Workpiece design method and device
By combining static and dynamic topology optimization algorithms and machine learning models, the topological features of aero-engine components are extracted and fused, solving the problem that existing technologies fail to effectively consider dynamic loads and achieving efficient component design.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-19
- Publication Date
- 2026-03-27
AI Technical Summary
Existing technologies only consider static loads in the design of aero-engine components, failing to effectively consider dynamic loads, resulting in unsatisfactory design results and lengthy design processes.
By combining static and dynamic topology optimization algorithms, a topology feature set is extracted through a machine learning model, and feature fusion is performed to construct an optimized topology model that meets multiple working conditions.
It improves the efficiency and effectiveness of aero-engine component design, reduces the number of iterations, and meets the requirements of static and dynamic loads.
Smart Images
Figure CN121744567A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present disclosure relates to the technical field of aero-engines, and particularly relates to a workpiece design method and device. BACKGROUND
[0002] The working load of an aero-engine has characteristics such as high temperature, high pressure and high speed. Moreover, when the aero-engine switches between different states, the working load thereof will change dramatically. In addition, the aero-engine also needs to withstand high-energy transient impact loads such as external object impact and internal structure failure impact. Such complex and variable loads bring difficulties to the design of parts of the aero-engine.
[0003] At present, when the parts of the aero-engine are structurally designed, the workpiece is generally structurally designed (or said to be modeled and optimized) by taking static working load as input, and then the designed structure is examined for impact dynamics limit load through test. By taking the above-mentioned manner, the following problems exist: since only static load conditions are considered when the workpiece is structurally designed, and dynamic load conditions are not considered, the structural design result of the part is not ideal. If a better structural design result is to be obtained, multiple rounds of iteration are needed, resulting in a long overall process. SUMMARY
[0004] The present disclosure provides a workpiece design method and device, which can solve the technical problems in the related art.
[0005] According to a first aspect of the present disclosure, a workpiece design method is provided, comprising: determining a design domain of a topological model of a workpiece; under the condition of loading static load conditions, optimizing the topological model of the workpiece in the design domain by using a static topology optimization algorithm to obtain a first optimization result; under the condition of loading dynamic load conditions, optimizing the topological model of the workpiece in the design domain by using a dynamic topology optimization algorithm to obtain a second optimization result; extracting a first set of topological features from the first optimization result and a second set of topological features from the second optimization result; fusing the first set of topological features and the second set of topological features to obtain a fused set of topological features; and constructing an optimized topological model of the workpiece according to the fused set of topological features.
[0006] In some embodiments, the extracting the first set of topological features from the first optimization result and the extracting the second set of topological features from the second optimization result comprises: performing feature recognition on the first optimization result and the second optimization result by using a machine learning model to obtain a first set of candidate features and a second set of candidate features; filtering out features in the first set of candidate features that are not included in a typical feature library to obtain the first set of topological features; and filtering out features in the second set of candidate features that are not included in the typical feature library to obtain the second set of topological features.
[0007] In some embodiments, the typical feature library is determined according to the following manner: selecting a feature library corresponding to the type of the workpiece from a plurality of feature libraries, and taking the feature library corresponding to the type of the workpiece as the typical feature library.
[0008] In some embodiments, the fusing the first set of topological features and the second set of topological features to obtain a fused set of topological features comprises: screening out matching features between the first set of topological features and the second set of topological features, the matching features being topological features located at the same position and of the same type; fusing a value of the matching features in the first set of topological features and a value of the matching features in the second set of topological features to obtain a final value of the matching features; and taking a set of matching features with the final value as the fused set of topological features.
[0009] In some embodiments, the determining a design domain of the topological model of the workpiece comprises: obtaining an engineering design requirement file of the workpiece; parsing the engineering design requirement file to obtain a plurality of design requirements of the workpiece; and determining the design domain of the topological model of the workpiece according to the plurality of design requirements of the workpiece.
[0010] In some embodiments, the design requirements comprise at least one of a size requirement, a weight requirement, a load requirement, a performance requirement, an installation requirement, a maintenance requirement, an economy requirement, a strength requirement, and a reliability requirement.
[0011] In some embodiments, the statics topological optimization algorithm is a variable density method or a level set method.
[0012] In some embodiments, the dynamics topological optimization algorithm is a cellular automaton method or an equivalent static load method.
[0013] In some embodiments, the workpiece is a part of an aero-engine.
[0014] According to a second aspect of the present disclosure, there is provided a workpiece design apparatus, comprising: a module for performing the workpiece design method as described above.
[0015] According to a third aspect of the present disclosure, there is provided an electronic device comprising: a memory; and a processor coupled to the memory, the processor being configured to perform the workpiece design method as previously described based on instructions stored in the memory.
[0016] According to a fourth aspect of the present disclosure, there is provided a computer-readable storage medium having stored thereon computer program instructions which, when executed by a processor, implement the workpiece design method as previously described.
