Method and device for designing compressor blade profile, electronic equipment, medium and product

Abstract

The embodiments of the present disclosure disclose a design method and device of a compressor blade profile, electronic equipment, medium and product. The design scene parameters of the compressor blade profile are determined. The design scene parameters are used to indicate the working environment of the compressor blade profile. A target model matched with the design scene parameters is determined. The optimization target of the compressor blade profile and the constraint condition corresponding to the optimization target are determined. The optimization target is used to indicate the aerodynamic performance improvement target of the compressor blade profile. The target optimization variable of the compressor blade profile is obtained by the target model based on the optimization target and the constraint condition, and the design scheme of the compressor blade profile is determined based on the target optimization variable. The target optimization variable is used to indicate the core geometric quantization parameter of the compressor blade profile.

Description

Technical Field

[0001] This disclosure relates to the technical field of compressor design, and more particularly to a method, apparatus, electronic equipment, medium, and product for designing compressor blade profiles. Background Technology

[0002] In the compressor design system, blade shape design is an important tool for converting aerodynamic parameters into geometric coordinates, playing a crucial role in the success of the design. Blade shape design, based on one-dimensional design and flow path design, uses meridional plane information as initial conditions and obtains three-dimensional blade solid coordinates through certain design methods, aiming to make the aerodynamic environment constrained by the obtained blade geometry as close as possible to the aerodynamic design objective (velocity triangle).

[0003] In related technologies, the design parameters of the compressor blades are typically adjusted manually and repeatedly modified to achieve the design goals. This method has a long design cycle, low efficiency, and is highly dependent on the experience of the designers. Summary of the Invention

[0004] This disclosure addresses some deficiencies mentioned in the background art by providing a method, apparatus, electronic device, medium, and product for designing compressor blades.

[0005] In a first aspect, embodiments of this disclosure provide a method for designing compressor blade profiles, comprising:

[0006] Determine the design scenario parameters for the compressor blade profile; wherein, the design scenario parameters are used to indicate the working environment of the compressor blade profile;

[0007] Determine the target model that matches the design scenario parameters;

[0008] Determine the optimization objective of the compressor airfoil and the corresponding constraints; wherein the optimization objective is used to indicate the aerodynamic performance improvement target of the compressor airfoil;

[0009] Based on the optimization objective and the constraints, the target optimization variables of the compressor blade profile are obtained through the target model, and the design scheme of the compressor blade profile is determined based on the target optimization variables; wherein, the target optimization variables are used to indicate the core geometric quantification parameters of the compressor blade profile.

[0010] In one embodiment of the first aspect, determining the target model that matches the design scenario parameters includes:

[0011] Obtain the training scenario parameters of the trained model;

[0012] The training scene parameters that meet the preset similarity requirements with the designed scene parameters are determined as the target scene parameters;

[0013] The trained model corresponding to the target scene parameters is determined as the target model.

[0014] In one embodiment of the first aspect, determining the optimization objective of the compressor blade profile and the constraints corresponding to the optimization objective includes:

[0015] Based on the operating parameters of the compressor airfoil, the optimization objectives of the compressor airfoil are determined; wherein, the optimization objectives include, but are not limited to: reducing the design point loss coefficient of the compressor airfoil and increasing the usable angle of attack range of the compressor airfoil;

[0016] When the optimization objective is to reduce the compressor blade design point loss coefficient, the constraint condition is that the available angle of attack range is not reduced.

[0017] When the optimization objective is to increase the available angle of attack range of the compressor blade, the constraint condition is that the design point loss coefficient should not increase.

[0018] In one embodiment of the first aspect, before passing the target model based on the optimization objective and the constraints, the method further includes:

[0019] Obtain the model training environment; wherein the model training environment is used to indicate the direction of model training;

[0020] Based on the model training environment, model training samples are generated;

[0021] The neural network model to be trained is trained based on the training samples of the model to obtain the trained model; wherein, the trained model includes the target model.

[0022] In one embodiment of the first aspect, training the neural network model to be trained based on the model training samples to obtain the trained model includes:

[0023] The linear rectified function is determined as the activation function of the neural network model to be trained, and the mean squared error function is determined as the loss function of the neural network model to be trained.

