire machining environment through geometric reasoning. It defines a series of "no-fly zones".

[0132] For example: Tool interference constraint: The system automatically generates the rule "No fixture element may intrude into the space of the envelope offset outward by X millimeters (safety margin)" based on the machining avoidance area (tool motion envelope) extracted from S200.

[0133] Machine tool interference constraints: Based on the machine tool model (such as spindle head, column, and protective door) in the processing equipment information, generate the following rules: "The total height of the fixture shall not exceed the minimum distance from the end face of the machine tool spindle to the worktable" or "The fixture profile shall not intersect with the movement trajectory of the machine tool protective door".

[0134] Workpiece loading and unloading interference constraints: The rule may be "The clamping mechanism (such as a pressure plate) should have sufficient opening stroke or turning radius so that the workpiece can be put in and taken out without obstruction".

[0135] The safety margin X can be dynamically adjusted according to the machining type (roughing / finishing). The rules can also include consideration of chip removal space to prevent the fixture structure from forming chip accumulation grooves.

[0136] Manufacturing and assembly constraints: This involves bringing the design to the actual workshop, ensuring that the solution is not just "drawn," but "can be made and assembled." It integrates manufacturing processes and assembly knowledge.

[0137] For example, the standard parts priority constraint states: "Under the premise of meeting the function, clamping elements (such as bolts, pressure plates, cylinders) should be selected from the enterprise or industry standard parts library first" to reduce costs and delivery time.

[0138] Machining constraints: "Deep holes, grooves, and other structures on the fixture body should be avoided, or special machining processes (such as electrical discharge machining) should be specified."

[0139] Assembly sequence and accessibility constraints: "Sufficient wrench operating space must be left at the threaded connection" and "The pressing direction of the locating pin should have sufficient guide length and there should be a tool relief hole on the opposite side".

[0140] Constraints can be linked to a company’s specific manufacturing resource library (existing steel plate specifications, general base models) and process capability library (maximum machinable size, existing welding processes) to achieve true “resource-based design”.

[0141] By systematically constructing and applying this fixture structure constraint rule system, this embodiment transforms the tacit experience and principles—traditionally hidden in engineers' minds, fragmented, and difficult to pass on—into an explicit knowledge base that is calculable, callable, and verifiable. This fundamentally constrains and guides large models to creatively conceive within reasonable engineering boundaries, ensuring that automatically generated solutions not only meet functional requirements but are also theoretically correct, spatially safe, and manufacturing-feasible. This transforms the generative potential of artificial intelligence into robust and reliable engineering design productivity, significantly improving the first-time success rate and practical engineering value of design results.

[0142] Optionally, after generating at least one set of fixture design schemes corresponding to the tooling fixture to be designed based on the fixture structure description data, the method further includes: The fixture design scheme is constrained and verified to obtain the verification results; the constraint verification includes at least semantic compliance verification based on the clamping requirement semantic model and structural compliance verification based on the fixture structure constraint rule system. If the verification result is unsuccessful, the verification result is fed back to the large language reasoning model, which then adjusts the fixture structure description data and regenerates the fixture design scheme until a design scheme that meets the verification requirements is obtained.

[0143] Specifically, constraint verification involves two independent, rule-based automated reviews of the generated fixture design.

[0144] Semantic compliance verification: This involves reverse mapping of the design scheme to check whether it 100% meets all instructions in the semantic model of the clamping requirements. This is a form of "functional acceptance testing".

[0145] Verification Example: The semantic model requires "constraining the rotational degree of freedom (RX) about the X-axis". The system checks if the design has a valid structure (such as a stop or a side support point) to provide a constraint on this degree of freedom. If the design relies only on three points of support on the bottom surface without any lateral restraints, this verification fails and reports "RX degree of freedom constraint missing".

[0146] Structural compliance verification: This involves checking whether the design scheme violates any clause in the fixture structure constraint rules system. This is a type of "engineering compliance audit".

[0147] Verification Example 1 (Spatial Interference): The system performs a 3D Boolean operation to check whether the solid model of the fixture intersects with the machining avoidance area (tool envelope). If there is an intersection, the verification fails, and the system accurately reports the name of the interfering part and the size of the interference volume.

[0148] Verification Example 2 (Clamping Constraint): Calculate the actual force direction output by the clamping mechanism (such as a hydraulic cylinder) and check whether the angle between it and the normal of the main positioning surface exceeds the threshold set in the rule base (such as 15°). If it exceeds, report "The clamping force direction is unreasonable, which may cause the workpiece to lift."

