An optimization method and processor for a plate shell structure

Abstract

The application provides a kind of plate shell structure optimization method and processor, it is related to structural design technical field, a kind of plate shell structure optimization method, including execution finite element analysis and strain stress analysis, fine shell surface model is divided into several optimization units, optimization unit mapping matrix is established, then stress ratio matrix is generated in combination with stress distribution result, obtain the stress distribution result of fine shell surface model, compare design allowable stress value range with each optimization unit stress value size, after the optimization unit of average stress value substandard is optimized, form optimization model, when optimizing, extract over-standard unit and underload unit, according to stress difference classification, with grid size as the thickness of coating thickness reference, apply positive coating to over-standard unit, apply reverse coating to underload unit, preset multi-working condition influence coefficient, give working condition weight to the thickness of unit after coating by weight calculation, take the thickness value of thicker after weighting as final unit thickness, obtain optimization model.

Description

Technical Field

[0001] This invention relates to the field of structural design technology, and more specifically, to an optimization method and processor for plate and shell structures. Background Technology

[0002] In the process of 3D structural design, designers need to optimize the dimensions of each structure in the original design model. In existing structural optimization techniques, the original 3D model is usually constructed first, then imported into pre-processing software for shell surface processing. After obtaining stress and deformation results through finite element analysis, structural adjustments are made to areas that do not meet the optimization objectives (such as weight reduction, stiffness, etc.).

[0003] While existing optimization techniques can achieve a certain degree of structural optimization, they are insufficient to meet the design requirements of high-efficiency, high-requirement engineering projects within a short design cycle. Conventional optimization techniques suffer from low modeling efficiency, typically employing non-parametric modeling methods. This requires manual model reconstruction for each optimization, such as altering model thickness and spatial relationships, hindering the rapid tracking of design changes and significantly reducing design efficiency. Furthermore, inherent flaws in the modeling and shell surface processing workflow lead to poor shell element mesh quality, prone to distortion and discontinuity during mesh generation, resulting in insufficient computational accuracy or excessively long computation cycles. Additionally, existing optimization algorithms have significant limitations, struggling to achieve an optimal balance between meeting structural stress requirements and reducing equipment weight, creating a significant conflict between refined analysis and design cycle time. These issues severely restrict the efficiency and accuracy of structural optimization, failing to meet the demands of efficient iterative engineering design. Summary of the Invention

[0004] This invention aims to solve the problem that insufficient efficiency and accuracy in structural optimization cannot meet the needs of efficient iteration in engineering design.

[0005] To address the above problems, this invention provides a method for optimizing a plate and shell structure, comprising: S1. Construct a parametric 3D model based on the plate and shell structure parameters, import it into preprocessing software to convert it into an initial shell surface model, assign material properties, set mesh parameters and connection relationships to obtain a refined shell surface model, and apply external loads and boundary conditions to the refined shell surface model. S2. Perform finite element analysis and strain-stress analysis, divide the refined shell model into several optimization elements, establish the optimization element mapping matrix, and then combine the stress distribution results to generate the stress ratio matrix to obtain the stress distribution results of the refined shell model. S3. Preset an allowable stress value range, compare the design allowable stress value range with the stress value of each optimization unit, optimize the optimization units whose average stress value does not meet the standard, form an optimization model and return to S2; During optimization, out-of-range and under-loaded elements are extracted, classified according to stress difference, and the mesh size is used as the coating thickness reference. Forward coating is applied to out-of-range elements and reverse coating is applied to under-loaded elements. Multiple working condition influence coefficients are preset, and working condition weights are assigned to the thickness of the coated unit through weight calculation. The thicker thickness value after weighting is taken as the final unit thickness to obtain the optimization model. S4. Export the optimization unit number and update the 3D model parameters, and mark the modified nodes; S5. Output the final optimized model, import the modified parameters into the solid model, back up the source geometric solid model, and modify the design parameters.