[0024] The neural network model to be trained is trained based on the activation function, the loss function, and the model training samples to obtain the trained model.

[0025] In one embodiment of the first aspect, the model training environment includes at least one of the following: a state space, a state transition probability, an action space, and a reward function; wherein the state space is used to indicate the compressor blade profile parameter variables to be optimized, and the action space is used to indicate the changes in the compressor blade profile parameter variables to be optimized.

[0026] In a second aspect, embodiments of this disclosure provide a compressor blade design apparatus, comprising:

[0027] The first determining module is used to determine the design scenario parameters of the compressor blade profile; wherein, the design scenario parameters are used to indicate the working environment of the compressor blade profile;

[0028] The second determining module is used to determine the target model that matches the design scenario parameters;

[0029] The third determining module is used to determine the optimization target of the compressor blade profile and the constraints corresponding to the optimization target; wherein, the optimization target is used to indicate the aerodynamic performance improvement target of the compressor blade profile;

[0030] The design module is used to obtain the target optimization variables of the compressor blade profile through the target model based on the optimization objective and the constraints, and to determine the design scheme of the compressor blade profile based on the target optimization variables; wherein, the target optimization variables are used to indicate the core geometric quantification parameters of the compressor blade profile.

[0031] In a third aspect, an electronic device is provided, comprising: a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is in operation, the processor communicates with the memory via the bus, and the machine-readable instructions, when executed by the processor, perform the steps of the first aspect above, or any possible implementation of the first aspect.

[0032] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, performs the steps of the first aspect or any possible implementation thereof.

[0033] In a fifth aspect, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the steps of the method described in the preceding aspects.

[0034] As will be described in detail below, a compressor airfoil design method, apparatus, electronic device, medium, and product according to embodiments of this disclosure are provided. By first clarifying the design scenario parameters of the compressor airfoil to match a suitable target model, then locking in the optimization objectives and corresponding constraints related to aerodynamic performance improvement, and finally outputting the core geometrically quantified target optimization variables based on the target model to form a design scheme, the practicality and reliability of compressor airfoil design are improved. A systematic design logic is established through precise matching of design scenario parameters and the target model. The core geometrically quantified target optimization variables are directly derived by combining clear optimization objectives and constraints, eliminating the need for designers to manually and repeatedly adjust styling parameters. This avoids the strong dependence on designer experience in related technologies, while shortening the design cycle and improving design efficiency, effectively solving the problems of long design cycles and low efficiency in traditional manual design methods. Attached Figure Description

[0035] Figure 1 A flowchart illustrating the compressor blade design method provided in this embodiment of the disclosure;

[0036] Figure 2 A geometrical schematic diagram of the compressor blade profile provided in the compressor blade profile design method of this disclosure embodiment;

[0037] Figure 3 A characteristic diagram of the loss coefficient of the compressor blade profile at the design inlet Mach number for the compressor blade profile design method provided in this embodiment of the disclosure;

[0038] Figure 4 A schematic diagram of the thickness distribution of a straight blade in the compressor blade design method provided in this embodiment of the present disclosure;

[0039] Figure 5 A schematic diagram of the dimensionless local structural angle distribution of the airfoil from the leading edge to the trailing edge in the compressor airfoil design method provided in the embodiments of this disclosure;

[0040] Figure 6 A schematic diagram illustrating the convergence of the training model in the compressor blade design method provided in this embodiment of the present disclosure;

[0041] Figure 7 A geometric comparison diagram of the optimized compressor blade profile and the prototype in the compressor blade profile design method provided in the embodiments of this disclosure;

[0042] Figure 8 A comparison diagram of the angle-of-attack loss characteristics of the optimized compressor airfoil and the original airfoil at the design inlet Mach number in the compressor airfoil design method provided in this embodiment of the present disclosure;

[0043] Figure 9A schematic diagram of a compressor blade design device provided in an embodiment of this disclosure;

[0044] Figure 10 This is a schematic diagram of an electronic device provided in an embodiment of the present disclosure. Detailed Implementation

[0045] The present disclosure will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present disclosure and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the present disclosure are shown in the drawings, not the entire structure.

