An AI technology-based component intelligent optimization design system and method
The AI-based intelligent optimization design system for parts has automated the process from user requirements to 2D design drawings, solving the problem of low efficiency in traditional design, improving design quality and innovation capabilities, expanding the design domain, and reducing costs and error rates.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- AECC COMML AIRCRAFT ENGINE CO LTD
- Filing Date
- 2025-01-08
- Publication Date
- 2026-07-10
AI Technical Summary
Traditional component design and optimization processes are inefficient, consume a lot of manpower and computing power, limit innovative design capabilities, and rely on proxy models built based on shape parameters with a narrow scope of application, making it impossible to perform extensive design optimization.
The AI-based intelligent optimization design system for parts includes an AI intelligent requirement processing module, a modeling module, an evaluation module, an optimization module, and a drafting module. Through generative AI algorithms and deep learning principles, it automates the process from user requirements to two-dimensional design drawings, and performs multidisciplinary performance evaluation and topology optimization.
It improved the efficiency of component design, expanded the configuration design domain, solved the computing power problem in high-dimensional design space, realized a high degree of automation in the design process, enhanced innovative design capabilities and overall design quality, and reduced development costs and error rates.
Smart Images
Figure CN122365796A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of component design optimization technology, specifically to an intelligent component optimization design system and method based on AI technology. Background Technology
[0002] The structural design and optimization of components are crucial aspects of all engineering fields, and also represent a core capability reflecting an engineer's independent R&D level. With continuous technological advancements, the demands on component structural design and optimization capabilities are constantly increasing. In the component structural design process, an initial structural scheme is first obtained through design calculations based on theoretical principles. Then, finite element analysis software is used to model the initial structure and conduct structural strength analysis to verify whether the structure meets the requirements. Based on this scheme, processing drawings are output, leading to the production of the physical component. In fact, a design that merely meets strength requirements is not the only solution. Even if the initial structural scheme meets the requirements, optimization design can still be carried out from many aspects to improve component performance while still meeting the requirements, resulting in a more suitable and economical design scheme. This requires a significant investment of manpower and computing power for optimization design.
[0003] Traditionally, the design and optimization process for components involves first determining the structure of the component through design calculations, then using extensive CAD / CAE co-simulation to optimize specific details and dimensions, and finally generating structural drawings for factory processing. This entire process requires a large amount of manpower and computing power for optimization design, resulting in low efficiency, high initial design costs, and limiting innovative design capabilities.
[0004] With the maturation of artificial intelligence (AI) technology, many fields are beginning to integrate AI to replace repetitive manual labor, allowing engineers to focus more on creative work. For example, the implementation of automatic text generation and code writing functions, even just providing initial outlines, has already significantly accelerated the development process. The efficiency of AI technology extends far beyond this; intelligentization is a key focus for many scholars today. How to apply AI technology to engineering design to improve development speed, reduce costs, and ultimately enhance innovative design capabilities has become an important research direction for engineers. Summary of the Invention
[0005] The following provides a brief overview of one or more aspects to offer a basic understanding of them. This overview is not an exhaustive summary of all conceived aspects, nor is it intended to identify key or decisive elements of all aspects, nor to define the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form to prepare for the more detailed descriptions that follow.
[0006] The purpose of this invention is to solve the above-mentioned problems and provide an intelligent optimization design system and method for parts based on AI technology. This system can generate a large number of innovative configurations at the beginning of the design process, accelerate iteration using existing data during simulation analysis, use intelligent optimization to search for optimization schemes during design optimization, and use artificial intelligence to generate engineering drawings based on existing experience. This will speed up the development process, reduce development costs, liberate productivity, and enhance innovative design capabilities.
[0007] The technical solution of this invention is as follows:
[0008] This invention provides an AI-based intelligent optimization design system for components, comprising: an AI intelligent requirements processing module, an AI intelligent modeling module, an AI intelligent evaluation module, an AI intelligent optimization module, and an AI intelligent drafting module; wherein,
[0009] The AI intelligent demand processing module is used to analyze user demand data, generate initial three-dimensional geometric models of parts, and transmit the generated initial three-dimensional geometric models to the AI intelligent modeling module.
[0010] The AI intelligent modeling module is used to model based on the graph structure data of the initial 3D geometric model, generate multiple similar candidate geometric model schemes, and transmit the generated candidate geometric model schemes to the AI intelligent evaluation module.