[0149] Verification Example 3 (Manufacturing Constraints): Check whether the non-standard parts used in the design contain features that are difficult to machine (such as deep inner right-angle grooves), or whether the standard part model has been discontinued. If so, report "Manufacturability Risk".

[0150] Feedback and Iterative Optimization: When the verification fails, the system does not simply report an error, but instead initiates an intelligent iterative cycle of "diagnosis-feedback-redesign".

[0151] Specific procedures and examples: Generate diagnostic feedback: The generated structured feedback includes not only the problem description (such as "interference exists"), but also the problem location ("pressure plate A and tool path conflict within the coordinate range [X1,Y1,Z1] to [X2,Y2,Z2]") and the violated rule ID (such as rule "RC_003: spatial interference constraint").

[0152] Guided Model Adjustment: This feedback, along with the current fixture structure description data (i.e., the problematic design blueprint), is sent back to the large language inference model. The model will be prompted: "Based on the current solution, but the spatial interference problem reported by 'Rule RC_003' must be addressed." This guides the model to make targeted corrections.

[0153] Alternatively, an optimization strategy can be chosen: local fine-tuning: for simple problems, the model may only adjust a few parameters. For example, for the interference mentioned above, the model may choose to offset the installation position of "pressure plate A" by 10mm in the X direction.

[0154] For complex or fundamental conflicts, the model may undergo more significant modifications. For example, if it is found that the direction of the clamping force cannot meet the constraints, the model may refactor the "hydraulic cylinder top pressure" scheme into a "hinge link side pressure" scheme.

[0155] Multiple iterations: The process continues until the solution passes all validations or reaches the preset maximum number of iterations (at which point the optimal candidate solution can be output for manual decision-making).

[0156] By introducing an automated constraint verification and iterative optimization closed loop, this approach not only proactively identifies and quantifies potential functional deficiencies and engineering defects in generated solutions, far surpassing the efficiency and comprehensiveness of manual review, but more importantly, it achieves continuous self-improvement based on engineering feedback by feeding specific, structured verification results back to the generative model. This allows the design solution to converge to the optimal solution through iteration. This mechanism significantly reduces the trial-and-error costs and engineering risks of intelligent design results, ensuring that the final output solution is a mature and feasible solution that has undergone rigorous testing. This deeply integrates the creativity of artificial intelligence with the rigor of engineering practice, greatly enhancing the practical value and reliability of the entire method.

[0157] Optionally, after generating at least one set of fixture design schemes corresponding to the tooling fixture to be designed based on the fixture structure description data, the method further includes: The optimal fixture design is obtained by screening multiple different fixture design schemes according to the preset optimization rules.

[0158] Specifically, based on the automated generation of multiple feasible solutions, the decision-making process of senior engineers is simulated to automatically select the best overall solution or the solution that best meets the needs of a specific scenario according to the clear engineering objectives, thereby elevating design automation from "generating possibilities" to a higher level of "delivering optimal solutions".

[0159] Setting Predefined Optimization Rules: Optimization rules are a set of predefined, quantifiable evaluation indicators and decision-making logic. They transform the vague concept of a "good solution" into specific, calculable, and comparable criteria. These rules typically originate from the company's core value objectives, such as cost, efficiency, quality, and reliability.

[0160] For example, the economic rule (lowest cost): evaluation indicators may include: estimated total cost = material cost + standard parts procurement cost + estimated processing time cost. The system automatically estimates based on the BOM (Bill of Materials) and processing complexity of the solution.

[0161] Technical performance rules (highest stiffness / minimum deformation): Evaluation metrics may be the overall static stiffness value or the maximum deformation under simulated cutting forces. This requires the system to call upon simplified mechanical simulations or empirical stiffness models for rapid evaluation.

[0162] Operational efficiency rule (shortest clamping time): The evaluation index may be the estimated clamping time of a single piece, which is estimated by analyzing the number of clamping mechanisms, operation mode (manual / automatic), accessibility, etc. in the scheme.

[0163] Reliability / safety margin rule: Evaluation metrics may be the satisfaction margin of critical constraints, such as the percentage by which the actual clamping force exceeds the minimum required value, or the safety margin of the closest distance in interference checks.

[0164] Users can dynamically configure the priority or weight of rules according to the production scenario. For example: Mass production mode: weighting is configured as follows: cost: 50%, efficiency: 30%, rigidity: 20%.

[0165] High-precision machining mode: weighting is set as follows: stiffness: 50%, reliability: 30%, cost: 20%.