[0017] According to a fifth aspect of the present disclosure, there is provided a computer program product having stored thereon computer program instructions which, when executed by a processor, implement the workpiece design method as previously described.
[0018] Other features and advantages of the present disclosure will become apparent from the following detailed description of exemplary embodiments thereof, taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS
[0019] The accompanying drawings, which constitute a part of this specification, illustrate embodiments of the present disclosure and serve to explain the principles of the present disclosure.
[0020] Figure 1 is a flowchart of a workpiece design method according to some embodiments of the present disclosure;
[0021] Figure 2 is a flowchart of a workpiece design method according to some other embodiments of the present disclosure;
[0022] Figure 3 is a schematic diagram of an FBO load spectrum according to some embodiments of the present disclosure;
[0023] Figure 4 is a schematic diagram of a fan shaft assembly relationship and force transmission path according to some embodiments of the present disclosure;
[0024] Figure 5a is a schematic diagram of a geometric model of a fan shaft according to some embodiments of the present disclosure;
[0025] Figure 5b is a schematic diagram of a finite element model of a fan shaft according to some embodiments of the present disclosure;
[0026] Figure 6a is a schematic diagram of maximum stress variation in a statics topology optimization process of a fan shaft according to some embodiments of the present disclosure;
[0027] Figure 6b is a schematic diagram of volume fraction variation in a statics topology optimization process of a fan shaft according to some embodiments of the present disclosure;
[0028] Figure 6cis a schematic diagram of maximum stress variation in a dynamics topology optimization process of a fan shaft according to some embodiments of the present disclosure;
[0029] Figure 6d is a schematic diagram of volume fraction variation in a dynamics topology optimization process of a fan shaft according to some embodiments of the present disclosure;
[0030] Figure 7 is a schematic diagram of a workpiece design device according to some embodiments of the present disclosure;
[0031] Figure 8 is a schematic diagram of a workpiece design device according to some embodiments of the present disclosure;
[0032] Figure 9 is a schematic diagram of an electronic device according to some embodiments of the present disclosure;
[0033] Figure 10 is a schematic diagram of a computer system according to some embodiments of the present disclosure.
[0034] The present disclosure can be more clearly understood with reference to the following detailed description in conjunction with the accompanying drawings. DETAILED DESCRIPTION
[0035] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that the relative arrangement of components and steps, numerical expressions, and numerical values set forth in these embodiments are not limiting to the scope of the present disclosure unless specifically stated otherwise.
[0036] It should be understood, however, that the sizes of the components shown in the drawings are chosen for convenience only, and thus, are not necessarily drawn to scale.
[0037] The following description of at least one exemplary embodiment is merely exemplary in nature and is in no way intended to limit the scope of the present disclosure, its application, or uses.
[0038] Techniques, methods, and devices known to those of ordinary skill in the relevant art can not be discussed in detail herein. However, where appropriate, such techniques, methods, and devices can be considered as part of the described technology.
[0039] In all of the examples shown and discussed herein, any specific values should be interpreted as merely exemplary, and not as a limitation. Thus, other examples of exemplary embodiments can have different values.
[0040] It should be noted that like numbers and letters refer to like items throughout the drawings, and thus, once an item is defined in one drawing, it is not necessary that it be further discussed in the subsequent drawings.
[0041] In order to make the objectives, technical solutions and advantages of the present disclosure clearer, the present disclosure will be further described in detail below with reference to specific embodiments and in conjunction with the accompanying drawings.
[0042] In the related art, the structure design method of the workpiece has problems such as long design process and unsatisfactory design effect.
[0043] In view of this, the present disclosure proposes a workpiece design method and device to improve the effect of workpiece modeling optimization while efficiently performing workpiece modeling optimization.
[0044] Figure 1 is a flowchart of a workpiece design method according to some embodiments of the present disclosure. As shown in Figure 1 the workpiece design method includes steps S11 to S16.
[0045] Step S11: Determine the design domain of the topological model of the workpiece.
[0046] In some examples, the design domain is determined according to the following manner: according to the engineering design requirement information of the workpiece, the design domain of the topological model of the workpiece is determined. Wherein, the design domain of the topological model can be understood as the initial optimization space of the topological model.
[0047] Specifically, in the above example, the engineering design requirement file of the workpiece can be obtained first; the engineering design requirement file is parsed to obtain a plurality of design requirements of the workpiece; according to the plurality of design requirements of the workpiece, the design domain of the topological model of the workpiece is determined.
[0048] Wherein, the design requirement can include one or more of size requirement, weight requirement, load requirement, performance requirement, installation requirement, maintenance requirement, economic requirement, strength requirement, and reliability requirement. In specific implementation, different engineering design requirement files can be configured for different workpieces to meet diversified workpiece design requirements.