[0006] The present invention provides an optimization method for plate and shell structures, which, compared with the prior art, has, but is not limited to, the following beneficial effects: Parametric modeling was adopted during the 3D model design stage and imported into pre-processing software to be converted into a plate and shell model (i.e., shell surface model). This not only greatly shortened the design cycle, but also more clearly showed the affected areas in the design-analysis. The stress distribution results obtained from finite element analysis were used to classify the optimization elements. The mesh size was used as the basis for the coating thickness to avoid the influence of mesh distortion and reduce calculation errors. This not only ensured the completeness and scientific nature of the analysis results, but also greatly reduced the computational complexity in the analysis and design.

[0007] The allowable stress in the design is extended from a single numerical value in existing technologies to a range of allowable stresses. Optimization elements can be divided into overloaded and underloaded elements, with corresponding coatings that can be positive or negative, meaning the coating thickness can have positive or negative values, thus ensuring the superposition of multiple optimization calculations. At the end of each optimization calculation, a working condition influence coefficient is added to ensure the model's safety and reliability under multiple working conditions. Each optimized model is then re-imported into the finite element analysis calculation until no more optimization elements remain. Finally, the final optimized model is returned to the 3D model, and design parameters are modified, thus ensuring a complete closed-loop process of design-analysis-design. This method not only makes the design settlement results clearer, more accurate, and more reliable but also effectively shortens the design cycle, meeting the high-efficiency iterative needs of engineering design.

[0008] Furthermore, in S2, each mesh cell is treated as an optimization cell. Each optimization cell is numbered, and an optimization cell mapping matrix is ​​established based on the mapping relationship between the optimization cells and the shell.

[0009] Furthermore, in S3, when comparing the allowable stress range with the stress values ​​of each optimized unit, it is necessary to determine whether the average stress value of each optimized unit meets the standard. When the average stress value of the optimized element meets the target, proceed to S4; When the average stress value of the optimized element does not meet the standard, parameter optimization is required. After optimization, return to S2 and repeat the finite element analysis on the optimized model until there are no unqualified elements, then proceed to S4.

[0010] Furthermore, if the average stress value of the optimized element exceeds the allowable design stress value range, the optimized element is considered an out-of-specification element, and the following steps are initiated: S31. Apply a positive coating to the substandard unit; S311. Number the areas of each unit that exceed the standard in sequence, that is, take 1 / m of the difference between the maximum stress value and the maximum allowable stress value, divide them into m parts according to the size of the shell surface stress value, each part is a level, and there are a total of m levels. S312. Arrange the elements in the order of 1 to k according to the stress difference in the stress grading region from small to large.

[0011] Furthermore, S31 also includes: S313. Extract the shell thickness within the stress grading region. On the basis of the original shell, cover it with a thickened film with a thickness of f, where f > 0, and increase from there. For each additional layer, kf is added to the region.

[0012] Furthermore, S31 also includes: S314. Multiply the coating thickness by the weighting coefficient under the current operating conditions; S315. When there is a coating overlap in the coating direction of the optimization unit, determine the overlap direction of the optimization unit, calculate the fillet radius of the cross-overlapping unit according to the overlap area and add fillets, and determine whether to merge or add ribs for the same-direction overlapping unit according to the area ratio, and return to S2.

[0013] Furthermore, when the average stress value of the optimized element is less than the allowable design stress value range, the optimized element is an underloaded element, and the following steps are performed: S32. Apply reverse coating to the underloaded unit; S321. Number the underloaded areas of each underloaded unit in sequence, that is, take 1 / n of the difference between the maximum stress value and the maximum allowable stress value, divide it into n parts according to the size of the shell surface stress value, each part is a level, and there are a total of n levels. S322. Arrange the elements in the order of (-1) to (-k) according to the stress difference in the stress grading region from large to small.

[0014] Furthermore, S32 also includes: S323. Extract the shell thickness within the stress grading region. On the basis of the original shell, cover it with a thickened film with a thickness of e, where e < 0. The thickness increases from e to e. For each additional layer, ne is added to the region.