[0011] The AI intelligent evaluation module is used to evaluate the performance of the components corresponding to each candidate geometric model scheme according to the multidisciplinary learning mechanism, generate the corresponding 3D model multidisciplinary performance evaluation data, and transmit the generated 3D model multidisciplinary evaluation data to the AI intelligent optimization module.
[0012] The AI intelligent optimization module is used to perform multidisciplinary topology optimization on the multidisciplinary performance evaluation data of the 3D model corresponding to each candidate geometric model scheme, generate an optimized geometric model scheme, and transmit the generated optimized geometric model scheme to the AI intelligent mapping module.
[0013] The AI intelligent drafting module is used to intelligently draft based on optimized geometric model schemes, thereby generating two-dimensional design drawings of parts that can be used for machining.
[0014] According to an embodiment of the AI-based intelligent optimization design system for parts according to the present invention, before the AI intelligent demand processing module performs user demand data analysis, it first collects a large number of 3D models of parts and corresponding modeling command code samples and text description samples. Then, based on the collected modeling command code samples and text description samples of the 3D models of parts, it establishes and trains the AI demand processing model, thereby obtaining an AI demand processing model for user demand data analysis. The input of the AI demand processing model is natural language text, and the output is code commands that can be executed by 3D modeling software.
[0015] According to an embodiment of the AI-based intelligent optimization design system for parts according to the present invention, when the AI demand processing model analyzes user demand data, it uses a generative AI algorithm to convert the input natural language text into code commands that can be executed by the 3D modeling software, and then transmits the generated code commands to the 3D modeling software for modeling, thereby generating an initial 3D geometric model.
[0016] According to an embodiment of the AI-based intelligent optimization design system for components of the present invention, before the AI intelligent modeling module models the initial three-dimensional geometric model, it first collects a large number of historical aircraft drawings and model samples of aero engines. Then, it uses deep learning principles to establish and train the AI intelligent modeling module based on the collected historical aircraft drawings and model sample data of aero engines, thereby obtaining the AI intelligent modeling module for establishing geometric model schemes. The input of the AI intelligent modeling module is the graph structure data of the initial three-dimensional geometric model, and the output is multiple similar candidate geometric model schemes with different features of the components.
[0017] According to an embodiment of the AI-based intelligent optimization design system for components of the present invention, before evaluating each candidate geometric model scheme, the AI intelligent evaluation module first collects a large number of structural space samples of aero-engines and obtains corresponding multidisciplinary data based on the structural space samples. Then, it uses graph deep learning principles and self-attention mechanisms to establish and train the AI intelligent evaluation model, thereby obtaining an AI intelligent evaluation model for multidisciplinary performance evaluation. The input of the AI intelligent evaluation model is similar candidate geometric model schemes with component features, and the output is the corresponding three-dimensional model multidisciplinary performance evaluation data.
[0018] According to an embodiment of the AI-based intelligent optimization design system for components of the present invention, before performing multidisciplinary topology optimization on multidisciplinary performance evaluation data, the AI intelligent optimization module first collects the corresponding key features and parametric geometric models of the three-dimensional geometric model based on the existing three-dimensional geometric model. Then, using deep learning principles and generative adversarial networks, the AI intelligent optimization model is established and trained based on the collected key features and parametric geometric models of the three-dimensional geometric model, thereby obtaining an AI intelligent optimization model for multidisciplinary topology optimization. The input of the AI intelligent optimization model is the multidisciplinary performance evaluation data of the three-dimensional model, and the output is the optimized geometric model scheme after topology optimization.
[0019] According to an embodiment of the AI-based intelligent optimization design system for parts according to the present invention, when the AI intelligent optimization model performs multidisciplinary topology optimization, it first extracts and fits the key features of the three-dimensional geometric model based on the input three-dimensional model multidisciplinary performance evaluation data to generate a three-dimensional parametric geometric model. Then, it evaluates the three-dimensional parametric geometric model by combining the corresponding multidisciplinary data in the AI intelligent evaluation module, and iteratively improves and optimizes the topology of the three-dimensional parametric geometric model based on the evaluation results, and finally generates an optimized geometric model scheme that meets the comprehensive multidisciplinary performance optimization index.
