Arrangement method of film holes of turbine blade

Abstract

The present disclosure relates to the field of aero-engines, and particularly to a turbine blade film hole arrangement method. The turbine blade film hole arrangement method comprises: constructing a design sample space based on a value range of a turbine blade film hole arrangement parameter; wherein the design sample space comprises a plurality of turbine blade sample models with different turbine blade film hole arrangements; establishing an objective function with a minimum stress prediction value of the turbine blade sample model as a target; wherein the stress prediction value is calculated according to a pre-trained stress prediction model; and establishing a constraint condition based on an upper limit value and a lower limit value of the design sample space; establishing a film hole arrangement optimization model according to the objective function and the constraint condition, and solving the film hole arrangement optimization model to determine a target film hole arrangement parameter. The method can improve the turbine blade film hole arrangement optimization efficiency.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of aero-engines, in particular to a turbine blade film hole arrangement method. BACKGROUND

[0002] Turbine blade film hole is a common blade cooling structure in modern aero-engines, and the arrangement on the turbine blade has the characteristics of small and dense. In the parameterization of the traditional turbine blade film hole arrangement design, the same column of film holes is usually selected as a group of continuous position and same interval and punching direction for parameterization design or manual interactive film hole modeling based on sketches, and the influence of the hole array arrangement on the adiabatic cooling efficiency and the film comprehensive cooling efficiency is obtained by comparing the flow and heat transfer characteristics of the film cooling method under different hole array arrangement through numerical simulation.

[0003] The disadvantages of the method are that, first, in the traditional turbine blade film hole parameterization design, each column of film holes on the blade surface is arranged as continuous and equal interval and same punching direction, but this cannot achieve the best cooling effect; second, the traditional film hole arrangement optimization usually needs to use a surrogate model to fit the simulation results, and the design of the surrogate model needs the knowledge of experts in the field and a lot of manual labor, and the surrogate model has weak generalization ability when dealing with complex and diversified data sets, which reduces the optimization efficiency of the turbine blade film hole arrangement; finally, the objective function of the traditional film hole arrangement optimization mainly focuses on the influence of the film hole arrangement on the blade cooling characteristics, and less considers the influence of the film hole arrangement on the stress of the blade surface in the optimization design.

[0004] It should be noted that the information disclosed in the above background section is only used to strengthen the understanding of the background of the present disclosure, and therefore can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0005] The purpose of the present disclosure is to provide a turbine blade film hole arrangement method, aiming to solve the problem of low optimization efficiency of turbine blade film hole arrangement.

[0006] Other characteristics and advantages of the present disclosure will become apparent from the following detailed description, or will be learned by practice of the present disclosure.

[0007] According to an aspect of an embodiment of the present disclosure, a turbine blade film hole arrangement method is provided, characterized in that it comprises:

[0008] A design sample space is constructed based on the value range of the turbine blade film hole arrangement parameters; wherein the design sample space comprises a plurality of turbine blade sample models with different film hole arrangements;

[0009] establish a target function with a minimum stress prediction value of the turbine blade sample model as a target, wherein the stress prediction value is calculated according to a pre-trained stress prediction model, and establish a constraint condition based on an upper limit value and a lower limit value of the design sample space;

[0010] A film hole arrangement optimization model is established according to the target function and the constraint condition, and the film hole arrangement optimization model is solved to determine target film hole arrangement parameters.

[0011] According to some embodiments of the present disclosure, based on the foregoing scheme, the film hole arrangement parameters include position parameters and rotation angle parameters; wherein the turbine blade includes multiple groups of film holes, the position parameters include one or more of the distance from the center position of a single group of film holes to the center axis, the distance from the starting film hole position to the center axis, the distance from the starting film hole position to the blade tip, the included angle between the single group of film hole connecting lines and the horizontal axis, the number of single group of film holes, and the distance between two film holes in a single group of film holes; and the rotation angle parameters include one or more of the rotation angle normal angle, the rotation angle circumferential angle, and the rotation angle circumferential angle of a single film hole.

[0012] According to some embodiments of the present disclosure, based on the foregoing scheme, the film holes located in the same column, with continuous positions, the same film hole spacing, and the same film hole rotation angle of the turbine blade are a group of film holes.

[0013] According to some embodiments of the present disclosure, based on the foregoing scheme, the method further comprises: pre-training the stress prediction model, wherein the pre-training the stress prediction model comprises:

[0014] extracting a turbine blade sample model in the design sample space as a training sample;

[0015] performing blade strength simulation on the training sample to obtain a stress simulation value corresponding to each training sample, respectively;

[0016] training the stress prediction model according to the stress simulation value corresponding to each training sample, respectively.

[0017] According to some embodiments of the present disclosure, based on the foregoing scheme, the stress prediction model is a CGAN model, the CGAN model includes a defined generator and a discriminator, and the training the stress prediction model according to the stress simulation value corresponding to each training sample, respectively, comprises:

[0018] two-dimensionally expanding the surface of the training sample to obtain a film hole distribution matrix;

[0019] calculating the stress prediction value corresponding to the training sample according to the film hole distribution matrix by using the defined generator;

[0020] determining, by the discriminator, whether the stress prediction value belongs to the generator or to the stress simulation value;

[0021] alternately training the defined generator and the discriminator to modify model parameters until a training stop condition is met.

[0022] According to some embodiments of the present disclosure, based on the foregoing scheme, the calculating, by the defined generator, of the stress value corresponding to the training sample based on the gas film hole distribution matrix comprises:

[0023] encoding the gas film hole distribution matrix of a training sample to obtain a first feature map;

[0024] adding the boundary condition converted by the first fully connected layer to the first feature map in an element manner to obtain a stress distribution result;

[0025] decoding the stress distribution result to obtain a stress value;

[0026] traversing the training sample to obtain a stress value corresponding to each training sample respectively.