[0166] The rule base itself can also be expanded to include "green manufacturing" rules (such as material utilization) or "modular" rules (the proportion of common parts in existing fixtures).

[0167] For multiple candidate solutions that pass the verification, the above-mentioned optimization rules are applied in parallel to score or rate them, and finally a decision is made based on the aggregate score or Pareto optimality principle.

[0168] In some embodiments, quantitative evaluation is used: each solution receives a quantitative score for each rule. For example, if the cost estimates of three solutions are 1200 yuan, 1500 yuan, and 1000 yuan, respectively, then they will receive corresponding scores under the "economic rule" (the lower the cost, the higher the score).

[0169] Comprehensive Ranking: The system calculates a comprehensive score for each scheme according to preset weights. For example, Scheme A (low cost but average stiffness), Scheme B (high stiffness but high cost), and Scheme C (balanced). The final ranking is derived based on the comprehensive score = Σ(individual score × weight).

[0170] Decision output: Directly recommend the solution with the highest overall score.

[0171] For multi-objective optimization, the output is a "non-dominated solution set", meaning that no single solution can be better at one metric without sacrificing others. For example, the output might be {Solution C, Solution D}, with the explanation "C is cost-optimal, D is stiffness-optimal; please choose based on your primary objective".

[0172] Visualized comparison report: Generates comparison charts to clearly show the differences in indicators of each option on radar charts or bar charts, assisting in the final human decision.

[0173] By introducing a multi-solution automatic screening and optimization decision-making mechanism based on quantitative rules, intelligent design is elevated from "solving the existence problem" to a new level of "solving the quality problem." It overcomes the limitations of local optima that may exist with a single generated solution, significantly improving the overall quality and value potential of solutions by generating and comparing multiple design paths. Simultaneously, it transforms the solution selection process, which relies on personal experience, into a transparent, objective, and traceable data-driven decision-making process. This not only greatly improves decision-making efficiency and scientific rigor, enabling design outputs to accurately match diverse enterprise production goals (such as cost reduction, efficiency improvement, and quality enhancement), but the comparative data itself also becomes a valuable accumulation of domain knowledge, continuously feeding back into and optimizing the entire intelligent design system.

[0174] In some specific embodiments, taking the tooling fixture design for the side milling process of the aluminum alloy box part as an example, the implementation process of the tooling fixture scheme generation method includes: Step 1: Receive the design task and obtain complete initial clamping requirements. Specifically, this includes: 3D model of the part (3D model data): A CAD model of an aluminum alloy box with an external dimension of approximately 220×160×90 mm.

[0175] Machining process information: Perform "side fine milling" on the outer right side of the part. After machining, the flatness of this surface must be ≤0.02mm.

[0176] Machining equipment information: Use a "vertical CNC milling machine", with the spindle vertical (Z-axis) and the tool moving horizontally (along the X or Y axis).

[0177] Clamping posture requirements: The bottom surface of the part should be placed downwards, and the right side wall to be machined should be exposed to the cutting tool.

[0178] Machining stability requirement: The maximum allowable displacement of the part during machining is ≤0.01 mm.

[0179] Step Two: Analysis of Clamping Requirements Characteristics: Automatically parse the above information and extract key design features: Positioning reference features: The large flat surface of the bottom is identified from the 3D model as the main positioning reference; the two Φ12 mm circular holes on the left are identified as auxiliary positioning references.

[0180] Degrees of freedom constraint requirements: After analysis, it is determined that all translational and rotational degrees of freedom of the part need to be constrained (complete positioning).

[0181] Stress-sensitive characteristics: Analysis of the side milling process indicates that the part mainly bears horizontal lateral cutting force and may warp around the bottom edge.

[0182] No clamping area: Mark the outer wall to be milled on the right and its adjacent area as no-access areas.

[0183] Machining avoidance area: Based on the tool size and horizontal motion path, calculate the horizontal cylindrical motion envelope space formed on the right side of the part after the tool extends from the spindle, and mark it as the area that the structure must avoid.

[0184] Step 3: Semantic modeling of clamping requirements: Based on the extracted features, a structured semantic model of clamping requirements is constructed (using unified labels and parameters): Location requirement semantics: {datum: [plane: bottom surface, hole group: left hole 1, left hole 2], constrained degrees of freedom: [TX, TY, TZ, RX, RY, RZ]}; Clamping requirement semantics: {Direction: (0, 0, -1) (vertically downward), Position: top non-processing area, Force range: [500N, 800N]}; Support requirement semantics: {Location: Bottom four corner areas, Type: Rigid auxiliary support}; Process constraint semantics: {No clamping zone: right side wall, clearance space: [tool horizontal motion envelope]}; Step 4: Constructing Constraint Rules and Guiding Reasoning: 1. Construct a constraint rule system for the fixture structure. In this example, the activated rules include: Positioning structural constraints: Planar positioning must include at least three non-collinear support points.

