CN116595874A - Impeller mechanical performance prediction model parameter optimization method and device and storage medium - Google Patents
Impeller mechanical performance prediction model parameter optimization method and device and storage medium Download PDFInfo
- Publication number
- CN116595874A CN116595874A CN202310558377.8A CN202310558377A CN116595874A CN 116595874 A CN116595874 A CN 116595874A CN 202310558377 A CN202310558377 A CN 202310558377A CN 116595874 A CN116595874 A CN 116595874A
- Authority
- CN
- China
- Prior art keywords
- optimization
- parameters
- performance prediction
- impeller
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 166
- 238000000034 method Methods 0.000 title claims abstract description 146
- 238000003860 storage Methods 0.000 title claims abstract description 21
- 238000012360 testing method Methods 0.000 claims abstract description 125
- 230000006870 function Effects 0.000 claims description 36
- 230000008569 process Effects 0.000 claims description 30
- 238000004422 calculation algorithm Methods 0.000 claims description 22
- 238000012935 Averaging Methods 0.000 claims description 15
- 230000002068 genetic effect Effects 0.000 claims description 8
- 239000002245 particle Substances 0.000 claims description 4
- 238000002922 simulated annealing Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 description 24
- 238000012545 processing Methods 0.000 description 7
- 238000011161 development Methods 0.000 description 6
- 230000035939 shock Effects 0.000 description 6
- 238000004364 calculation method Methods 0.000 description 5
- 238000009826 distribution Methods 0.000 description 5
- 238000005259 measurement Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 4
- 239000000523 sample Substances 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 3
- 230000002457 bidirectional effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000012530 fluid Substances 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 238000005272 metallurgy Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000013102 re-test Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
- 238000011524 similarity measure Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 230000007723 transport mechanism Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/08—Fluids
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Software Systems (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Algebra (AREA)
- Computing Systems (AREA)
- Fluid Mechanics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Structures Of Non-Positive Displacement Pumps (AREA)
Abstract
A method and a device for optimizing parameters of an impeller mechanical performance prediction model and a storage medium are provided, wherein the method comprises the following steps: determining model parameters to be optimized and an optimization range in the impeller mechanical property prediction model; acquiring test data; establishing a multi-objective optimization function and an optimization objective, wherein the multi-objective optimization function is used for representing the difference between predicted values of at least two parameters output by the impeller mechanical property prediction model and test values of at least two parameters in the test data, and the at least two parameters comprise: macroscopic performance parameters and flow field aerodynamic parameters, the optimization objective being to minimize the multi-objective optimization function; and carrying out iterative computation optimizing on the multi-objective optimization function according to the optimization range to obtain optimized model parameters.
Description
Technical Field
The embodiment of the disclosure relates to the technical field of impeller machinery, in particular to a method and a device for optimizing parameters of an impeller mechanical performance prediction model and a storage medium.
Background
The impeller machine is a power machine which takes continuous rotary blades as main components and consumes mechanical work or outputs mechanical work through the energy conversion between the blades and fluid working media. The impeller machinery consuming mechanical work is called a working machine and comprises a gas compressor, a fan, a water pump, a propeller and the like; impeller machines that output mechanical work are known as prime movers and include turbines, hydroturbines, steam turbines, and the like. Unlike reciprocating piston power machines, impeller machines achieve continuous intake, have greater flow rates, and are more easily balanced in the rotor, are well suited for high speed rotation, and thus can produce greater power in smaller size and weight. Impeller machinery plays a very important role in industrial systems and national economy systems, and is widely applied to various fields such as aviation, ships, vehicles, power stations, metallurgy, chemical industry and the like.
The impeller machinery has extremely complex shock wave structure, end wall flow, secondary flow, rotor-stator interaction, interaction of shock wave and boundary layer and other phenomena, and the internal flow of the impeller machinery presents strong three-dimensional, viscous and unsteady characteristics, so that the impeller machinery performance prediction becomes the key and difficult problem faced by the field for a long time, and particularly as the impeller machinery performance is continuously improved and the load is continuously increased, the problem becomes more remarkable. The method realizes quick and accurate prediction of the mechanical performance of the impeller, is important to improving the development level of the impeller machinery, shortening the development period and reducing the development cost, and promotes the development of the impeller machinery from the traditional iterative design adopting design-test verification-modification design-retest to the predictive design.
The current impeller mechanical performance prediction method is divided according to dimensions and mainly comprises a three-dimensional performance prediction method, a two-dimensional performance prediction method and a one-dimensional performance prediction method. The three-dimensional performance prediction method is based on a three-dimensional computational fluid dynamics tool, and usually calculates the mechanical performance of the impeller by solving a Reynolds average Navier-Stokes equation (Navier-Stokes equation), and can acquire detailed full three-dimensional flow field information, but has the advantages of large calculated amount, high calculated cost and long calculation period; the two-dimensional performance prediction method is based on reasonable assumption of the internal flow field of the impeller machinery, the three-dimensional flow problem is converted into a two-dimensional axisymmetric flow problem, the impeller machinery performance is calculated by solving a radial balance equation, and flow field information distributed along the radial direction can be obtained; the one-dimensional performance prediction method is to assume that the internal flow of the impeller machine is axisymmetric, set a calculation station on the average diameter of the inlet and outlet of each blade row of the impeller machine, simplify the complex three-dimensional flow in the impeller machine to one-dimensional flow based on the average diameter, further calculate the mechanical performance of the impeller, and acquire flow field information on a flow line along the average diameter. The three-dimensional performance prediction method needs to know detailed three-dimensional geometry of the impeller machinery, is not suitable for the design stage of the scheme of the impeller machinery, and is generally applied to the detailed design stage of the impeller machinery; compared with the three-dimensional performance prediction method, the two-dimensional performance prediction method and the one-dimensional performance prediction method have the advantages that the calculation time is greatly shortened, meanwhile, higher calculation precision can be realized by selecting an empirical or semi-empirical formula obtained through verification or fitting of a large amount of test data, the two-dimensional performance prediction method and the one-dimensional performance prediction method are very common performance prediction methods in engineering, are generally applied to the design stage of an impeller mechanical scheme, and allow a designer to develop a large amount of iterative optimization design.