[0015] Furthermore, S32 also includes: S324. Multiply the coating thickness by the weighting coefficient under the current operating conditions and return to S2.

[0016] The present invention also provides a processor applicable to the optimization method of the plate and shell structure as described above.

[0017] Since the technological improvements and beneficial effects of the processor are at least the same as those of the optimization method of the shell structure, the processor will not be described in detail here. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the finite element method for structural optimization in existing technologies. Figure 2 This is a flowchart illustrating the optimization method for a plate and shell structure according to an embodiment of the present invention under a single working condition. Figure 3 This is a structural diagram of a rectangular plate unit according to an embodiment of the present invention; Figure 4 for Figure 2 A detailed flowchart of step S31 is shown below; Figure 5 for Figure 2 A detailed flowchart of step S32 is shown below; Figure 6 This is a flowchart illustrating the optimization method for a plate and shell structure according to an embodiment of the present invention under multiple operating conditions. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings showing multiple embodiments according to this application. It should be understood that the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments described in this application without creative effort will fall within the scope of protection of this application.

[0020] Unless otherwise defined, all technical and scientific terms used in this application have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used in the description of this application is for the purpose of describing specific embodiments only and is not intended to limit this application; the terms "comprising," "including," "having," "containing," etc., in the description, claims, and accompanying drawings of this application are open-ended terms. Therefore, "comprising," "including," or "having" refers to, for example, a method or apparatus having one or more steps or elements, but is not limited to having only these one or more elements. The terms "first," "second," etc., in the description, claims, or accompanying drawings of this application are used to distinguish different objects, not to describe a specific order or hierarchy. 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 number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0021] It should be emphasized that when the term "comprising / including" is used in this specification, it is used to explicitly indicate the presence of the stated feature, integer, step, or component, but does not exclude the presence or addition of one or more other features, integers, steps, parts, or groups of features, integers, steps, or parts.

[0022] In this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, in this application, the character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0023] See Figure 1 Currently, the commonly used finite element structural optimization method generally obtains the displacement and stress results of the model through finite element calculation, imports them into the optimization environment, and then defines the objective function and constraints to optimize the model boundary, thereby converging to the minimum of structural flexibility or volume.

[0024] Existing technologies use preprocessing software to manually convert 3D models into analysis models. Furthermore, during topology optimization, they focus more on whether the overall structure of the model meets the requirements for stiffness, strength, and self-weight, and the optimization calculation cycle is long.

[0025] Specifically, existing technologies employ non-parametric modeling, requiring manual model reconstruction for each optimization, such as changing model thickness and spatial relationships. This makes it difficult to quickly track design change requirements, significantly reducing design efficiency. This "manual reconstruction" method cannot guarantee the consistency of model geometry: when manually adjusting thickness and spatial position, it is difficult to accurately control the boundary transitions (such as surface connections, rounded corners) and topological relationships (such as the connection accuracy of adjacent structures) of the 3D model, easily leading to "non-standard geometric boundaries" (such as local protrusions and boundary misalignments). Moreover, post-processing shell conversion, which involves building the 3D model first and then importing it into pre-processing software for shell surface processing, cannot match the original geometric topology of the 3D model based on parametric logic. When faced with irregular 3D boundaries, the pre-processing software is prone to generating abnormally shaped elements (such as elements with excessively large elongation ratios or acute-angled elements), i.e., mesh distortion, when automatically meshing.

[0026] Shell surface processing is performed only after the original 3D model has been constructed. It is a post-processing step that is independent of the modeling stage. This process design makes it difficult to guarantee topological continuity and results in a lack of linkage between mesh updates.

[0027] The difficulty in ensuring topological continuity is specifically manifested in the following ways: when a 3D model is converted into a shell surface, if there are geometric connection errors in the original model (such as the contact surfaces of adjacent parts not coinciding), the shell surface model is prone to topological breaks (such as the nodes of adjacent shell surfaces not coinciding or gaps existing), which in turn leads to discontinuous meshes (such as mesh gaps or local overlaps) during mesh generation.