[0027] According to some embodiments of the present disclosure, based on the foregoing scheme, the determining, by the discriminator, of whether the stress prediction value belongs to the generator or to the stress simulation value comprises:

[0028] connecting and encoding the gas film hole distribution matrix, the stress prediction value and the stress simulation value of a training sample to obtain a second feature map;

[0029] adding the boundary condition converted by the second fully connected layer to the second feature map in an element manner to obtain a feature result;

[0030] converting and evaluating the feature result to determine whether the feature result belongs to the generator or to the stress simulation value.

[0031] According to a second aspect of the embodiments of the present disclosure, a gas film hole arrangement device of a turbine blade is provided, comprising:

[0032] a sample module configured to construct a design sample space based on a value range of a gas film hole arrangement parameter of a turbine blade; wherein the design sample space comprises a plurality of turbine blade sample models having different gas film hole arrangements;

[0033] a model module configured to establish a target function with a minimum stress prediction value of the turbine blade sample model as a target; wherein the stress prediction value is calculated according to a pre-trained stress prediction model; and establish a constraint condition based on an upper limit value and a lower limit value of the design sample space;

[0034] A solving module is configured to establish an air film hole arrangement optimization model according to the target function and the constraint condition, and solve the air film hole arrangement optimization model to determine target air film hole arrangement parameters.

[0035] According to a third aspect of the embodiments of the present disclosure, a computer readable storage medium is provided, and the computer readable storage medium stores a computer program. The computer program is executed by a processor to implement the air film hole arrangement method of the turbine blade in the above embodiments.

[0036] According to a fourth aspect of the embodiments of the present disclosure, an electronic device is provided, and the electronic device comprises one or more processors and a storage device configured to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the air film hole arrangement method of the turbine blade in the above embodiments.

[0037] The example embodiments of the present disclosure can have the following partial or all beneficial effects:

[0038] In the technical solutions provided by some embodiments of the present disclosure, on the one hand, a target function is established with the minimum stress prediction value of the turbine blade sample model as the target, and the stress distribution on the surface of the turbine blade is taken into account in the optimization, so that the stress around the optimized air film hole is relieved, and the phenomenon of air film hole stress interference can be avoided; on the other hand, a pre-trained stress prediction model is used to calculate the stress value. Compared with simulating the turbine blade sample model to obtain the stress value, the time of numerical calculation in the optimization process is greatly reduced, and the optimization efficiency and accuracy are improved. Compared with the traditional air film hole optimization using the proxy model, the nonlinear modeling capability is strong, which can better adapt to complex data patterns and rules, and has stronger adaptability and parallel computing capability. When processing large-scale data, the advantages are obvious.

[0039] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0040] The drawings incorporated into the specification and forming a part thereof, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure. Obviously, the drawings in the following description are only some embodiments of the present disclosure, and other drawings can be obtained by those skilled in the art without creative labor. In the drawings:

[0041] Figure 1 a flowchart schematically showing an air film hole arrangement method of a turbine blade in an example embodiment of the present disclosure;

[0042] Figure 2A schematic diagram of an arrangement of film holes on a turbine blade surface in an exemplary embodiment of the present disclosure is shown schematically.

[0043] Figure 3 A schematic diagram of a film hole position parameter in an exemplary embodiment of the present disclosure is shown schematically.

[0044] Figure 4 A schematic diagram of a single set of film hole position parameters in an exemplary embodiment of the present disclosure is shown schematically.

[0045] Figure 5 A schematic diagram of a film hole rotation parameter in an exemplary embodiment of the present disclosure is shown schematically.

[0046] Figure 6 A schematic diagram of a single film hole rotation parameter in an exemplary embodiment of the present disclosure is shown schematically.

[0047] Figure 7 A model structure diagram of a CGAN model in an exemplary embodiment of the present disclosure is shown schematically.

[0048] Figure 8 A flowchart of a stress prediction model training method in an exemplary embodiment of the present disclosure is shown schematically.

[0049] Figure 9 A flowchart of obtaining a film hole distribution matrix in an exemplary embodiment of the present disclosure is shown schematically.

[0050] Figure 10 A composition diagram of a film hole arrangement device for a turbine blade in an exemplary embodiment of the present disclosure is shown schematically.

[0051] Figure 11 A schematic diagram of a computer readable storage medium in an exemplary embodiment of the present disclosure is shown schematically.

[0052] Figure 12 A structure diagram of a computer system of an electronic device in an exemplary embodiment of the present disclosure is shown schematically. DETAILED DESCRIPTION

[0053] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations may, however, be implemented in many different forms and should not be construed as limited to the examples set forth herein; rather, these implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of example implementations to those skilled in the art.

[0054] Moreover, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the disclosure. One skilled in the relevant art will recognize, however, that the

[0055] The block diagrams shown in the drawings are functional entities, and do not necessarily have to correspond to physically independent entities. That is, the functional entities can be implemented in software, or in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0056] The flowcharts shown in the drawings are only illustrative, and do not necessarily include all contents and operations / steps, nor are they necessarily executed in the order described. For example, some operations / steps can be further divided, and some operations / steps can be combined or partially combined, so the actual execution order can be changed according to the actual situation.

[0057] Turbine blade film hole is a common blade cooling structure in modern aero-engines, and the arrangement on the turbine blade has the characteristics of small and dense. The basic principle is to set small holes between the two layers of turbine blade walls, and to form a gas film by injecting high-temperature gas to reduce the surface temperature and thermal stress of the blade, thereby improving the service life and performance of the blade. Double-wall turbine blade film hole not only can effectively reduce the surface temperature of the blade, but also can improve the internal flow field structure of the blade, improve the working efficiency and thrust. The film hole arrangement aims to achieve passive cooling and active thermal management of the blade by arranging film holes on the surface of the blade, thereby improving the performance and life of the engine. It has the advantages of improving cooling effect, improving thrust effect, reducing thermal stress and thermal expansion, and reducing blade vibration.