Because the low-dimensional performance prediction method of the impeller machine is greatly simplified for complex three-dimensional flow in the impeller machine, an empirical or semi-empirical formula is generally required to be introduced for correction so as to meet the performance prediction accuracy requirement. These models include reference angle of attack models, lag angle models, loss models, blockage models, and the like. The accuracy of the low-dimensional performance prediction method of the impeller machinery is highly dependent on the prediction accuracy of an empirical formula, and the parameters of the empirical formula are usually determined through a great amount of experimental data and design experience accumulated in the past and are continuously optimized along with the development and progress of the impeller machinery technology. With the continuous improvement of the mechanical performance and the load level of the impeller, the existing many empirical formulas are not applicable to the high-speed high-load impeller machinery adopting the advanced modern blade profile, and bring great challenges to the accurate prediction of the mechanical performance of the impeller. Therefore, how to optimize the parameter items in the low-dimensional performance prediction model to meet the impeller mechanical performance prediction requirement, and the optimized low-dimensional performance prediction model has strong universality and becomes a key problem in the field.
Disclosure of Invention
The embodiment of the disclosure provides a method for optimizing parameters of an impeller mechanical property prediction model, which comprises the following steps:
Determining model parameters to be optimized and an optimization range in the impeller mechanical property prediction model;
acquiring test data;
establishing a multi-objective optimization function and an optimization objective, wherein the multi-objective optimization function is used for representing the difference between predicted values of at least two parameters output by the impeller mechanical property prediction model and test values of at least two parameters in the test data, and the at least two parameters comprise: macroscopic performance parameters and flow field aerodynamic parameters, the optimization objective being to minimize the multi-objective optimization function;
and carrying out iterative computation optimizing on the multi-objective optimization function according to the optimization range to obtain optimized model parameters.
The embodiment of the disclosure also provides a parameter optimization device for the impeller mechanical property prediction model, which comprises a memory; and a processor coupled to the memory, the memory for storing instructions, the processor configured to perform the steps of the method of optimizing parameters of the predicted model of impeller mechanical performance of any embodiment of the present disclosure based on the instructions stored in the memory.
The embodiments of the present disclosure also provide a storage medium having a computer program stored thereon, which when executed by a processor, implements the method for optimizing parameters of a predictive model for impeller mechanical performance according to any of the embodiments of the present disclosure.
Additional features and advantages of the disclosure will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the disclosure. Other advantages of the present disclosure may be realized and attained by the structure particularly pointed out in the written description and drawings.
Drawings
The accompanying drawings are included to provide an understanding of the technical aspects of the present disclosure, and are incorporated in and constitute a part of this specification, illustrate the technical aspects of the present disclosure and together with the embodiments of the disclosure, not to limit the technical aspects of the present disclosure.
FIG. 1 is a flow chart of a method for optimizing parameters of an impeller mechanical performance prediction model according to an exemplary embodiment of the present disclosure;
FIG. 2A illustrates an exemplary three stage axial flow compressor noon flow path schematic;
FIG. 2B illustrates an exemplary compressor pressure ratio-flow characteristic and efficiency-flow characteristic diagram;
FIG. 2C is a flow chart of another method for optimizing parameters of an impeller mechanical performance prediction model provided by an exemplary embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of an impeller mechanical performance prediction model parameter optimization device according to an exemplary embodiment of the present disclosure.
Detailed Description
The present disclosure describes several embodiments, but the description is illustrative and not limiting, and it will be apparent to those of ordinary skill in the art that many more embodiments and implementations are possible within the scope of the embodiments described in the present disclosure. Although many possible combinations of features are shown in the drawings and discussed in the detailed description, many other combinations of the disclosed features are possible. Any feature or element of any embodiment may be used in combination with or in place of any other feature or element of any other embodiment unless specifically limited.
The present disclosure includes and contemplates combinations of features and elements known to those of ordinary skill in the art. The embodiments, features and elements of the present disclosure that have been disclosed may also be combined with any conventional features or elements to form a unique inventive arrangement as defined by the claims. Any feature or element of any embodiment may also be combined with features or elements from other inventive arrangements to form another unique inventive arrangement as defined in the claims. Thus, it should be understood that any of the features shown and/or discussed in this disclosure may be implemented alone or in any suitable combination. Accordingly, the embodiments are not to be restricted except in light of the attached claims and their equivalents. Further, various modifications and changes may be made within the scope of the appended claims.
Furthermore, in describing representative embodiments, the specification may have presented the method and/or process as a particular sequence of steps. However, to the extent that the method or process does not rely on the particular order of steps set forth herein, the method or process should not be limited to the particular sequence of steps described. Other sequences of steps are possible as will be appreciated by those of ordinary skill in the art. Accordingly, the particular order of the steps set forth in the specification should not be construed as limitations on the claims. Furthermore, the claims directed to the method and/or process should not be limited to the performance of their steps in the order written, and one skilled in the art can readily appreciate that the sequences may be varied and still remain within the spirit and scope of the embodiments of the present disclosure.
Early studies were fitted through a large number of planar cascade test data in order to determine parameters in a low dimensional performance prediction model of the turbomachine. Because of the large difference between the flow conditions in the planar cascade and the impeller machine, and the simplified assumption generally introduced in the modeling process, many empirical formulas proposed in the period have certain limitations, and the model prediction result and the actual test result have certain deviation, which is particularly remarkable in predicting the performance of the impeller machine under the non-design working condition. After that, the annular blade cascade test and the low-speed low-load impeller mechanical test are continuously developed and data are accumulated, so that an empirical formula is continuously improved and perfected, and the influence caused by the three-dimensional flow effect can be gradually considered. However, as impeller mechanical performance and load levels continue to increase, the model parameters set forth earlier are no longer applicable. In recent years, a plurality of scholars have also put forward a plurality of new empirical formulas based on early empirical formulas and combined with the application of modern complex three-dimensional blade shapes and the latest knowledge of complex three-dimensional flow inside the impeller machinery. Whether the experimental formula is an old experimental formula or a new experimental formula, parameters of the experimental formulas such as reference attack angle, lag angle, loss, blockage and the like are optimized by using a physical test result or a high-precision numerical test result so as to meet the mechanical property prediction requirement of the high-speed high-load impeller.
In the current optimization method of the low-dimensional performance prediction model parameters of the impeller machinery, the thought of the method takes the empirical formula parameters as optimization variables, takes the minimization of the difference between the prediction result of the low-dimensional performance prediction model and the physical test result or the three-dimensional performance prediction model result as an optimization target, and adopts different optimization methods to carry out optimization. Since the optimization is only carried out for a specific impeller machine, and only the macroscopic performance of the impeller machine (for example, for a compressor, the macroscopic performance can be a pressure ratio-flow characteristic or an efficiency-flow characteristic; for a turbine, the macroscopic performance can be a flow-expansion ratio characteristic or an efficiency-expansion ratio characteristic) is considered to be the smallest in prediction error, and whether the prediction of the aerodynamic parameters of the key flow field inside the impeller machine such as attack angle, lag angle, total pressure loss and the like is accurate or not is ignored, the method is more a correction process on data and lacks of physical connotation, and larger performance prediction deviation can occur when the optimized low-dimensional performance prediction model is used for other impeller machines of the same type, namely the universality of the low-dimensional performance prediction model is poor.