[0020] According to an embodiment of the AI-based intelligent optimization design system for parts according to the present invention, before performing intelligent drawing, the AI intelligent drawing module first collects a large number of historical two-dimensional machining drawing samples of parts, and then establishes and trains an AI intelligent drawing model based on the principle of deep learning and the collected historical two-dimensional machining drawing samples of parts, thereby obtaining an AI intelligent drawing model that automatically draws two-dimensional design drawings of three-dimensional geometric models; wherein, the input of the AI intelligent drawing model is the optimized geometric model scheme after topology optimization, and the output is the two-dimensional design drawings of parts.
[0021] This invention also provides an intelligent optimization design method for components based on AI technology, comprising the following steps:
[0022] Step S1: Analyze user demand data using a pre-trained AI demand processing model to generate an initial three-dimensional geometric model of the parts.
[0023] Step S2: Based on the graph structure data of the initial 3D geometric model of a pre-trained AI intelligent modeling model, model multiple similar candidate geometric model schemes are generated.
[0024] Step S3: Utilize a pre-trained AI intelligent evaluation model to evaluate the performance of the components corresponding to each candidate geometric model scheme based on a multidisciplinary learning mechanism, and generate corresponding 3D model multidisciplinary performance evaluation data.
[0025] Step S4: Use a pre-trained AI intelligent optimization model to perform multidisciplinary topology optimization on the multidisciplinary performance evaluation data of the 3D models corresponding to each candidate geometric model scheme, and generate an optimized geometric model scheme.
[0026] Step S5: Use a pre-trained AI intelligent drawing model to intelligently draw the optimized geometric model scheme, thereby generating two-dimensional design drawings of parts that can be used for processing.
[0027] The present invention also provides a computer-readable medium storing computer program code that, when executed by a processor, implements the method described above.
[0028] The present invention also provides an intelligent optimization design device for parts based on AI technology, comprising:
[0029] Memory, used to store instructions that can be executed by a processor; and
[0030] A processor for executing the instructions to implement the method described above.
[0031] Compared with existing technologies, this invention offers the following advantages: Targeting the field of engineering design, it proposes an AI-based intelligent optimization design system and method for components, improving the design efficiency of typical aero-engine components and expanding the configuration design domain. Simultaneously, this invention overcomes the limitation of narrow applicability of surrogate models built based on shape parameters. It does not rely on geometric parameters to define the design, but rather uses the shape of the design itself as input for broader design optimization exploration. This solves the computational power problem of conventional refined CAE simulation in high-dimensional design spaces, releasing the potential for multidisciplinary design optimization. For needs requiring massive computation, it can significantly improve the efficiency of model performance prediction. Furthermore, this invention creates an automated process from topology optimization to parametric geometric models, achieving a high degree of automation in the design process. The optimization results can achieve a balance among the needs of various disciplines, improving the overall design quality. Through this invention, development speed can be accelerated, development costs reduced, the error rate of drafting design lowered, and the completeness of drafting design improved. This not only liberates productivity but also enhances innovative design capabilities, thereby accelerating intelligent optimization design of components, improving the innovative design capabilities of component structures, and raising the level of forward design for major equipment in my country. Attached Figure Description
[0032] The above-described features and advantages of the present invention will be better understood after reading the following detailed description of embodiments of the present disclosure in conjunction with the accompanying drawings. In the drawings, components are not necessarily drawn to scale, and components having similar related characteristics or features may have the same or similar reference numerals.
[0033] Figure 1 This is a system architecture diagram illustrating an embodiment of the AI-based intelligent optimization design system for components according to the present invention.
[0034] Figure 2 This is a flowchart illustrating the steps of an embodiment of the AI-based intelligent optimization design method for components according to the present invention. Detailed Implementation
[0035] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this application. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.
[0036] As indicated in this application and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" are not specifically singular and may include plural forms. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of explicitly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.
[0037] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values of the components and steps described in these embodiments do not limit the scope of this application. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following drawings denote similar items; therefore, once an item is defined in one drawing, it need not be further discussed in subsequent drawings.
[0038] In detailing the embodiments of the present invention, for ease of explanation, the cross-sectional views illustrating the device structure will be partially enlarged and not to scale. Furthermore, the schematic diagrams are merely examples and should not limit the scope of protection of the present invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.
[0039] In the description of this application, it should be understood that the orientation or positional relationship indicated by directional terms such as "front, back, up, down, left, right", "horizontal, vertical, horizontal" and "top, bottom" is usually based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing this application and simplifying the description. Unless otherwise stated, these directional terms do not indicate or imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on the scope of protection of this application; the directional terms "inner" and "outer" refer to the inner and outer contours relative to the outline of each component itself.
