An end region design method, device and optimization design platform

Abstract

The application discloses a method and device for designing an end region and an optimization design platform, and is applied to optimization design of a wide-length vaneless diffuser of a centrifugal compressor, and specifically to determining a modeling region of the wide-length vaneless diffuser. An initial database is established based on an experimental design method, and the initial database comprises multiple groups of optimization target variables. An initial prediction model is established by using a radial basis function proxy model based on the initial database. Hyperparameter optimization processing is performed on the initial prediction model to obtain an optimal end region prediction proxy model. A genetic algorithm is used to perform multi-objective optimization design based on the optimal end region prediction proxy model to obtain an optimal modeling design scheme of the end region. The technical scheme does not depend on artificial experience, and thus can effectively improve the stability margin of the centrifugal compressor.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial design, and more particularly to an end region design method and device and an optimization design platform. BACKGROUND

[0002] The main function of the vaneless diffuser of the centrifugal compressor is to reduce the speed of the high-speed airflow at the outlet of the impeller and increase the pressure. As the airflow flows radially in the vaneless diffuser, the flow area gradually increases, so that the airflow speed decreases and the pressure increases, thereby forming an adverse pressure gradient. When the adverse pressure gradient exceeds a certain limit, the low-speed airflow near the wall surface increases in resistance and decreases in speed, and when the kinetic energy is insufficient to overcome the adverse pressure, flow separation occurs. Flow separation is an important precursor to stall, and the separated airflow forms a vortex, which disturbs the normal flow of the main flow, thereby causing the diffuser of the centrifugal compressor to stall. Therefore, in order to improve the stability margin of the centrifugal compressor, flow control needs to be performed on the vaneless diffuser of the centrifugal compressor to delay or weaken flow separation.

[0003] The present inventors have found in practice that the non-axisymmetric end wall modeling technology can weaken the transverse pressure gradient in the end region and control the load distribution, and has significant effects on improving the performance and stability of the compressor. However, the traditional empirical non-axisymmetric end wall modeling technology relies too much on artificial experience, and the effect of expanding the stability is limited, and local optimization is easy to form, which cannot effectively improve the stability margin of the centrifugal compressor. SUMMARY

[0004] In view of this, the present application provides an end region design method, device and optimization design platform for optimizing the end region of a wide and long vaneless diffuser of a centrifugal compressor to improve the stability margin of the centrifugal compressor.

[0005] In order to achieve the above-mentioned purpose, the present scheme is as follows:

[0006] An end region design method is applied to the end region design method of a wide and long vaneless diffuser of a centrifugal compressor, and the end region design method comprises the following steps:

[0007] determining a modeling region of the wide and long vaneless diffuser;

[0008] establishing an initial database based on the modeling region, wherein the initial database comprises a plurality of groups of optimization target variables;

[0009] establishing an initial prediction model using a radial basis function proxy model based on the initial database;

[0010] performing hyperparameter optimization processing on the initial prediction model to obtain an optimal end region prediction proxy model;

[0011] Based on the optimal end region prediction agent model, a genetic algorithm is used for multi-objective optimization design, so as to obtain an optimal shaping design scheme of the end region.

[0012] Optionally, the method for determining the final shaping region of the wide-length bladeless diffuser comprises the following steps:

[0013] Parameterizing the wall surface of the wide-length bladeless diffuser;

[0014] Based on the wall surface parameters, shaping design is performed, and the shaping region is determined through sensitivity analysis.

[0015] Optionally, the method for establishing the initial database based on the experimental design method comprises the following steps:

[0016] The optimal Latin hypercube method in the experimental design method is used to obtain the geometric sample data of the wall surface parameterization;

[0017] According to the geometric sample data, the wall surface of the wide-length bladeless diffuser is shaped to obtain a plurality of different shaping design schemes;

[0018] A CFD simulation chain established based on Isight is used to simulate and process different shaping design schemes;

[0019] According to the simulation results and a stability margin formula, corresponding optimization target variables are obtained;

[0020] The initial database is established based on the optimization target variables.