As shown in fig. 1, an embodiment of the present disclosure provides a method for optimizing parameters of an impeller mechanical performance prediction model, including:
Step 101, determining model parameters to be optimized and an optimization range in an impeller mechanical property prediction model;
102, acquiring test data;
step 103, establishing a multi-objective optimization function and an optimization objective, wherein the multi-objective optimization function is used for representing the difference between the predicted value of at least two parameters output by the impeller mechanical property prediction model and the test value of at least two parameters in the test data, and the at least two parameters comprise: macroscopic performance parameters and flow field aerodynamic parameters, wherein the optimization targets are to minimize a multi-target optimization function;
and 104, carrying out iterative computation optimizing on the multi-objective optimization function according to the optimization range to obtain optimized model parameters.
According to the impeller mechanical performance prediction model parameter optimization method, the built multi-objective optimization function considers not only the difference between macroscopic performance parameters of the impeller machinery but also the difference between aerodynamic parameters of a flow field, so that mathematical accuracy and physical accuracy are considered, performance prediction accuracy is improved, and meanwhile universality of a performance prediction model is guaranteed.
In some exemplary embodiments, the impeller machine according to the embodiments of the present disclosure may be an impeller machine (i.e., a working machine) that consumes mechanical work such as a compressor, a fan, a water pump, a propeller, or an impeller machine (i.e., a prime mover) that outputs mechanical work such as a turbine, a water turbine, a steam turbine, or the like.
In some exemplary embodiments, the impeller machines described in the examples of the present disclosure may be configured in an axial flow configuration or a radial flow configuration.
In some exemplary embodiments, the number of stages of the turbomachine described in the examples of the present disclosure may be single stage or multi-stage.
In some exemplary embodiments, the dimension of the predicted model of the mechanical properties of the impeller to be optimized may be one-dimensional or two-dimensional.
In the embodiment of the disclosure, the impeller mechanical performance prediction model to be optimized may be a one-dimensional performance prediction model or a two-dimensional performance prediction model. The one-dimensional performance prediction model comprises an average streamline model, a cascade method model and the like; the two-dimensional performance prediction model comprises a streamline curvature method model, a circumferential average through flow method model based on a Navier-Stokes equation and the like.
In general, a large number of empirical or semi-empirical formulas are introduced into a low-dimensional performance prediction model (i.e., a one-dimensional performance prediction model or a two-dimensional performance prediction model) to improve performance prediction accuracy. In some exemplary implementations, the impeller mechanical performance prediction model of the embodiments of the present disclosure may use any of the empirical formulas in the prior art documents, for example, the impeller mechanical performance prediction model may include empirical formulas such as reference attack angle, reference lag angle, non-design point lag angle, vane loss, shock loss, secondary flow loss, etc., the non-design point lag angle empirical formulas may use Creveling empirical formulas, howell empirical formulas, etc., and the vane loss empirical formulas may use Creveling empirical formulas, aungier empirical formulas, etc. In other exemplary embodiments, the impeller mechanical performance prediction model in the disclosed embodiments may also use a newly proposed empirical formula, which is not limited by the disclosed embodiments.
In some exemplary embodiments, the number of model parameters to be optimized in the determined impeller mechanical performance prediction model may be 1 or more.
Illustratively, assume that the non-design point trailing angle empirical formula is the Creveling empirical formula:
wherein δ is the falling angle, x= (i-i) * )/ε * I is the actual angle of attack, i * For the reference angle of attack, ε may be calculated from the empirical formula of the reference angle of attack * For the turning angle of the air flow in the reference state epsilon * ＝θ-δ * +i * ，δ * For reference lag angle, it can be calculated from the empirical formula of reference lag angle, θ being the vane angle.
At this time, the model parameters to be optimized may be set to the following coefficient values: -0.000809, 0.5588, -0.2928, 0.0001191, 0.48 and 0.3452. In addition, the model parameters to be optimized may also include a reference angle of attack i * Reference lag angle delta * Is not shown in the present disclosure). In addition, the model parameters to be optimized may also include an index 1 in x and/or x 2 The number of the model parameters to be optimized is not limited by the index 2 in the embodiment of the disclosure, and any coefficient in any empirical formula can be used as the model parameters to be optimized.
Taking the above-mentioned empirical formula of the non-design point falling angle as an example, assume that the determined model parameters to be optimized are the following coefficient values: 0.000809, 0.5588, -0.2928, 0.0001191, 0.48 and 0.3452, the above non-design point lag angle empirical formula can be rewritten as:
Wherein C is 1 To C 6 Is the model parameter to be optimized.
In some exemplary embodiments, model parameters to be optimized in the predicted model of the impeller mechanical performance may be determined for the rotor blade row and the stator blade row, respectively.
For example, still taking the above-described empirical formula of non-design point lag angle as an example, the model parameter to be optimized for the non-design point lag angle empirical formula of a rotor blade row is C 1R To C 6R The method comprises the steps of carrying out a first treatment on the surface of the For a non-design point lag angle empirical formula of a stator blade row, the model parameter to be optimized is C 1S To C 6S And the non-design point lag angle empirical formula is increased from the original 6 model parameters to be optimized to 12 model parameters to be optimized.
In some exemplary embodiments, step 101 may further include: and determining initial values of model parameters to be optimized in the impeller mechanical property prediction model.
In the embodiment of the disclosure, the initial value of the model parameter to be optimized may be a parameter value in the prior art document, for example, C 1 To C 6 The initial value of (2) may be determined as: -0.000809, 0.5588, -0.2928, 0.0001191, 0.48 and 0.3452, in other exemplary embodimentsIn the formula, the initial value of the model parameter to be optimized can also be selected according to engineering experience.
In the actual optimization process, a smaller optimization range may be set first in step 101, and when no model parameter value meeting the prediction accuracy requirement is found in the smaller optimization range, the optimization range is modified, and then the optimization process is re-executed, so as to reduce the calculation amount of the optimization process as much as possible.
In some exemplary embodiments, step 101 may further include: and inputting geometric parameters of the impeller mechanical property prediction model.