[0058] In the design of traditional turbine blade film hole arrangement, the same column of film holes is usually selected as a group of continuous position and same pitch and hole direction for parameterized design or manual interactive film hole modeling based on sketches, and the flow and heat transfer characteristics of film cooling under different hole array arrangement are compared and studied by numerical simulation method to obtain the influence law of hole array arrangement on adiabatic cooling efficiency and film comprehensive cooling efficiency.

[0059] The disadvantages of the conventional turbine blade film hole arrangement optimization design are that, firstly, in the conventional turbine blade film hole parameterized design, each column of film holes on the blade surface is regarded as continuous and equidistantly arranged with the same punching direction. However, in order to achieve the best cooling effect, the punching direction and the hole spacing of each column of film holes can be different. The conventional film hole arrangement design method is difficult to parameterize the overall film hole arrangement of the turbine blade and cannot give the optimal film hole distribution strategy. Secondly, the conventional film hole arrangement optimization usually uses the DOE test design to construct the design space, selects the design sample points, performs numerical simulation calculation on each sample point, and then fits the simulation results by using the surrogate model. The surrogate model usually needs to manually design the feature extractor, rules or model structure, which may require the knowledge of domain experts and a large amount of manual labor. At the same time, the surrogate model has weak generalization ability when dealing with complex and diversified data sets, thus reducing the optimization efficiency of the turbine blade film hole arrangement. Finally, the conventional film hole arrangement optimization design objective function mainly focuses on the influence of the film hole arrangement on the blade cooling characteristics, and less considers the influence of the film hole arrangement on the blade stress.

[0060] Compared with the conventional turbine cooling blade, the turbine blade film hole diameter and the double-wall spacing of the double-wall turbine blade are only about 0.5 mm, which is much smaller than the turbine blade span height and flow direction length, and has a significant "cross-size" feature. At the same time, the cooling units are periodically arranged, and there are staggered and densely arranged film holes, impact holes and turbulence columns in each unit, and the structure is very complex, with small structure size, large number of holes / columns and periodic dense arrangement.

[0061] Therefore, in view of the characteristics that the conventional double-wall turbine blade film hole arrangement optimization method cannot fully consider the internal force of the blade, cannot parameterize the overall blade film hole, and cannot give the optimal film hole distribution strategy, a double-wall turbine blade film hole arrangement optimization design method is developed. The method combines parameterized modeling and CGAN neural network model to predict the turbine blade surface stress under different film hole arrangements when numerically analyzing and calculating the performance of the turbine blade. The optimization algorithm is used to quickly complete the turbine blade film hole arrangement optimization process. The method can greatly improve the efficiency of evaluating the film hole structural strength using numerical methods and reduce the calculation cost.

[0062] The implementation details of the technical solutions of the embodiments of the present disclosure are described in detail below.

[0063] Figure 1 A flowchart schematically showing a turbine blade film hole arrangement method in an exemplary embodiment of the present disclosure is shown. As shown in Figure 1 The turbine blade film hole arrangement method includes steps S101 to S103:

[0064] In step S101, a design sample space is constructed based on a value range of the film hole arrangement parameter of the turbine blade; the design sample space includes a plurality of turbine blade sample models with different film hole arrangements.

[0065] In step S102, a target function is established with a minimum stress prediction value of the turbine blade sample model as a target; the stress prediction value is calculated according to a pre-trained stress prediction model; and a constraint condition is established based on an upper limit value and a lower limit value of the design sample space.

[0066] In step S103, a film hole arrangement optimization model is established according to the target function and the constraint condition, and the film hole arrangement optimization model is solved to determine a target film hole arrangement parameter.

[0067] In the technical solution provided by some embodiments of the present disclosure, on the one hand, a target function is established with a minimum stress prediction value of the turbine blade sample model as a target, and the surface stress distribution of the turbine blade is taken into account in the optimization, so that the stress around the optimized film hole is relieved, and the phenomenon of film hole stress interference can be avoided; on the other hand, a pre-trained stress prediction model is used to calculate the stress value, which greatly reduces the time of numerical calculation in the optimization process compared with simulating and calculating the stress value of the turbine blade sample model, improves the optimization efficiency and accuracy, has strong nonlinear modeling capability compared with the traditional film hole optimization using a proxy model, can better adapt to complex data patterns and rules, has stronger adaptability and parallel computing capability, and has obvious advantages in processing large-scale data.

[0068] In the following, each step of the film hole arrangement method of the turbine blade in the example embodiment will be described in more detail with reference to the accompanying drawings and embodiments.

[0069] In step S101, a design sample space is constructed based on a value range of the film hole arrangement parameter of the turbine blade; the design sample space includes a plurality of turbine blade sample models with different film hole arrangements.

[0070] The film hole is an important structure in the double-walled turbine cooling blade, and the relative position of the film hole and the blade profile will significantly affect the outflow of the film hole and the flow characteristics of the cascade channel. At the same time, there is a complex compound angle between the film hole and the blade surface. Therefore, the position and angle of the film hole should be finely designed, and the parameterized modeling design should also consider these two aspects to determine appropriate parameterized design variables to facilitate the optimization of the model. The present disclosure proposes a parameterized design method related to the characteristic size, which can realize flexible adjustment of the shape and position of the film hole, and fully explore the design space on the basis of ensuring the correctness of the model.

[0071] In one embodiment of the present disclosure, the gas film hole arrangement parameters include a position parameter and a rotation angle parameter. The position parameter represents the arrangement mode of the gas film holes, and the rotation angle parameter represents the rotation angle of a single gas film hole.