[0028] The lack of linkage in mesh updates is specifically manifested in the following ways: Because it is a non-parametric model, if the mesh quality is substandard (such as distortion), it cannot be quickly optimized by adjusting parameters (such as linking the mesh size and cell type). It can only be manually re-meshed. However, when manually re-meshing, it is difficult to ensure the consistency of the global mesh. Even after local repairs, there may still be problems such as poor cell shape and discontinuous nodes.

[0029] The aforementioned mesh quality issues can directly lead to a chain reaction of insufficient computational accuracy and excessively long computation cycles. Distorted elements increase the error in solving the stiffness matrix of finite element calculations. Specifically, the more the element shape deviates from the ideal state (e.g., a quadrilateral element deviates from a square), the worse the accuracy of the simulated structural stress. The final output stress and deformation results deviate significantly from actual working conditions, causing accuracy problems. Discontinuous meshes can lead to difficulties in iterative convergence during finite element analysis. This means the software needs to handle computational conflicts caused by mesh gaps and overlaps, or increase the number of iterations due to failure to achieve convergence. Furthermore, in non-parametric models, mesh quality issues cannot be quickly resolved through parameter linkage; repeated manual model adjustments and re-meshing are required, further extending the computation cycle.

[0030] Furthermore, the optimization logic of existing technologies prioritizes unidirectional objectives and fails to establish a synergistic balance between achieving stress targets and reducing overall weight. This deficiency in the non-parametric modeling process directly limits the effectiveness of optimization algorithms. The core value of optimization algorithms lies in finding the optimal solution through a cycle of "iterative parameter adjustment → performance verification → readjustment." However, non-parametric modeling prevents algorithms from quickly and accurately adjusting structural parameters (such as element thickness). When manually reconstructing the model, it is difficult to make differentiated adjustments based on subtle differences in stress distribution (such as local high-stress areas and low-stress redundant areas), resulting in only a coarse-grained approach (such as uniformly thickening / thinning a certain area). For example, if only local elements in a certain area exceed the stress limit, but the entire area needs to be thickened, it leads to redundant weight of non-exceeding-stress elements, ultimately failing to find a precise balance between achieving local stress targets and reducing overall weight.

[0031] Existing optimization algorithms, in order to balance stress and weight, rely on "refined analysis," which directly leads to excessively long computation cycles. If refined optimization is chosen, the stress and thickness of a large number of elements must be individually matched and adjusted to find a state where stress meets the standard and weight is minimized. However, the dramatic increase in computational load causes the design cycle to far exceed engineering requirements, making it difficult to implement in practical applications. If simplified optimization is chosen, to control the computation cycle, the optimization dimensions can only be reduced (such as reducing the number of element analyses and simplifying stress judgment criteria), resulting in coarse optimization results. Either the stress is redundant (too heavy) or the weight is too light (stress not meeting the standard), failing to achieve the optimal balance.

[0032] This contradiction between precision and timeliness means that existing algorithms can only compromise between the two, rather than achieving a true optimal balance.

[0033] See Figure 2 An optimization method for a plate shell structure according to an embodiment of the present invention includes: S1. Construct a parametric 3D model based on the plate and shell structure parameters, import it into preprocessing software to convert it into an initial shell surface model, assign material properties, set mesh parameters and connection relationships to obtain a refined shell surface model, and apply external loads and boundary conditions to the refined shell surface model.

[0034] S2. Perform finite element analysis and strain-stress analysis, divide the refined shell model into several optimization elements, establish the optimization element mapping matrix, and then combine the stress distribution results to generate the stress ratio matrix to obtain the stress distribution results of the refined shell model.

[0035] In this process, each grid cell is treated as an optimization cell. Each optimization cell is numbered, and an optimization cell mapping matrix is ​​established based on the mapping relationship between the optimization cells and the plate and shell.

[0036] See Figure 3The local coordinate system of the rectangular plate element is established by taking the two axes of symmetry of the rectangular plate element as the x and y axes, and the normal direction of the plate surface as the z axis. The displacements along the x, y and z axes are assumed to be u, v and w, respectively.