Taking a compressor as an example, the input geometric parameters may include: the diameter of the blade root of the inlet and the outlet of each blade row, the diameter of the blade tip of the inlet and the outlet of each blade row, the geometric angle of the blades of the average diameter of the inlet and the outlet of each blade row, the consistency of each blade row and the like. In the embodiment of the disclosure, the geometric parameters of the impeller mechanical performance prediction model input in step 101 should be consistent with the geometric parameters of the impeller machinery corresponding to the test data obtained in step 102, so as to ensure the accuracy of the data in the optimization process.
In some exemplary embodiments, the test data obtained in step 102 may be numerical test data or physical test data, where the numerical test data may be obtained based on a high-dimensional performance prediction method of the turbomachine, which may be a three-dimensional performance prediction method or a two-dimensional performance prediction method for parameter optimization of a one-dimensional performance prediction model of the turbomachine, and may be a three-dimensional performance prediction method for parameter optimization of a two-dimensional performance prediction model of the turbomachine, after verification of a large number of physical test data. The following will describe in detail the case where the parameters of the one-dimensional performance prediction model of the impeller machine are optimized and the obtained test data may be three-dimensional numerical test data or physical test data.
In the embodiment of the disclosure, when a three-dimensional numerical test or a physical test is performed, a certain number of numerical probes or physical probes can be arranged on the inlet and outlet cross sections of each blade row along the radial direction and the circumferential direction so as to record aerodynamic parameter data at different rotating speeds, different working conditions and different three-dimensional flow field positions. And acquiring macroscopic performance data of the air compressor and pneumatic parameter data of the discharge field of each blade through a three-dimensional numerical test or a physical test.
In some exemplary embodiments, step 102 may further include: the test data is subjected to an averaging process, which illustratively includes, but is not limited to, at least one of: mass flow weighted average, area weighted average, power weighted average, and arithmetic average.
Since the data obtained by the three-dimensional numerical test or the physical test are data in three-dimensional distribution (distribution in the axial direction, radial direction, and circumferential direction), in order to compare them with the one-dimensional performance prediction data, the data needs to be subjected to an averaging process. Data averaging methods include, but are not limited to, mass flow weighted averaging, area weighted averaging, power weighted averaging, arithmetic averaging, and the like.
In the one-dimensional performance prediction model, the parameters at the average diameter of the inlet and outlet cross sections of each blade row represent the whole cross section parameters. In the three-dimensional numerical test or the physical test, the aerodynamic parameters of the flow field on the inlet and outlet cross sections of each blade row are distributed in two dimensions. And carrying out average processing on the two-dimensional distributed data along the radial direction and the circumferential direction to obtain average aerodynamic parameters on the inlet and outlet cross sections of each blade row, and representing the whole cross section parameters by the average aerodynamic parameters.
In the embodiment of the disclosure, the data on the inlet and outlet cross sections of each blade row obtained by the three-dimensional numerical test or the physical test can be subjected to average treatment along the radial direction and along the circumferential direction (average treatment is performed on the whole cross section), and the data on the average diameter of the inlet and outlet cross sections of each blade row obtained by the three-dimensional numerical test or the physical test can also be subjected to average treatment along the circumferential direction (average treatment is performed on the average diameter), so that the specific method is selected and determined according to the actual effect.
In an embodiment of the present disclosure, in step 102, the test data acquired may include test data for one or more of the turbomachine(s). Accordingly, in step 101, one or more geometric parameters of the turbomachine may be entered correspondingly. According to the embodiment of the disclosure, the model parameters to be optimized of the impeller mechanical performance prediction model are optimized by using the performance prediction values and the test data of the plurality of impeller machines, so that the universality of the optimization result can be further improved.
In the disclosed embodiment, in step 102, test data may be obtained from a three-dimensional numerical test database or a physical test database. By establishing a three-dimensional numerical test database or a physical test database, data can be accumulated continuously along with the continuous development of the three-dimensional numerical test or the physical test of the air compressor. The flow field aerodynamic parameter data such as attack angle, lag angle, total pressure loss coefficient and the like in the three-dimensional numerical test database or the physical test database can be used in the parameter optimization process of any one-time performance prediction model, and the better the universality of the performance prediction model obtained by optimization is along with the continuous increase of the data quantity in the database.
In some exemplary embodiments, the data of the three-dimensional numerical test database or the physical test database may be classified and stored according to a preset classification. For example, the load coefficients and the flow coefficients may be classified and stored. When the impeller mechanical performance prediction of different load levels and through-flow capacities is related, the database data under the corresponding category is selected for performance prediction model parameter optimization.
In some exemplary embodiments, in step 103, the multi-objective optimization function represents the difference between the predicted value of the macroscopic performance parameter and the experimental value of the macroscopic performance parameter using a distance metric method or a similarity metric method, and represents the difference between the predicted value of the aerodynamic parameter of the flow field and the experimental value of the aerodynamic parameter of the flow field using a distance metric method or a similarity metric method.
In the embodiment of the disclosure, a difference measurement method between performance prediction data (including a predicted value of a macroscopic performance parameter and a predicted value of a flow field aerodynamic parameter) of an impeller mechanical performance prediction model and three-dimensional numerical test or physical test data may use a distance measurement, or may use a similarity measurement, and an exemplary distance measurement may use a Hausdorff (Hausdorff) distance or the like, and a similarity measurement may use a similarity coefficient or the like.
In some exemplary embodiments, in step 103, the multi-objective optimization function calculates the difference between the predicted and trial values of at least two parameters using a multi-objective optimization method, which exemplary multi-objective optimization method includes, but is not limited to, at least one of: linear weighting, epsilon constraint, or evolutionary algorithms, including but not limited to at least one of: standard genetic algorithm, particle swarm algorithm, simulated annealing algorithm, second generation non-dominant ordered genetic algorithm.
The optimization problem of the embodiment of the disclosure belongs to a multi-objective optimization problem, and optimization is performed by adopting a multi-objective optimization method. The disclosed embodiments may employ conventional multi-objective optimization methods, such as linear weighting, epsilon constraint, etc., and evolutionary algorithm solutions, such as standard genetic algorithms, particle swarm algorithms, simulated annealing algorithms, second generation non-dominant ordering genetic algorithms, etc.
In some exemplary embodiments, in step 103, the macroscopic performance parameter comprises at least one of: pressure ratio, flow, power, efficiency, however, embodiments of the present disclosure are not limited in this regard.
In some exemplary embodiments, in step 103, the flow field aerodynamic parameters include at least one of: angle of attack, angle of fall, inlet mach number, total pressure loss coefficient, however, embodiments of the present disclosure are not so limited.