[0072] (1) Position parameter

[0073] Figure 2 A schematic diagram of the arrangement of a gas film hole on a turbine blade in an exemplary embodiment of the present disclosure is shown schematically. Figure 2 As shown in the figure, the arrangement of the gas film holes on the turbine blade has the property of dense arrangement in columns. When the distance between the holes is too wide, the peripheral area of each hole will bear a greater load, which will lead to a decrease in the overall strength of the material, thereby affecting the stability and service life of the structure. Conversely, when the hole spacing is narrowed, stress interference occurs around the holes, which can easily lead to the formation of micro-cracks, and further cause fatigue damage and fracture of the material. Therefore, the spacing of the gas film holes should be within a reasonable range.

[0074] Since the position of the single-column gas film holes is still irregular, the present disclosure proposes to divide the gas film holes on the turbine blade into multiple groups, and divide the gas film holes located in the same column, having continuous positions, the same hole spacing, and the same rotation angle of the gas film holes on the turbine blade into the same group of gas film holes. In this way, each column of gas film holes is composed of several groups of gas film holes, and the center position of the gas film holes in each group is calculated, and the center positions of the gas film holes in different groups in the single-row gas film holes are connected to obtain the arrangement trajectory curve of the single-row gas film holes.

[0075] In one embodiment of the present disclosure, the position parameter includes one or more of the distance from the center position of a single group of gas film holes to the center axis, the distance from the starting gas film hole position to the center axis, the distance from the starting gas film hole position to the blade tip, the angle between the connection line of the single group of gas film holes and the horizontal axis, the number of the single group of gas film holes, and the spacing between the two gas film holes in the single group of gas film holes.

[0076] Specifically, Figure 3 A schematic diagram of a gas film hole position parameter in an exemplary embodiment of the present disclosure is shown schematically. As shown in the figure, Figure 3 The single-column gas film holes are divided into three groups, namely group 1, group 2, and group 3. S is the distance from the center position of the gas film hole to the center axis, and the three groups correspond to S1, S2, and S3, respectively.

[0077] Figure 4 A schematic diagram of a single group of gas film hole position parameters in an exemplary embodiment of the present disclosure is shown schematically. Taking a certain group of gas film holes after grouping as the parameterized object as an example, taking the first gas film hole in each group as the starting gas film hole position, the single group of gas film hole position parameters specifically include:

[0078] x: the distance from the starting gas film hole position to the center axis;

[0079] y: distance from the initial gas film hole position to the blade tip;

[0080] θ: angle between the single group of gas film holes and the horizontal axis;

[0081] N: number of single group of gas film holes;

[0082] d: distance between each two gas film holes in a single group of gas film holes.

[0083] (2) Rotation angle parameter

[0084] In the double-wall turbine blade, the gas film hole angle has an important influence on the blade strength and stability. The hole angle can affect the rigidity and bending vibration characteristics of the blade. Due to the influence of gas flow, the gas film hole will produce slight vibration when the blade is running, which further affects the vibration characteristics of the blade. By optimizing the angle of the gas film hole, the stress level around the edge of the gas film hole and the surface of the blade can be reduced, and the durability and life of the blade can be improved.

[0085] In one embodiment of the present disclosure, the rotation angle parameter includes one or more of the rotation angle normal angle, the rotation angle circumferential angle, and the rotation angle circumferential angle of a single gas film hole.

[0086] Figure 5 A schematic diagram of a gas film hole rotation parameter in an exemplary embodiment of the present disclosure is schematically shown. Due to the complex three-dimensional bending and twisting features of the turbine blade aerodynamic shape, it is difficult to determine the deflection angle of the gas film hole. Therefore, a local coordinate system needs to be established to realize the parameterized design of the gas film hole. As shown in Figure 5 , a local coordinate system is established along the normal line of the blade surface normal to the gas film hole center point, the tangential line is the y-axis, and the blade height direction is the z-axis.

[0087] Figure 6 A schematic diagram of a single gas film hole rotation parameter in an exemplary embodiment of the present disclosure is schematically shown. The gas film hole is designed as a parameter positioning parameter with the rotation angle normal angle a of the gas film hole around the y-axis of the local coordinate system; the rotation angle circumferential angle β of the gas film hole around the x-axis of the local coordinate system; and the rotation angle circumferential angle γ of the gas film hole around the z-axis of the local coordinate system.

[0088] In this way, the gas film hole arrangement parameters specifically include: the distance S from the gas film hole center position of each group of gas film holes to the center axis, the distance x from the initial gas film hole position to the center axis, the distance y from the initial gas film hole position to the blade tip, the angle θ between the single group of gas film holes and the horizontal axis, the number N of single group of gas film holes, the distance d between each two gas film holes in a single group of gas film holes, and the rotation angle normal angle a, the rotation angle circumferential angle β, and the rotation angle circumferential angle γ of a single gas film hole.

[0089] Then in step S101, a design sample space is constructed based on the value range of the turbine blade film hole arrangement parameters. Specifically, after the film hole arrangement parameters are preset, a reasonable value range can be set for each film hole arrangement parameter. According to this value range, a plurality of turbine blade sample models with different film hole arrangements can be constructed as the design sample space by using the design of experiments (DOE) method according to the variation range of the design variables.

[0090] Wherein, DOE (Design of Experiments) is an experimental design method used to explore and verify the influence of factors on results. In DOE, experiments are usually divided into multiple combinations, each combination controls a factor and measures its influence on the result. In this way, the influence of factors on the result can be more comprehensively understood, and the best factor combination can be determined.

[0091] In step S102, a target function is established with the minimum stress prediction value of the turbine blade sample model as the target; wherein the stress prediction value is calculated according to the pre-trained stress prediction model; and a constraint condition is established based on the upper limit value and the lower limit value of the design sample space.

[0092] Specifically, in the optimization process, in order to balance the optimization efficiency and accuracy, the stress prediction value of the turbine blade sample model calculated by the stress prediction model is used as the stress analysis result of the model surface, which replaces the actual stress simulation value. In this way, it is not necessary to perform simulation calculation for each turbine blade sample model, which can greatly reduce the calculation process and improve the optimization efficiency of the film hole arrangement design.