[0021] Optionally, the stability margin formula is as follows:

[0022]

[0023] Wherein, m is the mass flow, π is the pressure ratio, the subscript NS represents the near stall point, PEW represents the end wall shaping, and SW represents the prototype.

[0024] Optionally, when the hyperparameter optimization processing is performed on the initial prediction model, a performance evaluation formula used is as follows:

[0025]

[0026] Wherein, NMSE is the normalized mean square error, x is the radial basis function agent model prediction result, and x0 is the simulation result.

[0027] An end region design device is applied to a wide-length bladeless diffuser of a centrifugal compressor, and the end region design device comprises:

[0028] A shaping determination module is configured to determine a shaping region of the wide-length bladeless diffuser.

[0029] The database establishing module is configured to establish an initial database based on the shaping area, and the initial database includes multiple groups of optimization target variables;

[0030] The model constructing module is configured to establish an initial prediction model by using a radial basis function proxy model based on the initial database;

[0031] The model optimizing module is configured to perform hyperparameter optimization processing on the initial prediction model to obtain an optimal end region prediction proxy model;

[0032] The design executing module is configured to perform multi-objective optimization design based on the optimal end region prediction proxy model to obtain an optimal shaping design scheme of the end region.

[0033] Optionally, the shaping determining module includes:

[0034] The parameterization processing unit is configured to parameterize the wall surface of the wide-long vaneless diffuser to obtain a near-wall surface parameter;

[0035] The determining executing unit is configured to perform shaping design based on the wall surface parameter and determine the shaping area by sensitivity analysis.

[0036] Optionally, the database establishing module includes:

[0037] The sample collecting unit is configured to obtain geometric sample data of the wall surface parameter by using an optimal Latin hypercube method in the experimental design method;

[0038] The shaping processing unit is configured to perform shaping processing on the wall surface of the wide-long vaneless diffuser according to the geometric sample data to obtain multiple different shaping design schemes;

[0039] The simulation processing unit is configured to perform simulation and emulation processing on different shaping design schemes;

[0040] The variable calculating unit is configured to obtain corresponding optimization target variables according to simulation results and a stability margin formula;

[0041] The constructing executing unit is configured to establish the initial database based on the optimization target variables by using the experimental design method.

[0042] An optimization design platform includes a CFD simulation chain and an optimization process, wherein:

[0043] The CFD simulation chain is used to store computer programs or instructions, and can automatically perform parameterized geometric modeling, mesh division, numerical simulation solving, and post-processing performance parameter extraction;

[0044] The optimization process is used to perform end wall contour optimization design to make the wide-length vaneless diffuser of the centrifugal compressor achieve the end region design method as described above.

[0045] From the above technical solution, the application discloses an end region design method and device and an optimization design platform. The method and device are applied to optimization design of a wide-length vaneless diffuser of a centrifugal compressor, and specifically, a contour region of the wide-length vaneless diffuser is determined. An initial database is established based on an experimental design method, and the initial database includes multiple groups of optimization target variables. An initial prediction model is established based on the initial database by using a radial basis function proxy model. The initial prediction model is subjected to hyperparameter optimization processing to obtain an optimal end region prediction proxy model. A genetic algorithm is used to perform multi-objective optimization design based on the optimal end region prediction proxy model to obtain an optimal contour design scheme of the end region. The technical solution does not depend on artificial experience, and thus can effectively improve the stability margin of the centrifugal compressor. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0047] Figure 1 FIG. 1 is a partial schematic view of a wide-length vaneless diffuser of an embodiment of the present application;

[0048] Figure 2 FIG. 3 is a flowchart of an end region optimization design method of an embodiment of the present application;

[0049] Figure 3 FIG. 5 is a parameterization schematic view of a contour region determined by an embodiment of the present application;

[0050] Figure 4 FIG. 7 is a schematic view of an optimal contour design scheme of an embodiment of the present application;

[0051] Figure 5 FIG. 9 is a schematic view of a CFD simulation chain of Isight software built by an embodiment of the present application;

[0052] Figure 6 FIG. 11 is a block diagram of an end region design device of an embodiment of the present application. DETAILED DESCRIPTION

[0053] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all the other embodiments obtained by a person of ordinary skill in the art without creative work fall within the scope of the present application.

