A design method for gears with three-periodic minimal surface structure

Abstract

The present invention discloses a design method for a gear with a three-periodic minimal surface structure. The method comprises obtaining multiple sets of gear parameter sequences and three-periodic minimal surface characteristic parameter sequences to establish a three-periodic minimal surface-gear composite structure; and performing force analysis to determine the maximum stress, the maximum stress position, the maximum deformation, and the maximum deformation position; training a Stacking ensemble learning algorithm, and using the Stacking ensemble learning algorithm to obtain the predicted values ​​of the output maximum stress, the maximum stress position, the maximum deformation, and the maximum deformation position when inputting the gear parameter sequence and the three-periodic minimal surface characteristic parameter sequence; and using the predicted values ​​and the target values ​​to determine the final gear parameter sequence and the three-periodic minimal surface characteristic parameter sequence of the gear with a three-periodic minimal surface structure. The scheme of the present invention can accurately predict the performance of a gear with a three-periodic minimal surface lattice composite structure, and realize the modeling of a gear with a three-periodic minimal surface lattice structure of any complex features.

Description

Technical Field

[0001] The present invention relates to the technical field of gear design, and in particular to a design method for a gear with a three-period minimal surface structure. Background Art

[0002] As aerospace vehicles continue to develop towards long-distance flight and high load-bearing capacity, the demand for lightweight transmission system structures is increasing. Gears are key components in the transmission system. Designing a gear that has both high strength and lightweight characteristics is crucial to improving the overall performance of traditional systems. The three-periodic minimal surface (TPMS) structure has shown great potential in the field of structural lightweighting due to its excellent mechanical properties and unique porous characteristics. Especially in gear design, compared with traditional gears, TPMS structural gears can not only significantly reduce weight, but also effectively reduce gear vibration and enhance stress bearing capacity. However, the research and application of TPMS lattice structure gears are still in their infancy and face two core challenges: First, how to cleverly integrate the complex geometric features of TPMS into gear design to achieve accurate modeling of any TPMS lattice structure gear is a technical bottleneck that needs to be solved urgently. Second, the manufacturing of TPMS structural gears relies on additive manufacturing technologies such as selective laser melting (SLM). Although these technologies can accurately construct complex structures, the cost of metal additive manufacturing is relatively high. Therefore, a method is urgently needed in the early stage of design to accurately predict the performance of TPMS lattice composite structure gears to avoid unnecessary later experiments and remanufacturing costs. Summary of the invention

[0003] In order to solve the technical problem of how to accurately design and predict the performance of TPMS lattice structure gears, an embodiment of the present invention provides a design method for a three-period minimal surface structure gear.

[0004] The technical solution of the embodiment of the present invention is achieved as follows:

[0005] An embodiment of the present invention provides a design method for a gear with a three-period minimal surface structure, the method comprising:

[0006] Acquire multiple sets of gear parameter sequences and three-periodic minimal surface characteristic parameter sequences; wherein the gear parameter sequences include gear characteristic parameters and gear material parameters, the gear characteristic parameters include the number of teeth, module, pressure angle and tooth width, the gear material parameters include density, Young's modulus and Poisson's ratio, and the three-periodic minimal surface characteristic parameter sequences include gradient function parameters, unit cell size and type parameters;

[0007] Establishing a corresponding three-periodic minimal surface-gear composite structure according to each group of the gear parameter sequence and the three-periodic minimal surface characteristic parameter sequence;

[0008] Performing a stress analysis on each of the three-periodic minimal surface-gear composite structures to determine the maximum stress, maximum stress position, maximum deformation, and maximum deformation position of each of the three-periodic minimal surface-gear composite structures;

[0009] Constructing a Stacking ensemble learning algorithm; using a sequence consisting of multiple groups of the gear parameter sequences and the three-period minimal surface feature parameter sequences as an input sequence of the Stacking ensemble learning algorithm; using a sequence consisting of multiple groups of the maximum stress, maximum stress position, maximum deformation and maximum deformation position of the three-period minimal surface-gear composite structure as an output sequence of the Stacking ensemble learning algorithm, training the constructed Stacking ensemble learning algorithm, and obtaining a trained Stacking ensemble learning algorithm;