[0037] Using the four vertices of the rectangular plate element as nodes, each node k has three linear displacement components u, v, and w, as well as the angular displacement of the normal at point k. , and The entire unit has a total of 24 degrees of freedom.

[0038] The nodes of the constructed shell model are numbered and summarized into a node coordinate matrix:

[0039] Where i represents the node with node number i. Let x, y, and y be the x, y, and y coordinates of the node.

[0040] The element matrix of the plate and shell model is:

[0041] Where k represents the number of the plate / shell unit, Indicates the starting node number of the plate and shell element. Indicates the node number of the end point of the plate and shell unit. Indicates the element type of the plate / shell element. This indicates the material properties of the plate and shell unit.

[0042] The plate and shell model is divided into several optimization elements, each mesh element is called an optimization element, and each optimization element is numbered. An optimization element mapping matrix is ​​established based on the mapping relationship between the optimization elements and the plate and shell.

[0043] The optimization units are numbered starting from 0, and k in the matrix represents the optimization unit numbered k. , , This indicates that the number of optimization units within the surface will vary depending on the surface's location.

[0044] The loads applied to the model are represented by the load matrix as follows:

[0045] in, Indicates the element number to which the load is applied. This represents the degree of freedom applied by the load on element k. This indicates the magnitude of the load.

[0046] The constraints imposed on the model are represented by the boundary constraint matrix:

[0047] Similar to the load matrix, This indicates the value that is constrained.

[0048] Based on the finite element analysis results, the stress ratios of all plates and shells are obtained, and a stress ratio matrix is ​​generated based on the optimized element mapping matrix.

[0049] The stress ratio matrix of the plate and shell element is obtained from the finite element method:

[0050] in, The value is the value in the optimization unit mapping matrix. The stress ratio is the stress ratio, where each value in the stress ratio matrix represents the stress value of the plate shell in an optimization element.

[0051] S3. Preset an allowable stress value range, compare the design allowable stress value range with the stress value of each optimized unit, and determine whether the average stress value of each optimized unit meets the standard.

[0052] When the average stress value of the optimized unit meets the standard, proceed to S4.

[0053] When the average stress value of the optimized element does not meet the standard, parameter optimization is required. After optimization, return to S2 and repeat the finite element analysis on the optimized model until there are no unqualified elements, then proceed to S4.

[0054] During optimization, out-of-range and under-loaded elements are extracted, classified according to stress difference, and the mesh size is used as the coating thickness reference. Forward coating is applied to out-of-range elements, and reverse coating is applied to under-loaded elements.

[0055] Specifically as follows: See Figure 2 , Figures 4-5 When the average stress value of the optimized unit exceeds the allowable design stress value range, the optimized unit is considered an out-of-specification unit, and the following steps are initiated: S31. Apply positive coating to units that exceed the standard.

[0056] S311. Number the out-of-standard areas of each out-of-standard unit in sequence, that is, take 1 / m of the difference between the maximum stress value and the maximum allowable stress value, divide it into m parts according to the size of the shell surface stress value, each part is a level, and there are a total of m levels.

[0057] S312. Arrange the elements in the order of 1 to k according to the stress difference in the stress grading region from small to large.

[0058] S313. Extract the shell thickness within the stress grading region. On the basis of the original shell, cover it with a thickened film with a thickness of f, where f > 0, and increase from there. For each additional layer, kf is added to the region.

[0059] S314. Multiply the coating thickness by the weighting coefficient under the current operating conditions; S315. When there is a coating overlap in the coating direction of the optimization unit, determine the overlap direction of the optimization unit, calculate the fillet radius of the cross-overlapping unit according to the overlap area and add fillets, and determine whether to merge or add ribs for the same-direction overlapping unit according to the area ratio, and return to S2.

[0060] When the average stress value of the optimized element is less than the allowable design stress value, the optimized element is an underloaded element, and the following steps are performed: S32. Apply reverse coating to the underloaded unit.