In some exemplary embodiments, the multi-objective optimization function includes differences between predicted data characteristic lines and test data characteristic lines of macroscopic performance parameters corresponding to different numbers of rotational speed lines, and differences between predicted data characteristic lines and test data characteristic lines corresponding to aerodynamic parameters of a flow field corresponding to different numbers of blade rows.
An exemplary three stage axial flow compressor noon flow path is shown in fig. 2A. In fig. 2A, 1 and 2 represent a rotor blade row and a stator blade row, respectively, 11, 12, 13 represent a first stage rotor blade row, a second stage rotor blade row, and a third stage rotor blade row, respectively, 21, 22, 23 represent a first stage stator blade row, a second stage stator blade row, and a third stage stator blade row, respectively, 3 represents a casing, 4 represents a hub, 5 represents a wheel disc, 6 represents a rotation axis, 7 represents a compressor inlet section, 8 represents a compressor outlet section, and 9 represents a streamline at an average diameter.
In the disclosed embodiment, the macroscopic performance parameters are calculated using the compressor outlet section 8 parameters and the compressor inlet section 7 parameters. The macroscopic performance of the compressor is generally represented by the pressure ratio-flow characteristic line and the efficiency-flow characteristic line shown in fig. 2B, with the abscissa representing flow, the ordinate representing pressure ratio or efficiency, and the parameter representing rotational speed.
In the disclosed embodiment, the aerodynamic parameters of the flow field refer to aerodynamic parameters on the inlet section and aerodynamic parameters on the outlet section of the rotor blade rows 11, 12, 13 and stator blade rows 21, 22, 23, and aerodynamic parameters calculated using the aerodynamic parameters. Similar to the representation method of macroscopic performance parameters shown in fig. 2B, the flow field pneumatic parameters can be represented by a lag angle-attack angle characteristic line, a total pressure loss coefficient-attack angle characteristic line, etc., the abscissa is attack angle, the ordinate is lag angle or total pressure loss coefficient, the parameter is inlet mach number, etc., or the flow field pneumatic parameters can be represented by a lag angle-flow characteristic line, a total pressure loss coefficient-flow characteristic line, etc., the abscissa is flow, the ordinate is lag angle or total pressure loss coefficient, the parameter is inlet mach number, etc. However, the flow field aerodynamic parameters may also be represented by other methods, which are not limited by embodiments of the present disclosure.
Since the macroscopic property of each rotation speed line is a set of numbers, for example, a set of correspondence between pressure ratio and flow rate can be obtained; alternatively, there may be a set of efficiency and flow correspondences. The optimization problem is therefore to minimize the difference between the two curves, the predicted data characteristic line and the test data characteristic line, corresponding to macroscopic performance parameters corresponding to different numbers of rotational speed lines. For example, if there are X rotational speed lines, there will be X sets of two curves (predicted data characteristic line and test data characteristic line) contrasted, where X is an integer greater than or equal to 1.
Since more than one aerodynamic parameter is selected for the flow field of each blade row, for example, the following equation (1.6) selects 2 aerodynamic parameters, one is the lag angle, one is the total pressure loss coefficient, for a three-stage axial flow compressor as shown in fig. 2A, there are a total of 6 blade rows, each blade row has the above 2 parameters, and there are a total of 6*2 =12 flow field parameters. And comparing the difference of two groups of curves of the predicted data characteristic line and the test data characteristic line of the flow field pneumatic parameters corresponding to different blade rows by adopting a similar method for comparing the difference between the predicted data characteristic line and the test data characteristic line of the macroscopic performance parameters corresponding to different numbers of rotating speed lines.
In some exemplary embodiments, step 104 specifically includes:
determining initial values of model parameters;
starting an optimization process according to the determined initial value and the optimization range;
detecting whether the current optimization process meets an optimization convergence criterion, if so, ending optimization, and determining the optimized model parameters and the optimized predicted values of the macroscopic performance parameters and the flow field aerodynamic parameters; if the optimization convergence criterion is not met, continuing the optimization process until the optimization convergence criterion is met;
Detecting whether the predicted values of the optimized macroscopic performance parameters and the flow field aerodynamic parameters meet the prediction precision requirement, and resetting at least one of the following if the predicted values do not meet the prediction precision requirement: the method comprises the steps of returning and circularly executing a step of starting an optimization process according to a determined initial value and an optimization range, wherein the step comprises the steps of an empirical formula of an impeller mechanical performance prediction model, the number of model parameters to be optimized of the impeller mechanical performance prediction model, the initial value of the model parameters and the optimization range; and if the prediction accuracy requirement is met, ending the optimization process.
In an embodiment of the present disclosure, optimizing the convergence criterion includes: and (5) reaching the upper limit of the iteration times or the convergence residual error smaller than a set value, and the like.
According to the parameter optimization method for the impeller mechanical property prediction model, the coefficients in the empirical or semi-empirical formula are optimized based on macroscopic property data and flow field aerodynamic parameter data obtained by an impeller mechanical three-dimensional numerical test or a physical test, so that the property prediction precision is improved, and meanwhile, the universality of the property prediction model is guaranteed.
Fig. 2C is a diagram for an exemplary description of an alternative embodiment of the method for optimizing parameters of the predictive model of mechanical performance of an impeller of the present disclosure, with respect to a predictive model of one-dimensional performance of a compressor based on the mean streamline method. As shown in fig. 2C, a method for optimizing parameters of an impeller mechanical performance prediction model according to an embodiment of the present disclosure includes the following steps:
(1) Inputting geometrical parameters of the compressor
The method comprises the steps of extracting required geometric parameters of the compressor from the three-dimensional geometry of the physical entity of the existing compressor, taking the geometric parameters as input of a one-dimensional performance prediction model of the compressor and input of a three-dimensional numerical test of the compressor, and ensuring that the geometric parameters input to the one-dimensional performance prediction model of the compressor correspond to the geometric parameters input to the three-dimensional numerical test of the compressor or the geometric parameters adopted by the physical test of the compressor.
(2) Selected model empirical formula and initial model parameters
In a one-dimensional performance prediction model of a compressor based on an average streamline method, empirical formulas such as a reference attack angle, a reference lag angle, a non-design point lag angle, leaf loss, shock loss, secondary flow loss and the like are generally adopted. The empirical formula is not unique, and can be selected according to requirements, for example, the non-design point falling angle empirical formula is a Creveling empirical formula, a Howell empirical formula and the like, and the leaf loss empirical formula is a Creveling empirical formula, an Aungier empirical formula and the like.
In the present embodiment, the optimization of the non-design point falling angle empirical formula coefficient and the leaf loss empirical formula coefficient is taken as an example.
