Vacuum equipment design method and system based on machine learning algorithm

Abstract

The invention provides a vacuum equipment design method and system based on a machine learning algorithm, and relates to the technical field of vacuum equipment design. According to the technical scheme provided by the invention, the vacuum equipment to be designed is divided into different types according to the actual demand parameters of the vacuum equipment to be designed, and optimization adjustment is performed in a targeted manner: for the cooling type vacuum equipment, the first initial optimization scheme is emphatically optimized to form the first special optimization scheme; for the intensity type vacuum equipment, the second initial optimization scheme is emphatically optimized to form a second special optimization scheme; and for the compatible vacuum equipment, the two initial optimization schemes are optimized at the same time. According to the design method for classifying and emphatically optimizing the special areas based on the demand parameters, the problem that the same design standard is adopted for all the areas is avoided. By identifying the special demand area of the vacuum equipment and performing targeted design, the excessive design of other areas is effectively reduced, and an adjustment design scheme can be generated for the special area, so that the design cost is reduced.

Description

Technical Field

[0001] This application relates to the field of vacuum equipment design technology, specifically to a vacuum equipment design method and system based on machine learning algorithms. Background Technology

[0002] Vacuum ion plating equipment is an industrial device that uses a vacuum environment to deposit thin films. During operation, the evaporation source inside the vacuum chamber generates a large amount of heat, and the equipment is also subjected to stress from atmospheric pressure. Therefore, the design of cooling water pipe layout and reinforcing rib layout is necessary for the vacuum equipment. Cooling water pipes are used to maintain stable equipment temperature and prevent deformation and performance degradation caused by localized overheating; reinforcing ribs are used to improve the structural strength of the equipment and ensure good sealing and structural stability under vacuum conditions.

[0003] The most common design approach in existing technologies is to adopt a holistic optimization approach, incorporating all surface areas of the vacuum equipment into a unified design scope and arranging water pipes and reinforcing ribs according to the same design standards. This approach prioritizes overall aesthetics but fails to differentiate between the actual needs of different areas, resulting in widespread over-design of the overall structure. This indiscriminate design not only leads to a bloated structural design, hindering the lightweighting of the equipment, but also significantly increases costs. Summary of the Invention

[0004] This application provides a vacuum equipment design method and system based on machine learning algorithms, which can generate and adjust design schemes for special areas, thereby reducing design costs.

[0005] In a first aspect, this application provides a vacuum device design method based on machine learning algorithms, the method comprising: Obtain historical design data and standard design models for vacuum equipment; The historical design data is used as training data to train the standard design model. With the optimization goal of optimizing the overall cooling effect of the vacuum equipment, a first initial design model for cooling requirements and a first initial optimization scheme corresponding to the first initial design model are generated. The historical design data and the output of the first initial design model are used as training data to train the standard design model. The overall compressive strength of the vacuum equipment is taken as the optimization objective to generate a second initial design model for the strength requirement and a second initial optimization scheme corresponding to the second initial design model. Obtain the cooling and strength requirements of the vacuum equipment to be designed; If the cooling requirement parameter is greater than a preset first threshold and the intensity requirement parameter is not greater than a preset second threshold, the vacuum device to be designed is determined to be a cooling type vacuum device, the first initial optimization scheme is adjusted to a first special optimization scheme, and the first special optimization scheme is combined with the second initial optimization scheme to obtain the cooling type design scheme corresponding to the cooling type vacuum device. If the cooling requirement parameter is not greater than a preset first threshold and the intensity requirement parameter is greater than a preset second threshold, the vacuum equipment to be designed is determined to be an intensity-type vacuum equipment, the second initial optimization scheme is adjusted to a second special optimization scheme, and the first initial optimization scheme and the second special optimization scheme are combined to obtain the intensity-type design scheme corresponding to the intensity-type vacuum equipment. If the cooling requirement parameter is greater than a preset first threshold and the intensity requirement parameter is greater than a preset second threshold, the vacuum device to be designed is determined as a balanced vacuum device, the first initial optimization scheme is adjusted to a first special optimization scheme, and the second initial optimization scheme is adjusted to a second special optimization scheme; and the first special optimization scheme and the second special optimization scheme are combined to obtain the balanced design scheme corresponding to the balanced vacuum device.

[0006] By employing the aforementioned technical solution, a first and second initial design model are generated from historical design data, establishing basic design frameworks for cooling effect and compressive strength, respectively. Then, based on the actual requirements of the vacuum equipment to be designed, it is categorized into different types and optimized accordingly: for cooling-type vacuum equipment, the first initial optimization scheme is optimized to form a first special optimization scheme; for strength-type vacuum equipment, the second initial optimization scheme is optimized to form a second special optimization scheme; for balanced vacuum equipment, both initial optimization schemes are optimized simultaneously. This design method, based on requirement parameters and focusing on optimizing specific areas, avoids the problem of applying the same design standards to all areas. By identifying the special requirement areas of the vacuum equipment and performing targeted design, over-design in other areas is effectively reduced, and adjusted design schemes can be generated for specific areas, thereby reducing design costs.

[0007] Optionally, adjusting the first initial optimization scheme to a first special optimization scheme includes: Obtain the surface structure parameters and water cooling pipe parameters of the cooling vacuum equipment; The area where the cooling demand parameter is greater than the first threshold is designated as the first special area, and based on the surface structure parameter and the water cooling pipe parameter, a first regular area adjacent to the first special area and whose water cooling pipe arrangement affects the cooling effect of the first special area is determined. Input the cooling requirement parameters into the first initial design model and output the first pipeline layout scheme for the first special area; Based on the first pipeline layout scheme as a constraint, multiple second pipeline layout schemes are generated for the first conventional area; The first pipeline layout scheme and each of the second pipeline layout schemes are input into the heat conduction simulation system, and the cooling effect index corresponding to each of the second pipeline layout schemes is output. The second pipeline layout scheme with the best cooling effect index is combined with the first pipeline layout scheme to generate the first special optimization scheme of the cooling vacuum equipment.

[0008] Optionally, determining the first conventional region adjacent to the first special region and whose water-cooling pipe arrangement affects the cooling effect of the first special region based on the surface structure parameters and the water-cooling pipe parameters includes: The boundary coordinates of the first special region are determined based on the surface structure parameters, and the heat conduction range parameters of the water-cooling pipe are determined based on the water-cooling pipe parameters. Based on the boundary coordinates, candidate regions adjacent to the first special region are determined, and the heat conduction influence range of the candidate regions is determined according to the heat conduction range parameters of the water-cooled pipe. Calculate the overlap between the heat conduction influence range and the first special region; If the overlap is greater than a preset overlap threshold, the corresponding candidate region is determined as the first regular region.

[0009] Optionally, adjusting the second initial optimization scheme to a second specific optimization scheme includes: Obtain the structural parameters and structural stiffener parameters of the high-strength vacuum equipment; The region where the strength requirement parameter is greater than the second threshold is designated as the second special region, and based on the structural parameters, a second regular region is determined that is structurally related to the second special region and whose structural changes affect the strength of the second special region. Input the strength requirement parameters into the second initial design model, and output the first structural reinforcement scheme for the second special area; Based on the first structural reinforcement scheme as a constraint, multiple second structural reinforcement schemes are generated for the second conventional region; The first structural reinforcement scheme and each of the second structural reinforcement schemes are input into the structural mechanics simulation system, and the strength effect index corresponding to each of the second structural reinforcement schemes is output. The second structural reinforcement scheme with the best strength performance index is combined with the first structural reinforcement scheme to generate the second special optimization scheme of the strength-type vacuum equipment.

[0010] Optionally, determining a second conventional region that is structurally related to the second special region and whose structural changes affect the strength of the second special region based on the structural parameters includes: Finite element meshes are generated for the second special region to obtain stress distribution data for each mesh element; Based on the stress distribution data, a stress transmission path diagram is constructed; Identify the main stress transmission paths in the stress transmission path diagram and determine the influence areas on the main stress transmission paths; Calculate the stress concentration factor of each of the affected regions, and determine the affected regions whose stress concentration factor is greater than a preset stress threshold as the second normal region.

[0011] Optionally, the step of using the historical design data as training data to train the standard design model, with the optimization objective of achieving the optimal overall cooling effect of the vacuum equipment, to generate a first initial design model for the cooling requirements and a first initial optimization scheme corresponding to the first initial design model, includes: Extract the design parameters of the water pipe layout and the corresponding historical cooling effects from the historical design data; Using the design parameters as input features and the historical cooling effect as output labels, a water pipe layout training dataset is constructed. The water pipe layout training dataset is input into the standard design model, and the model is trained using the stochastic gradient descent algorithm. The goal is to minimize the error between the cooling effect predicted by the model and the historical cooling effect, thereby obtaining the first initial design model. With the goal of achieving optimal cooling effect, the output structure of the first initial design model is optimized using the variable density method to generate the first initial optimization scheme corresponding to the first initial design model.

[0012] Optionally, the step of using the historical design data and the output of the first initial design model as training data to train the standard design model, with the highest overall compressive strength of the vacuum equipment as the optimization objective, to generate a second initial design model for the strength requirements and a second initial optimization scheme corresponding to the second initial design model, includes: Extract the stiffener design parameters and corresponding historical strength effects from the historical design data; Using the stiffener design parameters as input features and the historical strength effect as output labels, a stiffener training dataset is constructed. The training dataset of the reinforcing ribs is input into the standard design model for training, with the goal of minimizing the error between the strength effect predicted by the model and the historical strength effect, to obtain the second initial design model; An optimization function is set with the goal of maximizing the overall compressive strength of the vacuum device. The output of the first initial design model is used as the geometric constraint condition for the layout of the reinforcing ribs. A two-way progressive structural optimization method is used for iterative optimization to generate a second initial optimization scheme corresponding to the second initial design model.