[0046] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0047] In this document, the term "and / or" merely describes a relationship, indicating that three relationships can exist. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Furthermore, the term "at least one" in this document means any combination of at least two of any one or more elements. For example, including at least one of A, B, and C can mean including any one or more elements selected from the set consisting of A, B, and C.

[0048] Research has shown that blade design is a crucial tool in compressor design, transforming aerodynamic parameters into geometric coordinates and playing a key role in the success of the design. Blade design, based on one-dimensional and flow path design, uses meridional information as initial conditions and employs specific design methods to obtain the three-dimensional solid coordinates of the blade. The aim is to make the aerodynamic environment constrained by the obtained blade geometry as close as possible to the aerodynamic design objective (velocity triangle).

[0049] In related technologies, the design parameters of the compressor blades are typically adjusted manually and repeatedly modified to achieve the design goals. This method has a long design cycle, low efficiency, and is highly dependent on the experience of the designers.

[0050] Based on the above research, this disclosure provides a compressor airfoil design method. It first clarifies the design scenario parameters of the compressor airfoil to match a suitable target model, then locks in the optimization objectives and corresponding constraints related to aerodynamic performance improvement, and finally outputs the core geometrically quantified target optimization variables based on the target model to form a design scheme. This improves the practicality and reliability of compressor airfoil design. By establishing a systematic design logic through precise matching of design scenario parameters and the target model, and directly deriving the core geometrically quantified target optimization variables based on clear optimization objectives and constraints, it eliminates the need for designers to manually and repeatedly adjust styling parameters, avoiding the strong dependence on designer experience in related technologies. Simultaneously, it shortens the design cycle, improves design efficiency, and effectively solves the problems of long design cycles and low efficiency inherent in traditional manual design methods.

[0051] Reference Figure 2 The diagram shown is a geometrical schematic of the compressor blade profile of the compressor blade profile design method provided in this embodiment of the present disclosure, including the trailing edge and the leading edge.

[0052] The outer surfaces of the trailing and leading edges are suction surfaces, while the inner surfaces of the trailing and leading edges are pressure surfaces.

[0053] To facilitate understanding of this embodiment, a compressor blade design method disclosed in this disclosure will first be described in detail. The execution entity of the compressor blade design method provided in this disclosure is generally a computer device with certain computing capabilities, such as a terminal device, a server, or other processing devices. In some possible implementations, the compressor blade design method can be implemented by a processor calling computer-readable instructions stored in memory.

[0054] See Figure 1 The diagram shows a flowchart of a compressor blade design method provided in this embodiment of the present disclosure. The method includes steps S101 to S104, wherein:

[0055] S101. Determine the design scenario parameters of the compressor blade profile; wherein, the design scenario parameters are used to indicate the working environment of the compressor blade profile.

[0056] In the embodiments of this disclosure, the design scenario parameters are the engineering design specifications of the compressor blade profile, which can indicate the operating environment of the compressor blade profile. For example, the design scenario parameters can be that the compressor blade profile operates in an environment of "inlet Mach number 0.75 and airflow deflection angle 35°".

[0057] S102. Determine the target model that matches the design scenario parameters.

[0058] In embodiments of this disclosure, a target model that matches the working environment of the compressor blade profile can be selected from the trained model by designing scenario parameters.

[0059] Here, the target model that matches the working environment of the compressor blade can be understood as the training environment being the working environment of the compressor blade during the training process of the target model.

[0060] For example, if the design scenario parameters are "inlet Mach number 0.75, airflow turning angle 35°", the model labeled "inlet Mach number 0.7-0.8, airflow turning angle 30°-40°" in the trained model can be identified as the target model.

[0061] Here, the training objective of the target model is the same as the optimization objective of the compressor blade profile.

[0062] S103. Determine the optimization objective of the compressor blade profile and the corresponding constraints; wherein, the optimization objective is used to indicate the aerodynamic performance improvement target of the compressor blade profile.

[0063] In the embodiments of this disclosure, firstly, the optimization objective of the compressor blade profile can be determined. Then, after determining the optimization objective, the corresponding constraints can be determined.

[0064] Here, the constraints corresponding to different optimization objectives are different.

[0065] S104. Based on the optimization objective and constraints, the target optimization variables of the compressor blade profile are obtained through the target model, and the design scheme of the compressor blade profile is determined based on the target optimization variables; wherein, the target optimization variables are used to indicate the core geometric quantification parameters of the compressor blade profile.