[0054] The technical solutions provided in the present application are used for optimizing the end region of a wide and long vaneless diffuser of a centrifugal compressor, so as to improve the stability margin of the centrifugal compressor. The wide and long vaneless diffuser is shown in FIG. 1, which has a width ratio b / r1>0.1 and a radial ratio r2 / r1>1.8. The specific solutions of the present application are described as follows. Figure 1

[0055] Figure 2 The flow chart of the end region optimization design method of an embodiment of the present application.

[0056] As shown in FIG. 2, the end region optimization design method provided in the present embodiment is applied to the end region optimization design of a wide and long vaneless diffuser of a centrifugal compressor, so as to improve the stability margin of the centrifugal compressor. The optimization design platform can be understood as a computer, a server or a cloud platform with data calculation and information processing capabilities. The end region optimization design method of the present application includes the following steps: Figure 2 S1, determining the modeling region of the wide and long vaneless diffuser.

[0057] Specifically, the modeling region of the wide and long vaneless diffuser is determined through parameterization operation, and the specific process is as follows:

[0058] Firstly, the single passage disc side of the wide and long vaneless diffuser of the centrifugal compressor is parameterized to obtain the wall surface parameters.

[0059] Then, the single variable method is used for modeling design, and the final modeling region is determined through sensitivity analysis.

[0060] In the present example, the modeling region is the region along the flow direction in the front 80% of the disc side of the wide and long vaneless diffuser (starting from the inlet of the diffuser), as shown in FIG. 3. Among them, 6 control lines are uniformly arranged along the circumference, and 6 control points are evenly distributed on each control line. In order to ensure the smooth connection between the control points and the control points and the flat wall, the boundary control point offset is 0, and there are 20 free control points.

[0061] Figure 3 S2, establishing an initial database based on the experimental design method.

[0062] Specifically, the initial database is established by determining the modeling region of the wide and long vaneless diffuser, and the specific process is as follows:

[0063] ​​​

[0064] First, in order to ensure that the randomly generated geometric sample data is uniformly distributed in the set space, the method of optimal Latin hypercube in DOE (Design of Experiments) is used to obtain the geometric sample data of the disc side parameters. DOE is a statistical method for exploring the influence of multiple factors on the results by systematically arranging experiments, aiming to identify key factors and their interactions with the least number of experiments to achieve optimal results.

[0065] Then, according to the geometric sample data, the disc side of the wide-length vaneless diffuser is designed, and each control point offset is connected by a B-spline curve to obtain a variety of different design schemes.

[0066] Then, based on the CFD simulation chain established by Isight, the different design schemes are simulated and processed to obtain the corresponding simulation results.

[0067] Then, according to the simulation results and the stability margin formula, the corresponding optimization target variables are obtained.

[0068] Finally, based on the free variables and their corresponding optimization target variables, the initial database is established.

[0069] The stability margin formula is as follows:

[0070]

[0071] Where m is the mass flow, π is the pressure ratio, the subscript NS represents the near stall point, PEW represents the end wall modeling, and SW represents the prototype.

[0072] S3, an initial prediction model is established using a radial basis function surrogate model.

[0073] According to the initial database obtained above, an initial prediction model is established using a radial basis function surrogate model.

[0074] S4, the initial prediction model is optimized for super parameters.

[0075] Multiple optimization processes can be implemented, each using k-fold cross-validation to obtain different training sets and test sets for super parameter optimization. The numerical range of the optimization parameter variables is set, and a genetic algorithm is used to perform global optimization during the establishment of the initial prediction model to obtain the optimal value of the parameter variables in the radial basis function surrogate model, so that the prediction accuracy of the radial basis function surrogate model is the highest, thereby obtaining an optimal end region prediction surrogate model with satisfactory accuracy.

[0076] During the optimization process, it is determined whether the parameter variables converge or not. If they do not converge, the optimization process continues, and if they have converged, the subsequent steps are executed.