[0093] In an embodiment of the present disclosure, the method further comprises pre-training the stress prediction model, wherein the pre-training the stress prediction model comprises: extracting the turbine blade sample models in the design sample space as training samples; performing blade strength simulation on the training samples to obtain stress simulation values corresponding to each training sample, respectively; and training the stress prediction model according to the stress simulation values corresponding to each training sample, respectively.

[0094] Specifically, the stress prediction model can be a CGAN (Conditional Generative Adversarial Network) model. The CGAN simply puts the idea of supervised learning on the generation model, which can be expected to generate corresponding output according to the label input by the network.

[0095] Figure 7 The model structure diagram of a CGAN model in an exemplary embodiment of the present disclosure is schematically shown. As shown in Figure 7 The CGAN model includes a defined generator and a discriminator.

[0096] Figure 8Fig. 1 schematically shows a flowchart of a stress prediction model training method according to an example embodiment of the present disclosure. As shown in Fig. 1, the stress prediction model is trained according to the stress simulation values corresponding to each of the training samples, including: Figure 8

[0097] In step S801, the surface of the training sample is two-dimensionally unfolded to obtain a gas film hole distribution matrix.

[0098] In step S802, the definition generator is used to calculate the stress prediction value corresponding to the training sample according to the gas film hole distribution matrix.

[0099] In step S803, the discriminator is used to determine whether the stress prediction value belongs to the generator or belongs to the stress simulation value.

[0100] In step S804, the definition generator and the discriminator are alternately trained to modify the model parameters until the training stop condition is met.

[0101] Specifically, in step S801, the surface of the training sample is two-dimensionally unfolded to obtain a gas film hole distribution matrix.

[0102] Figure 9 Fig. 2 schematically shows a flowchart of obtaining a gas film hole distribution matrix according to an example embodiment of the present disclosure. Each training sample is a turbine blade sample model. The surface of the turbine blade is two-dimensionally unfolded, and then the surface is divided into grids to extract coordinate data of each grid part, and then a gas film hole distribution matrix is obtained. As shown in Fig. 2, the boxes in the matrix indicate that there are gas film holes in the part, and a, b, and g are rotation angle parameters of the gas film holes. The finally output stress distribution result is also in the form of a matrix. Figure 9

[0103] In step S802, the definition generator is used to calculate the stress prediction value corresponding to the training sample according to the gas film hole distribution matrix.

[0104] In an example embodiment of the present disclosure, the specific process of step S802 is as follows: for a training sample, the gas film hole distribution matrix of the training sample is encoded to obtain a first feature map; the boundary condition is converted through a first fully connected layer and then added to the first feature map in an element manner to obtain a stress distribution result; the stress distribution result is decoded to obtain a stress value; and the training sample is traversed to obtain the stress value corresponding to each of the training samples.

[0105] As shown in Fig. 1, the stress prediction model is trained according to the stress simulation values corresponding to each of the training samples, including: Figure 7 ​​As shown, the definition generator first encodes Input1 into a feature map, where Input1 is the air film pore distribution matrix, which is the geometric information of the air film pore arrangement; then, Input2 is transformed through the first fully connected layer (FC1) and added to the feature map element by element, where Input2 is the boundary condition for training the model; finally, the result is decoded into the output result to obtain the stress value, which is the stress prediction value generated by the definition generator.

[0106] In step S803, the discriminator is used to determine whether the predicted stress value belongs to the generator or the simulated stress value;

[0107] In one embodiment of this disclosure, step S803 is specifically as follows: For a training sample, the air film pore distribution matrix, stress prediction value, and stress simulation value of the training sample are connected and encoded to obtain a second feature map; the boundary conditions are transformed by the second fully connected layer and added to the second feature map element by element to obtain feature results; the feature results are transformed and evaluated to determine whether the feature results belong to the generator generated or to the stress simulation value.

[0108] like Figure 7 As shown, after the generator converts geometric information and boundary conditions into stress distribution, a discriminator evaluates the predicted data. The discriminator first uses an encoding block to concatenate and encode the geometric information and stress distribution into a feature map. The stress distribution includes the stress predictions generated by the generator and the simulated stress values ​​obtained from the simulation. Then, Input2 (boundary conditions) is added element-wise to this feature map through a second fully connected layer (FC2). The result is finally transformed using a "Conv" block to identify whether the stress distribution belongs to the generator-generated value or the simulated stress value.

[0109] In step S804, the definition generator and the discriminator are trained alternately to modify the model parameters until the training stopping condition is met.

[0110] Through model training, the stress prediction values ​​generated by the definition generator can ultimately be verified by the discriminator. Therefore, the definition generator in the trained stress prediction model can calculate the stress prediction values ​​of the turbine blade sample model. These stress prediction values ​​are close to the stress simulation values ​​obtained through blade strength simulation, without the need for complex simulation experiments, which can greatly improve the optimization efficiency of film cooling hole arrangement.

[0111] Once the stress prediction model is trained, an optimization model for the air film perforation arrangement can be constructed. This is based on the surface stress values ​​of the turbine blade. The minimum objective function is constructed with the distance S of the center of the film hole to the center axis, the distance x of the starting film hole position to the center axis, the distance y of the starting film hole position to the blade tip, the angle θ of the single group of film hole connecting lines to the horizontal axis, the number N of the single group of film holes, the distance d between the two film holes of the single group of film holes, and the rotation angle normal angle α, the rotation angle circumferential angle β and the rotation angle circumferential angle γ of the single film hole as the decision variables, and the upper limit value and the lower limit value of the design sample space are set as the constraint conditions, and the film hole arrangement optimization model is constructed as follows:

[0112] (5)

[0113] (6)

[0114] (7)

[0115] Wherein, X is a turbine blade sample model, U and V are the upper limit value and the lower limit value of the design sample space.