Assuming that the non-design point lag angle empirical formula is selected as the Creveling empirical formula, the lag angle delta is calculated using the following equation:
Wherein x= (i-i) * )/ε * I is the actual angle of attack, i * For the reference angle of attack, ε may be calculated from the empirical formula of the reference angle of attack * For the turning angle of the air flow in the reference state epsilon * ＝θ-δ * +i * ，δ * For reference relief angle, it can be calculated from the empirical formula of reference relief angle,θ is the leaf angle.
Assuming that the leaf pattern loss empirical formula is selected as the Creveling empirical formula, the leaf pattern lossCalculated using the following formula:
in the method, in the process of the invention,for reference leaf loss, M can be calculated from the empirical formula of reference leaf loss in Inlet mach number for the blade row.
The coefficients of-0.000809, 0.5588, -0.2928, 0.0001191, 0.48 and 0.3452 in formula (1.1), and the coefficients of 0.0005, -0.005, 0.01, -0.0594 and 0.0667 in formula (1.2) are 11 parameters in total, namely the initial model parameters before optimization.
(3) One-dimensional performance prediction
Based on the one-dimensional performance prediction model of the air compressor and the model empirical formula and initial model parameters selected in the step (2), performance prediction is carried out aiming at different rotating speeds and different working conditions, and macroscopic performance data of the air compressor and aerodynamic parameters of the drainage fields of all blades are obtained. Wherein the macro performance data includes pressure ratio-flow characteristic line data and efficiency-flow characteristic line data; the aerodynamic parameters of the flow field include attack angle, lag angle, inlet Mach number, total pressure loss coefficient, etc.
(4) Performing three-dimensional numerical test or physical test
And (3) carrying out a three-dimensional numerical test or a physical test according to the geometric parameters of the air compressor input in the step (1). The three-dimensional numerical test is developed based on a three-dimensional performance prediction method of the air compressor. When a three-dimensional numerical test or a physical test is carried out, a certain number of numerical probes or physical probes are required to be arranged on the inlet and outlet cross sections of each blade row along the radial direction and the circumferential direction so as to record aerodynamic parameter data at different rotating speeds, different working conditions and different three-dimensional flow field positions. And acquiring macroscopic performance data of the air compressor and aerodynamic parameters of the discharge field of each blade through a three-dimensional numerical test or a physical test.
(5) Average processing of three-dimensional numerical test or physical test data
Since the data obtained by the three-dimensional numerical test or the physical test are data in three-dimensional distribution (distribution in the axial direction, radial direction, and circumferential direction), in order to compare them with the one-dimensional performance prediction data, the data needs to be subjected to an averaging process. The data averaging methods include mass flow weighted averaging, area weighted averaging, power weighted averaging, arithmetic averaging, etc.
In the one-dimensional performance prediction model, the parameters at the average diameter of the inlet and outlet cross sections of each blade row represent the whole cross section parameters. In the three-dimensional numerical test or the physical test, the aerodynamic parameters of the flow field on the inlet and outlet cross sections of each blade row are distributed in two dimensions. The two-dimensional distribution data are subjected to average treatment along the radial direction and the circumferential direction, so that the average aerodynamic parameters on the inlet and outlet cross sections of each blade row can be obtained, and the parameters represent the whole cross section parameters.
(6) Method for determining difference measurement of performance prediction data and test data
When the performance prediction model parameter optimization is carried out, a multi-objective optimization function is constructed according to the difference between the one-dimensional performance prediction data and the three-dimensional numerical test or the physical test data, and the minimization of the multi-objective optimization function is used as an optimization target.
In constructing the multi-objective optimization function, a measure of the difference between the performance prediction data and the test data is involved. To evaluate the difference between the two sets of data sets, a distance metric or similarity metric method may be employed, the greater the distance between the two sets of data sets, the less similarity. The distance measure may be Hausdorff distance, and the like, and the similarity measure may be similarity coefficient, and the like.
The difference between Hausdorff distance metric performance prediction data and test data is described below as an example.
Hausdorff distance is a distance defined between any two sets in the metric space, is the maximum distance from one set to the nearest point in the other set, and is expressed by the mathematical formula:
H(A,B)＝max(h(A,B),h(B,A))
wherein H (A, B) is a bidirectional Hausdorff distance; h (A, B) is a unidirectional Hausdorff distance, i.e., the distance from the point set A to the point set B; ||a i -b j I represents the point a in the point set a i From point B, point B j Is a distance of (3). h (A, B) is all points a in the point set A i The maximum distance from the closest point in the set of points B.
According to the definition of the bidirectional Hausdorff distance, the Hausdorff distance between the data such as the compressor pressure ratio, efficiency, lag angle, total pressure loss coefficient and the like obtained by one-dimensional performance prediction and the compressor pressure ratio, efficiency, lag angle and total pressure loss coefficient obtained by a three-dimensional numerical test or a physical test is H (eta) test ,η 1D )、H(π test ,π 1D )、H(δ test ,δ 1D )、
Wherein eta, pi, delta andrespectively representing efficiency, pressure ratio, lag angle and total pressure loss coefficient, and subscripts test and 1D respectively representing three-dimensional numerical test or physical test result and one-dimensional performance prediction result, lag angle delta 1D Calculated by using the formula (1.1),leaf loss->Calculated by the formula (1.2), the secondary flow is lost +.>Calculated by using a secondary flow loss empirical formula, shock loss is calculatedMalnutrition of the heart>And calculating by using a shock loss empirical formula.
(7) Multi-objective optimization is performed by taking model parameters (i.e. empirical formula coefficients) as optimization variables
The empirical formula coefficients in the one-dimensional performance prediction model are selected as optimization variables (namely model parameters to be optimized), and specifically are 11 coefficients in a non-design point lag angle empirical formula (1.1) and a leaf type loss empirical formula (1.2).
In the optimization process, the non-design point trailing angle empirical formula and the leaf type loss empirical formula respectively adopt the following forms:
wherein C is 1 To C 6 C is an optimization variable in a non-design point lag angle empirical formula 7 To C 11 Is an optimization variable in the leaf loss empirical formula.
The optimization target is to minimize the difference among the data such as the compressor pressure ratio, efficiency, lag angle, total pressure loss coefficient and the like obtained by one-dimensional performance prediction and the compressor pressure ratio, efficiency, lag angle and total pressure loss coefficient obtained by a three-dimensional numerical test or a physical test. From the above, the optimization problem belongs to a multi-objective optimization problem, and the optimization is performed by adopting a multi-objective optimization method. Aiming at the multi-objective optimization problem, a traditional multi-objective optimization method such as a linear weighting method, an Epsilon constraint method and the like can be adopted, and an evolutionary algorithm such as a standard genetic algorithm, a particle swarm algorithm, a simulated annealing algorithm, a second-generation non-dominant ordering genetic algorithm and the like can be adopted for solving.