[0077] S5, performing multi-objective optimization design based on the optimal end region prediction proxy model.

[0078] The radial basis function proxy model after hyperparameter optimization in the above steps obtains the nonlinear relationship between the design variables with the highest prediction accuracy and the optimization objectives. Taking the stability margin improvement and the maximum design operating condition efficiency as the optimization objectives, a non-dominated sorting genetic algorithm II (NSGA-II) optimization algorithm is used for multi-objective optimization to obtain the optimal modeling design variables. The true values are obtained by CFX numerical simulation calculation, and the true results are added to the database for continuous iteration optimization to obtain a new radial basis function proxy model and a new optimal modeling design scheme. In turn, the calculation is iteratively optimized until the optimization stopping condition (optimization convergence) is met, and the final optimal modeling design scheme is obtained, as shown in FIG. 6. Figure 4

[0079] The final optimal modeling design scheme obtained in this example improves the stability margin of the centrifugal compressor by 6.89%, and the design operating condition efficiency is improved by 0.63%.

[0080] As can be seen from the above technical solutions, the embodiment provides an end region design method, which is applied to the optimization design of a wide-length bladeless diffuser of a centrifugal compressor, specifically for determining a modeling region of the wide-length bladeless diffuser; an initial database is established based on an experimental design method, and the initial database includes multiple groups of optimization objective variables; an initial prediction model is established based on the initial database by using a radial basis function proxy model; the initial prediction model is subjected to hyperparameter optimization processing to obtain an optimal end region prediction proxy model; and a genetic algorithm is used to perform multi-objective optimization design based on the optimal end region prediction proxy model to obtain an optimal modeling design scheme of the end region. The technical solution does not rely on artificial experience, thereby effectively improving the stability margin and efficiency of the centrifugal compressor.

[0081] In this application, the modeling design scheme is large in calculation amount and the process is repetitive and tedious due to the establishment of the initial sample database and the subsequent optimization iteration calculation. Therefore, in order to fully utilize the computing resources and save the calculation time, a CFD simulation chain is built with the help of Isight software, including automatic modeling, mesh drawing, simulation, result post-processing, and saving according to the obtained geometric sample data.

[0082] The performance evaluation formula in the optimization system in this application is:

[0083]

[0084] where NMSE is the normalized mean square error, which is used to measure the gap between the prediction result x of the radial basis function proxy model and the simulation result x0.​

[0085] The application is realized by a high-level programming language, and geometric data of a modeling design scheme optimized by a genetic algorithm based on a radial basis function proxy model and a predicted stability margin; the high-level programming language adopts Python or MATLAB;

[0086] The model training and optimization of the modeling design scheme in the application are realized by a radial basis function proxy model, simulated and calculated by a CFD simulation chain built by Isight software, and a real value x0 is obtained. The CFD simulation chain is as shown in Figure 5 .

[0087] The application adopts an automatic performance evaluation and optimization algorithm module, builds an optimization design platform by a high-level programming language and Isight software, and continuously iterates and updates optimization of design variables by an optimization search method until an optimization stop condition is met, so that an optimal modeling design scheme of a wide-length vaneless diffuser of a centrifugal compressor is obtained.

[0088] The application proposes an end region design method suitable for a wide-length vaneless diffuser of a centrifugal compressor for the case of boundary layer separation near a wall surface of a vaneless diffuser of a centrifugal compressor, which can solve or delay compressor stall caused by flow separation of the vaneless diffuser of the centrifugal compressor, and fills the research gap of non-axisymmetric end wall modeling design of the vaneless diffuser of the centrifugal compressor.

[0089] To solve the problems of excessive dependence on artificial experience of traditional empirical non-axisymmetric end wall modeling technology, limited stability expansion effect, easy formation of local optimum, and high calculation cost and low efficiency of global optimization calculation by directly using a single optimization algorithm, the application adopts a dynamic radial basis function proxy model, which is suitable for small sample point optimization, and improves optimization calculation efficiency while ensuring calculation accuracy.