[0049] For example, when designing a fan shaft in an aero-engine, the size requirement, weight requirement, load requirement, and performance requirement are considered. Wherein, the size requirement can include requirements for parameters such as the outer diameter, length, and thickness of the fan shaft, the weight requirement can include requirements for the overall weight of the fan shaft; the load requirement can include requirements for parameters such as the type, size, and direction of the load applied on the fan shaft; the performance requirement can include requirements for the aerodynamic performance of the fan shaft. For example, some parts of the fan shaft cannot leak air, so when designing the structure of the fan shaft, although some positions in the optimization result can have no material, the wall thickness of the fan shaft at these positions cannot be 0.
[0050] In the embodiments of the present disclosure, the design domain of the topological model of the workpiece is determined according to the analysis result of the engineering design requirement file, so that the design domain can be determined automatically and intelligently, which helps to improve the efficiency of the workpiece design process compared with manually determining the design domain.
[0051] In some examples, the design domain is determined according to the type of the workpiece by querying a pre-set corresponding relationship between the workpiece and the design domain.
[0052] In some embodiments, the workpiece design method further comprises: constructing a topological model of the workpiece. For example, the topological model of the workpiece can be constructed according to the following manner: first, simplifying the geometric structure of the workpiece to obtain a simplified geometric model; and then, establishing a finite element model of the workpiece according to the simplified geometric model, and taking the finite element model as the topological model of the workpiece.
[0053] Taking a fan shaft as an example, when the geometric structure of the fan shaft is simplified, the influence of the bolt holes at the mounting position of the fan shaft and the fan disc and the actual structural features of the toothed sleeve at the rear end of the fan shaft can be ignored to obtain a simplified geometric model. Then, the simplified geometric model and the key parameters of the geometric model can be input into a finite element analysis module to obtain a finite element model of the fan shaft.
[0054] Step S12: under the condition of loading the static load case, a static topology optimization algorithm is used to optimize the topological model of the workpiece in the design domain to obtain a first optimization result.
[0055] The static load case can include a working load and a limiting load. For example, for an aero-engine, the working load is the load acting on the aero-engine during the flight of the airplane, mainly including the gravity of the engine rotor, the engine thrust and the rotor unbalanced vibration load. The flight process of the airplane can be divided into the engine starting and preheating stage, the taxiing stage, the take-off stage, the climbing stage, the cruising stage, the descending stage, the hovering stage, the taxiing and descending stage, the landing taxiing and shutdown stage. The rotating speed and thrust level are different in different stages, and the load acting on the aero-engine also has great difference, so the working load is not a single load, but a load spectrum.
[0056] The load spectrum, such as an airplane landing load and a limit climb load, is the maximum load allowed to be borne by a part during flight of the airplane, and these loads are defined as limit loads. The limit load still belongs to the category of normal loads, but the load magnitude is large, the occurrence frequency is low, and the duration is short. Generally, the strength criterion requirements for the working load and the limit load are that the maximum stress in the structure under the load is generally not allowed to be greater than the yield stress of the material, and no residual deformation is allowed after unloading. The limit load is usually used as the static strength design load of the part. Considering that the limit load is often difficult to accurately measure, a maximum load factor that can be borne by the workpiece structure can be used to define the limit load. The limit load is the basis for designing the maximum bearing capacity of the aero-engine, and needs to ensure that the aero-engine meets the-3σ failure probability.
[0057] Topology optimization is a mathematical method for optimizing material distribution in a given area according to given load conditions, constraint conditions and performance indicators. Topology optimization belongs to a kind of structural optimization.
[0058] In this step, the static load condition can be determined first, and then the topology optimization algorithm such as the variable density method, the level set method or the intelligent optimization algorithm is used to optimize the topology model of the workpiece in the design domain under the condition of loading the load condition, to obtain the static topology optimization result (i.e. the first optimization result).
[0059] Step S13: Under the condition of loading the dynamic load condition, the dynamic topology optimization algorithm is used to optimize the topology model of the workpiece in the design domain to obtain the second optimization result.
[0060] The dynamic load condition includes the limit load. For example, for the parts of the aero-engine, the limit load is an extreme load that the airplane will inevitably encounter with a very small probability, such as a bird strike load and a fan blade out (FBO) load. The limit load is usually characterized by transient vibration, that is, the vibration peak value is very large (which can reach several hundred g) but the duration is very short (generally within 100 ms).
[0061] For example, for the design of a fan blade, the static load generated by the rotating centrifugal force and the aerodynamic force needs to be loaded in step S12, and the impact dynamic load caused by bird strike needs to be loaded in step S13; for the design of an intermediate casing, the working load of the engine needs to be loaded in step S12, and the FBO load needs to be loaded in step S13.