[0116] In step S103, the film hole arrangement optimization model is established according to the objective function and the constraint condition, and the film hole arrangement optimization model is solved to determine the target film hole arrangement parameters.

[0117] Specifically, after the film hole arrangement optimization model is established according to the objective function and the constraint condition in step S102, the model can be solved by using an optimization algorithm. The optimization algorithm is not limited in the present disclosure, for example, genetic algorithm, ant colony algorithm, simulated annealing algorithm and the like can be solved.

[0118] In an embodiment of the present disclosure, the overall film hole arrangement design optimization system is completely described, and the specific steps are as follows:

[0119] (1) The distance S of the center of the film hole to the center axis, the distance x of the starting film hole position to the center axis, the distance y of the starting film hole position to the blade tip, the angle θ of the single group of film hole connecting lines to the horizontal axis, the number N of the single group of film holes, the distance d between the two film holes of the single group of film holes, the normal angle α of the film hole, the circumferential angle β and the circumferential angle γ of the rotation angle are taken as the design variables, and the value range is given to establish the design space;

[0120] (2) A series of design sample spaces can be constructed in the design space by using the DOE test design method ;

[0121] (3) Based on the film hole parameterized design method, the response of each sample point (i.e. a turbine blade sample model) is calculated by using the blade strength simulation method​​​ i.e. the turbine blade surface stress distribution under each design sample point;

[0122] (4) Based on the existing DOE test design sample data, the CGAN neural network model is trained and the stress distribution prediction value is output;

[0123] (5) The multi-island genetic algorithm is used for optimization calculation to determine the gas film hole arrangement parameters under the optimal stress distribution.

[0124] Based on the above method, the disclosure takes the distance S from the densely arranged gas film hole center position to the center axis, the distance x from the starting gas film hole position to the center axis, the distance y from the starting gas film hole position to the blade tip, the angle θ of the single group of gas film holes to the horizontal axis, the number N of the single group of gas film holes, the distance d between the two gas film holes of the single group of gas film holes, the gas film hole normal angle , the circumferential angle and the circumferential angle as design variables; the DOE test design method and numerical simulation technology are used to obtain the gas film hole arrangement design sample space and its response; the objective function is established by the numerical distribution of stress, the CGAN neural network model is trained based on the simulation calculation real result, and the turbine blade surface stress distribution is predicted; finally, the gas film hole arrangement design optimization method is established by combining the optimization algorithm, and the gas film hole arrangement mode with good mechanical properties can be obtained through optimization calculation. The main technical effects are:

[0125] (1) Based on the influence of gas film hole arrangement on the strength of turbine blade in the complex service environment of turbine blade, a gas film hole arrangement parameterization technology is developed. The gas film hole arrangement mode and rotation angle are considered in the turbine blade gas film hole arrangement optimization design, and the design variables associated with it will be automatically coordinated and changed, ensuring the integrity of the overall geometric model and local cooling features. The stress of the optimized gas film hole is relieved, and the phenomenon of gas film hole stress interference is avoided. Compared with the traditional gas film hole arrangement which only considers the local gas film hole spacing or regards each column of gas film holes as a continuous equal spacing and the same punching direction arrangement, it is more comprehensive.

[0126] (2) Combining DOE experimental design, a conditional generative adversarial neural network (CGAN) model and optimization algorithm were used to construct an optimization system with the arrangement and rotation angle of the film film holes as parameters and the stress distribution on the turbine blade surface as the objective function. Compared with the traditional film film hole arrangement optimization design, the stress distribution on the turbine blade surface was incorporated into the optimization consideration. At the same time, the use of the neural network model greatly reduced the time of numerical calculation in the optimization process, improved the optimization efficiency and accuracy, and compared with the traditional film film hole optimization using a surrogate model, it has strong nonlinear modeling ability, can better adapt to complex data patterns and rules; automatic feature extraction simplifies the model construction process; stronger adaptability and parallel computing ability, and has obvious advantages in processing large-scale data.

[0127] Figure 10 This schematic diagram illustrates the composition of a turbine blade film perforation arrangement device according to an exemplary embodiment of the present disclosure, such as... Figure 10 As shown, the turbine blade film pore arrangement device 1000 may include a sample module 1001, a model module 1002, and a solution module 1003. Wherein:

[0128] The sample module 1001 is used to construct a design sample space based on the range of values ​​of the air film hole arrangement parameters of the turbine blade; wherein, the design sample space includes multiple turbine blade sample models with different air film hole arrangements.

[0129] Model module 1002 is used to establish an objective function with the goal of minimizing the stress prediction value of the turbine blade sample model; wherein the stress prediction value is calculated based on a pre-trained stress prediction model; and to establish constraints based on the upper and lower limits of the design sample space.

[0130] The solver module 1003 is used to establish an optimization model for the arrangement of air film pores based on the objective function and the constraints, and to solve the optimization model for the arrangement of air film pores to determine the target air film pore arrangement parameters.

[0131] According to an exemplary embodiment of this disclosure, the air film perforation arrangement parameters include position parameters and rotation angle parameters; wherein, the turbine blade includes multiple sets of air film perforations, and the position parameters include one or more of the following: the distance from the center position of a single set of air film perforations to the central axis, the distance from the starting position of the air film perforation to the central axis, the distance from the starting position of the air film perforation to the blade tip, the angle between the line connecting the single sets of air film perforations and the transverse axis, the number of single sets of air film perforations, and the spacing between any two air film perforations in a single set; the rotation angle parameters include one or more of the following: the normal angle of rotation of a single air film perforation, the circumferential angle of rotation, and the circumferential angle of rotation.

[0132] According to an example embodiment of the present disclosure, the film holes in the same column, position continuous, same film hole hole spacing and same film hole rotation angle of the turbine blade are a group of film holes.