[0010] The trained Stacking ensemble learning algorithm is used to process the gear parameter sequence to be calculated and the three-period minimal surface feature parameter sequence to obtain the maximum stress prediction value, the maximum stress position prediction value, the maximum deformation prediction value and the maximum deformation position prediction value; it is judged whether the maximum stress prediction value exceeds the corresponding preset range, whether the maximum deformation prediction value exceeds the corresponding preset range, whether the maximum stress position prediction value is at the structural critical position and whether the maximum deformation position prediction value is at the structural critical position; if the maximum stress prediction value exceeds the preset range, or the maximum deformation prediction value exceeds the preset range, or the maximum stress position prediction value is at the structural critical position If the gear parameter sequence is at a critical position of the structure, or the predicted value of the maximum deformation position is at a critical position of the structure, the three-period minimal surface feature parameter sequence is modified until the maximum stress prediction value corresponding to the modified three-period minimal surface feature parameter sequence does not exceed a preset range, and the maximum deformation prediction value does not exceed a preset range, and the maximum stress position prediction value is not at a critical position of the structure, and the maximum deformation position prediction value is not at a critical position of the structure; the last modified gear parameter sequence and three-period minimal surface feature parameter sequence are obtained, and the last modified gear parameter sequence and three-period minimal surface feature parameter sequence are used as the final design of the three-period minimal surface structure gear.

[0011] In one embodiment, a corresponding three-periodic minimal surface-gear composite structure is established according to each group of the gear parameter sequence and the three-periodic minimal surface feature parameter sequence, including: establishing a gear object according to the gear parameter sequence in each group; establishing a three-periodic minimal surface object according to the three-periodic minimal surface feature parameter sequence in each group; and establishing a corresponding three-periodic minimal surface-gear composite structure according to the gear object and the three-periodic minimal surface object.

[0012] In one embodiment, a gear object is established according to the gear parameter sequence in each group, including: constructing a gear geometric model using geometric modeling software according to the gear parameter sequence; and outputting the gear geometric model into an STL format.

[0013] In one embodiment, a three-period minimal surface object is established according to a three-period minimal surface feature parameter sequence in each group, including: determining the corresponding three-period minimal surface construction function type according to the type parameter in the three-period minimal surface feature parameter sequence; based on the three-period minimal surface construction function type, establishing a three-period minimal surface construction function using the unit cell size in the three-period minimal surface feature parameter sequence; establishing a gradient function using the gradient function parameters in the three-period minimal surface feature parameter sequence; and constructing a three-period minimal surface modeling inequality according to the three-period minimal surface construction function and the gradient function.

[0014] In one embodiment, the three-period minimal surface construction function includes: if the three-period minimal surface construction function type is a Gyroid type, then the three-period minimal surface construction function established is:

[0015]

[0016]

[0017] If the three-periodic minimal surface construction function type is IWP type, the three-periodic minimal surface construction function established is:

[0018]

[0019]

[0020] in, is the unit cell size, x, y and z are spatial coordinates;

[0021] The gradient function is:

[0022]

[0023] in, is the gradient function parameter;

[0024] The three-periodic minimal surface modeling inequality is:

[0025]

[0026] in, is the gradient function, Construct functions for triperiodic minimal surfaces.

[0027] In one embodiment, a corresponding three-periodic minimal surface-gear composite structure is established according to the gear object and the three-periodic minimal surface object, including: defining a regional inequality according to the regional range in which the three-periodic minimal surface is embedded in the gear; restricting the three-periodic minimal surface modeling inequality within the regional range to obtain a regional three-periodic minimal surface modeling formula, and saving the regional three-periodic minimal surface geometric structure obtained by using the regional three-periodic minimal surface modeling formula in STL format; importing the gear geometric model and the regional three-periodic minimal surface geometric structure in STL format into a preset modeling software, and constructing the three-periodic minimal surface-gear composite structure using Boolean operations in the preset modeling software.

[0028] In one embodiment, the area inequality is:

[0029]

[0030] in, represents the embedding area, is the lower boundary of the embedding, close to the axle side, It is the upper boundary of the embedding, close to the gear tooth side;

[0031] The modeling formula of the regional three-periodic minimal surface is:

[0032]

[0033] in, is the gradient function, Construct functions for triperiodic minimal surfaces.