[0040] This document discloses an embodiment of an AI-based intelligent optimization design system for parts. Figure 1 This is a system architecture diagram illustrating an embodiment of the AI-based intelligent optimization design system for components according to the present invention. Figure 1 As shown in this embodiment, the AI-based intelligent optimization design system for parts includes: an AI intelligent requirement processing module, an AI intelligent modeling module, an AI intelligent evaluation module, an AI intelligent optimization module, and an AI intelligent drawing module. The AI intelligent requirement processing module analyzes user requirement data, generates an initial 3D geometric model of the parts, and transmits the generated initial 3D geometric model to the AI intelligent modeling module. The AI intelligent modeling module models based on the graph structure data of the initial 3D geometric model, generates multiple similar candidate geometric model schemes, and transmits the generated candidate geometric model schemes to the AI intelligent evaluation module. The AI intelligent evaluation module evaluates the performance of the parts corresponding to each candidate geometric model scheme using a multidisciplinary learning mechanism, generates corresponding 3D model multidisciplinary performance evaluation data, and transmits the generated 3D model multidisciplinary evaluation data to the AI intelligent optimization module. The AI intelligent optimization module performs multidisciplinary topology optimization on the 3D model multidisciplinary performance evaluation data corresponding to each candidate geometric model scheme, generates an optimized geometric model scheme, and transmits the generated optimized geometric model scheme to the AI intelligent drawing module. The AI intelligent drawing module performs intelligent drawing based on the optimized geometric model scheme, thereby generating 2D design drawings of the parts that can be processed.
[0041] Furthermore, in this embodiment, before the AI intelligent demand processing module analyzes user demand data, it first collects a large number of 3D models of parts and corresponding modeling command code samples and text description samples. Then, based on the collected modeling command code samples and text description samples of the 3D models of parts, it establishes and trains the AI demand processing model, thereby obtaining an AI demand processing model for user demand data analysis. The input of the AI demand processing model is natural language text, and the output is code commands executable by the 3D modeling software. When analyzing user demand data, the AI demand processing model uses a generative AI algorithm to parse the input natural language text, i.e., the obtained user demand data, converting it into code commands executable by the 3D modeling software. The generated code commands are then transmitted to the 3D modeling software for modeling, thereby generating the initial 3D geometric model.
[0042] Specifically, in this embodiment, after collecting a large number of 3D models and corresponding modeling command code samples and text description samples, the data is preprocessed, including data cleaning, data transformation, data normalization, and standardization. The 3D geometric model targeted by the AI requirement processing model should ideally include all components of the aero-engine. Based on different modeling steps and methods, the collected data is divided into training and validation sets. Then, the AI requirement processing model is established using semantic recognition technology, data analysis, and data mining principles and methods, and the collected samples are used for AI model training. After training and validation, the AI requirement processing model can process natural language and generate executable code commands for 3D modeling software. Finally, based on the trained AI requirement processing model, the parameters and constraints of the engineering components to be modeled are defined. Users can describe specific requirements and design elements within the provided input interface. The input can be keywords, a description, or a model sample. Generative AI algorithms process the input natural language, converting it into executable code commands for 3D modeling software, thereby generating the initial 3D geometric model of the component.
[0043] Furthermore, in this embodiment, before the AI intelligent modeling module models the initial 3D geometric model, it first collects a large number of historical aircraft model drawings and model samples of aero engines. Then, using deep learning principles, the AI intelligent modeling module is established and trained based on the collected historical aircraft model drawings and model sample data, thereby obtaining the AI intelligent modeling module for establishing geometric model schemes. The input of the AI intelligent modeling module is the graph structure data of the initial 3D geometric model, and the output is multiple similar candidate geometric model schemes with different features of the components. After generating the initial 3D geometric model of the components, large-scale AI intelligent modeling is performed using the AI intelligent modeling module. Through graph representation learning and training on historical component drawings and models, a large number of candidate geometric model schemes are quickly generated for the initial 3D geometric model using a deep learning framework.