[0013] Secondly, this application provides a vacuum device design system based on machine learning algorithms, the system comprising: The initialization module is used to acquire historical design data and standard design models of vacuum equipment. The first training module is used to train the standard design model with the historical design data as training data, and to generate a first initial design model for cooling requirements and a first initial optimization scheme corresponding to the first initial design model with the optimization goal of optimizing the overall cooling effect of the vacuum equipment. The second training module is used to train the standard design model with the historical design data and the output results of the first initial design model as training data, with the overall compressive strength of the vacuum equipment as the optimization objective, to generate a second initial design model for the strength requirements and a second initial optimization scheme corresponding to the second initial design model. The input module is used to obtain the cooling and strength requirements of the vacuum equipment to be designed. A cooling-type scheme output module is used to determine the vacuum device to be designed as a cooling-type vacuum device if the cooling requirement parameter is greater than a preset first threshold and the intensity requirement parameter is not greater than a preset second threshold, adjust the first initial optimization scheme to a first special optimization scheme, and combine the first special optimization scheme with the second initial optimization scheme to obtain the cooling-type design scheme corresponding to the cooling-type vacuum device. The intensity-type scheme output module is used to determine the vacuum equipment to be designed as an intensity-type vacuum equipment if the cooling requirement parameter is not greater than a preset first threshold and the intensity requirement parameter is greater than a preset second threshold, adjust the second initial optimization scheme to a second special optimization scheme, and combine the first initial optimization scheme and the second special optimization scheme to obtain the intensity-type design scheme corresponding to the intensity-type vacuum equipment. The balanced solution output module is used to determine the vacuum device to be designed as a balanced vacuum device if the cooling requirement parameter is greater than a preset first threshold and the intensity requirement parameter is greater than a preset second threshold, adjust the first initial optimization scheme to a first special optimization scheme, adjust the second initial optimization scheme to a second special optimization scheme according to the output result of the first special optimization scheme, and combine the first special optimization scheme and the second special optimization scheme to obtain the balanced design scheme corresponding to the balanced vacuum device.

[0014] Thirdly, this application provides a computer storage medium storing a plurality of instructions adapted for loading by a processor and executing any of the methods described above.

[0015] Fourthly, this application provides an electronic device including a processor, a memory, and a transceiver, wherein the memory is used to store instructions, the transceiver is used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform any of the methods described above.

[0016] In summary, the beneficial effects of the technical solution of this application include: By employing the aforementioned technical solution, a first and second initial design model are generated from historical design data, establishing basic design frameworks for cooling effect and compressive strength, respectively. Then, based on the actual requirements of the vacuum equipment to be designed, it is categorized into different types and optimized accordingly: for cooling-type vacuum equipment, the first initial optimization scheme is optimized to form a first special optimization scheme; for strength-type vacuum equipment, the second initial optimization scheme is optimized to form a second special optimization scheme; for balanced vacuum equipment, both initial optimization schemes are optimized simultaneously. This design method, based on requirement parameters and focusing on optimizing specific areas, avoids the problem of applying the same design standards to all areas. By identifying the special requirement areas of the vacuum equipment and performing targeted design, over-design in other areas is effectively reduced, and adjusted design schemes can be generated for specific areas, thereby reducing design costs. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a vacuum device design method based on machine learning algorithms according to an embodiment of this application. Figure 2 This is a schematic diagram of the structure of a vacuum device design system based on machine learning algorithms according to an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0018] Explanation of reference numerals in the attached drawings: 300, electronic device; 301, processor; 302, communication bus; 303, user interface; 304, network interface; 305, memory. Detailed Implementation

[0019] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0020] In the description of the embodiments of this application, words such as "illustrative," "for example," or "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "illustrative," "for example," or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design solutions. Rather, the use of words such as "illustrative," "for example," or "for example" is intended to present the relevant concepts in a specific manner.

[0021] In the description of the embodiments of this application, the term "multiple" means two or more. For example, multiple systems means two or more systems, and multiple screen terminals means two or more screen terminals. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature. The terms "comprising," "including," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0022] Please see Figure 1 This document presents a flowchart illustrating a vacuum device design method based on machine learning algorithms, as provided in an embodiment of this application. This method can be implemented using a computer program, a microcontroller, or run on a machine learning-based vacuum device design system based on the von Neumann architecture. The computer program can be integrated into the application or run as a standalone tool application. The specific steps of the machine learning-based vacuum device design method are described in detail below.

[0023] S101: Obtain historical design data and standard design models for vacuum equipment; Historical design data represents the parameters and performance data of various vacuum furnace equipment (including vacuum heat treatment furnaces, vacuum sintering furnaces, vacuum brazing furnaces, etc.) that have been successfully designed and used in the past, including actual operating data such as structural parameters, cooling system parameters, and strength parameters. The standard design model refers to the basic model framework used for designing vacuum furnace equipment, which processes and optimizes the input design parameters. Vacuum equipment refers to specialized equipment used for heat treatment, sintering, brazing, and other processes in a vacuum environment, mainly composed of a vacuum chamber, heating system, cooling system, control system, and supporting structure. The heating system represents the device that provides the heat required for the process. The cooling system represents the device used to control the equipment temperature. The vacuum chamber refers to the sealed cavity that provides a vacuum environment.

[0024] Before starting the design of a new vacuum furnace, it is necessary to first obtain historical data and standard models as the design basis. Specifically, the system will extract historical data such as design parameters, operating data, and performance indicators of existing vacuum furnace equipment from the database, including detailed information such as vacuum chamber dimensions, heating element arrangement, cooling water circuit configuration, sealing structure, support structure, material properties, vacuum level, temperature distribution, cooling efficiency, and structural strength. At the same time, it will call upon a pre-established standard design model, which contains core elements such as the basic principles, process requirements, structural constraints, performance indicators, and optimization objectives for different types of vacuum furnace equipment design.

[0025] In some embodiments, historical data and standard models can be acquired in a variety of ways: Optionally, firstly, a dedicated vacuum furnace equipment design database is constructed to systematically store and manage information such as historical design schemes, process parameters, operating data, and performance evaluations of different types of vacuum furnaces; then, a data extraction interface is established, and data filtering rules are formulated according to design requirements and equipment type, focusing on historical cases similar to the target design; finally, through data cleaning and preprocessing, abnormal data is removed, and missing information is supplemented to ensure the completeness, accuracy, and usability of the acquired historical data.

[0026] Optionally, a standardized design model library can be established to collect and organize general design models for various types of vacuum furnace equipment, including vacuum chamber structure models, heating system models, cooling system models, etc. Then, these models are classified and evaluated, and the most suitable basic model for the current design task is selected based on factors such as equipment type, process requirements, and operating environment. Finally, the selected model is modified and customized as necessary, and key parameters and constraints are adjusted to meet specific design requirements.

[0027] S102: Use historical design data as training data to train the standard design model, with the optimization goal of optimizing the overall cooling effect of the vacuum equipment, and generate a first initial design model for cooling requirements and a first initial optimization scheme corresponding to the first initial design model. Here, training data refers to the historical design dataset used for training the machine learning model. Overall cooling effect refers to the temperature control capability and cooling system performance indicators of the vacuum furnace equipment during operation. The first initial design model represents a machine learning model optimized for cooling requirements. The first initial optimization scheme refers to the optimized cooling system design scheme generated based on the trained model.

[0028] After acquiring historical data and a standard model, optimization training is required to optimize the cooling effect. Specifically, firstly, parameters related to the cooling system are extracted from historical design data, including water-cooled pipe layout, pipe diameter, flow distribution, cooling medium type, and heat exchange efficiency. This data is then input into the standard design model for training. During training, the goal is to improve the overall cooling effect, and the prediction results are continuously optimized by adjusting the model parameters. After training, the model can automatically generate an optimized water circuit layout scheme based on the input cooling requirements. Simultaneously, the generated scheme is simulated and iteratively optimized to ultimately form an initial optimized scheme that meets the cooling requirements.

[0029] S103: Use historical design data and the output of the first initial design model as training data to train the standard design model. With the highest overall compressive strength of the vacuum equipment as the optimization objective, generate a second initial design model for the strength requirements and a second initial optimization scheme corresponding to the second initial design model. The first initial design model outputs the optimized vacuum furnace structural layout. Overall compressive strength refers to the vacuum furnace's ability to withstand external atmospheric pressure during operation. The second initial design model represents a machine learning model for structural strength optimization. The second initial optimization scheme is a structural reinforcement design scheme generated based on the trained model. The stiffener parameters represent the dimensions, location, shape, and other characteristics of the supporting components used to enhance the structural strength of the equipment.

[0030] After optimizing the cooling system, the structural strength of the equipment needs to be considered. Specifically, firstly, structural strength-related parameters from historical data are combined with the optimized cooling system layout to construct a complete training dataset. This data includes information such as vacuum chamber wall thickness, stiffener arrangement, support structure design, material properties, deformation, and stress distribution. Then, the data is input into a standard design model for training, aiming to improve overall compressive strength through iterative optimization of model parameters. During training, the impact of the cooling system layout on structural strength must be carefully considered to ensure that the two do not interfere with each other.

[0031] S104: Obtain the cooling and strength requirements of the vacuum equipment to be designed; The cooling requirement parameters represent the specific temperature control requirements of the vacuum furnace equipment during operation, including indicators such as maximum operating temperature, cooling rate, and temperature uniformity. The strength requirement parameters represent the pressure load and structural stress requirements that the equipment must withstand under vacuum conditions. The preset first threshold is a standard value for judging the cooling requirement level. The preset second threshold is a standard value for judging the strength requirement level. Temperature distribution represents the actual operating temperature of various parts of the equipment.

[0032] After completing model training, it is necessary to clarify the specific requirements parameters of the newly designed vacuum furnace. Specifically, firstly, based on the user's actual application scenario and process requirements, determine the temperature control parameters such as the equipment's operating temperature range, heating rate, holding time, and cooling rate. Simultaneously, based on the equipment's dimensions and vacuum requirements, calculate the external atmospheric pressure, thermal stress, and other loads that the equipment needs to withstand during operation. These parameters will serve as crucial bases for subsequent design scheme selection, determining the equipment's type, positioning, and optimization direction.