[0066] In embodiments of this disclosure, constraints can be defined as constraints of the target model to optimize the optimization objective.

[0067] Here, the parameters to be optimized, constraints, and design scenario parameters can be input into the target model for optimization to obtain the optimized parameters.

[0068] Here, after obtaining the optimized parameter variables, the "MISES Blade Aerodynamic Analysis Program" can be used to analyze the optimized parameter variables and obtain a performance evaluation report. Each iteration corresponds to one performance evaluation report.

[0069] Here, if the optimized parameter variables are determined to have reached the optimization target based on the performance evaluation report, the iteration stops and the optimized parameter variables obtained from the last iteration are output.

[0070] Afterwards, the optimized parameter variables obtained from the last iteration can be compared with the parameter variables to be optimized to obtain a performance comparison chart. The performance comparison chart can then be displayed to the user through the display module to intuitively show the improvement effect of the optimized parameter variables.

[0071] In the embodiments of this disclosure, firstly, design scenario parameters for the compressor airfoil are determined; wherein, the design scenario parameters are used to indicate the working environment of the compressor airfoil; secondly, a target model matching the design scenario parameters is determined; thirdly, optimization objectives and corresponding constraints for the compressor airfoil are determined; wherein, the optimization objectives are used to indicate the aerodynamic performance improvement objectives of the compressor airfoil; finally, based on the optimization objectives and constraints, target optimization variables for the compressor airfoil are obtained through the target model, and the design scheme of the compressor airfoil is determined based on the target optimization variables; wherein, the target optimization variables are used to indicate the core geometric quantification parameters of the compressor airfoil.

[0072] In the above implementation, by first clarifying the design scenario parameters of the compressor blade profile to match the suitable target model, then locking in the optimization objectives and corresponding constraints related to aerodynamic performance improvement, and finally outputting the core geometrically quantified target optimization variables based on the target model to form a design scheme, the practicality and reliability of compressor blade profile design are improved. By establishing a systematic design logic through precise matching of design scenario parameters and the target model, and directly deriving the core geometrically quantified target optimization variables based on clear optimization objectives and constraints, designers are not required to manually and repeatedly adjust the styling parameters, avoiding the strong dependence on designer experience in related technologies. This also shortens the design cycle, improves design efficiency, and effectively solves the problems of long design cycles and low efficiency in traditional manual design methods.

[0073] In an optional embodiment, determining the target model that matches the design scenario parameters specifically includes the following steps:

[0074] First, obtain the training scenario parameters of the trained model;

[0075] Secondly, the training scene parameters that meet the preset similarity requirements with the design scene parameters are determined as the target scene parameters;

[0076] Finally, the trained model corresponding to the target scene parameters is determined as the target model.

[0077] In embodiments of this disclosure, training scenario parameters used by each trained model during training are determined. Then, first training scenario parameters, including design scenario parameters, are determined.

[0078] Then, a second training scene parameter that meets the preset similarity requirement with the design scene parameter is determined from the first training scene parameter.

[0079] The preset similarity requirement is the first training scene parameter that has the highest similarity to the design scene parameter.

[0080] Finally, the second scene parameters are determined as the target scene parameters, and the trained model corresponding to the target scene parameters is determined as the target model.

[0081] In an optional embodiment, determining the optimization objective of the compressor blade profile and the corresponding constraints includes the following steps:

[0082] First, based on the operating parameters of the compressor airfoil, the optimization objectives of the compressor airfoil are determined; among them, the optimization objectives include, but are not limited to: reducing the design point loss coefficient of the compressor airfoil and increasing the usable angle of attack range of the compressor airfoil.

[0083] Secondly, with the optimization objective being to reduce the compressor blade design point loss coefficient, the constraint condition is that the available angle of attack range should not be reduced.

[0084] Finally, with the optimization objective being to increase the available angle of attack range of the compressor blade profile, the constraint condition is that the design point loss coefficient should not increase.

[0085] In the embodiments of this disclosure, the compressor airfoil design point loss coefficient and the available angle of attack range of the compressor airfoil are aerodynamic performance evaluation parameters.