[0034] In one embodiment, the constructed Stacking ensemble learning algorithm includes a first-level basic model and a second-level meta-model: the first-level basic model includes a first random forest model, a first deep neural network, and a second deep neural network;

[0035] The first random forest model is:

[0036]

[0037] in, is the predicted output of the first random forest, corresponding to the actual output performance sequence , , the subscript i is the grouping number, indicating the i-th group of data in the data set, and N is the number of data sets, that is, the number of data sets. is the x-coordinate of the maximum stress position, is the y coordinate of the maximum stress position, is the maximum stress, is the x-coordinate of the maximum deformation position, The y coordinate of the maximum deformation position, is the maximum deformation, is the first random forest model, for , is the number of teeth, is the modulus, is the pressure angle, is the tooth width, is the density, is Young's modulus, is Poisson's ratio, for , is the gradient function parameter, is the unit cell size, is a type parameter;

[0038] The first deep neural network is:

[0039]

[0040] in, The first deep neural network predictions, The first deep neural network The training model, The first deep neural network predictions, The first deep neural network The training model, The first deep neural network predictions, The first deep neural network The training model, The first deep neural network predictions, The first deep neural network The training model, The first deep neural network predictions, The first deep neural network The training model, The first deep neural network predictions, The first deep neural network Training model of

[0041] The second deep neural network is:

[0042]

[0043] in, is the predicted output of the second deep neural network, corresponding to the actual output performance sequence , It is the second deep neural network model.

[0044] In one embodiment, the second level meta-model is:

[0045]

[0046] in, is the predicted value of the second-level meta-model for the i-th sample, , , is the weight coefficient of the second-level meta-model, is the predicted value of the first random forest model for the i-th sample, is the predicted value of the first deep neural network for the i-th sample, is the predicted value of the second deep neural network model for the i-th sample, is the bias of the meta-model.

[0047] In one embodiment, the training process of the Stacking ensemble learning algorithm is: separately training the first random forest model, the first deep neural network, and the second deep neural network in the first-level basic model to obtain the prediction values ​​output by the first random forest model, the first deep neural network, and the second deep neural network respectively; inputting the prediction values ​​output by the first random forest model, the first deep neural network, and the second deep neural network respectively into the second-level meta-model, comparing the prediction values ​​output by the second-level meta-model with the target output sequence, training the second-level meta-model, and determining the weight coefficient in the second-level meta-model to complete the training process of the Stacking ensemble learning algorithm.

[0048] This embodiment has the following beneficial effects:

[0049] This embodiment scheme can realize the modeling of TPMS lattice structure gears with arbitrarily complex features, and through Stacking integrated learning technology, based on a variety of TPMS and gear parameters, it can realize accurate prediction of the stiffness, strength, fatigue life and other performance of the TPMS lattice structure gears. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic flow chart of a design method of a gear with a periodic minimal curved surface structure according to a third embodiment of the present invention;

[0051] Figure 2 A schematic diagram of the application process of the TPMS lattice structure gear design and performance prediction method based on Stacking ensemble learning according to an embodiment of the present invention;

[0052] Figure 3 Schematic diagram of the geometric structure of two TPMS lattice structure gears according to the embodiment of the present invention; wherein, Figure 3 (a) Schematic diagram of the geometric structure of the IWP type TPMS lattice structure gear; Figure 3 (b) Schematic diagram of the geometric structure of the Gyroid type TPMS lattice structure gear;

[0053] Figure 4 Schematic diagram of the finite element analysis results of the TPMS lattice structure gear strength according to an embodiment of the present invention; wherein: Figure 4 (a) is a schematic diagram of the maximum stress and the maximum stress position in the finite element analysis results of the TPMS lattice structure gear strength. Figure 4 (b) Schematic diagram of the maximum deformation and the maximum deformation position in the finite element analysis results of the TPMS lattice structure gear strength;

[0054] Figure 5 1 is a diagram showing the internal structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The present invention will be described in further detail below with reference to the accompanying drawings and embodiments.