[0033] S105: If the cooling requirement parameter is greater than the preset first threshold and the strength requirement parameter is not greater than the preset second threshold, the vacuum equipment to be designed is determined to be a cooling type vacuum equipment, the first initial optimization scheme is adjusted to the first special optimization scheme, and the first special optimization scheme is combined with the second initial optimization scheme to obtain the cooling type design scheme corresponding to the cooling type vacuum equipment. Among these, "cooling-type vacuum furnace equipment" refers to equipment designed primarily to optimize cooling performance. "First special optimization scheme" refers to a water circuit layout specifically optimized for areas with high cooling demands. "Special area" refers to a key area of ​​focus with high cooling requirements. "Regular area" refers to areas with ordinary cooling demands adjacent to the special area. "Cooling-type design scheme" refers to a complete design scheme combining the first special optimization scheme and the second initial optimization scheme. "Water-cooled pipe parameters" refers to specific parameters related to the cooling system, including pipe diameter, wall thickness, and layout. "Heat conduction influence range" indicates the area that the cooling system's heat dissipation effect can cover.

[0034] After obtaining the equipment requirement parameters, if the cooling requirement parameters exceed a preset first threshold while the strength requirement parameters do not exceed a preset second threshold, the equipment needs to be classified as a cooling-type vacuum furnace. Specifically, the equipment first needs to be divided into zones to identify special areas requiring enhanced cooling and related regular areas. Then, for the high-intensity cooling requirements of the special areas, the first initial optimization scheme is deeply adjusted and optimized to generate a more targeted first special optimization scheme. This process needs to consider multiple factors such as heat conduction characteristics and fluid dynamics characteristics. Finally, the optimized cooling scheme is integrated with the original second initial optimization scheme to ensure that the strength requirements are basically met while focusing on improving the cooling performance of the equipment, forming a complete cooling-type design scheme.

[0035] Optionally, thermal imaging technology and temperature field analysis are first used to identify areas that require focused cooling; then, the required cooling capacity for each area is determined based on thermodynamic calculations, and the cooling effect is improved by adjusting parameters such as water circuit layout, pipe diameter, and flow distribution; finally, the scheme is verified and modified in combination with structural constraints to ensure that the optimized cooling system does not affect the basic strength requirements of the equipment.

[0036] S106: If the cooling requirement parameter is not greater than the preset first threshold and the strength requirement parameter is greater than the preset second threshold, the vacuum equipment to be designed is determined as a strength-type vacuum equipment, the second initial optimization scheme is adjusted to the second special optimization scheme, and the first initial optimization scheme and the second special optimization scheme are combined to obtain the strength-type design scheme corresponding to the strength-type vacuum equipment. Among them, "strength-type vacuum furnace equipment" refers to equipment types whose primary design objective is to enhance compressive strength. The second special optimization scheme represents a structural reinforcement scheme specifically optimized for areas with high strength requirements. Structural stiffener parameters indicate the dimensions, location, shape, and other characteristics of the supporting components used to enhance equipment strength. The stress concentration factor represents the multiple of the stress level in a specific area relative to the average stress. The strength-type design scheme refers to the complete design scheme combining the first initial optimization scheme and the second special optimization scheme. Structurally related areas represent adjacent areas that influence each other in terms of mechanical properties. The stress transmission path represents the direction and route of the main force transmission within the equipment.

[0037] After obtaining the equipment requirement parameters, if the strength requirement parameter exceeds a preset second threshold while the cooling requirement parameter does not exceed a preset first threshold, the equipment needs to be classified as a strength-type vacuum furnace. Specifically, structural stress analysis is first required to identify high-stress areas requiring focused reinforcement and related stress transfer areas. Then, for the high strength requirements of these areas, the second initial optimization scheme is deeply adjusted and optimized, including adjusting the layout of reinforcing ribs, optimizing the support structure, and improving stress distribution, generating a more targeted second special optimization scheme. Finally, the optimized strength scheme is integrated with the original first initial optimization scheme to ensure that the cooling requirements are basically met while focusing on improving the compressive strength of the equipment, forming a complete strength-type design scheme.

[0038] S107: If the cooling requirement parameter is greater than the preset first threshold and the strength requirement parameter is greater than the preset second threshold, the vacuum equipment to be designed is determined as a balanced vacuum equipment, the first initial optimization scheme is adjusted to the first special optimization scheme, and the second initial optimization scheme is adjusted to the second special optimization scheme; and the first special optimization scheme and the second special optimization scheme are combined to obtain the balanced design scheme corresponding to the balanced vacuum equipment.

[0039] Among them, the "combined-use vacuum furnace equipment" refers to the equipment to be designed that simultaneously meets high cooling and high strength requirements. The first special optimization scheme represents a water circuit layout scheme specifically optimized for areas with high cooling requirements. The second special optimization scheme represents a structural reinforcement scheme specifically optimized for areas with high strength requirements under the constraints of the water circuit layout. A combined-use design scheme refers to a complete scheme in which the water cooling system layout is determined first, and then structural reinforcement design is carried out based on that layout.

[0040] After obtaining the equipment requirement parameters, if both the cooling and intensity requirements exceed their respective preset thresholds, the equipment needs to be positioned as a balanced vacuum furnace. Specifically, firstly, for areas with high cooling requirements, the first initial optimization scheme is deeply adjusted to generate a first special optimization scheme, determining the specific layout of the water-cooling pipes. Then, using this water circuit layout as a constraint, for areas with high intensity requirements, the second initial optimization scheme is deeply adjusted. Without affecting the already determined water circuit system, a suitable reinforcing rib layout is designed, generating a second special optimization scheme. Finally, these two sequentially optimized schemes are combined to form a complete balanced design scheme.

[0041] Optionally, first, a detailed design of the water cooling system is carried out to determine specific parameters such as water pipe layout, pipe diameter, and connection method; then, based on the determined water circuit layout, the space and location available for arranging reinforcing ribs are analyzed; finally, the shape, size, and layout of the reinforcing ribs are designed within these available spaces to ensure that the arrangement of the reinforcing ribs does not interfere with the installation and operation of the water circuit system.

[0042] Based on the above embodiments, as an optional implementation method, step S102 generates a first initial design model for cooling requirements and a first initial optimization scheme corresponding to the first initial design model, specifically including steps S201-S203.

[0043] S201: Extract the design parameters of water pipe layout and the corresponding historical cooling effect from historical design data; The design parameters for water pipe layout describe various technical indicators of the water-cooled pipe arrangement, including specific values ​​such as the number of pipes, pipe diameter, installation location, routing path, and distribution density. Historical cooling performance represents the temperature control performance indicators during actual operation in the past design phase, including data records such as maximum temperature, temperature uniformity, and cooling rate. For example, the water pipe design parameters for a vacuum furnace are recorded as follows: main water pipe diameter 50mm, branch pipe diameter 25mm, pipe spacing 200mm, and surface coverage 65%; the corresponding cooling performance is recorded as: maximum temperature not exceeding 80℃, temperature difference less than 15℃, and cooling rate 0.5℃ / s.

[0044] Specifically, this step involves extracting two key types of information from historical design documents through data filtering and organization: the specific parameters of the water pipe layout scheme and the cooling effect data of the scheme in actual application. First, the water cooling system design drawings, process parameter tables, and installation instructions recorded in historical design documents are retrieved to extract various parameters of the water pipe layout, including pipe specifications, location coordinates, connection methods, and pipe routing. Simultaneously, actual cooling effect data for the corresponding scheme is extracted from equipment operation records, test reports, and performance evaluation documents, including temperature values ​​at each measuring point, temperature field distribution diagrams, and cooling rate curves. This data is then organized according to a one-to-one correspondence between design schemes and effects, forming a complete historical dataset.

[0045] S202: Use design parameters as input features and historical cooling effects as output labels to construct a water pipe layout training dataset; Here, the input features represent a vector of parameters describing the water pipe layout scheme. The output labels represent the cooling effect index values ​​corresponding to the input features. The training dataset represents a collection of data consisting of multiple pairs of input features and output labels. For example, parameters such as the location coordinates, pipe diameter, and layout density of the water pipes can be used as input features, and the corresponding temperature control indexes can be used as output labels.

[0046] Specifically, this step transforms the extracted historical data into a standardized training data format. First, feature engineering is performed on the pipe layout parameters, converting different types of parameters into numerical forms. For example, pipe locations are converted to standardized coordinate values, pipe diameters to cross-sectional area ratios, and layout density to area coverage. Then, the cooling effect data is standardized, converting indicators such as temperature and rate into evaluation scores with uniform dimensions. Next, each set of pipe layout parameters is paired with its corresponding cooling effect indicator in a feature-label pair format to construct a standard-format training dataset. Missing values ​​in the data are imputed using the average of similar cases; outliers are identified and processed using statistical analysis methods.

[0047] S203: Input the water pipe layout training dataset into the standard design model, and train it using the stochastic gradient descent algorithm. The goal is to minimize the error between the cooling effect predicted by the model and the historical cooling effect, and obtain the first initial design model. Here, stochastic gradient descent represents a training method that iteratively optimizes model parameters. Prediction error represents the deviation between the model's output value and the actual value. The first initial design model represents a predictive tool used to predict the cooling effect of a water pipe layout scheme after training.

[0048] Specifically, this step involves inputting the constructed training data into the standard design model for training. The training process employs the stochastic gradient descent algorithm, which is executed as follows: first, the model parameters are randomly initialized; then, a batch of training data is randomly selected each time and input into the model to obtain the prediction result; the error between the prediction result and the actual cooling effect is calculated; the parameter gradient is calculated based on the error value; and the model parameters are updated along the opposite direction of the gradient. This process is repeated until the model prediction error is reduced to below a set threshold. Mini-batch training with a batch size of 32 is used, the initial learning rate is set to 0.01 with a dynamic adjustment strategy, and the training epochs are set to 1000.