[0086] Here, the aerodynamic performance evaluation parameters can be solved using the "MISES blade aerodynamic analysis program" (i.e., the S1 forward problem analysis program).

[0087] Reference Figure 3 The figure shown is a characteristic graph of the compressor blade design method provided in this embodiment of the compressor blade design method at the design inlet Mach number, where the horizontal axis is the angle of attack value and the vertical axis is the total pressure loss coefficient of the blade at that angle of attack.

[0088] Here, as Figure 3 As shown, the minimum loss coefficient on the angle-of-attack loss characteristic curve of the imported Mach number blade is denoted as ω0. Here, the two angle-of-attack values ​​θ1 and θ2 (θ2>θ1) corresponding to 2ω0 are the minimum usable angle of attack and the maximum usable angle of attack, respectively. Here, we can define... θ = θ2 - θ1 represents the available angle of attack range of the compressor blade.

[0089] Here, the loss coefficient can be obtained by solving the airflow field of the airfoil under different inlet airflow angles using the "MISES Airfoil Aerodynamic Analysis Program".

[0090] The compressor blade profile employs a parametric modeling method: first, a thickness distribution is generated on the mid-arc line of a straight line, and then the mid-arc line is modeled and the thickness is superimposed.

[0091] Here, regardless of the optimization objective, it must meet fixed design conditions. These fixed design conditions include at least one of the following: the outlet airflow angle of the compressor blade profile at the design point must meet design requirements, and the strength of the compressor blade profile must meet design requirements.

[0092] In an optional embodiment, before proceeding through the target model based on the optimization objective and constraints, the following steps are further included:

[0093] First, obtain the model training environment; the model training environment is used to indicate the direction of model training.

[0094] Secondly, based on the model training environment, model training samples are generated;

[0095] Finally, the neural network model to be trained is trained based on the model training samples to obtain the trained model; the trained model includes the target model.

[0096] In the disclosed embodiments, the neural network model to be trained can be trained in multiple model training environments. The model training environment is determined based on historical implementation experience. The neural network model to be trained is a fully connected neural network model.

[0097] Here, it is determined that the simulation training environment can transform the compressor blade optimization problem into a reinforcement learning problem.

[0098] Here, the model training environment includes at least one of the following: state space, state transition probability, action space, and reward function; wherein, the state space is used to indicate the compressor blade profile parameter variables to be optimized, and the action space is used to indicate the changes in the compressor blade profile parameter variables to be optimized.

[0099] Here, the state transition probability is the probability of transitioning to another state after taking an action in a certain state. The reward function is the immediate reward given when taking an action in a certain state, used to guide the algorithm to find the optimal solution, and is usually designed as a function positively correlated with the optimization objective.

[0100] The parameters involved in the reinforcement learning parameter tuning process include learning optimization parameters, exploration strategy parameters, experience replay parameters, network architecture parameters, and training process parameters. These parameters need to be adjusted and optimized in combination with task characteristics and algorithm features to ensure model training convergence. Compressor blade profiles with different training scenario parameters (inlet Mach number, airflow turning angle, etc.) can be trained and converged individually and then saved to form a model library.

[0101] Reference Figure 4The diagram shown illustrates the thickness distribution of a straight airfoil in the compressor airfoil design method provided in this embodiment. The thickness design divides the airfoil from front to back into a leading edge small circle, a leading half curve, a trailing half curve, and a trailing edge small circle. The leading and trailing half curves are cubic curves. Six design parameters are included to ensure airfoil strength constraints: leading edge thickness, leading half curve thickness coefficient, maximum thickness, maximum thickness location, trailing half curve thickness coefficient, and trailing edge thickness.

[0102] Here, the three design parameters—the thickness coefficient of the first half of the curve, the location of the maximum thickness, and the thickness coefficient of the second half of the curve—can be determined as the parameters to be optimized.

[0103] Wherein, φ1 is the thickness distribution parameter of the leading edge segment; φ2 is the thickness distribution parameter of the first half of the curve; φ3 is the thickness distribution parameter of the second half of the curve; and φ4 is the thickness distribution parameter of the trailing edge segment.