[0044] Specifically, in this embodiment, after collecting a large number of historical aircraft engine drawings and model samples, data preprocessing is performed, and the collected data is divided into training and validation sets. Then, based on generative AI technology, an AI intelligent modeling model is established and trained through deep learning, and graph representation learning is carried out. After training and validation, the resulting AI intelligent modeling model can quickly generate a large number of candidate geometric model schemes. Then, combined with the initial 3D geometric model of the component generated by the AI requirement processing model, the graph data structure of the initial 3D geometric model is analyzed through generative AI technology, thereby quickly generating a large number of similar component candidate geometric model schemes with different characteristics, expanding the component configuration design domain for subsequent optimization design.
[0045] Furthermore, in this embodiment, before evaluating each candidate geometric model scheme, the AI intelligent evaluation module first collects a large number of structural space samples of aero-engines and obtains corresponding multidisciplinary data based on the structural space samples. Then, it uses graph deep learning principles and self-attention mechanisms to establish and train the AI intelligent evaluation model, thereby obtaining an AI intelligent evaluation model for multidisciplinary performance evaluation. The input of the AI intelligent evaluation model is similar candidate geometric model schemes with component features, and the output is the corresponding 3D model multidisciplinary performance evaluation data. After generating a large number of similar component candidate geometric model schemes with different features, the AI intelligent evaluation model performs rapid multidisciplinary evaluation on all generated components. Combined with multidisciplinary data, graph deep learning is performed, thereby utilizing a graph neural network framework and self-attention mechanism to complete the rapid evaluation of component performance.
[0046] Specifically, in this embodiment, after collecting a large number of structural space samples, data preprocessing is first carried out, including data cleaning and structuring, etc., to remove irrelevant or redundant information. After dividing the collected data into training and validation sets, the ubiquitous structural space samples are used to acquire multidisciplinary data sources such as aerodynamics, heat transfer, strength, and acoustics. Graph deep learning is then conducted, employing the transformer principle, and using a graph neural network framework and self-attention mechanism to establish and train an AI intelligent evaluation model. The trained and validated AI intelligent evaluation model can achieve rapid evaluation of component performance. Based on candidate geometric model schemes generated by the AI intelligent modeling module, generative AI technology is used to rapidly evaluate the model's multidisciplinary performance, thereby forming multidisciplinary performance evaluation data for the three-dimensional model of the components.
[0047] Furthermore, in this embodiment, before performing multidisciplinary topology optimization on the multidisciplinary performance evaluation data, the AI intelligent optimization module first collects the corresponding key features and parametric geometric models of the existing 3D geometric models. Then, using deep learning principles and generative adversarial networks, it establishes and trains an AI intelligent optimization model based on the collected key features and parametric geometric models of the 3D geometric models, thereby obtaining an AI intelligent optimization model for multidisciplinary topology optimization. The input of the AI intelligent optimization model is the multidisciplinary performance evaluation data of the 3D model, and the output is the optimized geometric model scheme after topology optimization. Multidisciplinary topology optimization is carried out using the multidisciplinary evaluation data results obtained through the AI intelligent optimization model. The AI intelligent optimization model is used to perform deep learning on a large number of historical topology optimization schemes. After training, for the existing 3D geometric models, combined with multidisciplinary performance indicators, AI technology is used to realize the conversion from topological structure to parametric model, thereby completing the automated process of topology-shape parameter co-optimization and outputting the optimized geometric model scheme.
[0048] Specifically, in this embodiment, deep learning is first conducted based on the multidisciplinary data obtained in the AI intelligent evaluation module for training the AI model, combined with the corresponding topology scheme. After initial training and verification, the AI intelligent optimization model can identify and extract key features of the fitted 3D geometric model, completing the reconstruction of the geometric model, i.e., automatically generating a 3D parametric geometric model. Then, based on the generated parametric geometric model, the multidisciplinary data in the AI intelligent evaluation module is used to perform another multidisciplinary rapid evaluation of the topology model. Through the principle of adversarial networks, the evaluation result is used as feedback to continuously improve and optimize the 3D geometric model of the components, forming an automated process from topology optimization to parametric geometric model reconstruction, ultimately generating an optimized geometric model scheme that meets comprehensive multidisciplinary performance optimization indicators.