[0056] The embodiment of the present invention provides a design method for a gear with a three-periodic minimal surface structure. Figure 1 As shown, the method includes:

[0057] Step 101: Acquire multiple sets of gear parameter sequences and three-periodic minimal surface characteristic parameter sequences; wherein the gear parameter sequence includes gear characteristic parameters and gear material parameters, the gear characteristic parameters include the number of teeth, module, pressure angle and tooth width, the gear material parameters include density, Young's modulus and Poisson's ratio, and the three-periodic minimal surface characteristic parameter sequence includes gradient function parameters, unit cell size and type parameters;

[0058] Step 102: establishing a corresponding three-periodic minimal surface-gear composite structure according to each group of the gear parameter sequence and the three-periodic minimal surface characteristic parameter sequence;

[0059] Step 103: performing a stress analysis on each of the three-periodic minimal surface-gear composite structures to determine the maximum stress, maximum stress position, maximum deformation and maximum deformation position of each of the three-periodic minimal surface-gear composite structures;

[0060] Step 104: constructing a Stacking ensemble learning algorithm; using a sequence consisting of multiple groups of the gear parameter sequences and the three-period minimal surface feature parameter sequences as an input sequence of the Stacking ensemble learning algorithm; using a sequence consisting of multiple groups of the maximum stress, maximum stress position, maximum deformation and maximum deformation position of the three-period minimal surface-gear composite structure as an output sequence of the Stacking ensemble learning algorithm, training the constructed Stacking ensemble learning algorithm, and obtaining a trained Stacking ensemble learning algorithm;

[0061] Step 105: Use the trained Stacking ensemble learning algorithm to process the gear parameter sequence to be calculated and the three-period minimal surface feature parameter sequence to obtain the maximum stress prediction value, the maximum stress position prediction value, the maximum deformation prediction value and the maximum deformation position prediction value; determine whether the maximum stress prediction value exceeds the corresponding preset range, whether the maximum deformation prediction value exceeds the corresponding preset range, whether the maximum stress position prediction value is at the structural critical position and whether the maximum deformation position prediction value is at the structural critical position; if the maximum stress prediction value exceeds the preset range, or the maximum deformation prediction value exceeds the preset range, or the maximum stress position prediction value exceeds the preset range, or the maximum stress position prediction value exceeds the preset range, If the maximum stress prediction value corresponding to the modified three-period minimal surface feature parameter sequence does not exceed the preset range, and the maximum deformation prediction value does not exceed the preset range, and the maximum stress position prediction value is not at the structural critical position, and the maximum deformation position prediction value is not at the structural critical position; obtain the last modified gear parameter sequence and three-period minimal surface feature parameter sequence, and use the last modified gear parameter sequence and three-period minimal surface feature parameter sequence as the final design of the three-period minimal surface structure gear.

[0062] The method of this embodiment can realize the modeling of TPMS lattice structure gears with arbitrarily complex features. Through Stacking integrated learning technology, based on a variety of TPMS and gear parameters, accurate prediction of the stiffness, strength, fatigue life and other performance of the TPMS lattice structure gears can be achieved.

[0063] Specifically, see Figure 2 The complete process of this embodiment includes the following steps:

[0064] S1. First, establish the gear object and obtain the gear characteristic parameters and material parameters, including the number of teeth, module, pressure angle, and tooth width. The material parameters include density, Young's modulus, and Poisson's ratio.

[0065] Define gear characteristic parameters and material parameter sequences :

[0066] in, is the number of teeth, is the modulus, is the pressure angle, is the tooth width, is the density, is Young's modulus, is Poisson's ratio.

[0067] The gear geometry model is constructed using the geometry modeling software NX and exported in STL format.

[0068] S2. Establish a TPMS object and obtain TPMS characteristic parameters, including thickness-related gradient parameters, cell size, and type parameters.

[0069] Constructing TPMS modeling inequalities:

[0070]

[0071] in, is the gradient function parameter related to TPMS thickness, is the unit cell size, is a type parameter, which can be defined by One-Hot encoding. is the gradient function, Build functions for TPMS.