[0062] In step S13, the dynamic load case can be determined first, and then the topological optimization algorithm such as cellular automaton, equivalent static load method, or intelligent optimization algorithm is used to optimize the topological model of the workpiece in the design domain under the condition of loading the load case, so as to obtain the impact dynamics optimization result (i.e., the second optimization result).
[0063] In some embodiments, in order to improve the efficiency of topological optimization, step S12 and step S13 are executed in parallel. In addition, in specific implementation, step S12 and step S13 can also be executed in sequence.
[0064] Step S14: Extracting the first topological feature set from the first optimization result and the second topological feature set from the second optimization result.
[0065] The first topological feature set and the second topological feature set can each include one or more topological features. The topological feature can be a basic topological structure in geometry. For example, the topological feature can be a flat plate, a circular arc plate, a circular hole, a square hole, a circular arc, a rib structure, or other structures. These structures can be controlled by parameterization to control the shape and size features. The topological feature can be further subdivided. For example, for a circular arc plate, the topological feature can be further divided into the thickness and density of the circular arc plate.
[0066] Step S14 can be implemented in various ways. The following describes two exemplary implementations.
[0067] In the first exemplary implementation, step S14 includes: using a machine learning model to perform feature recognition on the first optimization result to obtain the first topological feature set; and using the machine learning model to perform feature recognition on the second optimization result to obtain the second topological feature set. The machine learning model can be an artificial neural network model obtained by pre-training. By using the machine learning model to extract the topological feature, the precision extraction demand of various workpieces, especially complex workpieces, can be met, and the applicability of topological feature extraction is improved.
[0068] In the second exemplary implementation, step S14 includes: extracting the first topological feature set and the second topological feature set from the first optimization result according to a typical feature library corresponding to the type of the workpiece. For example, different typical feature libraries are set in advance for different workpieces. The typical feature library has one or more topological features. When step S14 is executed, the typical feature library corresponding to the type of the workpiece is directly used to accurately extract the topological feature from the optimization result of the workpiece. By using the pre-set typical feature library to extract the topological feature, the extraction step is simple, and the real-time performance of topological feature extraction is improved.
[0069] For example, for the design of the fan shaft, the first circular arc plate, the second circular arc plate, and the third circular arc plate are extracted from the first optimization result by step S14, and the fourth circular arc plate, the fifth circular arc plate, and the sixth circular arc plate are extracted from the second optimization result. The sizes and shapes of the circular arc plates extracted from the first optimization result and the second optimization result are different.
[0070] In the embodiments of the present disclosure, by performing topology feature extraction on the first optimization result and the second optimization result, the geometry model of the workpiece can be discretized, and preparation is made for subsequent efficient feature fusion processing.
[0071] Step S15: Fusion is performed on the first topology feature set and the second topology feature set to obtain a fused topology feature set.
[0072] In some embodiments, step S15 includes: screening out matching features between the first topology feature set and the second topology feature set, the matching features being topology features located at the same position and of the same type; fusing the values of the matching features in the first topology feature set and the values of the matching features in the second topology feature set to obtain final values of the matching features; and taking a set of the matching features with the final values as the fused topology feature set.
[0073] The values of the matching features can include the values of the description parameters of the topology features in one or more dimensions. For example, for the design of the fan shaft, the first circular arc plate and the fourth circular arc plate are matching features, and the values of the two matching features include the thickness and the density of the circular arc plate.
[0074] In specific implementation, before the values of the matching features are fused, a fusion rule can be set to make the fused topology feature meet the requirements of static load and dynamic load. For example, the fusion rule is to take the topology feature with a larger value of thickness or density in the matching features as the fused topology feature.
[0075] For example, for the design of the fan shaft, it is assumed that in the first topology feature set, the thickness of the first circular arc plate is 3 mm, the thickness of the second circular arc plate is 3.5 mm, and the thickness of the third circular arc plate is 4 mm; in the second topology feature set, the thickness of the fourth circular arc plate is 2 mm, the thickness of the fifth circular arc plate is 4 mm, and the thickness of the sixth circular arc plate is 3 mm. Moreover, the first circular arc plate and the fourth circular arc plate are matching features, the second circular arc plate and the fifth circular arc plate are matching features, and the third circular arc plate and the sixth circular arc plate are matching features. Then, the fused feature set can include the first circular arc plate, the fifth circular arc plate, and the third circular arc plate.
[0076] In the embodiments of the present disclosure, the above manner of topological feature fusion helps to improve the fusion effect of the topological features, so that the fused topological features can meet the requirements of both static loads and dynamic loads.
[0077] Step S16: constructing an optimized topological model of the workpiece according to the fused topological feature set.
[0078] In step S16, the first optimization result obtained in step S12 or the second optimization result obtained in step S13 can be adjusted in topological structure according to the fused topological feature set, so as to obtain a final optimized topological model.