When the multi-objective optimization function is constructed, only whether macroscopic performance data prediction is accurate is concerned, modification processing on data is considered in a form, physical connotation is lacked, and larger performance prediction deviation is likely to occur when an optimized performance prediction model is used for other types of impeller machines, namely model universality is poor. The construction of the multi-objective optimization function in the embodiment of the disclosure considers both the difference between macroscopic characteristics of the compressor and the difference between aerodynamic parameters of the flow field, thereby taking into account mathematical accuracy and physical accuracy, and ensuring the universality of the performance prediction model while improving the performance prediction accuracy.
The following gives a typical multi-objective optimization function Obj form obtained using the linear weighting method:
wherein n and m respectively represent the number of rotation speed lines and the number of blade rows, w η 、w π 、w δ Andweight coefficients representing efficiency term, pressure ratio term, falling angle term and total pressure loss coefficient term, w at different rotation speeds η And w π W, which may be the same or different, for different blade rows δ And->May be the same or different. For a three stage axial flow compressor as shown in fig. 2A, m=6. For the compressor macro performance characteristic as shown in fig. 2B, n=5.
The optimization objective is to minimize the multi-objective optimization function Obj equation (1.6).
(8) Judging whether the optimization process meets the optimization convergence criterion
If the optimization process meets the optimization convergence criterion, for example, the upper limit of the iteration times is reached or the convergence residual error is smaller than a set value, the optimization is finished, and the optimized performance prediction model parameters, the optimized macroscopic performance data and the optimized flow field aerodynamic parameters are determined; if the optimization convergence criterion is not met, continuing to perform optimization until the optimization convergence criterion is met.
(9) Judging whether the performance prediction result meets the prediction precision requirement
If the prediction precision of the optimized macroscopic performance data and the flow field aerodynamic parameters does not meet the set value, reselecting the model empirical formula and the initial model parameters, and repeating the steps (2) to (8) (in the process, when the model empirical formula is reselected, the formulas (1.1) and (1.2) may change in form and coefficient) until the prediction precision requirement is met; if the set value is met, ending the optimization flow.
In the above optimization method, the input geometric parameters of the compressors may be geometric parameters of one compressor or geometric parameters of a plurality of compressors. If the geometric parameters of the compressors are input, the three-dimensional numerical test or the physical test data of the compressors are correspondingly required to be developed, so that a multi-objective optimization function of the multi-objective optimization problem is constructed according to the difference between the performance prediction data and the test data of different compressors.
In the above-mentioned optimization method, a three-dimensional numerical test or physical test database may be established, which continuously accumulates data as the three-dimensional numerical test or physical test of the compressor is continuously developed. The flow field aerodynamic parameter data such as attack angle, lag angle, total pressure loss coefficient and the like in the database can be used in the parameter optimization process of any one-time performance prediction model, and the better the universality of the performance prediction model obtained by optimization is along with the continuous increase of the data quantity in the database.
In the optimization method, the data in the three-dimensional numerical test or physical test database can be classified, for example, the data can be classified by adopting a load coefficient and a flow coefficient, and when the mechanical performance prediction of the impeller with different load levels and through-flow capacities is related, the database data in the corresponding category is selected for the performance prediction model parameter optimization.
In the above-described optimization method, macroscopic performance parameters are not limited to pressure ratio, flow rate, efficiency, power, etc. The aerodynamic parameters of the flow field are not limited to attack angle, lag angle, total pressure loss coefficient and the like, and can be obtained by a performance prediction model, a three-dimensional numerical test or a physical test.
In the above-described optimization method, the optimization variables are not limited to 11 coefficients in the non-design-point trailing angle empirical formula (1.1) and the leaf loss empirical formula (1.2), and any coefficient in any empirical formula may be used as the optimization variables.
In the optimization method, if the optimized macroscopic performance data and the flow field aerodynamic parameter prediction precision do not meet the set values, the model parameters to be optimized, the model parameter optimization range and the like of the impeller mechanical performance prediction model can be reselected besides the model empirical formula and the initial model parameters.
In the above-described optimization method, the multi-objective optimization function form is not limited to the formula (1.6), and the multi-objective optimization function form may be determined according to the needs of researchers. For the multi-objective optimization function of equation (1.6), a variety of optimization algorithms may be employed.
The method for optimizing the parameters of the low-dimensional performance prediction model of the impeller machinery based on the macroscopic performance data and the flow field aerodynamic parameters obtained by the three-dimensional numerical test or the physical test has the advantages that the accuracy in performance prediction and the accuracy in physics are considered, the prediction accuracy of the macroscopic performance and the flow field aerodynamic parameters of the impeller machinery can be improved on the basis of keeping the original low-dimensional performance prediction model structure unchanged, and meanwhile, the optimized low-dimensional performance prediction model has good universality.
The embodiment of the disclosure also provides a parameter optimization device for the impeller mechanical property prediction model, which comprises a memory; and a processor coupled to the memory, the memory for storing instructions, the processor configured to perform the steps of the method of optimizing parameters of the predicted model of impeller mechanical performance according to any embodiment of the present disclosure based on the instructions stored in the memory.
As shown in fig. 3, in one example, the impeller mechanical performance prediction model parameter optimization apparatus may include: processor 310, memory 320, and bus system 330, wherein processor 310 and memory 320 are coupled via bus system 330, memory 320 is configured to store instructions, and processor 310 is configured to execute the instructions stored by memory 320. Specifically, the processor 310 determines model parameters and an optimization range to be optimized in the impeller mechanical performance prediction model; acquiring test data; establishing a multi-objective optimization function and an optimization objective, wherein the multi-objective optimization function is used for representing the difference between predicted values of at least two parameters output by the impeller mechanical property prediction model and test values of at least two parameters in the test data, and the at least two parameters comprise: macroscopic performance parameters and flow field aerodynamic parameters, the optimization objective being to minimize the multi-objective optimization function; and carrying out iterative computation optimizing on the multi-objective optimization function according to the optimization range to obtain optimized model parameters.
It should be appreciated that the processor 310 may be a central processing unit (Central Processing Unit, CPU), and the processor 310 may also be other general purpose processors, digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), off-the-shelf programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Memory 320 may include read only memory and random access memory and provide instructions and data to processor 310. A portion of memory 320 may also include non-volatile random access memory. For example, the memory 320 may also store information of a device type.