[0104] Reference Figure 5 The diagram shown is a schematic diagram of the dimensionless local structural angle distribution of the airfoil from the leading edge to the trailing edge in the compressor airfoil design method provided in this embodiment of the present disclosure. The mid-arc shape is formed by giving the inlet and outlet airflow angles, angle of attack and lag angle, and the dimensionless local structural angle distribution of the airfoil from the leading edge to the trailing edge.

[0105] Here, a 6th-order Bezier curve fitting method can be adopted on the angular distribution of the mid-arc construction. Among them, the five control points in the middle are variable, and each control point contains two variables in the x and y directions; together with the angle of attack and the lag angle as variables, the above 12 variables can be determined as the parameters to be optimized.

[0106] Here, we can determine the thickness coefficient of the first half of the curve, the location of the maximum thickness, the thickness coefficient of the second half of the curve, and 5 control points that are variable, each control point containing 2 variables in the x and y directions; plus 15 parameters to be optimized, namely the angle of attack and the lag angle.

[0107] Among them, the deep reinforcement learning algorithm continuously generates new compressor blade samples by modifying 15 parameters to be optimized in the above-mentioned blade thickness distribution and mid-curve shape during model training.

[0108] For example, during training, the state transition probability can be set to 1; the state space can be set to 15 parameters to be optimized; the action space can be defined as the adjustment of the 15 parameters to be optimized; the reward function can be set as follows: when the deviation of the airflow angle at the design point exit is greater than or equal to 1 degree, a penalty value is fed back; when the deviation of the airflow angle at the design point exit is less than 1 degree and the loss coefficient increases, a penalty value is fed back; when the deviation of the airflow angle at the design point exit is less than 1 degree and the loss coefficient does not decrease, a reward value is fed back, and the reward value is inversely proportional to the loss coefficient corresponding to the minimum / maximum usable angle of attack of the prototype.

[0109] In an optional embodiment, the neural network model to be trained is trained based on model training samples to obtain a trained model, specifically including the following steps:

[0110] First, the linear rectified function is determined as the activation function of the neural network model to be trained, and the mean squared error function is determined as the loss function of the neural network model to be trained.

[0111] Then, the neural network model to be trained is trained based on the activation function, loss function, and model training samples to obtain the trained model.

[0112] Here, the activation function can be the Rectified Linear Activation Function (ReLU), the mean squared error can be the Mean Squared Error Function (MSE), and the weights can be updated using the Stochastic Gradient Descent (SGD) method. The main parameters of the neural network model after convergence are shown in Table 1 below.

[0113] Table 1

[0114]

[0115] Reference Figure 6 The diagram shown is a convergence schematic of the training model in the compressor blade design method provided in this embodiment of the present disclosure.

[0116] The trained, converged, and saved model (i.e., the trained model mentioned above) can be called multiple times. For new compressor blade optimization problems, a trained model with similar design requirements can be directly called from the model library. After loading the trained, converged model, compressor blade optimization can be automatically performed and the optimization design results can be saved.

[0117] Reference Figure 7 The figure shown is a geometric comparison diagram between the optimized compressor blade profile and the prototype in the compressor blade profile design method provided in the embodiments of this disclosure. The optimized compressor blade profile increases the usable angle of attack range from 10.65° of the prototype to 12.5° at the design inlet Mach number, which is an increase of 17.37%.

[0118] The compressor blade design method provided in the embodiments of this disclosure can be experimentally verified through planar blade cascade tests.

[0119] Reference Figure 8 The figure shown is a comparison of the angle of attack loss characteristics of the optimized compressor blade and the original blade at the design inlet Mach number in the compressor blade design method provided in this embodiment of the present disclosure. The optimized compressor blade increases the usable angle of attack range at the design inlet Mach number from 10.5° of the prototype to 12.5°, an increase of 19.05%.

[0120] In the above implementation, except for the environment definition and parameter tuning, everything is automatically trained and optimized by the algorithm, which effectively reduces the reliance on the professionalism of the designers. In this embodiment, the blade optimization time is 10 minutes, which significantly shortens the design time compared with traditional manual adjustment. In this embodiment, the optimization goal is effectively achieved, and the usable angle of attack range of the target compressor blade at the design inlet Mach number is expanded.

[0121] Based on the same inventive concept, this disclosure also provides a compressor blade design device corresponding to the compressor blade design method. Since the principle of the device in this disclosure for solving the problem is similar to the compressor blade design method described above in this disclosure, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be described again.