[0079] In the embodiments of the present disclosure, by respectively performing static topological optimization and impact dynamic topological optimization on the workpiece, the topological model optimization results of the workpiece under various load conditions can be obtained. Then, by topological feature extraction and fusion processing, and according to the fused topological features, a final optimized topological model can be constructed, which can meet the requirements of both static loads and impact dynamic loads. In this way, the effect of topological optimization can be improved, and the number of repeated topological optimization and tests can be reduced, and the efficiency of the entire workpiece design process can be improved.
[0080] Figure 2 is a flowchart of a workpiece design method according to other embodiments of the present disclosure. As shown in Figure 2 , the workpiece design method includes steps 201 to 211.
[0081] Step 201: determining a design domain.
[0082] In some examples, in step 201, an engineering design requirement file of the workpiece is first obtained, and then the design domain of the workpiece is determined according to the engineering design requirement file of the workpiece.
[0083] In other examples, a user instruction carrying the design domain information of the workpiece configured by the user is first obtained, and then the design domain of the workpiece is determined according to the user instruction.
[0084] In some embodiments, the workpiece design method further includes constructing an initial topological model of the workpiece and determining the static loads and impact dynamic loads of the workpiece.
[0085] The static loads include working loads and limiting loads. The impact dynamic loads include extreme loads. For example, for the design of parts of an aero-engine, the extreme loads are extreme loads that the aircraft will inevitably encounter with a very small probability, such as bird impact loads, FBO loads, etc. The FBO load spectrum is as shown in Figure 3 .
[0086] The following takes the fan shaft as an example to illustrate the process of constructing the initial workpiece topological model.
[0087] The fan shaft on an aero-engine is usually used to connect the fan rotor and the low-pressure shaft of the aero-engine. The front end of the fan shaft is connected to the fan disc through bolts, the rear end of the fan shaft is connected to the low-pressure shaft through a spline, and the fan shaft is the main force transmission component between the fan rotor and the low-pressure shaft and is connected to the intermediate casing through a fulcrum to provide front support for the low-pressure rotor. In the normal working state, the fan shaft mainly transmits torque, and under extreme loads such as FBO loads, the fan shaft will bear the rotor unbalance load after the fan disc transmits the blade loss. Figure 4 An exemplary fan shaft assembly relationship and force transmission path of FBO load are shown. In Figure 4 , the fan shaft 401 is located between the fan rotor and the low-pressure shaft.
[0088] The construction of the initial topological model of the fan shaft includes: first simplifying the geometric structure of the fan shaft to obtain a simplified geometric model; and establishing a finite element model of the fan shaft according to the simplified geometric model, and taking the finite element model as the topological model of the fan shaft.
[0089] For example, when the geometric structure of the fan shaft is simplified, the influence of the fan shaft and the fan disc mounting position bolt hole is ignored, and the actual structural features of the spline at the rear end of the fan shaft are ignored, thereby obtaining the simplified geometric model of the fan shaft as shown in Figure 5a . Then, the geometric model of the fan shaft is input into a finite element analysis module to construct a finite element model of the fan shaft. Hexahedral solid elements can be used to construct the finite element model. The constructed finite element model is as shown in Figure 5b .
[0090] Step 202: Statics topology optimization.
[0091] In this step, the topological model of the workpiece is subjected to statics topology optimization under the condition of loading statics load cases to obtain a statics topology optimization result. When performing statics topology optimization, a target function and a constraint condition of statics topology optimization need to be set.
[0092] For example, for the design of the fan shaft, the main part of the fan shaft can be taken as the design domain, and the rigid shaft section and the spline region of the fan shaft can be taken as the non-design domain. The maximum load in each stage of the flight envelope of the aircraft can be selected as the statics load. In addition, the constraint condition of statics topology optimization can include that the peak stress of the fan shaft is less than the yield strength of the material and the volume fraction is less than or equal to 0.7, and the target function of statics topology optimization is to minimize the flexibility of the fan shaft.
[0093] In addition, in order to simulate the actual load state during the static topology optimization, the following processing can also be performed on the finite element model of the fan shaft: rigid shaft segments 501 are added at the fan disc and fan shaft mounting positions, axial forces, torques, bending moments, etc. are applied to the loading surfaces 502 of the rigid shaft segments; the corresponding rotating speed in the calculation state is applied; spring damping units are applied at the positions of the No. 1 support point 503 and the No. 2 support point 504 to simulate the support stiffness; since the No. 2 support point is a ball bearing, an axial displacement constraint is added at the No. 2 support point; the circumferential displacement is constrained at the toothed sleeve 505.