[0049] S204: With the goal of achieving optimal cooling effect, the output structure of the first initial design model is optimized using the variable density method to generate the first initial optimization scheme corresponding to the first initial design model.

[0050] Among them, the variable density method represents a method of optimization by adjusting the density distribution of different parts of the structure. Topology optimization represents the process of finding the optimal material distribution under given constraints. The first initial optimization scheme represents the specific water pipe layout scheme obtained after optimization.

[0051] Specifically, this step further optimizes the output of the trained model. When using the variable density method for optimization, the water pipe layout area is first divided into grid cells; then, a density variable is assigned to each cell, with the density value varying between 0 and 1, indicating whether a water pipe needs to be placed there; an optimization function is established with cooling effect as the objective, considering indicators such as temperature uniformity and cooling rate; under the conditions of satisfying the overall volume constraint and manufacturing constraint, the density value of each cell is updated through iterative calculation. When the optimization objective converges, the specific layout location and form of the water pipes are determined based on the final density distribution, forming a complete water pipe layout scheme.

[0052] Based on the above embodiments, as an optional implementation, step S204 can be further improved when generating the first initial optimization scheme. This is because if the optimal overall cooling effect of the vacuum equipment is taken as the sole optimization objective, the model may tend to generate extremely complex and dense cooling water pipe layout schemes when using the variable density method for topology optimization. Although such "over-optimized" design can theoretically achieve the ultimate cooling performance, in practical engineering applications it will lead to a sharp increase in manufacturing costs, an overly complex and difficult-to-implement water system, and an unnecessary increase in the total weight of the equipment. To solve this problem, it is necessary to introduce consideration of design costs during the optimization process to seek a balance between cooling performance and economy.

[0053] Specifically, during topology optimization, the original single optimization objective can be adjusted to a composite optimization objective. This new objective not only aims for better cooling performance but also aims to prevent excessive growth in the total amount of water pipes used. To achieve this, a cost penalty term positively correlated with the total amount of water pipes can be introduced into the optimization function. Under this mechanism, the optimization algorithm will make trade-offs at each step of the iterative optimization process: if a layout adjustment scheme can significantly improve the cooling performance, then even if the amount of water pipes used increases, the scheme may be accepted; conversely, if a scheme only slightly improves the cooling performance but leads to a significant increase in the amount of water pipes used, then the cost penalty term will reduce the overall evaluation value of the scheme, thus discarding it by the algorithm. As another implementation method, a clear upper limit on the total amount of water pipes can also be set as a hard constraint, forcing the algorithm to find the optimal solution for cooling performance only within the scheme space that meets the material budget.

[0054] By adopting the improved technical solution described above, this invention can automatically perform a cost-benefit analysis during the generation of the first initial optimized solution. The final output is no longer a theoretically optimal solution regardless of cost, but a first initial optimized solution that balances cooling effect and economy. This solution effectively avoids redundant and inefficient water pipe layouts while ensuring that the cooling needs of critical areas are met, thereby significantly reducing manufacturing costs and system complexity while guaranteeing reliable equipment operation.

[0055] Based on the above embodiments, as an optional implementation method, step S103 generates a second initial design model for strength requirements and a second initial optimization scheme corresponding to the second initial design model, specifically including steps S301-S303.

[0056] S301: Extract the stiffener design parameters and corresponding historical strength effects from historical design data; Among them, the stiffener design parameters describe various technical indicators of the structural stiffeners, including specific values ​​such as the shape, size, arrangement, cross-sectional characteristics, and material properties of the stiffeners. Historical strength performance represents the structural load-bearing performance indicators during actual operation in the past design phase, including data records such as maximum stress, deformation, and stress distribution. For example, the stiffener design parameters for a vacuum furnace are recorded as: T-shaped stiffeners, height 100mm, thickness 12mm, circumferential spacing 300mm; the corresponding strength performance records are: maximum stress 120MPa, maximum deformation 0.8mm, and stress distribution uniformity coefficient 0.85.

[0057] Specifically, this step involves extracting two key types of information from historical design documents through data filtering and organization: the specific parameters of the stiffener arrangement scheme and the strength effect data of the scheme in actual application. First, the structural design drawings, engineering parameter tables, and fabrication instructions recorded in historical design documents are retrieved to extract various parameters of the stiffener arrangement, including the specific specifications of the stiffeners, installation location coordinates, connection node design, and material selection. Simultaneously, actual strength effect data of the corresponding scheme is extracted from equipment operation records, stress test reports, and structural analysis reports, including stress measurement values, deformation measurement results, and structural response curves. This data is then organized according to a one-to-one correspondence between design schemes and their effects, forming a complete historical dataset.

[0058] S302: Construct a stiffener training dataset by using stiffener design parameters as input features and historical strength effects as output labels; Here, the input features represent the parameter vectors describing the stiffener arrangement scheme. The output labels represent the strength performance index values ​​corresponding to the input features. The training dataset represents a collection of data consisting of multiple pairs of input features and output labels. For example, parameters such as the location coordinates, cross-sectional parameters, and spacing of the stiffeners can be used as input features, and the corresponding structural strength indices can be used as output labels.

[0059] Specifically, this step transforms the extracted historical data into a standardized training data format. First, feature engineering is performed on the stiffener placement parameters, converting different types of parameters into numerical forms. For example, stiffener positions are converted to standardized coordinate values, cross-sectional features are converted to moments of inertia values, and placement spacing is converted to relative size proportions. Then, the strength effect data is standardized, converting stress, deformation, and other indicators into evaluation scores with uniform dimensions. Finally, each set of stiffener placement parameters is paired with its corresponding strength effect indicator in a feature-label pair format to construct a standard-format training dataset.

[0060] S303: Input the reinforcement training dataset into the standard design model for training, with the goal of minimizing the error between the strength effect predicted by the model and the historical strength effect, to obtain the second initial design model; The prediction error represents the deviation between the strength effect output by the model and the actual strength effect. The second initial design model represents the prediction tool used to predict the strength effect of the stiffener layout scheme after training. The strength effect includes mechanical performance indicators such as stress distribution and deformation of the structure under external pressure.

[0061] Specifically, this step involves inputting the constructed training data into the standard design model for training. During training, the input stiffener parameters are first used to calculate the predicted strength effect, including indicators such as maximum stress, stress distribution, and deformation. Then, the predicted results are compared with the actual strength effects recorded in historical data to calculate the prediction error for each indicator. Based on the error value, the model parameters are adjusted and optimized, and the prediction error is continuously reduced through multiple iterations of training. Data is input in batches during training, with each batch containing 50 sets of data, for a total of 2000 training rounds, until the prediction error is reduced below a set value, resulting in the completed second initial design model.

[0062] S304: Set an optimization function with the goal of maximizing the overall compressive strength of the vacuum equipment. Use the output of the first initial design model as the geometric constraint condition for the stiffener layout. Use the two-way progressive structural optimization method to iteratively optimize and generate the second initial optimization scheme corresponding to the second initial design model.

[0063] The optimization function represents the mathematical expression used to evaluate the strength performance of the design scheme. Geometric constraints represent the spatial limitations imposed by the water system on the placement of stiffeners. The two-way progressive structural optimization method represents a method for optimizing the structure by gradually adding and removing materials. The second initial optimization scheme represents the specific stiffener layout scheme obtained after optimization.

[0064] Specifically, this step optimizes the arrangement of stiffeners under the constraints of the water channel layout. First, an optimization function is established based on the stress characteristics of the vacuum equipment, including multiple evaluation indicators such as maximum stress, stress uniformity, and total mass. Then, the water channel layout output from the first initial design model is used as a geometric constraint to determine the spatial range suitable for arranging stiffeners. When using a two-way progressive structural optimization method, the structure is divided into mesh elements. In each iteration: the stress level of each element is calculated, identifying regions with high and low stress; material is added to high-stress regions to add stiffeners, while material is reduced and redundant stiffeners are removed in low-stress regions; the constraints of the water channel layout are checked; if the constraints are met, the current modification is accepted; otherwise, the previous state is reverted. Through multiple iterations of optimization, a stiffener layout scheme that meets strength requirements without affecting the water channel system is finally obtained.

[0065] Based on the above embodiments, as an optional implementation, step S304 can also be optimized and improved when generating the second initial optimization scheme. The reason is that if the sole objective during structural optimization is to maximize the overall compressive strength of the vacuum equipment, and a two-way progressive structural optimization method is used for iterative optimization, the algorithm might indiscriminately add reinforcing ribs to all stress areas in pursuit of absolute structural strength. This "over-optimization" strategy would result in an abnormally heavy equipment structure, significantly increasing material costs and placing higher demands on equipment transportation, installation, and foundation support, which is inconsistent with the modern industrial equipment lightweight design trend. Therefore, it is essential to effectively control the total weight of the structure while ensuring structural safety.

[0066] Specifically, when performing structural topology optimization, the original optimization problem can be redefined. The optimization objective remains maximizing the overall compressive strength (or equivalently, minimizing the overall structural flexibility), but a key constraint needs to be imposed on this optimization problem: setting an upper limit constraint on the total weight of the stiffened structure. Within this framework, the behavior of the two-way progressive structural optimization method changes during the iteration process. When the algorithm identifies a high-stress area and attempts to enhance its strength by "adding" material, it must first check whether the total weight of the current structure has reached the preset weight limit. If adding the material would lead to overweight, the operation will be prohibited. If the algorithm wants to continue enhancing the structure at this critical location, it must first "remove" material in other areas with lower stress levels and less contribution to overall strength to release sufficient "weight quota."

[0067] By implementing this optimization process with a weight upper limit constraint, the present invention ensures that the final generated second initial optimized scheme is an intelligent design that achieves a high balance between strength and weight. It is no longer a product of blindly piling up materials, but rather a precise allocation of limited material resources to the most needed locations in the structure. The resulting second initial optimized scheme, which balances strength and lightweight, not only fully meets the strength and stability requirements for the equipment to operate in a vacuum environment, but also optimizes its structural weight to the maximum extent, thereby effectively reducing material costs.