[0122] Reference Figure 9 The diagram shown is a schematic of a compressor blade design device according to an embodiment of this disclosure. The device includes: a first determining module 91, a second determining module 92, a third determining module 93, and a design module 94; wherein:

[0123] The first determining module 91 is used to determine the design scenario parameters of the compressor blade profile; wherein, the design scenario parameters are used to indicate the working environment of the compressor blade profile;

[0124] The second determining module 92 is used to determine the target model that matches the design scenario parameters;

[0125] The third determining module 93 is used to determine the optimization target of the compressor blade profile and the constraint conditions corresponding to the optimization target; wherein, the optimization target is used to indicate the aerodynamic performance improvement target of the compressor blade profile;

[0126] Design module 94 is used to obtain the target optimization variables of the compressor blade profile through the target model based on the optimization objective and the constraints, and to determine the design scheme of the compressor blade profile based on the target optimization variables; wherein, the target optimization variables are used to indicate the core geometric quantification parameters of the compressor blade profile.

[0127] This disclosure improves the practicality and reliability of compressor blade design by first defining the design scenario parameters of the compressor blade to match a suitable target model, then locking in the optimization objectives and corresponding constraints related to aerodynamic performance improvement, and finally outputting the core geometrically quantified target optimization variables based on the target model to form a design scheme. By establishing a systematic design logic through precise matching of design scenario parameters and the target model, and directly deriving the core geometrically quantified target optimization variables based on clear optimization objectives and constraints, this eliminates the need for designers to manually and repeatedly adjust styling parameters, avoiding the strong reliance on designer experience in related technologies. It also shortens the design cycle, improves design efficiency, and effectively solves the problems of long design cycles and low efficiency inherent in traditional manual design methods.

[0128] The processing flow of each module in the device and the interaction flow between each module can be referred to the relevant descriptions in the above method embodiments, and will not be detailed here.

[0129] Corresponding to Figure 1 This disclosure also provides an electronic device 1000 for detecting the operational status of computing centers, as described in the following embodiments. Figure 10 The diagram shown is a structural schematic of an electronic device 1000 provided in an embodiment of this disclosure, including:

[0130] The system includes a processor 101, a memory 102, and a bus 103. The memory 102 stores execution instructions and includes main memory 1021 and external memory 1022. The main memory 1021, also called internal memory, temporarily stores computational data in the processor 101, as well as data exchanged with external memory such as a hard disk. The processor 101 exchanges data with the external memory 1022 through the main memory 1021. When the electronic device 1000 is running, the processor 101 communicates with the memory 102 through the bus 103, causing the processor 101 to execute the following instructions:

[0131] Determine the design scenario parameters for the compressor blade profile; wherein, the design scenario parameters are used to indicate the working environment of the compressor blade profile;

[0132] Determine the target model that matches the design scenario parameters;

[0133] Determine the optimization objective of the compressor airfoil and the corresponding constraints; wherein the optimization objective is used to indicate the aerodynamic performance improvement target of the compressor airfoil;

[0134] Based on the optimization objective and the constraints, the target optimization variables of the compressor blade profile are obtained through the target model, and the design scheme of the compressor blade profile is determined based on the target optimization variables; wherein, the target optimization variables are used to indicate the core geometric quantification parameters of the compressor blade profile.

[0135] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0136] The block diagrams of devices, apparatuses, devices, and systems disclosed herein are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0137] Additionally, as used herein, the "or" used in a list of items beginning with "at least one" indicates a separate list, such that a list of, for example, "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not imply that the described example is preferred or better than other examples.

[0138] It should also be noted that in the systems and methods of this disclosure, the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions to this disclosure.

[0139] Various changes, substitutions, and modifications can be made to the technology described herein without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of this disclosure is not limited to the specific aspects of the processes, machines, manufactures, events, means, methods, and actions described above. Currently existing or later-developed processes, machines, manufactures, events, means, methods, or actions that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein can be utilized. Therefore, the appended claims include such processes, machines, manufactures, events, means, methods, or actions within their scope.