[0094] The stress change of the fan shaft during the dynamic topology optimization is shown in FIG. 12, and the volume fraction change is shown in FIG. 13. As shown in the figures, during the dynamic topology optimization, the maximum stress of the fan shaft tends to be the ultimate stress 1800 MPa, and the volume fraction tends to be 0.7. Figure 6a Figure 6b As shown in the figures, during the dynamic topology optimization, the maximum stress of the fan shaft tends to be the ultimate stress 1800 MPa, and the volume fraction tends to be 0.7.
[0095] Step 203: feature recognition is performed on the static topology optimization result.
[0096] In this step, a machine learning model can be used to extract features from the static topology optimization result to obtain a first candidate feature set; features not included in the typical feature library in the first candidate feature set are filtered out to obtain a first topology feature set.
[0097] Step 204: impact dynamics topology optimization.
[0098] In this step, the impact dynamics topology optimization is performed on the topology model of the workpiece under the condition of loading the impact dynamics load case to obtain a dynamics topology optimization result. During the impact dynamics topology optimization, the objective function and the constraint condition of the impact dynamics topology optimization need to be set.
[0099] For example, for the design of the fan shaft, the main part of the fan shaft can be taken as the design domain, and the part of the fan shaft in contact with the fan disc, the part of the fan shaft in contact with the low-pressure disc, and the part of the fan shaft in contact with the No. 1 support point and the No. 2 support point can be taken as the non-design domain. In addition, the FBO load can be selected as the impact dynamics load. Furthermore, the constraint condition of the impact dynamics topology optimization can include that the maximum stress of the fan shaft does not exceed the ultimate strength of the material during the loading of the FBO load. The objective function of the impact dynamics topology optimization is the minimum internal energy.
[0100] The stress change of the fan shaft during the dynamic topology optimization is shown in FIG. 12, and the volume fraction change is shown in FIG. 13. As shown in the figures, during the dynamic topology optimization, the maximum stress of the fan shaft tends to be the ultimate stress 1800 MPa, and the volume fraction tends to be 0.7. Figure 6c Figure 6d As shown in the figures, during the dynamic topology optimization, the maximum stress of the fan shaft tends to be the ultimate stress 1800 MPa, and the volume fraction tends to be 0.7.
[0101] Step 205: feature recognition is performed on the dynamic topology optimization result.
[0102] In this step, a machine learning model can be used to extract features from the dynamic topology optimization result to obtain a second candidate feature set; features in the second candidate feature set that are not included in the typical feature library are filtered out to obtain a second topology feature set.
[0103] The typical feature library can be determined according to the following manner: selecting a feature library corresponding to the type of the workpiece from a plurality of feature libraries, and taking the feature library corresponding to the type of the workpiece as the typical feature library.
[0104] In the embodiments of the present disclosure, on the one hand, the machine learning model is used to perform feature recognition on the static topology optimization result and the dynamic topology optimization result, so that the topology features in the workpiece structure can be accurately extracted; on the other hand, by comparing the extracted topology features with the typical feature library, the correctness of the extracted topology features can be further ensured, the probability of missing identification of key topology features can be reduced, and the number of extracted topology features can be effectively controlled, thereby reducing the computational amount of subsequent processing, and improving the timeliness of workpiece topology optimization.
[0105] Step 206: feature comparison.
[0106] In this step, the features in the first topology feature set and the second topology feature set can be compared.
[0107] Step 207: determining matching features.
[0108] In this step, according to the comparison result of the previous step, the matching features are determined. The matching features refer to the topology features located at the same position and having the same type.
[0109] Step 208: model reconstruction.
[0110] In this step, the matching features are fused to obtain a fused topology feature set; then, according to the fused topology feature set, the topology model of the workpiece is reconstructed to obtain a final optimized topology model. For how to fuse the features, reference can be made to the related descriptions of the foregoing embodiments.
[0111] Step 209: performance evaluation.
[0112] In this step, the performance of the optimized topology model can be evaluated by using a simulation method, and the performance of the optimized topology model can be evaluated by using a test method.
[0113] Step 210: determining whether the requirement is met. If yes, step 211 is performed; otherwise, step 201 is performed again.
[0114] Step 211: output the optimized configuration.
[0115] In this step, for the optimized topological model passing the performance test, it is outputted.
[0116] In the embodiments of the present disclosure, through the above process, the workpiece modeling optimization can be performed efficiently, and the effect of the workpiece modeling optimization is improved.
[0117] Figure 7 is a structural schematic diagram of a workpiece design device according to some embodiments of the present disclosure. As shown in Figure 7 The workpiece design device 70 is configured to perform the workpiece design method as described above, including a design domain determination module 71, a statics topology optimization module 72, a dynamics topology optimization module 73, a feature extraction module 74, a feature fusion module 75, and a construction module 76.
[0118] The design domain determination module 71 is configured to determine a design domain of a topological model of a workpiece.
[0119] The statics topology optimization module 72 is configured to, in a case of loading a statics load condition, utilize a statics topology optimization algorithm to optimize the topological model of the workpiece in the design domain to obtain a first optimization result.