[0072] TPMS build function Can be of various types, see Figure 3 , such as Gyroid type or IWP type TPMS function, in this case, the Gyroid type can be defined as 0,1 through One-Hot encoding, and the IWP type can be defined as 1,0 through One-Hot encoding:

[0073]

[0074]

[0075]

[0076]

[0077] Where x, y and z are spatial coordinates.

[0078] Gradient function It can also be defined in different types, such as along the radial direction of the gear:

[0079]

[0080] Among them, x and y are spatial coordinates.

[0081] Define TPMS characteristic parameter sequence : .

[0082] S3, establishing a TPMS lattice structure gear, specifically including the following steps:

[0083] S31. Control the area range of the TPMS embedded in the gear and define the area inequality:

[0084]

[0085] Among them, x and y represent the coordinate axes, indicating that the gear is built on the xoy plane (o is the coordinate origin), and the normal direction of the gear is the z axis. Indicates that the embedded region is a ring. is the lower boundary of the embedding, close to the axle side, It is the upper boundary of embedding, close to the gear tooth side.

[0086] S32, restrict the TPMS modeling inequality of S2 to the area defined by S31, and obtain the regional TPMS modeling formula:

[0087]

[0088] The regional TPMS geometry obtained using the regional TPMS modeling formula described above is saved in STL format.

[0089] S33, import the gear and regional TPMS STL geometry format files obtained in S1 and S32 into nTopology, and use the Boolean operation of nTopology to construct the TPMS-gear composite structure. In this way, N groups (N is the number of data sets) with different sequences are constructed. and sequence TPMS lattice structure gear and its geometric model.

[0090] S4: Build a Stacking ensemble learning algorithm to realize the characteristic parameters and material parameter sequence of gears based on S1 , TPMS characteristic parameter sequence of S2 The performance prediction of the TPMS lattice structure gear includes the following steps:

[0091] S41, establish input sequence. Corresponding to the N groups of TPMS-gear composite structures constructed in S33, define the input (feature) sequence of the ensemble learning algorithm as , from the characteristic parameters and material parameter sequence of the gear and TPMS characteristic parameter sequence composition:

[0092]

[0093] Among them, i is the grouping number, indicating the i-th group of data (sample) in the data set, and N is the number of data sets, that is, the number of data sets.

[0094] S42, establish output sequence. Through finite element secondary development, automatically complete the force analysis of the N groups of TPMS lattice structure gear geometric models obtained in S33, such as Figure 4 As shown, the maximum stress of the TPMS-gear composite structure is obtained and the maximum stress position , Maximum deformation and the maximum deformation position , define the performance sequence of TPMS lattice structure gears , which is the output (target) sequence of the ensemble learning algorithm:

[0095]

[0096] Among them, the subscript i is the grouping number, indicating the i-th group of data in the data set, and N is the number of data sets, that is, the number of data sets. is the x-coordinate of the maximum stress position, is the y coordinate of the maximum stress position, is the maximum stress, is the x-coordinate of the maximum deformation position, The y coordinate of the maximum deformation position, is the maximum deformation.

[0097] S43. Construct the first-level basic model of the Stacking ensemble learning algorithm. The first-level basic model includes three machine learning models, random forest model 1 (RF) (i.e., the first random forest model), deep neural network 2 (Deep1) (i.e., the first deep neural network) and deep neural network 3 (Deep2) (i.e., the second deep neural network). For the random forest model:

[0098]

[0099] in, is the predicted output of the random forest, corresponding to the actual output performance sequence , is a random forest model. The number of random forest trees is H, which is determined by the number of data sets N.

[0100] For Deep 1, the performance sequence In , , , Implement separate training predictions:

[0101]

[0102] in, For Deep 1 neural network predictions, For Deep1 The training model, For Deep1 predictions, For Deep1 The training model, For Deep1 predictions, For Deep1 The training model, For Deep1 predictions, For Deep1 The training model, For Deep1 predictions, For Deep1 The training model, For Deep1 predictions, For Deep1 The training model.