[0049] Furthermore, in this embodiment, before performing intelligent drawing, the AI intelligent drawing module first collects a large number of historical 2D machining drawings of parts. Then, based on deep learning principles and the collected historical 2D machining drawings, it establishes and trains an AI intelligent drawing model to automatically draw 2D design drawings of 3D geometric models. The input to the AI intelligent drawing model is an optimized geometric model scheme after topology optimization, and the output is the 2D design drawings of the parts. Intelligent drawing of the topology-optimized 3D geometric model is performed using the AI intelligent drawing model to learn graphical representations from historical part drawings and models. After training, for the optimized geometric model scheme, rapid PMI intelligent drawing is performed through a deep learning framework to output the final 2D design drawings of the parts that can be used for machining.
[0050] Specifically, in this embodiment, after collecting the design and manufacturing requirements of engineering components, and considering factors such as manufacturing and inspection, an engineering drawing atlas is established based on the drafting experience of engineering designers. These drawings should include all elements related to processing, such as dimensioning, tolerance markings, and manufacturing requirements. Then, a large number of historical two-dimensional processing drawing samples of components are collected, and the data is preprocessed, including data cleaning, data transformation, data normalization, and standardization. Then, based on text recognition technology, data analysis, and data mining principles and methods, an AI intelligent drawing model is established through deep learning. The collected samples and the engineering drawing atlas are used to train the AI model. After training and verification, the resulting AI intelligent drawing model can achieve PMI intelligent drawing and generate design drawings for processing. Therefore, based on the optimized geometric model scheme after topology optimization, rapid PMI intelligent drawing can be performed through generative AI technology, ultimately generating two-dimensional design drawings of components that can be used for processing.
[0051] An embodiment of an AI-based intelligent optimization design method for components is also disclosed herein. Figure 2 This is a flowchart illustrating an embodiment of the AI-based intelligent optimization design method for components according to the present invention. Please refer to... Figure 2 The following is a detailed explanation of each step in the AI-based intelligent optimization design method for components.
[0052] Step S1: Analyze user demand data using a pre-trained AI demand processing model to generate an initial three-dimensional geometric model of the parts.
[0053] Step S2: Based on the graph structure data of the initial 3D geometric model of a pre-trained AI intelligent modeling model, model multiple similar candidate geometric model schemes are generated.
[0054] Step S3: Utilize a pre-trained AI intelligent evaluation model to evaluate the performance of the components corresponding to each candidate geometric model scheme based on a multidisciplinary learning mechanism, and generate corresponding 3D model multidisciplinary performance evaluation data.
[0055] Step S4: Use a pre-trained AI intelligent optimization model to perform multidisciplinary topology optimization on the multidisciplinary performance evaluation data of the 3D models corresponding to each candidate geometric model scheme, and generate an optimized geometric model scheme.
[0056] Step S5: Use a pre-trained AI intelligent drawing model to intelligently draw the optimized geometric model scheme, thereby generating two-dimensional design drawings of parts that can be used for processing.
[0057] Therefore, through the above steps, an AI-powered demand processing model can extract key structural features from user requirements, and an AI-powered intelligent modeling model based on graph representation learning and generative AI technology can rapidly generate a large number of candidate innovative design solutions. Then, using an AI-powered intelligent evaluation model that integrates multidisciplinary data sources, based on aerodynamics, heat transfer, strength, acoustics, and other multidisciplinary data sources from ubiquitous structural space samples, it can quickly and accurately predict the multidisciplinary and multiphysical fields of components. Furthermore, an AI-powered intelligent optimization model is used for boundary feature identification and fitting, and geometric model reconstruction. Through efficient topology optimization and intelligent geometric feature identification / fitting, an automated process from topology optimization to parametric geometric model reconstruction is achieved, providing a new method for innovative optimization design of components. Finally, for the structural design and drawing of typical components, especially those with high design tolerance accuracy requirements and high design-manufacturing collaboration, an AI-powered intelligent drawing model based on component structural features and reinforcement learning, considering factors such as processing, manufacturing, and inspection, uses the PMI intelligent drawing method to generate two-dimensional design drawings of components suitable for processing, thereby shortening the design cycle and reducing human error.
[0058] This specification also provides a computer-readable medium storing computer program code that, when executed by a processor, implements the AI-based intelligent optimization design method for components as described above.
[0059] This specification also provides an AI-based intelligent optimization design method for components, including a memory storing instructions executable by a processor, and a processor for executing instructions in the instruction memory to implement the AI-based intelligent optimization design method for components as described above.