[0120] The dynamics topology optimization module 73 is configured to, in a case of loading a dynamics load condition, utilize a dynamics topology optimization algorithm to optimize the topological model of the workpiece in the design domain to obtain a second optimization result.
[0121] The feature extraction module 74 is configured to extract a first set of topological features from the first optimization result and a second set of topological features from the second optimization result.
[0122] The feature fusion module 75 is configured to fuse the first set of topological features and the second set of topological features to obtain a fused set of topological features.
[0123] The construction module 76 is configured to construct an optimized topological model of the workpiece according to the fused set of topological features.
[0124] In the embodiments of the present disclosure, through the above device, the workpiece modeling optimization can be performed efficiently, and the effect of the workpiece modeling optimization is improved.
[0125] Figure 8 is a structural schematic diagram of a workpiece design device according to some embodiments of the present disclosure. As shown in Figure 8As shown, the workpiece design device comprises a user interface 81, a design domain determination module 82, a statics topology optimization module 83, a dynamics topology optimization module 84, a feature extraction and fusion module 85, a construction module 86, and a verification module 87.
[0126] The user interface 81 is used for interacting with the user, facilitating the user to configure parameters in the topology optimization process and input control instructions. In addition, the user interface 81 can also integrate the design domain determination module 82, the statics topology optimization module 83, etc. to visually display each module.
[0127] The design domain determination module 82 can comprise a requirement analysis sub-module and a requirement definition sub-module. The module can take a top-level requirement file as input, parse the file through the requirement analysis sub-module, and further process the parsed results through the requirement definition sub-module to obtain the design domain of the workpiece.
[0128] The statics topology optimization module 83 can comprise a statics topology optimization modeling sub-module, a statics topology optimization calculation sub-module, and a statics topology optimization post-processing sub-module. Through modeling by the modeling sub-module, optimization solving of the model by the calculation sub-module, and further processing of the optimization solving results by the post-processing sub-module, the final statics optimization results (i.e. the first optimization results) are obtained.
[0129] The dynamics topology optimization module 84 can comprise a dynamics topology optimization modeling sub-module, a dynamics topology optimization calculation sub-module, and a dynamics topology optimization post-processing sub-module. Through modeling by the modeling sub-module, optimization solving of the model by the calculation sub-module, and further processing of the optimization solving results by the post-processing sub-module, the final dynamics optimization results (i.e. the second optimization results) are obtained.
[0130] The feature extraction and fusion module 85 comprises a statics optimization result feature recognition sub-module, a dynamics optimization result feature recognition sub-module, a feature comparison sub-module, and a matching feature determination sub-module. Through extracting the first topology feature set and the second topology feature set by the two feature recognition sub-modules, determining the matching features by the feature comparison sub-module and the matching feature determination sub-module, and performing fusion processing according to the matching features, the fused topology feature set is obtained.
[0131] The construction module 86 is used for constructing the optimized topology model according to the fused topology feature set.
[0132] The verification module 87 comprises a simulation verification sub-module and a test verification sub-module, and is used for simulating and testing the performance of the optimized topology model. If the optimized topology model passes the verification, it is output as the optimized configuration. Otherwise, topology optimization is performed again.
[0133] In the embodiments of the present disclosure, the above device can improve the effect of workpiece modeling optimization while efficiently performing workpiece modeling optimization.
[0134] Figure 9 is a structural schematic diagram of an electronic device according to some embodiments of the present disclosure. As shown in Figure 9 The electronic device 90 includes a memory 91 and a processor 92 coupled to the memory 91. The memory 91 is configured to store instructions for performing corresponding embodiments of the workpiece design method. The processor 92 is configured to execute the workpiece design method in any of some embodiments of the present disclosure based on the instructions stored in the memory 91.
[0135] Figure 10 is a structural schematic diagram of a computer system according to some embodiments of the present disclosure. As shown in Figure 10 The workpiece design device can be in the form of a general-purpose computing device. The computer system 100 includes a memory 101, a processor 102, and a bus 103 connecting different system components.
[0136] The memory 101 may, for example, include system memory, non-volatile storage media, and the like. The system memory may, for example, store an operating system, application programs, a boot loader, and other programs. The system memory can include volatile storage media, such as random access memory (RAM) and / or cache memory. The non-volatile storage media may, for example, store instructions for performing at least one of the workpiece design methods in corresponding embodiments. The non-volatile storage media includes, but is not limited to, magnetic disk storage media, optical storage media, flash memory, and the like.
[0137] The processor 102 can be implemented in the form of a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a discrete hardware component, or the like. Accordingly, each of the modules, such as the design domain determination module, the statics topology optimization module, the dynamics topology optimization module, and the like, can be implemented by a central processing unit (CPU) running instructions in the memory for performing corresponding steps, or by a dedicated circuit for performing corresponding steps.