[0103] The final prediction result of deep neural network 2 (Deep1) is:

[0104]

[0105] For Deep1, the activation function is Sigmoid, and the number of neurons in the input layer is The number of elements in the same, the number of neurons in the output layer is 1, the number of hidden layers is M, which is determined by the number of data sets N, and the loss function is the mean square error MSE:

[0106]

[0107] Where N is the number of data sets, and the subscript i represents the i-th group of data in the data.

[0108] For Deep2, the performance sequence Implement overall training prediction:

[0109]

[0110] in, is the predicted output of the deep neural network 3 (Deep2), corresponding to the actual output performance sequence , It is the Deep2 model.

[0111] For Deep2, the activation function is Sigmoid, and the number of neurons in the input layer is The number of elements in the output layer is the same as The number of elements in is the same, the number of hidden layers is K, which is determined by the number of data sets N, and the loss function is the mean square error MSE:

[0112]

[0113] Where N is the number of data sets, and the subscript i represents the i-th group of data in the data.

[0114] S44. Construct the second-level meta-model of the Stacking ensemble learning algorithm:

[0115]

[0116] in, is the predicted value of the meta-model for the i-th sample, which is also the final predicted value of the Stacking ensemble learning algorithm. , , is the weight coefficient of the meta-model, is the predicted value of the random forest model RF for the i-th sample, for The model's predicted value for the i-th sample, for The model's predicted value for the i-th sample, is the bias of the meta-model.

[0117] S45, training the Stacking ensemble learning algorithm. First, train the first-level basic model and train RF separately. , ,get , , Then, the base model is obtained , , As new features, it is input into the second-level meta-model. Finally, the meta-model finally predicts Performance sequence with objectives Compare, train meta-model, and learn the best , , and The Stacking ensemble learning algorithm finally realizes the characteristic parameters and material parameter sequence based on the gear. , TPMS characteristic parameter sequence TPMS lattice structure gear performance prediction.

[0118] S5. Analyze the maximum stress and maximum stress position, maximum deformation and maximum deformation position of the TPMS lattice structure gear predicted by the Stacking ensemble learning algorithm. If the maximum stress and maximum deformation exceed the allowable range, or the position of the maximum stress and maximum deformation is a critical position of the structure (the stress and strain at this position should not be too large), modify the TPMS structure at the corresponding position and adjust the parameters in S2. , so that the TPMS lattice structure gear as a whole meets the strength requirements.

[0119] This embodiment scheme can accurately estimate the performance of TPMS lattice composite structure gears, avoid unnecessary post-experimentation and remanufacturing costs, and can realize the modeling of three-periodic minimal surface lattice structure gears with arbitrarily complex features.

[0120] In order to implement the method of the embodiment of the present invention, the embodiment of the present invention also provides a computer program product, which includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device performs the steps of the above method.

[0121] Based on the hardware implementation of the above program modules, and in order to implement the method of the embodiment of the present invention, the embodiment of the present invention further provides an electronic device (computer device). Specifically, in one embodiment, the computer device may be a terminal, and its internal structure diagram may be as follows: Figure 5As shown. The computer device includes a processor A01, a network interface A02, a display screen A04, an input device A05 and a memory (not shown in the figure) connected through a system bus. Among them, the processor A01 of the computer device is used to provide computing and control capabilities. The memory of the computer device includes an internal memory A03 and a non-volatile storage medium A06. The non-volatile storage medium A06 stores an operating system B01 and a computer program B02. The internal memory A03 provides an environment for the operation of the operating system B01 and the computer program B02 in the non-volatile storage medium A06. The network interface A02 of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor A01, the method of any one of the above embodiments is implemented. The display screen A04 of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device A05 of the computer device can be a touch layer covered on the display screen, or a key, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0122] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0123] The device provided by the embodiment of the present invention includes a processor, a memory, and a program stored in the memory and executable on the processor. When the processor executes the program, the method of any one of the above embodiments is implemented.

[0124] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0125] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0126] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0127] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0128] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0129] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0130] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0131] It can be understood that the memory of the embodiment of the present invention can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); the magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), synchronous static random access memory (SSRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), and direct RAMbus random access memory (DRRAM, Direct Rambus Random Access Memory).The memories described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable types of memories.