[0060] The prior description of this disclosure is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to this disclosure will be apparent to those skilled in the art, and the general principles defined herein may be applied to other variations without departing from the spirit or scope of this disclosure. Therefore, this disclosure is not intended to be limited to the examples and designs described herein, but should be accorded the widest scope consistent with the principles and novel features disclosed herein.
[0061] Those skilled in the art will further appreciate that the various illustrative logic blocks, modules, circuits, and algorithm steps described in conjunction with the embodiments disclosed herein can be implemented as electronic hardware, computer software, or a combination of both. To clearly illustrate this interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps are described above in a generalized manner in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system. Those skilled in the art may implement the described functionality in different ways for each specific application, but such implementation decisions should not be construed as departing from the scope of the invention.
[0062] The various illustrative logic blocks, modules, and circuits described in conjunction with the embodiments disclosed herein can be implemented or performed using a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. The general-purpose processor may be a microprocessor, but in alternatives, it may be any conventional processor, controller, microcontroller, or state machine. The processor may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors cooperating with a DSP core, or any other such configuration.
[0063] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be embodied directly in hardware, in a software module executed by a processor, or in a combination of both. The software module may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor such that the processor can read and write information to / from the storage medium. In an alternative, the storage medium may be integrated into the processor. The processor and storage medium may reside in an ASIC. The ASIC may reside in a user terminal. In an alternative, the processor and storage medium may reside as discrete components in the user terminal.
[0064] In one or more exemplary embodiments, the described functionality may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software as a computer program product, the functionality may be stored or transmitted as one or more instructions or code on or through a computer-readable medium. A computer-readable medium includes both computer storage media and communication media, encompassing any medium that facilitates the transfer of a computer program from one location to another. A storage medium may be any available medium accessible to a computer. By way of example and not limitation, such a computer-readable medium may include RAM, ROM, EEPROM, CD-ROM or other optical disc storage, disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and is accessible to a computer. Any connection is also legitimately referred to as a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of a medium. As used in this article, disk and disc include compact discs (CDs), laser discs, optical discs, digital multi-purpose discs (DVDs), floppy disks, and Blu-ray discs. Disks typically reproduce data magnetically, while discs reproduce data optically using lasers. Combinations of these should also be included within the scope of computer-readable media.
Claims
1. A component intelligent optimization design system based on AI technology, characterized in that, include: The AI-powered intelligent demand processing module includes an AI-powered intelligent modeling module, an AI-powered intelligent evaluation module, an AI-powered intelligent optimization module, and an AI-powered intelligent mapping module; among them, The AI intelligent demand processing module is used to analyze user demand data, generate initial three-dimensional geometric models of parts, and transmit the generated initial three-dimensional geometric models to the AI intelligent modeling module. The AI intelligent modeling module is used to model based on the graph structure data of the initial 3D geometric model, generate multiple similar candidate geometric model schemes, and transmit the generated candidate geometric model schemes to the AI intelligent evaluation module. The AI intelligent evaluation module is used to evaluate the performance of the components corresponding to each candidate geometric model scheme according to the multidisciplinary learning mechanism, generate the corresponding 3D model multidisciplinary performance evaluation data, and transmit the generated 3D model multidisciplinary evaluation data to the AI intelligent optimization module. The AI intelligent optimization module is used to perform multidisciplinary topology optimization on the multidisciplinary performance evaluation data of the 3D model corresponding to each candidate geometric model scheme, generate an optimized geometric model scheme, and transmit the generated optimized geometric model scheme to the AI intelligent mapping module. The AI intelligent drafting module is used to intelligently draft based on optimized geometric model schemes, thereby generating two-dimensional design drawings of parts that can be used for machining.
2. The AI-based intelligent optimization design system for components according to claim 1, characterized in that, Before performing user demand data analysis, the AI intelligent demand processing module first collects a large number of 3D models of parts and corresponding modeling command code samples and text description samples. Then, based on the collected modeling command code samples and text description samples of the 3D models of parts, the AI demand processing model is established and trained to obtain the AI demand processing model used for user demand data analysis. The input of the AI demand processing model is natural language text, and the output is code commands that can be executed by 3D modeling software.
3. The AI-based intelligent optimization design system for components according to claim 2, characterized in that, When analyzing user demand data, the AI demand processing model uses a generative AI algorithm to convert the input natural language text into code commands that can be executed by 3D modeling software. The generated code commands are then transmitted to the 3D modeling software for modeling, thereby generating an initial 3D geometric model.