[0068] Based on the above embodiments, as an optional implementation method, the first initial optimization scheme is adjusted to a first special optimization scheme, specifically including steps S401-S403.

[0069] S401: Obtain the surface structure parameters and water cooling pipe parameters of the cooled vacuum equipment; Among them, surface structure parameters represent technical indicators describing the geometric features of the outer surface of vacuum equipment, including specific values ​​such as the equipment's external dimensions, wall thickness, and surface irregularities. Water cooling pipe parameters represent technical indicators describing the basic characteristics of the water cooling system, including specific parameters such as pipe material, pipe diameter, and wall thickness.

[0070] Specifically, this step involves measuring and extracting two types of key data: the geometric parameters of the equipment surface and the basic parameters of the water cooling system. For obtaining surface structure parameters, specific dimensional data of various parts of the equipment are recorded through 3D scanning or on-site measurement, including overall outline dimensions, surface undulations, and local feature dimensions. For obtaining water cooling pipe parameters, the specifications of the water pipes are determined by consulting design specifications and product manuals, including standard pipe diameter, material performance parameters, and connection requirements.

[0071] S402: Define the area where the cooling demand parameter is greater than the first threshold as the first special area, and determine the first regular area that is adjacent to the first special area and whose water cooling pipe arrangement affects the cooling effect of the first special area based on the surface structure parameter and water cooling pipe parameter. The first special area represents the region requiring focused cooling optimization. The first regular area represents the ordinary area adjacent to the special area and affecting its cooling effect. Cooling effect impact indicates the degree to which the water-cooling pipe arrangement in adjacent areas affects the temperature control of the target area. The first threshold represents the standard value used to determine whether special cooling treatment is required.

[0072] Specifically, this step involves dividing and classifying the equipment surface into regions. First, based on the distribution of cooling demand parameters, regions where parameter values ​​exceed a first threshold are identified and marked as first special regions. Then, the areas surrounding the first special regions are analyzed, and heat conduction analysis is used to determine the range and extent of the impact of the water-cooling pipe arrangement on the temperature control of the first special regions. Adjacent regions whose impact reaches a set level are designated as first normal regions.

[0073] S403: Input the cooling requirement parameters into the first initial design model and output the first piping layout scheme for the first special area; S404: Based on the first pipeline layout scheme as a constraint, generate multiple second pipeline layout schemes for the first conventional area; The first pipeline layout scheme represents the water pipe arrangement designed for the first special area. The second pipeline layout scheme represents multiple alternative water pipe arrangement schemes designed for the first regular area. The basic constraints represent the restrictions imposed by the first pipeline layout on the second pipeline layout.

[0074] Specifically, these two steps involve calculating water pipe layout schemes for different regions using a model. First, the cooling requirement parameters for the first special region are input into the first initial design model. The model calculates and outputs the water pipe layout scheme for that region based on the input parameters, including details such as pipe routing and spacing. Then, using this scheme as constraints, multiple different water pipe layout schemes are generated for the first regular region. During the generation process, it is necessary to ensure that these schemes match the pipe layout of the first special region, including requirements such as reasonable pipe connection point locations and coordinated flow channel arrangement.

[0075] S405: Input the first pipeline layout scheme and each of the second pipeline layout schemes into the heat conduction simulation system, and output the cooling effect index corresponding to each of the second pipeline layout schemes; S406: Combine the second pipeline layout scheme with the first pipeline layout scheme to generate the first special optimization scheme for the cooling vacuum equipment.

[0076] The heat conduction simulation system represents a numerical analysis tool used to simulate and calculate the temperature field distribution. Cooling performance indicators represent technical metrics used to evaluate the performance of the water cooling scheme, including parameters such as temperature distribution uniformity and maximum temperature. The first special optimization scheme represents the final, determined complete water pipe layout scheme.

[0077] Specifically, these two steps involve selecting the optimal solution and integrating the solutions through simulation analysis. First, all generated layout schemes are input into the heat conduction simulation system, and temperature field calculations are performed under identical operating conditions. The calculation results include temperature values ​​at each measuring point, temperature distribution cloud maps, temperature gradients, and other data. By comparing these data, the cooling effect of each second pipeline layout scheme is evaluated. Evaluation indicators include: maximum temperature not exceeding the set value, temperature uniformity meeting requirements, and a smooth temperature transition with the first special area. The second pipeline layout scheme with the best overall evaluation is selected and integrated with the first pipeline layout scheme to form a complete water pipe layout scheme, i.e., the first special optimized scheme.

[0078] Based on the above embodiments, as an optional implementation method, step S402 determines a first conventional region that is adjacent to the first special region and whose water cooling pipe arrangement affects the cooling effect of the first special region according to the surface structure parameters and water cooling pipe parameters, specifically including steps S4021-S4024.

[0079] S4021: Determine the boundary coordinates of the first special region based on the surface structure parameters, and determine the heat conduction range parameters of the water-cooling pipe based on the water-cooling pipe parameters; The boundary coordinates represent the spatial limits of the first special region, using a three-dimensional coordinate system (x, y, z) to represent the location of the boundary points. The heat conduction range parameter represents the heat dissipation range of the water-cooled pipe, including the effective heat transfer distance along the pipe wall and perpendicular to the pipe wall. For example, the boundary coordinates of a certain region are: upper left corner (0, 0, 0), lower right corner (500, 300, 0) mm; the heat conduction range parameter of the water-cooled pipe is: 150 mm along the pipe wall and 80 mm perpendicular to the pipe wall.

[0080] Specifically, this step involves two calculations: First, based on the surface structure parameters, the boundary tracing method is used to determine the specific range of the first special region, and the spatial positions of the region's boundary points are represented by coordinate values, forming a complete set of boundary coordinates. Then, based on parameters such as the thermal conductivity of the water-cooled pipe material, pipe wall thickness, and cooling water flow rate, heat conduction calculations are used to determine the effective heat dissipation range of a single water pipe, obtaining the heat conduction range parameters. The calculations consider the material's thermal conductivity and convective heat transfer coefficient to determine the distance required for the temperature to drop to the set value under steady-state conditions.

[0081] S4022: Determine the candidate region adjacent to the first special region based on the boundary coordinates, and determine the heat conduction influence range of the candidate region according to the heat conduction range parameter of the water-cooled pipe; Here, the candidate region represents the area directly adjacent to the first special region in terms of spatial location. The heat conduction influence range represents the spatial range within the candidate region where the arrangement of water-cooled pipes affects the temperature of the first special region. For example, the heat conduction influence range of a certain candidate region is a strip-shaped area extending 200 mm inward from the boundary.

[0082] Specifically, this step first identifies all regions adjacent to the first special region through spatial adjacency analysis and marks these regions as candidate regions. Then, for each candidate region, its potential heat conduction influence range is calculated based on the heat conduction range parameter of the water-cooled pipe. The calculation method is as follows: taking the common boundary between the candidate region and the first special region as a reference, the distance defined by the heat conduction range parameter is extended to both sides to form a three-dimensional influence space. For irregularly shaped boundaries, a discrete point segment-by-segment calculation method is used to determine the complete influence range.

[0083] S4023: Calculate the overlap between the heat conduction influence range and the first special region; The overlap ratio indicates the degree to which the thermal conduction influence range of the candidate region overlaps spatially with the first special region, and is expressed as the ratio of the overlapping volume to the total volume of the first special region. For example, an overlap ratio of 0.3 indicates that the thermal conduction influence range covers 30% of the volume of the first special region.

[0084] Specifically, this step calculates the overlap value through spatial analysis. First, both the first special region and the heat conduction influence range are discretized into grid cells; then, the number of grid cells belonging to both regions simultaneously is counted and recorded as the overlapping cell number; the overlapping cell number is divided by the total number of cells in the first special region to obtain the overlap value. The grid size used in the calculation needs to be small enough to ensure calculation accuracy, usually taken as 1 / 10 of the smallest feature size.

[0085] S4024: If the overlap is greater than the preset overlap threshold, the corresponding candidate region is determined as the first regular region.

[0086] The preset overlap threshold represents the standard value used to determine whether a candidate region should be included in the first regular region. The first regular region refers to the region that requires collaborative design after the overlap judgment. For example, setting the preset overlap threshold to 0.2 means that when the overlap exceeds 20%, the candidate region will be included in the first regular region.

[0087] Specifically, this step involves region selection by comparing overlap with a preset threshold. The calculated overlap value for each candidate region is compared to the preset overlap threshold. When the overlap is greater than the threshold, it indicates that the water-cooled pipe arrangement in that candidate region has a significant impact on the temperature control of the first special region, and it needs to be included in the unified design scope. Therefore, this candidate region is designated as the first regular region. Candidate regions with an overlap less than the threshold are not included in the first regular region and can be designed independently.

[0088] Specifically, after completing the topology optimization of the cooling system and obtaining the first initial optimization scheme including the three-dimensional geometric layout of the cooling water pipes, the method will enter the second stage of optimizing the stiffener layout based on this model. The reason for introducing a special collaborative optimization mechanism at this stage is that if the strength optimization algorithm freely arranges stiffeners in a "blank" geometric space, it will cause two fatal problems. First, there is physical interference and overlap; the stiffeners generated by the algorithm may directly pass through or occupy the already planned cooling water pipe positions, making the final design physically impossible to manufacture. Second, even without direct overlap, the stiffeners and the hollow cooling water pipe structure will form sharp geometric transitions at the interface or adjacent areas. This will inevitably lead to severe stress concentration when the equipment is subjected to external pressure, forming inherent weak points in the design and weakening the overall reliability of the structure, contrary to the original intention of strengthening the design.