[0132] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0133] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A design method for a three-period minimal surface structure gear, characterized in that: The method comprises: Acquire multiple sets of gear parameter sequences and three-periodic minimal surface characteristic parameter sequences; wherein the gear parameter sequences include gear characteristic parameters and gear material parameters, the gear characteristic parameters include the number of teeth, module, pressure angle and tooth width, the gear material parameters include density, Young's modulus and Poisson's ratio, and the three-periodic minimal surface characteristic parameter sequences include gradient function parameters, unit cell size and type parameters; Establishing a corresponding three-periodic minimal surface-gear composite structure according to each group of the gear parameter sequence and the three-periodic minimal surface characteristic parameter sequence; Performing a stress analysis on each of the three-periodic minimal surface-gear composite structures to determine the maximum stress, maximum stress position, maximum deformation, and maximum deformation position of each of the three-periodic minimal surface-gear composite structures; Constructing a Stacking ensemble learning algorithm; using a sequence consisting of multiple groups of the gear parameter sequences and the three-period minimal surface feature parameter sequences as an input sequence of the Stacking ensemble learning algorithm; using a sequence consisting of multiple groups of the maximum stress, maximum stress position, maximum deformation and maximum deformation position of the three-period minimal surface-gear composite structure as an output sequence of the Stacking ensemble learning algorithm, training the constructed Stacking ensemble learning algorithm, and obtaining a trained Stacking ensemble learning algorithm; The trained Stacking ensemble learning algorithm is used to process the gear parameter sequence to be calculated and the three-period minimal surface feature parameter sequence to obtain the maximum stress prediction value, the maximum stress position prediction value, the maximum deformation prediction value and the maximum deformation position prediction value; it is judged whether the maximum stress prediction value exceeds the corresponding preset range, whether the maximum deformation prediction value exceeds the corresponding preset range, whether the maximum stress position prediction value is at the structural critical position and whether the maximum deformation position prediction value is at the structural critical position; if the maximum stress prediction value exceeds the preset range, or the maximum deformation prediction value exceeds the preset range, or the maximum stress position prediction value is at the structural critical position If the gear parameter sequence is at a critical position of the structure, or the predicted value of the maximum deformation position is at a critical position of the structure, the three-period minimal surface feature parameter sequence is modified until the maximum stress prediction value corresponding to the modified three-period minimal surface feature parameter sequence does not exceed a preset range, and the maximum deformation prediction value does not exceed a preset range, and the maximum stress position prediction value is not at a critical position of the structure, and the maximum deformation position prediction value is not at a critical position of the structure; the last modified gear parameter sequence and three-period minimal surface feature parameter sequence are obtained, and the last modified gear parameter sequence and three-period minimal surface feature parameter sequence are used as the final design of the three-period minimal surface structure gear.

2. The design method of a three-period minimal surface structure gear according to claim 1, characterized in that: A corresponding three-periodic minimal surface-gear composite structure is established according to each set of the gear parameter sequence and the three-periodic minimal surface characteristic parameter sequence, including: Create gear objects according to the gear parameter sequence in each group; Establish a three-periodic minimal surface object according to the three-periodic minimal surface characteristic parameter sequence in each group; A corresponding three-periodic minimal surface-gear composite structure is established according to the gear object and the three-periodic minimal surface object.

3. The design method of a three-period minimal surface structure gear according to claim 2, characterized in that: Gear objects are created based on the sequence of gear parameters in each group, including: A gear geometric model is constructed using geometric modeling software according to the gear parameter sequence; and the gear geometric model is output in an STL format.

4. The design method of a three-period minimal surface structure gear according to claim 3 is characterized in that: A three-periodic minimal surface object is established according to the three-periodic minimal surface feature parameter sequence in each group, including: Determine the corresponding three-periodic minimal surface construction function type according to the type parameter in the three-periodic minimal surface characteristic parameter sequence; Based on the three-periodic minimal surface construction function type, a three-periodic minimal surface construction function is established using the unit cell size in the three-periodic minimal surface characteristic parameter sequence; Establishing a gradient function using the gradient function parameters in the three-periodic minimal surface characteristic parameter sequence; A three-periodic minimal surface modeling inequality is constructed according to the three-periodic minimal surface construction function and the gradient function.