[0138] The bus 103 can use any of a variety of bus structures. For example, the bus structure includes, but is not limited to, an industry standard architecture (ISA) bus, a microchannel architecture (MCA) bus, a peripheral component interconnect (PCI) bus.
[0139] The interfaces 104, 105, 106, the memory 101 and the processor 102 in the computer system 100 can be connected through the bus 103. The input and output interface 104 can provide a connection interface for display, mouse, keyboard and other input and output devices. The network interface 105 provides a connection interface for various networking devices. The storage interface 106 provides a connection interface for external storage devices such as floppy disk, U disk, SD card and the like.
[0140] Here, various aspects of the disclosure are described with reference to flowcharts and / or block diagrams of methods, apparatuses and computer program products according to embodiments of the disclosure. It should be understood that each block of the flowchart and / or block diagram can be implemented by computer readable program instructions.
[0141] These computer readable program instructions can be provided to a processor of a general purpose computer, a special purpose computer, or other programmable apparatus to produce a machine, so that the instructions executed by the processor produce the apparatus that implements the functions specified in one or more blocks of the flowchart and / or block diagram.
[0142] These computer readable program instructions can also be stored in a computer readable storage medium that causes a computer to work in a specific manner, thereby producing a manufactured product including instructions that implement the functions specified in one or more blocks of the flowchart and / or block diagram.
[0143] The disclosure can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects.
[0144] So far, the workpiece design method and apparatus according to the disclosure have been described in detail. In order to avoid obscuring the concept of the disclosure, some details known in the art are not described. Those skilled in the art can fully understand how to implement the technical solutions disclosed herein according to the above description.
Claims
1. A workpiece design method, comprising: Determine the design domain of the workpiece's topological model; Under static load conditions, the topology model of the workpiece is optimized within the design domain using a static topology optimization algorithm to obtain a first optimization result. Under dynamic load conditions, the topology model of the workpiece is optimized within the design domain using a dynamic topology optimization algorithm to obtain a second optimization result. Extract a first topological feature set from the first optimization result and extract a second topological feature set from the second optimization result; The first topological feature set and the second topological feature set are fused to obtain a fused topological feature set; Based on the fused topological feature set, an optimized topological model of the workpiece is constructed.
2. The workpiece design method according to claim 1, wherein, The steps of extracting a first topological feature set from the first optimization result and extracting a second topological feature set from the second optimization result include: Using a machine learning model, feature recognition is performed on the first optimization result and the second optimization result to obtain a first candidate feature set and a second candidate feature set; Features not included in the typical feature library from the first candidate feature set are filtered out to obtain the first topological feature set; Features not included in the typical feature library from the second candidate feature set are filtered out to obtain the second topological feature set.
3. The workpiece design method according to claim 2, wherein, The typical feature library is determined according to the following method: Select a feature library from multiple feature libraries that corresponds to the type of the workpiece, and use the feature library that corresponds to the type of the workpiece as the typical feature library.
4. The workpiece design method according to claim 1, wherein, The process of fusing the first topological feature set and the second topological feature set to obtain the fused topological feature set includes: Filter out matching features between the first topological feature set and the second topological feature set, wherein the matching features are topological features located in the same position and of the same type; The values of the matching feature in the first topological feature set and the values of the matching feature in the second topological feature set are fused to obtain the final value of the matching feature. The set of matching features that have the final value is used as the fused topological feature set.
5. The workpiece design method according to claim 1, wherein, The design domain for determining the topology model of the workpiece includes: Obtain the engineering design requirements document for the workpiece; The engineering design requirements document is parsed to obtain multiple design requirements for the workpiece; Based on the multiple design requirements of the workpiece, the design domain of the workpiece's topology model is determined.
6. The workpiece design method according to claim 5, wherein, The design requirements include at least one of the following: size requirements, weight requirements, load requirements, performance requirements, installation requirements, maintenance requirements, economic requirements, strength requirements, and reliability requirements.
7. The workpiece design method according to claim 1, wherein, The static topology optimization algorithm is either the variable density method or the level set method.
8. The workpiece design method according to claim 1, wherein, The dynamic topology optimization algorithm is either the cellular automata method or the equivalent static load method.
9. The workpiece design method according to claim 1, wherein, The workpiece is a component of an aircraft engine.
10. A workpiece design apparatus, comprising: A module for performing the workpiece design method as described in any one of claims 1 to 9.
11. An electronic device, comprising: Memory; as well as A processor coupled to the memory, the processor being configured to execute the workpiece design method as described in any one of claims 1 to 9 based on instructions stored in the memory.
12. A computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the workpiece design method as described in any one of claims 1 to 9.
13. A computer program product having stored computer program instructions thereon, which, when executed by a processor, implement the workpiece design method as described in any one of claims 1 to 9.