[0089] To fundamentally address the aforementioned problems, this invention proposes a collaborative optimization method that inputs multiphysics constraints as control parameters into a machine learning model. Specifically, before starting the topology optimization of the stiffeners, the 3D model of the cooling water pipe determined by the first initial optimization scheme is input as a key geometric boundary condition into the machine learning model used for strength optimization. During each iteration of the optimization algorithm, the model performs a dual constraint check: First, it defines the geometric model of the cooling water pipe as a set of "prohibited layout regions." Any attempt to "grow" material within these regions to form stiffeners will be directly rejected by the algorithm, thus eliminating the possibility of physical interference and overlap from the outset. Second, and more crucially, for the layout of the stiffeners in the vicinity of the cooling water pipe, the algorithm quantifies the geometric relationship between them (such as minimum distance, curvature of the contact boundary, intersection angle, etc.) into a set of dynamic control parameters. These control parameters are input in real-time into a built-in evaluation function of the machine learning model. This function predicts and evaluates the potential stress concentration based on these parameters in real time. If a rib-reinforcing layout can improve the macroscopic stiffness of the structure, but its corresponding control parameters indicate that it will cause high-risk stress concentration near the water pipe, then the overall fitness score of the scheme will be significantly reduced by a penalty function.

[0090] By adopting the aforementioned collaborative optimization technology, the final result is that the entire optimization process is no longer a simple superposition of two isolated steps, but an intelligent design process with inherent logical connections. The algorithm is "taught" how to "see" and "understand" the existing cooling structure, and on this basis, to implement a "benefit-oriented and harm-avoidance-avoidance" reinforcement layout. The final output of the second initial optimization scheme allows the reinforcements and cooling water pipes to coexist harmoniously and support each other mechanically, ensuring both the manufacturability of the design scheme and actively avoiding the risk of stress concentration.

[0091] Based on the above embodiments, as an optional implementation method, the second initial optimization scheme is adjusted to a second special optimization scheme, specifically including steps S501-S503.

[0092] S501: Obtain the structural parameters and structural stiffener parameters of the high-strength vacuum equipment; Among them, structural parameters represent technical indicators describing the basic structural characteristics of vacuum equipment, including specific values ​​such as shell shape, wall thickness, joint location, and opening size. Structural stiffener parameters represent technical indicators describing the basic characteristics of stiffening components, including specific parameters such as stiffener profile specifications, material properties, and arrangement.

[0093] Specifically, this step involves measuring and extracting two types of key data: the basic structural parameters of the equipment and the basic parameters of the reinforcing ribs. For obtaining the structural parameters, the specific structural dimensions of each part of the equipment are recorded through 3D measurement or drawing analysis, including the shape and dimensions of the main structure, local structural features, and the location of connection nodes. For obtaining the reinforcing rib parameters, the specifications of the reinforcing ribs are determined by consulting design specifications and material manuals.

[0094] S502: Define the area where the strength requirement parameter is greater than the second threshold as the second special area, and determine the second regular area that is structurally related to the second special area and whose structural changes affect the strength of the second special area based on the structural parameters. The second special region refers to areas requiring focused structural reinforcement. The second regular region refers to ordinary regions structurally related to the special regions and influencing their strength. Structural association indicates a structural connection between the two regions where force transfer or deformation influence exists. The second threshold represents a standard value used to determine whether special structural reinforcement is needed.

[0095] Specifically, this step involves dividing and classifying the equipment structure into regions. First, based on the distribution of strength requirement parameters, regions where parameter values ​​exceed a second threshold are identified and marked as second special regions. Then, the force transmission paths of these second special regions are analyzed, and adjacent regions with force transmission relationships are determined through structural mechanics analysis. For structural changes in these regions that significantly affect the strength of the second special regions, they are designated as second regular regions. The division process must consider the integrity of the structure and the continuity of force transmission to ensure that the zoning results conform to the principles of structural mechanics.

[0096] S503: Input the strength requirement parameters into the second initial design model and output the first structural reinforcement scheme for the second special area; S504: Based on the first structural reinforcement scheme as a constraint, generate multiple second structural reinforcement schemes for the second conventional region; Here, the first structural reinforcement scheme represents the reinforcement arrangement scheme designed for the second special area. The second structural reinforcement scheme represents multiple alternative reinforcement arrangement schemes designed for the second regular area. Foundation constraints represent the restrictions imposed by the first reinforcement scheme on the second reinforcement scheme.

[0097] Specifically, these two steps involve calculating stiffener placement schemes for different regions using a model. First, the strength requirement parameters for the second special region are input into the second initial design model. The model calculates and outputs the stiffener placement scheme for that region based on the input parameters, including details such as the type, quantity, and location of the stiffeners. Then, using this scheme as a constraint, multiple different stiffener placement schemes are generated for the second regular region. During the generation process, it is necessary to ensure that these schemes match the first structural reinforcement scheme, including requirements such as reasonable stiffener connection methods and continuous force transmission paths.

[0098] S505: Input the first structural reinforcement scheme and each of the second structural reinforcement schemes into the structural mechanics simulation system, and output the strength effect index corresponding to each of the second structural reinforcement schemes; S506: Combine the second structural reinforcement scheme with the first structural reinforcement scheme to generate a second special optimization scheme for a strength-type vacuum device.

[0099] Among them, the structural mechanics simulation system represents the numerical analysis tool used to simulate and calculate stress and strain distribution. The strength performance index represents the technical indicators used to evaluate the performance of the strengthening scheme, including parameters such as maximum stress value, deformation, and stress distribution. The second special optimization scheme represents the final determined complete stiffener layout scheme.

[0100] Specifically, these two steps involve selecting the optimal solution and integrating the solutions through simulation analysis. First, all generated reinforcement schemes are input into the structural mechanics simulation system, and stress-strain calculations are performed under the same load conditions. The calculation results include stress values ​​at each measuring point, deformation contour maps, stress distribution, and other data. By comparing these data, the strength effect of each secondary structural reinforcement scheme is evaluated. Evaluation indicators include: maximum stress below the allowable value, deformation meeting requirements, and reasonable stress transition with the second special region. The secondary structural reinforcement scheme with the best overall evaluation is selected and integrated with the primary structural reinforcement scheme to form a complete stiffener layout scheme, i.e., the second special optimized scheme.

[0101] Based on the above embodiments, as an optional implementation method, step S502, which generates a first initial design model for cooling requirements and a first initial optimization scheme corresponding to the first initial design model, specifically includes steps S5021-S5024.

[0102] S5021: Perform finite element mesh generation on the second special region and obtain stress distribution data for each mesh element; In this context, the finite element mesh represents a computational grid system that discretizes a continuous structure into a finite number of small elements. Each mesh element represents a basic computational unit in finite element analysis, including element type, node coordinates, and material properties. Stress distribution data represents the stress state values ​​in each mesh element, including calculated results such as principal stresses and von Mises stresses. For example, using hexahedral mesh elements with a size of 10 mm, each element containing 8 nodes, the nodal stress values ​​and the stress values ​​at the element center are recorded.

[0103] Specifically, this step involves stress analysis of the second special region using the finite element method. First, an appropriate element type is selected based on the region's geometry: quadrilateral shell elements are used for plate and shell structures, while hexahedral or tetrahedral elements are used for solid structures. Then, meshing parameters are set, employing a denser mesh in regions with large stress gradients and a coarser mesh in regions with gradual stress changes. Mesh quality control indicators include: element aspect ratio not exceeding 5, adjacent element size ratio not exceeding 1.5, and element orthogonality deviation not exceeding 30 degrees. After meshing, boundary conditions and loads are applied, and static analysis calculations are performed to obtain the stress components and equivalent stress values ​​for each mesh element.

[0104] S5022: Construct a stress transfer path diagram based on stress distribution data; The stress transfer path diagram is a network diagram describing the direction and magnitude of stress transfer in a structure, including information such as node locations, path directions, and stress intensity. Stress transfer represents the process and path distribution of load transmission within the structure. For example, the direction of stress transfer is indicated by arrows, and the intensity of stress transfer is indicated by line thickness.

[0105] Specifically, this step involves constructing a stress transfer network structure through data analysis. First, based on the stress distribution data obtained from finite element analysis, the principal stress directions and magnitudes of each element are extracted. Then, based on the stress continuity between adjacent elements, the stress transfer connections are established. The specific method is as follows: calculate the stress transfer amount between adjacent elements and determine the transfer direction; mark connections where the stress transfer amount exceeds a set threshold as valid paths; and set weighting coefficients based on the transfer intensity of each path. Finally, a network diagram containing node locations, path directions, and transfer intensities is formed to characterize the stress transfer features in the structure.

[0106] S5023: Identify the main stress transmission paths in the stress transmission path diagram and determine the influence area on the main stress transmission paths; The primary stress transfer path represents the critical path in the structure that bears the main load transfer function. The influence region represents the area along the primary stress transfer path that has a significant impact on the strength of the second special region. For example, in a cylindrical structure, the axial force transfer path between the circumferential stiffeners is the primary stress transfer path.

[0107] Specifically, this step identifies key stress transmission areas through path analysis. First, an importance analysis is performed on the stress transmission path map, calculating the transmission intensity value for each path. Transmission intensity is scored based on the average stress value along the path and the continuity of the transmission direction; a higher score indicates a more important path. An intensity threshold is set to filter out the main stress transmission paths, typically selecting the top 20% of paths by transmission intensity. Then, the influence area is determined by expanding outwards along the main stress transmission paths. The expansion distance is determined based on the stress attenuation law, generally taking the point where the stress decreases to 30% of its peak value as the boundary of the influence area.

[0108] S5024: Calculate the stress concentration factor of each affected area, and determine the affected area with a stress concentration factor greater than the preset stress threshold as the second normal area.

[0109] The stress concentration factor represents the ratio of the local maximum stress to the nominal stress, and is used to measure the degree of stress concentration. The preset stress threshold is a standard value used to determine whether the affected area needs to be included in the design scope. For example, a stress concentration factor of 2.5 means that the local maximum stress is 2.5 times the nominal stress.

[0110] Specifically, this step involves identifying areas requiring collaborative design through stress analysis. First, the stress concentration factor is calculated for each affected area by extracting the maximum stress value within the area and dividing it by the nominal stress value in the far field. Then, the calculated stress concentration factor is compared to a preset threshold. When the stress concentration factor is greater than the threshold, it indicates that structural changes in this affected area will significantly impact the strength state of the second special area, requiring its inclusion in the unified design scope; therefore, it is designated as the second regular area. Affected areas with stress concentration factors less than the threshold are not included in the second regular area and can be designed independently.