[0133] According to an example embodiment of the present disclosure, the turbine blade film hole arrangement device 1000 further comprises a training model module for pre-training the stress prediction model, comprising: extracting a turbine blade sample model in the design sample space as a training sample; performing blade strength simulation on the training sample to obtain a stress simulation value corresponding to each training sample; and training the stress prediction model according to the stress simulation value corresponding to each training sample.

[0134] According to an example embodiment of the present disclosure, the stress prediction model is a CGAN model, the CGAN model comprises a defined generator and a discriminator, and the training model module is further configured to perform two-dimensional expansion on the surface of the training sample to obtain a film hole distribution matrix; calculate the stress prediction value corresponding to the training sample according to the film hole distribution matrix by using the defined generator; use the discriminator to determine whether the stress prediction value belongs to the generator or belongs to the stress simulation value; and alternately train the defined generator and the discriminator to modify the model parameters until a training stop condition is met.

[0135] According to an example embodiment of the present disclosure, the training model module is further configured to, for a training sample, encode the film hole distribution matrix of the training sample to obtain a first feature map; add the boundary condition converted by a first fully connected layer to the first feature map in an element manner to obtain a stress distribution result; decode the stress distribution result to obtain a stress value; and traverse the training sample to obtain a stress value corresponding to each training sample.

[0136] According to an example embodiment of the present disclosure, the training model module is further configured to, for a training sample, connect and encode the film hole distribution matrix, the stress prediction value and the stress simulation value of the training sample to obtain a second feature map; add the boundary condition converted by a second fully connected layer to the second feature map in an element manner to obtain a feature result; and convert and evaluate the feature result to determine whether the feature result belongs to the generator or belongs to the stress simulation value.

[0137] The specific details of each module in the above turbine blade film hole arrangement device 1000 have been described in detail in the corresponding turbine blade film hole arrangement method, and therefore will not be described here.

[0138] It should be noted that although several modules or units of the devices for action execution are mentioned in the foregoing detailed description, such division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functions of one module or unit described above can be further divided into embodied by multiple modules or units.

[0139] In the exemplary embodiments of the present disclosure, a storage medium capable of implementing the above method is also provided. Figure 11 The schematic diagram of a computer readable storage medium in the exemplary embodiments of the present disclosure is schematically shown as Figure 11 As shown, a program product 1100 for implementing the above method according to the embodiments of the present disclosure is described, which can adopt a portable compact disc read-only memory (CD-ROM) and include program codes, and can run on a terminal device such as a mobile phone. However, the program product of the present disclosure is not limited thereto, and in this document, the readable storage medium can be any tangible medium containing or storing a program which can be used by or in conjunction with an instruction execution system, apparatus or device.

[0140] In the exemplary embodiments of the present disclosure, an electronic device capable of implementing the above method is also provided. Figure 12 The structural schematic diagram of a computer system of an electronic device in the exemplary embodiments of the present disclosure is schematically shown.

[0141] It should be noted that, Figure 12 The computer system 1200 of the electronic device shown is only an example, and should not bring any limitation to the functions and use range of the embodiments of the present disclosure.

[0142] As Figure 12 shown, the computer system 1200 includes a central processing unit (CPU) 1201, which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM) 1202 or loaded from a storage portion 1208 into a random access memory (RAM) 1203. In the RAM 1203, various programs and data required for system operation are also stored. The CPU 1201, the ROM 1202 and the RAM 1203 are connected to each other through a bus 1204. An input / output (I / O) interface 1205 is also connected to the bus 1204.

[0143] The following components are connected to the I / O interface 1205: an input part 1206 including a keyboard, a mouse, etc.; an output part 1207 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), etc., and a speaker, etc.; a storage part 1208 including a hard disk, etc.; and a communication part 1209 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication part 1209 performs communication processing via a network such as the Internet. A drive 1210 is also connected to the I / O interface 1205 as necessary. A removable medium 1211 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 1210 as necessary, so that a computer program read out therefrom is installed in the storage part 1208 as necessary.

[0144] In particular, according to embodiments of the present disclosure, the processes described below with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by the communication part 1209, and / or installed from the removable medium 1211. When the computer program is executed by the central processing unit (CPU) 1201, various functions defined in the system of the present disclosure are executed.

[0145] It should be noted that the computer-readable medium in the present disclosure can be a computer-readable signal medium or a computer-readable storage medium or any combination thereof. The computer-readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination thereof. More specific examples of the computer-readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (Compact Disc Read-Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, device or apparatus. In the present disclosure, the computer-readable signal medium can include a data signal carrying a computer-readable program code in a baseband or as a part of a carrier wave. Such a propagated data signal can take various forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, which can send, propagate or transmit a program for use by or in conjunction with an instruction execution system, device or apparatus. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, or the like, or any suitable combination thereof.

[0146] The flowcharts and block diagrams in the drawings illustrate the possible implementation architectures, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowcharts or block diagrams can represent a module, a program segment, or a portion of code that contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different orders than that shown in the drawings. For example, two blocks that are shown in succession can actually be executed substantially concurrently, or they can sometimes be executed in reverse order, depending on the functionality involved. It should also be noted that each block in the block diagrams or flowcharts, and combinations of blocks in the block diagrams or flowcharts, can be implemented by a dedicated hardware-based system that performs specified functions or operations, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0147] The units described in the embodiments of the present disclosure can be implemented by software, or by hardware, or by a combination of software and hardware. The units described can also be located in a processor.

[0148] As another aspect, the present disclosure also provides a computer readable medium, which can be included in the electronic device described in the above embodiments, or can exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to implement the method described in the above embodiments.

[0149] It should be noted that although several modules or units for performing actions are mentioned in the above detailed description, the division into the modules or units is not mandatory. In fact, according to the embodiments of the present disclosure, features and functions of two or more modules or units described above can be embodied in one module or unit. Conversely, features and functions of one module or unit described above can be further divided into a plurality of modules or units.