5. The design method of a three-period minimal surface structure gear according to claim 4, characterized in that: The three-periodic minimal surface construction function includes: If the three-periodic minimal surface construction function type is a Gyroid type, the three-periodic minimal surface construction function established is: If the three-periodic minimal surface construction function type is IWP type, the three-periodic minimal surface construction function established is: in, is the unit cell size, x, y and z are spatial coordinates; The gradient function is: in, is the gradient function parameter; The three-periodic minimal surface modeling inequality is: in, is the gradient function, Construct functions for triperiodic minimal surfaces.

6. The design method of a three-periodic minimal surface structure gear according to claim 4, characterized in that: A corresponding three-periodic minimal surface-gear composite structure is established according to the gear object and the three-periodic minimal surface object, including: According to the range of the area where the three-periodic minimal surface is embedded in the gear, the regional inequality is defined; Limiting the three-periodic minimal surface modeling inequality within the region to obtain a regional three-periodic minimal surface modeling formula, and saving the regional three-periodic minimal surface geometric structure obtained by using the regional three-periodic minimal surface modeling formula in an STL format; The gear geometry model and the regional three-periodic minimal surface geometry in STL format are imported into the preset modeling software, and the three-periodic minimal surface-gear composite structure is constructed using the Boolean operation in the preset modeling software.

7. The design method of a three-period minimal surface structure gear according to claim 6, characterized in that: The area inequality is: in, represents the embedding area, is the lower boundary of the embedding, close to the axle side, It is the upper boundary of the embedding, close to the gear tooth side; The modeling formula of the regional three-periodic minimal surface is: in, is the gradient function, Construct functions for triperiodic minimal surfaces.

8. The design method of a three-period minimal surface structure gear according to claim 1, characterized in that: The constructed Stacking ensemble learning algorithm includes a first-level basic model and a second-level meta-model: the first-level basic model includes a first random forest model, a first deep neural network and a second deep neural network; The first random forest model is: in, is the predicted output of the first random forest, corresponding to the actual output performance sequence , , the subscript i is the grouping number, indicating the i-th group of data in the data set, and N is the number of data sets, that is, the number of data sets. is the x-coordinate of the maximum stress position, is the y coordinate of the maximum stress position, is the maximum stress, is the x-coordinate of the maximum deformation position, The y coordinate of the maximum deformation position, is the maximum deformation, is the first random forest model, for , is the number of teeth, is the modulus, is the pressure angle, is the tooth width, is the density, is Young's modulus, is Poisson's ratio, for , is the gradient function parameter, is the unit cell size, is a type parameter; The first deep neural network is: in, The first deep neural network predictions, The first deep neural network The training model, The first deep neural network predictions, The first deep neural network The training model, The first deep neural network predictions, The first deep neural network The training model, The first deep neural network predictions, The first deep neural network The training model, The first deep neural network predictions, The first deep neural network The training model, The first deep neural network predictions, The first deep neural network Training model of The second deep neural network is: in, is the predicted output of the second deep neural network, corresponding to the actual output performance sequence , It is the second deep neural network model.

9. The design method of a three-period minimal surface structure gear according to claim 8, characterized in that: The second level metamodel is: in, is the predicted value of the second-level meta-model for the i-th sample, , , is the weight coefficient of the second-level meta-model, is the predicted value of the first random forest model for the i-th sample, is the predicted value of the first deep neural network for the i-th sample, is the predicted value of the second deep neural network model for the i-th sample, is the bias of the meta-model.

10. The design method of a three-periodic minimal surface structure gear according to claim 9, characterized in that: The training process of the Stacking ensemble learning algorithm is as follows: Separately train the first random forest model, the first deep neural network, and the second deep neural network in the first-level basic model to obtain prediction values ​​output by the first random forest model, the first deep neural network, and the second deep neural network, respectively; The predicted values ​​output by the first random forest model, the first deep neural network, and the second deep neural network are respectively input into the second-level meta-model, the predicted values ​​output by the second-level meta-model are compared with the target output sequence, the second-level meta-model is trained, and the weight coefficients in the second-level meta-model are determined to complete the training process of the Stacking ensemble learning algorithm.

Images