4. The AI-based intelligent optimization design system for components according to claim 1, characterized in that, Before modeling the initial 3D geometric model, the AI intelligent modeling module first collects a large number of historical aircraft drawings and model samples of aero engines. Then, using deep learning principles, the AI intelligent modeling module is established and trained based on the collected historical aircraft drawings and model sample data of aero engines, thereby obtaining the AI intelligent modeling module for establishing geometric model schemes. The input of the AI intelligent modeling module is the graph structure data of the initial 3D geometric model, and the output is multiple similar candidate geometric model schemes with different features of the corresponding components.
5. The AI-based intelligent optimization design system for parts according to claim 1, characterized in that, Before evaluating each candidate geometric model scheme, the AI intelligent evaluation module first collects a large number of structural space samples of aero-engines and obtains corresponding multidisciplinary data based on the structural space samples. Then, it uses graph deep learning principles and self-attention mechanisms to build and train the AI intelligent evaluation model, thereby obtaining an AI intelligent evaluation model for multidisciplinary performance evaluation. The input of the AI intelligent evaluation model is similar candidate geometric model schemes with component features, and the output is the corresponding 3D model multidisciplinary performance evaluation data.
6. The AI-based intelligent optimization design system for components according to claim 1, characterized in that, Before performing multidisciplinary topology optimization on multidisciplinary performance evaluation data, the AI intelligent optimization module first collects the key features and parametric geometric models of the existing 3D geometric models. Then, using deep learning principles and generative adversarial networks, it establishes and trains an AI intelligent optimization model based on the collected key features and parametric geometric models of the 3D geometric models, thereby obtaining an AI intelligent optimization model for multidisciplinary topology optimization. The input of the AI intelligent optimization model is the multidisciplinary performance evaluation data of the 3D model, and the output is the optimized geometric model scheme after topology optimization.
7. The AI-based intelligent optimization design system for components according to claim 6, characterized in that, When performing multidisciplinary topology optimization, the AI intelligent optimization model first extracts and fits the key features of the 3D geometric model to the multidisciplinary performance evaluation data of the input 3D model, generating a 3D parametric geometric model. Then, it evaluates the 3D parametric geometric model by combining the corresponding multidisciplinary data in the AI intelligent evaluation module, and iteratively improves and optimizes the 3D parametric geometric model based on the evaluation results, ultimately generating an optimized geometric model scheme that meets the comprehensive multidisciplinary performance optimization indicators.
8. The AI-based intelligent optimization design system for parts according to claim 1, characterized in that, Before performing intelligent drawing, the AI intelligent drawing module first collects a large number of historical 2D machining drawings of parts. Then, based on the principles of deep learning and the collected historical 2D machining drawings of parts, it establishes and trains an AI intelligent drawing model, thereby obtaining an AI intelligent drawing model that automatically draws 2D design drawings of 3D geometric models. The input of the AI intelligent drawing model is the optimized geometric model scheme after topology optimization, and the output is the 2D design drawings of parts.
9. A component intelligent optimization design method based on AI technology, characterized in that, Includes the following steps: Step S1: Analyze user demand data using a pre-trained AI demand processing model to generate an initial three-dimensional geometric model of the parts. Step S2: Based on the graph structure data of the initial 3D geometric model of a pre-trained AI intelligent modeling model, model multiple similar candidate geometric model schemes are generated. Step S3: Utilize a pre-trained AI intelligent evaluation model to evaluate the performance of the components corresponding to each candidate geometric model scheme based on a multi-disciplinary learning mechanism, and generate corresponding 3D model multi-disciplinary performance evaluation data. Step S4: Use a pre-trained AI intelligent optimization model to perform multidisciplinary topology optimization on the multidisciplinary performance evaluation data of the 3D models corresponding to each candidate geometric model scheme, and generate an optimized geometric model scheme. Step S5: Use a pre-trained AI intelligent drawing model to intelligently draw the optimized geometric model scheme, thereby generating two-dimensional design drawings of parts that can be used for processing.
10. A computer-readable medium storing computer program code, characterized in that, The computer program code implements the method as described in any one of claims 9 when executed by a processor.
11. A component intelligent optimization design device based on AI technology, characterized in that, include: Memory is used to store instructions that can be executed by the processor; as well as A processor for executing the instructions to implement the method as described in claim 9.