[0061] S321. Number the underloaded areas of each underloaded unit in sequence, that is, take 1 / n of the difference between the maximum stress value and the maximum allowable stress value, divide it into n parts according to the size of the shell surface stress value, each part is a level, and there are a total of n levels.

[0062] S322. Arrange the elements in the order of (-1) to (-k) according to the stress difference in the stress grading region from large to small.

[0063] S323. Extract the shell thickness within the stress grading region. On the basis of the original shell, cover it with a thickened film with a thickness of e, where e < 0. The thickness increases from e to e. For each additional layer, ne is added to the region.

[0064] S324. Multiply the coating thickness by the weighting coefficient under the current operating conditions and return to S2.

[0065] In the specific implementation of step S3, it is still based on Figure 3 Taking the rectangular plate element as an example, the magnitudes of each stress value in the stress matrix are compared with the design stress value (i.e., the preset allowable stress range). The design stress value is: [ , ].

[0066]

[0067] Extracting more than of The corresponding optimization unit number is the "coating" unit thickness optimization unit in the plate and shell structure model.

[0068]

[0069] in, It is expressed as the stress difference for each stress level.

[0070] Each unit level is numbered sequentially from 1 to kk, such as ,in The most basic level, that is The cell in question. Extract cell thickness. The finite element mesh size is defined as the coating thickness f, and the updated mesh size is... The unit thickness is :

[0071] And so on, when... When calculating the coating thickness of a unit, it can be expressed as: :

[0072] The coating optimization of each optimization unit on the surface is completed sequentially.

[0073] Conversely, extract stress values ​​below the standard value from the matrix. of The corresponding optimization unit number is the "coating" unit thickness optimization unit in the plate and shell structure model.

[0074]

[0075] in, Represented as the stress difference at each stress level.

[0076] Each unit level is numbered sequentially from -1 to -kk, such as ,in The most basic level, that is The cell in question. Extract cell thickness. The finite element mesh size is defined as the coating thickness e, and the updated mesh size is... The unit thickness is :

[0077] And so on, when... When calculating the coating thickness of a unit, it can be expressed as: :

[0078] The coating optimization of each optimization unit on the surface is completed sequentially.

[0079] It should be noted that, 'e' and 'e' are scalars of the same type and can be superimposed. For example, Unit in the Secondary optimization of coating thickness The unit thickness is:

[0080] The required coating thickness in the i-th optimization calculation Therefore, the thickness of the coating on the unit this time is:

[0081] Therefore, the coating calculation method can overcome the limitation of previous incremental structures that could not modify the optimized areas again, increasing the flexibility of structural optimization.

[0082] The above describes the coating process under single operating conditions.

[0083] Under multiple operating conditions, it is necessary to preset the multi-operating condition influence coefficient, assign operating condition weights to the thickness of the coated unit through weight calculation, and take the thicker thickness value after weighting as the final unit thickness to obtain the optimization model.

[0084] See Figure 6 Finally, in this embodiment, multi-condition analysis is performed. During the initial parametric modeling, the influence coefficients of multiple operating conditions on the calculation model are designed. Effective weights. The calculation formula is:

[0085] in, This represents the input weighting coefficients for the analytical calculation of the working conditions.

[0086] After multiplying the thickness of each element that has undergone previous coating optimization by a weighting factor, the thicknesses are compared, and the thicker thickness value is selected. For example, in an element... corresponding Operating conditions and Operating conditions:

[0087] in, This is represented as the unit thickness value after weight calculation.

[0088] Specifically, when the coating of the optimized unit overlaps in the coating direction, this situation exists only in the out-of-standard unit.

[0089] First, based on the coordinate direction of the optimized elements, determine the overlap direction of the optimized elements. If the optimized elements overlap intersectingly, then the fillet radius between the two surfaces needs to be determined based on the size of the overlap area, and fillets are added accordingly to effectively reduce stress concentration. When the optimized elements overlap in the same direction in the coating direction, then the area M of the surface where the element is located and the overlap area M of the optimized elements within this surface need to be considered. kThe proportion of the surface determines whether it needs to be integrated into a single surface or reinforced between two surfaces.