[0090] The application optimizes hyperparameters of the radial basis function proxy model by a global optimization method, so that the prediction model obtained has the highest precision and is closest to the actual simulation result, reduces the number of optimization iterations, and improves optimization efficiency. The real simulation result of the optimal design modeling of each iteration is added to the database, and the sample points are updated and the proxy model is reconstructed, so that the proxy model and the sample points change in the whole optimization process, which effectively accelerates the optimization convergence speed.

[0091] In the optimization process, the non-dominated sorting genetic algorithm II (NSGA-II) optimization algorithm is used for multi-objective optimization with the optimization objectives of maximizing the stability margin and efficiency, so that the stable working range of the compressor is expanded, and the compressor efficiency and flow capacity are considered.

[0092] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

[0093] Although the operations are depicted in a particular, sequential order, this should not be understood as requiring or

[0094] It is to be understood that the steps of the methods recited in the method embodiments of the present disclosure can be carried out in a different order, or concurrently, than described. Furthermore, the method embodiments can include additional steps or omit described steps. The scope of the present disclosure is not limited in this regard.

[0095] Computer program code for carrying out operations of the present disclosure can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server.

[0096] Figure 6 A block diagram of an end region design apparatus according to an embodiment of the present application.

[0097] As Figure 6As shown, the end region design device provided by the embodiment is applied to the end region optimization design of the wide-length bladeless diffuser of the centrifugal compressor to improve the stability margin and efficiency of the centrifugal compressor. The optimization design platform can be understood as a computer, a server or a cloud platform with data calculation capability and information processing capability. The end region design device of the application includes a modeling region confirmation module 10, a database establishment module 20, a prediction model construction module 30, a prediction model optimization module 40 and a global optimization module 50.

[0098] The modeling region confirmation module is used to determine the modeling region of the wide-length bladeless diffuser.

[0099] Specifically, the modeling region of the wide-length bladeless diffuser is determined through parameterization operation. The module specifically includes a parameterization processing unit and a determination execution unit.

[0100] The parameterization processing unit parameterizes the single-flow channel wall surface of the wide-length bladeless diffuser of the centrifugal compressor.

[0101] The determination execution unit is used to perform modeling design by using the single variable method, and to determine the final modeling region through sensitivity analysis.

[0102] In the example, the modeling region is the 80% region along the flow direction of the disc side of the wide-length bladeless diffuser (starting from the diffuser inlet), as shown in the figure. Figure 3 Among them, 6 control lines are uniformly arranged along the circumference, and 6 control points are evenly distributed on each control line. In order to ensure the smooth connection between the control points and the control points and the flat wall, the boundary control point offset is 0, and there are 20 free control points.

[0103] The database establishment module is used to establish an initial database based on the experimental design method.

[0104] Specifically, the initial database is established based on the CFD simulation chain built by Isight by determining the modeling region of the wide-length bladeless diffuser. The module specifically includes a sample collection unit, a modeling processing unit, a simulation processing unit, a variable calculation unit and a construction execution unit.

[0105] In order to ensure that the randomly generated geometric sample data is uniformly distributed in the set space, the sample collection unit is used to obtain the geometric sample data of the wall surface parameters by using the optimal Latin hypercube method in DOE (Design of Experiments). DOE is a statistical method for exploring the influence of multiple factors on the results by systematically arranging experiments, aiming to identify key factors and their interactions to achieve optimal results with the least number of experiments.

[0106] The shaping processing unit is configured to design a wall surface of the wide-length vaneless diffuser according to the geometric sample data, each control point offset is connected by a B-spline curve, and a plurality of different design schemes are obtained.

[0107] The simulation processing unit is configured to simulate the different design schemes to obtain simulation results.

[0108] The variable calculation unit is configured to obtain corresponding optimization target variables according to the simulation results and a stability margin formula.

[0109] The construction execution unit is configured to establish the initial database based on the optimization target variables.

[0110] The stability margin formula is as follows.

[0111]

[0112] Wherein, m is mass flow, pi is pressure ratio, the subscript NS represents a near stall point, PEW represents an end wall shaping, and SW represents a prototype.