[0185] Clamping direction constraint: The clamping force should have its main component pointing vertically downwards towards the positioning bottom surface.

[0186] Structural interference constraints: No fixture component may enter the machining avoidance area.

[0187] Manufacturing and assembly constraints: Standard pressure plates, support blocks, and locating pins should be used preferentially.

[0188] 2. Large-scale model hierarchical inference generates descriptive data: Clamping Functional Layer: Output Functional Unit Set: {Plane Positioning Unit, Lateral Positioning Unit, Top Clamping Unit, Bottom Support Unit}.

[0189] Structural topology layer: Under the guidance of rules, output topology scheme: arrange three-point support bottom surface on the base plate, install two cylindrical positioning pins at the corresponding left hole positions, design a rotatable pressure plate mechanism at the top, and set support blocks at the four corners of the base plate to support the bottom surface of the parts.

[0190] Parameters and Layout Layer: Key Output Parameters: Clamping force 700N, support block height flush with main positioning point, positioning pin diameter 12h7, pressure plate rotation center position coordinates, etc. The above summarizes the fixture structure description data.

[0191] Step 5: Fixture design scheme generation, verification, and optimization: 1. Solution Generation: Based on the description data, the system automatically calls the 3D model of the standard parts library for assembly and generates a complete 3D design solution for the fixture, including the base plate, three support nails, two positioning pins, one adjustable pressure plate, four support blocks and their assembly relationship.

[0192] 2. Constraint verification: Perform automated verification of the solution.

[0193] Semantic compliance check: The check confirms that all required degrees of freedom (such as rotation about the Y-axis) are effectively constrained by the two locating pins, and passes.

[0194] Structural compliance verification: A three-dimensional interference check was performed, and it was found that when the pressure plate arm was in the lowered position, its end slightly encroached on the horizontal clearance space reserved for the tool (i.e., interference constraint violation).

[0195] 3. Feedback Iteration: The system feeds back the verification results ("Interference between pressure plate A and tool clearance space, coordinate range: [X1…]") to the large model inference module. Based on this, the large model adjusts its description data, moving the pressure plate's installation position 15mm further back (away from the machining side). The system regenerates the solution based on the adjusted data and verifies it again. All verifications pass this time, and the output is the final feasible fixture design.

[0196] like Figure 3 As shown, an embodiment of the present invention provides a tooling fixture solution generation system, comprising: The acquisition unit is used to acquire the initial clamping requirement information corresponding to the tooling fixture to be designed; the initial clamping requirement information includes at least the three-dimensional model data of the part to be processed or assembled, the processing procedure information, the processing equipment information, the clamping posture requirements, and the processing accuracy and stability requirements; The extraction unit is used to parse and process the initial clamping requirement information and extract key feature information for fixture design. The construction unit is used to construct a structured clamping requirement semantic model based on the key feature information; construct a fixture structure constraint rule system; and generate and reason through a large language reasoning model according to the clamping requirement semantic model and the fixture structure constraint rule system to obtain fixture structure description data; the fixture structure description data includes fixture structure topology, functional unit configuration and key parameters. The processing unit is used to generate at least one set of fixture design schemes corresponding to the fixture to be designed based on the fixture structure description data; wherein, the fixture design scheme includes the overall structural layout of the fixture, the configuration of positioning elements, the form of clamping mechanism, the setting of support structure and the assembly relationship between various structural elements.

[0197] An embodiment of the present invention provides a tooling fixture solution generation device, comprising a memory and a processor; the memory is used to store a computer program; the processor is used to implement the tooling fixture solution generation method as described above when the computer program is executed.

[0198] This invention provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the tooling fixture scheme generation method described above.