[0090]

[0091] This modification applies to all simulated operating conditions.

[0092] Finally, the optimized elements with the "coating" are re-imported into the finite element method for calculation. A new structural stress ratio matrix is ​​obtained, and it is checked whether there are still elements in the model that exceed the design stress range. If so, the above optimization process is repeated for the model until the plate and shell cannot be optimized again or all optimized elements meet the design requirements.

[0093] S4. Export the optimization unit number and update the 3D model parameters, and mark the modified nodes.

[0094] Export the optimization unit numbers from each step and modify the structural parameters corresponding to the optimization unit positions in the structure. This facilitates modification and marking of the parametric modeling regions corresponding to the optimization unit positions, making subsequent modifications and process documentation easier.

[0095] S5. Output the final optimized model, import the modified parameters into the solid model, back up the source geometric solid model, and modify the design parameters.

[0096] Specifically, modifying parameters such as the element thickness and fillet radius after coating, and using these as marker parameters, allows for the modification of design parameters in the source geometric solid model (i.e., the initial parametric 3D model), facilitating later modifications and verification by designers. Furthermore, these marker parameters also represent stress-sensitive areas of the model structure and should be given special attention as design parameters.

[0097] In this embodiment, parametric modeling is employed during the 3D model design stage and imported into pre-processing software to transform it into a shell model (i.e., a shell surface model). This not only significantly shortens the design cycle but also more clearly demonstrates the affected areas in the design-analysis process. The stress distribution results obtained from finite element analysis are used to classify the optimized elements. Using the mesh size as the basis for coating thickness not only avoids the influence of mesh distortion and reduces calculation errors but also allows for precise control of the thickness of each element at the microscale of the shell surface. Coating is performed according to the stress value of the elements, ensuring the completeness and scientific validity of the analysis results while significantly reducing the computational complexity in the analysis and design process.

[0098] The allowable stress in the design is extended from a single value in existing technologies to a range of allowable stresses. Optimization elements can be divided into overloaded and underloaded elements, with corresponding coatings that can be positive or negative, meaning the coating thickness can have positive or negative values, thus ensuring the superposition of multiple optimization calculations. At the end of each optimization calculation, a working condition influence coefficient is added to ensure the model's safety and reliability under multiple working conditions, broadening the applicability of the optimization results. Each optimized model is then re-imported into the finite element analysis calculation until no more optimization elements remain. Finally, the final optimized model is returned to the 3D model, and newly created design parameters are modified, ensuring a complete closed-loop design-analysis-design process. This method not only makes the design settlement results clearer, more accurate, and more reliable but also effectively shortens the design cycle, achieving an optimal balance between structural stress compliance and equipment weight reduction, meeting the needs of efficient iterative engineering design.

[0099] After each iteration of optimization is completed, the analysis and calculation are returned. The optimization process ends when the region that has reached the optimization target is reached. The modified points in the optimization model are used as model parameter variables to facilitate later verification and learning optimization from similar models.

[0100] Another embodiment of the present invention also provides a processor, including the aforementioned method for optimizing the shell structure.

[0101] Since the technological improvements and beneficial effects of the processor are at least the same as those of the optimization methods for the shell structure, the application methods of the optimization methods for the shell structure will not be elaborated further.