[0113] The prediction model construction module is configured to establish an initial prediction model by using a radial basis function surrogate model.

[0114] According to the initial database obtained above, an initial prediction model is established by using a radial basis function surrogate model.

[0115] The prediction model optimization module is configured to perform hyperparameter optimization processing on the initial prediction model.

[0116] The optimization processing can be performed multiple times, each time the k-fold cross-validation is used to obtain different training sets and test sets for hyperparameter optimization. The numerical range of the optimization parameter variable is set, and the genetic algorithm is used to perform global optimization in the process of establishing the initial prediction model to obtain the optimal value of the parameter variable in the radial basis function surrogate model, so that the prediction accuracy of the radial basis function surrogate model is the highest, thereby obtaining an optimal end region prediction surrogate model with required accuracy.

[0117] The global optimization module is configured to perform multi-objective optimization design based on the optimal end region prediction surrogate model.

[0118] The radial basis function proxy model after hyperparameter optimization in the scheme obtains the nonlinear relationship between the design variables with the highest prediction accuracy and the optimization target. And taking the improvement of the stability margin and the maximization of the efficiency as the optimization target, the non-dominated sorting genetic algorithm II (NSGA-II) optimization algorithm is used for multi-objective optimization to obtain the optimal shape design variable. The real value is obtained by CFX numerical simulation calculation, and the real result is added to the database for continuous iteration optimization to obtain a new radial basis function proxy model and a new optimal shape design scheme. In turn, the calculation is iterated and optimized until the optimization stopping condition (optimization convergence) is met, and the final optimal shape design scheme is obtained, as shown in Figure 4 wherein the parameter h represents the relative height of the diffuser end wall, and a positive value represents a protrusion and a negative value represents a depression.

[0119] The final optimal shape design scheme obtained in this example improves the stability margin of the centrifugal compressor by 6.89% and increases the design condition efficiency by 0.63%.

[0120] As can be seen from the above technical scheme, the embodiment provides an end region design device, which is applied to the optimization design of a wide-length bladeless diffuser of a centrifugal compressor, and specifically determines the shape region of the wide-length bladeless diffuser. An initial database is established based on the design of experiments, and the initial database includes multiple groups of optimization target variables. An initial prediction model is established based on the initial database by using a radial basis function proxy model. The initial prediction model is subjected to hyperparameter optimization processing to obtain an optimal end region prediction proxy model. The genetic algorithm is used for multi-objective optimization design based on the optimal end region prediction proxy model to obtain an optimal shape design scheme of the end region. The technical scheme does not depend on artificial experience, and thus can effectively improve the stability margin of the centrifugal compressor.

[0121] The embodiment also provides an embodiment of an optimization design platform.

[0122] The above computer-readable storage medium is applied to the optimization design platform and carries one or more computer programs, which, when executed by the optimization design platform, cause the optimization design platform to determine the shape region of the wide-length bladeless diffuser; establish an initial database based on the design of experiments, and the initial database includes multiple groups of optimization target variables; establish an initial prediction model based on the initial database by using a radial basis function proxy model; perform hyperparameter optimization processing on the initial prediction model to obtain an optimal end region prediction proxy model; and perform multi-objective optimization design based on the optimal end region prediction proxy model by using the genetic algorithm to obtain an optimal shape design scheme of the end region. The technical scheme does not depend on artificial experience, and thus can effectively improve the stability margin of the centrifugal compressor.

[0123] Various embodiments of the present specification are described in progressive manner, and each embodiment focuses on the difference from other embodiments, and the same or similar parts between various embodiments can be referred to each other.

[0124] Although the preferred embodiments of the embodiments of the present application have been described, those skilled in the art can make further changes and modifications to the embodiments once they know the basic inventive concept. Therefore, the appended claims are intended to cover all the preferred embodiments and all the changes and modifications falling within the scope of the embodiments of the present application.

[0125] Finally, it should also be noted that, in this document, the relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or terminal device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or terminal device. Without more limitations, the element defined by the statement "including a" does not exclude the presence of other identical elements in the process, method, article or terminal device including the element.