[0150] From the above description of the embodiments, those skilled in the art will readily appreciate that the example embodiments described herein can be implemented by software and / or by hardware coupled with software. Accordingly, the technical solutions of the embodiments of the present disclosure can be embodied in the form of a software product. The software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, or the like) or a network, and includes a number of instructions for causing a computing device (which can be a personal computer, a server, a terminal, or a network device, etc.) to perform the methods according to the embodiments of the present disclosure.

[0151] Other embodiments of the present disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the application disclosed herein. The present disclosure is intended to cover any variations, uses, or adaptations of the present disclosure following the general principles thereof and including such departures from the present disclosure that come within known use or custom in the art to which the present disclosure pertains.

[0152] It should be understood that the present disclosure is not limited to the precise structures described and illustrated in the above description and accompanying drawings, and that various modifications and changes can be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method of arranging film holes of a turbine blade, characterized by, The method comprises the following steps: constructing a design sample space based on the value range of the turbine blade film hole arrangement parameters; wherein the design sample space comprises a plurality of turbine blade sample models with different film hole arrangements; establishing an objective function with the minimum stress prediction value of the turbine blade sample model as the target; wherein the stress prediction value is calculated according to a pre-trained stress prediction model; and establishing a constraint condition based on the upper limit and lower limit of the design sample space; establishing a film hole arrangement optimization model according to the objective function and the constraint condition, and solving the film hole arrangement optimization model to determine the target film hole arrangement parameters; the stress prediction model is a CGAN model, and the CGAN model comprises a definition generator and a discriminator. The steps of pre-training the stress prediction model comprise: extracting the turbine blade sample models in the design sample space as training samples; performing blade strength simulation on the training samples to obtain the stress simulation values corresponding to each training sample, respectively; two-dimensionally expanding the surface of the training sample to obtain a film hole distribution matrix; calculating the stress prediction value corresponding to the training sample according to the film hole distribution matrix by using the definition generator; using the discriminator to distinguish whether the stress prediction value belongs to the generator or belongs to the stress simulation value; alternately training the definition generator and the discriminator to modify the model parameters until the training stop condition is met.

2. The turbine blade film hole arrangement method according to claim 1, characterized by, The film hole arrangement parameters comprise position parameters and rotation angle parameters; wherein the turbine blade comprises a plurality of groups of film holes, the position parameters comprise one or more of the distance from the center position of a single group of film holes to the center axis, the distance from the starting film hole position to the center axis, the distance from the starting film hole position to the blade tip, the included angle between the single group of film hole connecting lines and the horizontal axis, the number of single group of film holes, and the distance between two film holes in a single group of film holes; and the rotation angle parameters comprise one or more of the rotation angle normal angle, the rotation angle circumferential angle, and the rotation angle circumferential angle of a single film hole.

3. The turbine blade film hole arrangement method according to claim 2, characterized by, The film holes located in the same column, with continuous positions, the same film hole spacing, and the same film hole rotation angle on the turbine blade are a group of film holes.

4. The turbine blade film hole arrangement method according to claim 1, characterized by, The calculation of the stress value corresponding to the training sample according to the film hole distribution matrix by using the definition generator comprises: encoding the film hole distribution matrix of a training sample to obtain a first feature map; adding the boundary conditions to the first feature map in an element manner after conversion through a first fully connected layer to obtain a stress distribution result; decoding the stress distribution result to obtain a stress value; traversing the training samples to obtain the stress values corresponding to each training sample, respectively.

5. The turbine blade film hole arrangement method according to claim 1, characterized by, The use of the discriminator to distinguish whether the stress prediction value belongs to the generator or belongs to the stress simulation value comprises: connecting and encoding the film hole distribution matrix, the stress prediction value, and the stress simulation value of a training sample to obtain a second feature map; adding the boundary conditions to the second feature map in an element manner after conversion through a second fully connected layer to obtain a feature result; The feature results are converted and evaluated to determine whether the feature results belong to the generator generation or to the stress simulation value.

6. An arrangement of film holes for a turbine blade, characterized in that The method comprises the following steps: a sample module, configured to construct a design sample space based on a value range of a turbine blade film hole arrangement parameter; wherein the design sample space comprises a plurality of turbine blade sample models with different film hole arrangements; a model module, configured to establish an objective function with a minimum stress prediction value of the turbine blade sample model as a target; wherein the stress prediction value is calculated according to a pre-trained stress prediction model; and establish a constraint condition based on an upper limit value and a lower limit value of the design sample space; a solving module, configured to establish a film hole arrangement optimization model according to the objective function and the constraint condition, and solve the film hole arrangement optimization model to determine a target film hole arrangement parameter; a training model module, wherein the stress prediction model is a CGAN model, the CGAN model comprises a defined generator and a discriminator, the training model module is configured to extract turbine blade sample models in the design sample space as training samples; perform blade strength simulation on the training samples to obtain respective stress simulation values corresponding to each of the training samples; perform two-dimensional expansion on surfaces of the training samples to obtain film hole distribution matrices; calculate stress prediction values corresponding to the training samples according to the film hole distribution matrices by using the defined generator; use the discriminator to determine whether the stress prediction values belong to the generator generation or to the stress simulation value; and alternately train the defined generator and the discriminator to modify model parameters until a training stop condition is met. 7.A computer readable storage medium having stored thereon a computer program, the program being executed by a processor to implement the turbine blade film hole arrangement method according to any one of claims 1 to 5.

8. An electronic device, comprising: The method comprises the following steps: one or more processors; a storage device configured to store one or more programs, when the one or more programs are executed by the one or more processors, the one or more programs cause the one or more processors to implement the turbine blade film hole arrangement method according to any one of claims 1 to 5.

Images