[0102] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. An optimization method for a plate and shell structure, characterized in that, include: S1. Construct a parametric 3D model based on the plate and shell structure parameters, import it into preprocessing software to convert it into an initial shell surface model, assign material properties, set mesh parameters and connection relationships to obtain a refined shell surface model, and apply external loads and boundary conditions to the refined shell surface model. S2. Perform finite element analysis and strain-stress analysis, divide the refined shell model into several optimization elements, establish the optimization element mapping matrix, and then combine the stress distribution results to generate the stress ratio matrix to obtain the stress distribution results of the refined shell model. S3. Preset an allowable stress value range, compare the design allowable stress value range with the stress value of each optimization unit, optimize the optimization units whose average stress value does not meet the standard, form an optimization model and return to S2; During optimization, out-of-range and under-loaded elements are extracted, classified according to stress difference, and the mesh size is used as the coating thickness reference. Forward coating is applied to out-of-range elements and reverse coating is applied to under-loaded elements. Multiple working condition influence coefficients are preset, and working condition weights are assigned to the thickness of the coated unit through weight calculation. The thicker thickness value after weighting is taken as the final unit thickness to obtain the optimization model. S4. Export the optimization unit number and update the 3D model parameters, and mark the modified nodes; S5. Output the final optimized model, import the modified parameters into the solid model, back up the source geometric solid model, and modify the design parameters.

2. The optimization method for the plate and shell structure according to claim 1, characterized in that, In S2, each mesh element is treated as an optimization element. Each optimization element is numbered, and an optimization element mapping matrix is ​​established based on the mapping relationship between the optimization elements and the shell.

3. The optimization method for the plate and shell structure according to claim 1, characterized in that, In S3, when comparing the allowable stress range with the stress values ​​of each optimized unit, it is necessary to determine whether the average stress value of each optimized unit meets the standard. When the average stress value of the optimized element meets the target, proceed to S4; When the average stress value of the optimized element does not meet the standard, parameter optimization is required. After optimization, return to S2 and repeat the finite element analysis on the optimized model until there are no unqualified elements, then proceed to S4.

4. The optimization method for the plate and shell structure according to claim 3, characterized in that, When the average stress value of the optimized element exceeds the allowable design stress value, the optimized element is considered an out-of-range element, and the following steps are initiated: S31. Apply a positive coating to the substandard unit; S311. Number the areas of each unit that exceed the standard in sequence, that is, take 1 / m of the difference between the maximum stress value and the maximum allowable stress value, divide them into m parts according to the size of the shell surface stress value, each part is a level, and there are a total of m levels. S312. Arrange the elements in the order of 1 to k according to the stress difference in the stress grading region from small to large.

5. The optimization method for the plate and shell structure according to claim 4, characterized in that, S31 also includes: S313. Extract the shell thickness within the stress grading region. On the basis of the original shell, cover it with a thickened film with a thickness of f, where f > 0, and increase from there. For each additional layer, kf is added to the region.

6. The optimization method for the plate and shell structure according to claim 5, characterized in that, S31 also includes: S314. Multiply the coating thickness by the weighting coefficient under the current operating conditions; S315. When there is a coating overlap in the coating direction of the optimization unit, determine the overlap direction of the optimization unit, calculate the fillet radius of the cross-overlapping unit according to the overlap area and add fillets, and determine whether to merge or add ribs for the same-direction overlapping unit according to the area ratio, and return to S2.

7. The optimization method for the plate and shell structure according to claim 3, characterized in that, When the average stress value of the optimized element is less than the allowable design stress value, the optimized element is an underloaded element, and the following steps are performed: S32. Apply reverse coating to the underloaded unit; S321. Number the underloaded areas of each underloaded unit in sequence, that is, take 1 / n of the difference between the maximum stress value and the maximum allowable stress value, divide it into n parts according to the size of the shell surface stress value, each part is a level, and there are a total of n levels. S322. Arrange the elements in the order of (-1) to (-k) according to the stress difference in the stress grading region from large to small.

8. The optimization method for the plate and shell structure according to claim 7, characterized in that, S32 also includes: S323. Extract the shell thickness within the stress grading region. On the basis of the original shell, cover it with a thickened film with a thickness of e, where e < 0. The thickness increases from e to e. For each additional layer, ne is added to the region.

9. The optimization method for the plate and shell structure according to claim 8, characterized in that, S32 also includes: S324. Multiply the coating thickness by the weighting coefficient under the current operating conditions and return to S2.

10. A processor, characterized in that, The optimization method applicable to the plate and shell structure as described in any one of claims 1-9.

Images