[0126] The above describes the technical solutions provided by the present application in detail, and the principles and implementation manners of the present application are described by applying specific examples; the above embodiment description is only for helping to understand the method of the present application and its core idea; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation manner and application range will be changed, and the above description should not be understood as the limitation of the present application.

Claims

1. A method for designing an end section, applied to an optimization design platform, for optimizing the design of a wide and long vaneless diffuser for a centrifugal compressor, characterized in that: The end zone design method comprises the steps of: determining a final shaping region of the wide and long vaneless diffuser, specifically by parameterizing a wall surface of the wide and long vaneless diffuser to obtain wall parameters, performing shaping design based on the wall parameters, and determining a final shaping region through sensitivity analysis, wherein the final shaping region is an area 80% of the way forward along the flow path on the disc side of the wide and long vaneless diffuser; Establishing an initial database based on an experimental design method, the initial database including multiple sets of optimization target variables, specifically, using the optimal Latin hypercube method in the experimental design method to obtain geometric sample data of the wall parameters, performing shape design on the wall of the wide and long vaneless diffuser based on the geometric sample data to obtain multiple different shape design schemes, simulating the multiple shape design schemes based on a CFD simulation chain constructed by Isight, obtaining corresponding optimization target variables based on the simulation results and a stability margin formula, and establishing the initial database based on the free variables and their corresponding optimization target variables; Based on the initial database, an initial prediction model is established using a radial basis function proxy model; Performing hyperparameter optimization on the initial prediction model to obtain an optimal end zone prediction agent model; Based on the optimal end area prediction agent model, a genetic algorithm is used to perform multi-objective optimization design to obtain the optimal shape design scheme of the end area.

2. The terminal region design method according to claim 1, wherein: The stability margin formula is as follows: Where m is the mass flow rate, π is the pressure ratio, the subscript NS stands for near-stall point, PEW stands for endwall shaping, and SW stands for prototype.

3. The terminal region design method according to claim 1, wherein: When performing hyperparameter optimization on the initial prediction model, the performance evaluation formula used is: Among them, NMSE is the normalized root mean square error, x is the prediction result of the radial basis function proxy model, and x0 is the simulation result.

4. An end zone design device, applied to an optimization design platform, for optimizing the design of a wide and long bladeless diffuser of a centrifugal compressor, characterized in that: The end zone design device comprises: a shape determination module configured to determine a final shape region of the wide and long vaneless diffuser, wherein the final shape region is an area 80% of the way forward along the flow path on the disc side of the wide and long vaneless diffuser; A database establishment module is configured to establish an initial database based on an experimental design method, wherein the initial database includes multiple groups of optimization target variables; A model building module is configured to establish an initial prediction model based on the initial database using a radial basis function proxy model; A model optimization module is configured to perform hyperparameter optimization processing on the initial prediction model to obtain an optimal end zone prediction agent model; A design execution module is configured to perform a multi-objective optimization design using a genetic algorithm based on the optimal end area prediction agent model to obtain an optimal shape design scheme for the end area; The shape determination module includes: a parameterization processing unit configured to parameterize the wall surface of the wide and long vaneless diffuser to obtain wall surface parameters; determining an execution unit, configured to perform shape design based on the wall parameters and determine a final shape area through sensitivity analysis; The database establishment module includes: a sample collection unit configured to obtain geometric sample data of the wall parameters by adopting the optimal Latin hypercube method in the experimental design method; a shape processing unit configured to perform shape design on the wall surface of the wide and long vaneless diffuser according to the geometric sample data to obtain a plurality of different shape design schemes; A simulation processing unit, configured to simulate and process different styling design schemes based on a CFD simulation chain built by Isight; a variable calculation unit, configured to obtain a corresponding optimization target variable according to the simulation result and the stability margin formula; The construction execution unit is configured to establish the initial database based on the free variables and their corresponding optimization target variables.

5. The end zone design device according to claim 4, characterized in that: The stability margin formula is as follows: Where m is the mass flow rate, π is the pressure ratio, the subscript NS stands for near-stall point, PEW stands for endwall shaping, and SW stands for prototype.

Images