[0111] Another key embodiment of the present invention is to provide a graphical user interface (UI) that integrates the aforementioned complete collaborative optimization process and core algorithm, aiming to solve the problems of complex operation, high professional barriers and low iteration efficiency in the traditional multiphysics optimization process.

[0112] The reason for developing such an integrated UI is that, in traditional R&D models, executing a process that includes co-optimization of cooling and strength typically requires designers to switch between multiple independent professional software programs, manually export and import geometric models and data, and may require calling machine learning algorithms via complex command lines. This process is not only extremely tedious and time-consuming, and prone to errors due to human operation, but more importantly, it places extremely high demands on designers' cross-disciplinary skills, requiring them to be proficient in structural design, thermal fluid simulation, and algorithm debugging simultaneously, which undoubtedly creates a very high technical barrier to entry. Furthermore, when fine-tuning the design objectives is needed (for example, reducing the target cooling temperature from 100℃ to 95℃), designers must manually repeat the entire lengthy process, resulting in extremely low efficiency in design iteration and solution debugging.

[0113] To overcome the above pain points, the UI of this invention encapsulates and integrates the entire optimization process in the following way. First, the UI encapsulates the aforementioned cooling optimization, strength optimization, and the collaborative constraint algorithms between the two in the background. On the front end, it provides designers with a series of intuitive, parameterized input controls. For example, designers do not need to concern themselves with the internal details of the algorithms; they only need to fill in macroscopic design goals such as the desired cooling temperature, desired maximum deformation, allowable maximum stress, and upper limit of structural weight in the corresponding input boxes. After the designer sets these target parameters, they only need to click the "One-Click Optimization" or "Generate Scheme" button on the interface, and the system will automatically execute the complete collaborative optimization process in the background, including calling machine learning models, automatically transferring geometric models and control parameters between optimization stages, and finally generating an optimization result that meets all input requirements.

[0114] The following are system embodiments of this application, which can be used to execute the method embodiments of this application. For details not disclosed in the system embodiments of this application, please refer to the method embodiments of the application.

[0115] Please see Figure 2 This illustration shows a schematic diagram of a vacuum device design system based on machine learning algorithms provided in an exemplary embodiment of this application. The system can be implemented entirely or partially through software, hardware, or a combination of both. The vacuum device design system based on machine learning algorithms includes: The initialization module is used to acquire historical design data and standard design models of vacuum equipment. The first training module is used to train the standard design model with historical design data as training data, with the optimization goal of optimizing the overall cooling effect of the vacuum equipment, and to generate a first initial design model for cooling requirements and a first initial optimization scheme corresponding to the first initial design model. The second training module is used to train the standard design model with historical design data and the output results of the first initial design model as training data. The optimization objective is to maximize the overall compressive strength of the vacuum equipment, and to generate a second initial design model for the strength requirements and a second initial optimization scheme corresponding to the second initial design model. The input module is used to obtain the cooling and strength requirements of the vacuum equipment to be designed. The cooling-type scheme output module is used to determine the vacuum equipment to be designed as a cooling-type vacuum equipment if the cooling requirement parameter is greater than a preset first threshold and the strength requirement parameter is not greater than a preset second threshold, adjust the first initial optimization scheme to a first special optimization scheme, and combine the first special optimization scheme with the second initial optimization scheme to obtain the cooling-type design scheme corresponding to the cooling-type vacuum equipment. The intensity-type scheme output module is used to determine the vacuum equipment to be designed as an intensity-type vacuum equipment if the cooling requirement parameter is not greater than a preset first threshold and the intensity requirement parameter is greater than a preset second threshold, adjust the second initial optimization scheme to a second special optimization scheme, and combine the first initial optimization scheme and the second special optimization scheme to obtain the intensity-type design scheme corresponding to the intensity-type vacuum equipment. The balanced solution output module is used to determine the vacuum equipment to be designed as a balanced vacuum equipment if the cooling requirement parameter is greater than a preset first threshold and the strength requirement parameter is greater than a preset second threshold. It adjusts the first initial optimization scheme to a first special optimization scheme, adjusts the second initial optimization scheme to a second special optimization scheme based on the output result of the first special optimization scheme, and combines the first special optimization scheme and the second special optimization scheme to obtain the balanced design scheme corresponding to the balanced vacuum equipment.

[0116] Based on the above embodiments, as an optional embodiment, the first training module is further used to extract the design parameters of the water pipe layout and the corresponding historical cooling effects from the historical design data; to construct a water pipe layout training dataset by using the design parameters as input features and the historical cooling effects as output labels; to input the water pipe layout training dataset into the standard design model and train it using the stochastic gradient descent algorithm, with the goal of minimizing the error between the cooling effect predicted by the model and the historical cooling effect, thereby obtaining the first initial design model; and to optimize the output structure of the first initial design model using the variable density method with the goal of achieving the optimal cooling effect, thereby generating the first initial optimization scheme corresponding to the first initial design model.

[0117] Based on the above embodiments, as an optional embodiment, the second training module is further used to extract the stiffener design parameters and corresponding historical strength effects from historical design data; construct a stiffener training dataset by using the stiffener design parameters as input features and the historical strength effects as output labels; input the stiffener training dataset into the standard design model for training, with the goal of minimizing the error between the strength effect predicted by the model and the historical strength effect, to obtain a second initial design model; set an optimization function with the goal of maximizing the overall compressive strength of the vacuum equipment, use the output results of the first initial design model as the geometric constraints of the stiffener layout, and use a bidirectional progressive structural optimization method for iterative optimization to generate a second initial optimization scheme corresponding to the second initial design model.

[0118] Based on the above embodiments, as an optional embodiment, the cooling-type scheme output module or the balanced scheme output module is further used to obtain the surface structure parameters and water-cooling pipe parameters of the cooling-type vacuum equipment; define the area where the cooling demand parameters are greater than a first threshold as a first special area, and determine the first regular area adjacent to the first special area and where the water-cooling pipe arrangement affects the cooling effect of the first special area based on the surface structure parameters and water-cooling pipe parameters; input the cooling demand parameters into the first initial design model and output the first pipeline layout scheme of the first special area; generate multiple second pipeline layout schemes for the first regular area based on the first pipeline layout scheme as a basic constraint; input the first pipeline layout scheme and each second pipeline layout scheme into the heat conduction simulation system and output the cooling effect index corresponding to each second pipeline layout scheme; combine the second pipeline layout scheme with the optimal cooling effect index with the first pipeline layout scheme to generate the first special optimization scheme of the cooling-type vacuum equipment.

[0119] Based on the above embodiments, as an optional embodiment, the cooling-type solution output module or the combined solution output module is further configured to determine the boundary coordinates of the first special region according to the surface structure parameters, and determine the heat conduction range parameters of the water-cooling pipe according to the water-cooling pipe parameters; determine the candidate region adjacent to the first special region based on the boundary coordinates, and determine the heat conduction influence range of the candidate region according to the heat conduction range parameters of the water-cooling pipe; calculate the overlap between the heat conduction influence range and the first special region; if the overlap is greater than a preset overlap threshold, then the corresponding candidate region is determined as the first regular region.

[0120] Based on the above embodiments, as an optional embodiment, the strength-type scheme output module or the balanced scheme output module is further used to obtain the structural parameters and structural stiffener parameters of the strength-type vacuum equipment; to define the region where the strength requirement parameter is greater than the second threshold as the second special region, and to determine the second regular region that is structurally related to the second special region and whose strength is affected by structural changes, based on the structural parameters; to input the strength requirement parameter into the second initial design model and output the first structural reinforcement scheme for the second special region; to generate multiple second structural reinforcement schemes for the second regular region based on the first structural reinforcement scheme as the basic constraint; to input the first structural reinforcement scheme and each second structural reinforcement scheme into the structural mechanics simulation system and output the strength effect index corresponding to each second structural reinforcement scheme; and to combine the second structural reinforcement scheme with the first structural reinforcement scheme with the optimal strength effect index to generate the second special optimization scheme for the strength-type vacuum equipment.

[0121] Based on the above embodiments, as an optional embodiment, the strength-type scheme output module or the balanced scheme output module is further used to perform finite element meshing on the second special region, obtain stress distribution data of each mesh unit; construct a stress transmission path diagram based on the stress distribution data; identify the main stress transmission paths in the stress transmission path diagram, determine the influence areas on the main stress transmission paths; calculate the stress concentration coefficient of each influence area, and determine the influence areas with stress concentration coefficients greater than the preset stress threshold as the second regular region.

[0122] This application also provides a computer storage medium that can store multiple instructions. The instructions are adapted to be loaded and executed by a processor as described in the above embodiments of the vacuum device design method based on machine learning algorithms. For details of the execution process, please refer to the specific description of the embodiments, which will not be repeated here.

[0123] Please see Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 3 As shown, the electronic device 300 may include: at least one processor 301, at least one network interface 304, user interface 303, memory 305, and at least one communication bus 302.

[0124] The communication bus 302 is used to enable communication between these components.

[0125] The user interface 303 may include a display screen and a camera.

[0126] The network interface 304 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0127] The processor 301 may include one or more processing cores. The processor 301 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 305, and by calling data stored in the memory 305. Optionally, the processor 301 may be implemented using at least one hardware form of digital signal processing, field-programmable gate array, or programmable logic array. The processor 301 may integrate one or more of the following: a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 301 and may be implemented as a separate chip.

[0128] The memory 305 may include random access memory (RAM) or read-only memory (ROM). Optionally, the memory 305 may include a non-transitory computer-readable medium. The memory 305 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 305 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch functionality, sound playback functionality, image playback functionality, etc.), instructions for implementing the various method embodiments described above, etc.; the data storage area may store data involved in the various method embodiments described above, etc. Optionally, the memory 305 may also be at least one storage device located remotely from the aforementioned processor 301. Figure 3 As shown, the memory 305, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for a vacuum device design method based on machine learning algorithms.