[0199] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A method for generating a tooling fixture scheme, characterized in that, include: Obtain the initial clamping requirements information corresponding to the tooling fixture to be designed; The initial clamping requirements information includes at least the three-dimensional model data of the part to be processed or assembled, the processing procedure information, the processing equipment information, the clamping posture requirements, and the processing accuracy and stability requirements; The initial clamping requirements information is parsed and processed to extract key feature information for fixture design; Based on the aforementioned key feature information, a structured semantic model of clamping requirements is constructed. A fixture structure constraint rule system is constructed, and based on the clamping requirement semantic model and the fixture structure constraint rule system, a large language reasoning model is used to generate reasoning to obtain fixture structure description data; the fixture structure description data includes fixture structure topology, functional unit configuration and key parameters; Based on the fixture structure description data, at least one set of fixture design schemes corresponding to the tooling fixture to be designed is generated; The fixture design scheme includes the overall structural layout of the fixture, the configuration of positioning elements, the form of the clamping mechanism, the setting of the support structure, and the assembly relationship between the various structural elements.

2. The method for generating tooling and fixture solutions according to claim 1, characterized in that, The large language reasoning model includes a clamping function reasoning layer, a structural topology reasoning layer, and a parameter and layout reasoning layer; the clamping structure description data is obtained by generating reasoning through the large language reasoning model based on the clamping requirement semantic model and the clamping structure constraint rule system, including: Based on the clamping function inference layer, the set of functional units is determined according to the clamping requirement semantic model; Based on the structural topology reasoning layer, the structural topology scheme is determined according to the set of functional units and the fixture structure constraint rule system; Based on the parameters and layout inference layer, the fixture structure description data is determined according to the structural topology scheme.

3. The method for generating tooling and fixture solutions according to claim 1, characterized in that, The key feature information includes at least: positioning reference features identified from the three-dimensional model data, freedom constraint-related features and force-sensitive features analyzed from the processing procedure information, and prohibited clamping areas and processing avoidance areas determined based on the processing procedure information and the processing equipment information.

4. The method for generating a tooling fixture scheme according to claim 1, characterized in that, The clamping requirement semantic model is represented by a unified semantic tag system and a parameterized description method; wherein, the semantic tag system includes positioning requirement semantic tags for describing the positioning reference type and constraint degrees of freedom, clamping requirement semantic tags for describing the clamping direction and force range, and support requirement semantic tags for describing the support position and load-bearing requirements.

5. The method for generating a tooling fixture scheme according to claim 1, characterized in that, The fixture structure constraint rule system includes at least one of the following: positioning structure constraints for limiting the number and arrangement of positioning elements, clamping constraints for limiting the direction and range of clamping force, spatial interference constraints for preventing interference with parts, cutting tools or machine tools, and manufacturing and assembly constraints for satisfying manufacturability and assemblability.

6. The method for generating a tooling fixture scheme according to claim 1, characterized in that, After generating at least one set of fixture design schemes corresponding to the tooling fixture to be designed based on the fixture structure description data, the process further includes: The fixture design scheme is constrained and verified to obtain the verification results; the constraint verification includes at least semantic compliance verification based on the clamping requirement semantic model and structural compliance verification based on the fixture structure constraint rule system. If the verification result is unsuccessful, the verification result is fed back to the large language reasoning model, which then adjusts the fixture structure description data and regenerates the fixture design scheme until a design scheme that meets the verification requirements is obtained.

7. The method for generating a tooling fixture scheme according to claim 1, characterized in that, After generating at least one set of fixture design schemes corresponding to the tooling fixture to be designed based on the fixture structure description data, the process further includes: The optimal fixture design is obtained by screening multiple different fixture design schemes according to the preset optimization rules.

8. A tooling fixture solution generation system, characterized in that, include: The acquisition unit is used to acquire the initial clamping requirements information corresponding to the tooling fixture to be designed. The initial clamping requirements information includes at least the three-dimensional model data of the part to be processed or assembled, the processing procedure information, the processing equipment information, the clamping posture requirements, and the processing accuracy and stability requirements; The extraction unit is used to parse and process the initial clamping requirement information and extract key feature information for fixture design. The construction unit is used to construct a structured clamping requirement semantic model based on the key feature information; construct a fixture structure constraint rule system; and generate and reason through a large language reasoning model according to the clamping requirement semantic model and the fixture structure constraint rule system to obtain fixture structure description data; the fixture structure description data includes fixture structure topology, functional unit configuration and key parameters. The processing unit is used to generate at least one set of fixture design schemes corresponding to the fixture to be designed based on the fixture structure description data; wherein, the fixture design scheme includes the overall structural layout of the fixture, the configuration of positioning elements, the form of clamping mechanism, the setting of support structure and the assembly relationship between various structural elements.

9. A tooling fixture generation device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for generating tooling fixture schemes as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method for generating the tooling fixture scheme as described in any one of claims 1 to 7.

Images