[0140] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0141] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A method for designing compressor blade profiles, characterized in that, include: Determine the design scenario parameters for the compressor blade profile; wherein, the design scenario parameters are used to indicate the working environment of the compressor blade profile; Determine the target model that matches the design scenario parameters; Determine the optimization objective of the compressor airfoil and the corresponding constraints; wherein the optimization objective is used to indicate the aerodynamic performance improvement target of the compressor airfoil; Based on the optimization objective and the constraints, the target optimization variables of the compressor blade profile are obtained through the target model, and the design scheme of the compressor blade profile is determined based on the target optimization variables; wherein, the target optimization variables are used to indicate the core geometric quantification parameters of the compressor blade profile; The constraints are defined as the constraints of the target model in order to optimize the optimization objective; The parameters to be optimized, constraints, and design scenario parameters are input into the target model for optimization to obtain the optimized parameters. Before proceeding through the target model based on the optimization objective and the constraints, the process further includes: Obtain the model training environment; wherein the model training environment is used to indicate the direction of model training; Based on the model training environment, model training samples are generated; The neural network model to be trained is trained based on the training samples of the model to obtain the trained model; wherein, the trained model includes the target model; The model training environment includes at least one of the following: a state space, a state transition probability, an action space, and a reward function; wherein the state space is used to indicate the parameter variables to be optimized for the compressor blade profile, and the action space is used to indicate the changes in the parameter variables to be optimized for the compressor blade profile.

2. The method according to claim 1, characterized in that, The process of determining the target model that matches the design scenario parameters includes: Obtain the training scenario parameters of the trained model; The training scene parameters that meet the preset similarity requirements with the designed scene parameters are determined as the target scene parameters; The trained model corresponding to the target scene parameters is determined as the target model.

3. The method according to claim 1, characterized in that, The process of determining the optimization objective of the compressor blade profile and the corresponding constraints includes: Based on the operating parameters of the compressor airfoil, the optimization objectives of the compressor airfoil are determined; wherein, the optimization objectives include, but are not limited to: reducing the design point loss coefficient of the compressor airfoil and increasing the usable angle of attack range of the compressor airfoil; When the optimization objective is to reduce the compressor blade design point loss coefficient, the constraint condition is that the available angle of attack range is not reduced. When the optimization objective is to increase the available angle of attack range of the compressor blade, the constraint condition is that the design point loss coefficient should not increase.

4. The method according to claim 1, characterized in that, The process of training the neural network model to be trained based on the model training samples to obtain the trained model includes: The linear rectified function is determined as the activation function of the neural network model to be trained, and the mean squared error function is determined as the loss function of the neural network model to be trained. The neural network model to be trained is trained based on the activation function, the loss function, and the model training samples to obtain the trained model.

5. A compressor blade design device, characterized in that, include: The first determining module is used to determine the design scenario parameters of the compressor blade profile; wherein, the design scenario parameters are used to indicate the working environment of the compressor blade profile; The second determining module is used to determine the target model that matches the design scenario parameters; The third determining module is used to determine the optimization target of the compressor blade profile and the constraints corresponding to the optimization target; wherein, the optimization target is used to indicate the aerodynamic performance improvement target of the compressor blade profile; The design module is used to obtain the target optimization variables of the compressor blade profile based on the optimization objective and the constraints through the target model, and to determine the design scheme of the compressor blade profile based on the target optimization variables; wherein, the target optimization variables are used to indicate the core geometric quantification parameters of the compressor blade profile; the constraints are determined as the constraints of the target model to optimize the optimization objective; the parameter variables to be optimized, the constraints, and the design scenario parameters are input into the target model for optimization to obtain the optimized parameter variables; The third determining module is further configured to acquire a model training environment; wherein the model training environment is used to indicate the model training direction; generate model training samples based on the model training environment; train the neural network model to be trained based on the model training samples to obtain a trained model; wherein the trained model includes the target model; The model training environment includes at least one of the following: a state space, a state transition probability, an action space, and a reward function; wherein the state space is used to indicate the parameter variables to be optimized for the compressor blade profile, and the action space is used to indicate the changes in the parameter variables to be optimized for the compressor blade profile.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the compressor blade design method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the compressor blade design method according to any one of claims 1 to 4.

8. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the compressor blade design method according to any one of claims 1 to 4.

Images