[0129] exist Figure 3 In the electronic device 300 shown, the user interface 303 is mainly used to provide an input interface for the user and to obtain the user input data; while the processor 301 can be used to call an application program stored in the memory 305 that is a vacuum device design method based on a machine learning algorithm. When executed by one or more processors, the electronic device performs one or more methods as described in the above embodiments.

[0130] An electronic device readable storage medium stores instructions that, when executed by one or more processors, cause the electronic device to perform one or more methods as described in the above embodiments.

[0131] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0132] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0133] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the shown or discussed mutual couplings or direct couplings or communication connections may be through some service interfaces; indirect couplings or communication connections between apparatuses or units may be electrical or other forms.

[0134] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0135] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0136] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0137] The above are merely exemplary embodiments of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of other embodiments of this disclosure upon considering the specification and the disclosure of practical truths. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure.

Claims

1. A vacuum device design method based on machine learning algorithms, characterized in that, The method includes: Obtain historical design data and standard design models for vacuum equipment; The historical design data is used as training data to train the standard design model. With the optimization goal of optimizing the overall cooling effect of the vacuum equipment, a first initial design model for cooling requirements and a first initial optimization scheme corresponding to the first initial design model are generated. The historical design data and the output of the first initial design model are used as training data to train the standard design model. The overall compressive strength of the vacuum equipment is taken as the optimization objective to generate a second initial design model for the strength requirement and a second initial optimization scheme corresponding to the second initial design model. Obtain the cooling and strength requirements of the vacuum equipment to be designed; If the cooling requirement parameter is greater than a preset first threshold and the intensity requirement parameter is not greater than a preset second threshold, the vacuum device to be designed is determined to be a cooling type vacuum device, the first initial optimization scheme is adjusted to a first special optimization scheme, and the first special optimization scheme is combined with the second initial optimization scheme to obtain the cooling type design scheme corresponding to the cooling type vacuum device. If the cooling requirement parameter is not greater than a preset first threshold and the intensity requirement parameter is greater than a preset second threshold, the vacuum equipment to be designed is determined to be an intensity-type vacuum equipment, the second initial optimization scheme is adjusted to a second special optimization scheme, and the first initial optimization scheme and the second special optimization scheme are combined to obtain the intensity-type design scheme corresponding to the intensity-type vacuum equipment. If the cooling requirement parameter is greater than a preset first threshold and the intensity requirement parameter is greater than a preset second threshold, the vacuum device to be designed is determined as a balanced vacuum device, the first initial optimization scheme is adjusted to a first special optimization scheme, and the second initial optimization scheme is adjusted to a second special optimization scheme; and the first special optimization scheme and the second special optimization scheme are combined to obtain the balanced design scheme corresponding to the balanced vacuum device.

2. The method according to claim 1, characterized in that, The step of adjusting the first initial optimization scheme to a first special optimization scheme includes: Obtain the surface structure parameters and water cooling pipe parameters of the cooling vacuum equipment; The area where the cooling demand parameter is greater than the first threshold is designated as the first special area, and based on the surface structure parameter and the water cooling pipe parameter, a first regular area adjacent to the first special area and whose water cooling pipe arrangement affects the cooling effect of the first special area is determined. Input the cooling requirement parameters into the first initial design model and output the first pipeline layout scheme for the first special area; Based on the first pipeline layout scheme as a constraint, multiple second pipeline layout schemes are generated for the first conventional area; The first pipeline layout scheme and each of the second pipeline layout schemes are input into the heat conduction simulation system, and the cooling effect index corresponding to each of the second pipeline layout schemes is output. The second pipeline layout scheme with the best cooling effect index is combined with the first pipeline layout scheme to generate the first special optimization scheme of the cooling vacuum equipment.

3. The method according to claim 2, characterized in that, The step of determining a first conventional region adjacent to the first special region and whose water-cooling pipe arrangement affects the cooling effect of the first special region based on the surface structure parameters and the water-cooling pipe parameters includes: The boundary coordinates of the first special region are determined based on the surface structure parameters, and the heat conduction range parameters of the water-cooling pipe are determined based on the water-cooling pipe parameters. Based on the boundary coordinates, candidate regions adjacent to the first special region are determined, and the heat conduction influence range of the candidate regions is determined according to the heat conduction range parameters of the water-cooled pipe. Calculate the overlap between the heat conduction influence range and the first special region; If the overlap is greater than a preset overlap threshold, the corresponding candidate region is determined as the first regular region.

4. The method according to claim 1, characterized in that, The step of adjusting the second initial optimization scheme to a second special optimization scheme includes: Obtain the structural parameters and structural stiffener parameters of the high-strength vacuum equipment; The region where the strength requirement parameter is greater than the second threshold is designated as the second special region, and based on the structural parameters, a second regular region is determined that is structurally related to the second special region and whose structural changes affect the strength of the second special region. Input the strength requirement parameters into the second initial design model, and output the first structural reinforcement scheme for the second special area; Based on the first structural reinforcement scheme as a constraint, multiple second structural reinforcement schemes are generated for the second conventional region; The first structural reinforcement scheme and each of the second structural reinforcement schemes are input into the structural mechanics simulation system, and the strength effect index corresponding to each of the second structural reinforcement schemes is output. The second structural reinforcement scheme with the best strength performance index is combined with the first structural reinforcement scheme to generate the second special optimization scheme of the strength-type vacuum equipment.

5. The method according to claim 4, characterized in that, The step of determining a second conventional region that is structurally related to the second special region and whose structural changes affect the strength of the second special region based on the structural parameters includes: Finite element meshes are generated for the second special region to obtain stress distribution data for each mesh element; Based on the stress distribution data, a stress transmission path diagram is constructed; Identify the main stress transmission paths in the stress transmission path diagram and determine the influence areas on the main stress transmission paths; Calculate the stress concentration factor of each of the affected regions, and determine the affected regions whose stress concentration factor is greater than a preset stress threshold as the second normal region.

6. The method according to claim 1, characterized in that, The step of training the standard design model using the historical design data as training data, with the optimization objective of achieving the optimal overall cooling effect of the vacuum equipment, generates a first initial design model for the cooling requirements and a first initial optimization scheme corresponding to the first initial design model, including: Extract the design parameters of the water pipe layout and the corresponding historical cooling effects from the historical design data; Using the design parameters as input features and the historical cooling effect as output labels, a water pipe layout training dataset is constructed. The water pipe layout training dataset is input into the standard design model, and the model is trained using the stochastic gradient descent algorithm. The goal is to minimize the error between the cooling effect predicted by the model and the historical cooling effect, thereby obtaining the first initial design model. With the goal of achieving optimal cooling effect, the output structure of the first initial design model is optimized using the variable density method to generate the first initial optimization scheme corresponding to the first initial design model.

7. The method according to claim 1, characterized in that, The step of using the historical design data and the output of the first initial design model as training data to train the standard design model, with the goal of maximizing the overall compressive strength of the vacuum equipment, generates a second initial design model and a second initial optimization scheme corresponding to the strength requirements, including: Extract the stiffener design parameters and corresponding historical strength effects from the historical design data; Using the stiffener design parameters as input features and the historical strength effect as output labels, a stiffener training dataset is constructed. The training dataset of the reinforcing ribs is input into the standard design model for training, with the goal of minimizing the error between the strength effect predicted by the model and the historical strength effect, to obtain the second initial design model; An optimization function is set with the goal of maximizing the overall compressive strength of the vacuum device. The output of the first initial design model is used as the geometric constraint condition for the layout of the reinforcing ribs. A two-way progressive structural optimization method is used for iterative optimization to generate a second initial optimization scheme corresponding to the second initial design model.

8. A vacuum device design system based on machine learning algorithms, characterized in that, The system includes: The initialization module is used to acquire historical design data and standard design models of vacuum equipment. The first training module is used to train the standard design model with the historical design data as training data, and to generate a first initial design model for cooling requirements and a first initial optimization scheme corresponding to the first initial design model with the optimization goal of optimizing the overall cooling effect of the vacuum equipment. The second training module is used to train the standard design model with the historical design data and the output results of the first initial design model as training data, with the overall compressive strength of the vacuum equipment as the optimization objective, to generate a second initial design model for the strength requirements and a second initial optimization scheme corresponding to the second initial design model. The input module is used to obtain the cooling and strength requirements of the vacuum equipment to be designed. A cooling-type scheme output module is used to determine the vacuum device to be designed as a cooling-type vacuum device if the cooling requirement parameter is greater than a preset first threshold and the intensity requirement parameter is not greater than a preset second threshold, adjust the first initial optimization scheme to a first special optimization scheme, and combine the first special optimization scheme with the second initial optimization scheme to obtain the cooling-type design scheme corresponding to the cooling-type vacuum device. The intensity-type scheme output module is used to determine the vacuum equipment to be designed as an intensity-type vacuum equipment if the cooling requirement parameter is not greater than a preset first threshold and the intensity requirement parameter is greater than a preset second threshold, adjust the second initial optimization scheme to a second special optimization scheme, and combine the first initial optimization scheme and the second special optimization scheme to obtain the intensity-type design scheme corresponding to the intensity-type vacuum equipment. The balanced solution output module is used to determine the vacuum device to be designed as a balanced vacuum device if the cooling requirement parameter is greater than a preset first threshold and the intensity requirement parameter is greater than a preset second threshold, adjust the first initial optimization scheme to a first special optimization scheme, adjust the second initial optimization scheme to a second special optimization scheme according to the output result of the first special optimization scheme, and combine the first special optimization scheme and the second special optimization scheme to obtain the balanced design scheme corresponding to the balanced vacuum device.

9. A computer storage medium, characterized in that, The computer storage medium stores a plurality of instructions, which are adapted to be loaded by a processor and executed as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, The device includes a processor, a memory, and a transceiver, wherein the memory is used to store instructions, the transceiver is used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 7.

Images