The bus system 330 may include a power bus, a control bus, a status signal bus, and the like in addition to a data bus. But for clarity of illustration the various buses are labeled as bus system 330 in fig. 3.
In implementation, the processing performed by the processing device may be performed by integrated logic circuitry in hardware or instructions in software in processor 310. That is, the method steps of the embodiments of the present disclosure may be embodied as hardware processor execution or as a combination of hardware and software modules in a processor. The software modules may be located in random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, and other storage media. The storage medium is located in the memory 320, and the processor 310 reads the information in the memory 320, and in combination with its hardware, performs the steps of the above method. To avoid repetition, a detailed description is not provided herein.
The disclosed embodiments also provide a computer readable storage medium having stored thereon a computer program which when executed by a processor implements the method of optimizing parameters of an impeller mechanical performance prediction model according to any of the embodiments of the present disclosure. The method for optimizing parameters of the impeller mechanical performance prediction model driven by executing the executable instructions is basically the same as the method for optimizing parameters of the impeller mechanical performance prediction model provided in the above embodiment of the present disclosure, and will not be described in detail herein.
In some possible implementations, aspects of the method for optimizing parameters of an impeller mechanical performance prediction model provided by the embodiments of the present disclosure may also be implemented in the form of a program product, which includes a program code for causing a computer device to perform the steps in the method for optimizing parameters of an impeller mechanical performance prediction model according to the various exemplary embodiments of the present disclosure described above, when the program product is run on the computer device, for example, the computer device may perform the method for optimizing parameters of an impeller mechanical performance prediction model described in the embodiments of the present disclosure.
The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
Those of ordinary skill in the art will appreciate that all or some of the steps, systems, functional modules/units in the apparatus, and methods disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. In a hardware implementation, the division between the functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be performed cooperatively by several physical components. Some or all of the components may be implemented as software executed by a processor, such as a digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as known to those skilled in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer. Furthermore, as is well known to those of ordinary skill in the art, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media.
While the embodiments disclosed in the present disclosure are described above, the embodiments are only employed for facilitating understanding of the present disclosure, and are not intended to limit the present disclosure. Any person skilled in the art will recognize that any modifications and variations can be made in the form and detail of the present disclosure without departing from the spirit and scope of the disclosure, which is defined by the appended claims.
Claims (10)
1. A method for optimizing parameters of a predictive model of impeller mechanical properties, comprising:
determining model parameters to be optimized and an optimization range in the impeller mechanical property prediction model;
acquiring test data;
establishing a multi-objective optimization function and an optimization objective, wherein the multi-objective optimization function is used for representing the difference between predicted values of at least two parameters output by the impeller mechanical property prediction model and test values of at least two parameters in the test data, and the at least two parameters comprise: macroscopic performance parameters and flow field aerodynamic parameters, the optimization objective being to minimize the multi-objective optimization function;
and carrying out iterative computation optimizing on the multi-objective optimization function according to the optimization range to obtain optimized model parameters.
2. The method according to claim 1, wherein the test data is numerical test data or physical test data, the numerical test data is obtained based on a high-dimensional performance prediction method of the turbomachine, which is a three-dimensional performance prediction method or a two-dimensional performance prediction method for parameter optimization of a one-dimensional performance prediction model of the turbomachine, and which is a three-dimensional performance prediction method for parameter optimization of a two-dimensional performance prediction model of the turbomachine, after verification of the physical test data.
3. The method of claim 1, wherein after the step of acquiring test data, the method further comprises: averaging the test data, the averaging including at least one of: mass flow weighted average, area weighted average, power weighted average, and arithmetic average.
4. The method of claim 1, wherein the multi-objective optimization function represents a difference between the predicted value of the macro performance parameter and the test value of the macro performance parameter using a distance metric method or a similarity metric method, and represents a difference between the predicted value of the flow field aerodynamic parameter and the test value of the flow field aerodynamic parameter using a distance metric method or a similarity metric method.
5. The method of claim 1, wherein the multi-objective optimization function calculates a difference between the predicted and trial values of the at least two parameters using a multi-objective optimization method comprising at least one of: a linear weighting method, an Epsilon constraint method, or an evolutionary algorithm, the evolutionary algorithm comprising at least one of: standard genetic algorithm, particle swarm algorithm, simulated annealing algorithm, second generation non-dominant ordered genetic algorithm.
6. The method of claim 1, wherein iteratively optimizing the multi-objective optimization function according to the optimization range comprises:
determining initial values of model parameters;
starting an optimization process according to the determined initial value and the optimization range;
detecting whether the current optimization process meets an optimization convergence criterion, if so, ending optimization, and determining the optimized model parameters and the optimized predicted values of the macroscopic performance parameters and the flow field aerodynamic parameters; if the optimization convergence criterion is not met, continuing the optimization process until the optimization convergence criterion is met; the optimization convergence criterion includes: reaching the upper limit of iteration times or the convergence residual error is smaller than a set value;
Detecting whether the predicted values of the optimized macroscopic performance parameters and the flow field aerodynamic parameters meet the prediction precision requirement, and resetting at least one of the following if the predicted values do not meet the prediction precision requirement: the step of starting the optimization process according to the determined initial value and optimization range is returned and circularly executed; and if the prediction accuracy requirement is met, ending the optimization process.
7. The method of claim 1, wherein the macroscopic performance parameter comprises at least one of: pressure ratio, flow, power, efficiency; the flow field aerodynamic parameters include at least one of: angle of attack, angle of fall, total pressure loss coefficient.
8. The method of claim 1, wherein the turbomachine is a turbomachine working machine or a turbomachine prime mover, the turbomachine being of an axial flow type or a radial flow type, the turbomachine being of a single stage or a multiple stage.
9. The impeller mechanical performance prediction model parameter optimization device is characterized by comprising a memory; and a processor connected to the memory, the memory for storing instructions, the processor configured to perform the steps of the impeller mechanical performance prediction model parameter optimization method of any one of claims 1 to 8 based on the instructions stored in the memory.
10. A storage medium having stored thereon a computer program which when executed by a processor implements the method of optimizing the parameters of the predicted model of impeller mechanical performance according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310558377.8A CN116595874A (en) | 2023-05-17 | 2023-05-17 | Impeller mechanical performance prediction model parameter optimization method and device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310558377.8A CN116595874A (en) | 2023-05-17 | 2023-05-17 | Impeller mechanical performance prediction model parameter optimization method and device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116595874A true CN116595874A (en) | 2023-08-15 |
Family
ID=87598658
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310558377.8A Pending CN116595874A (en) | 2023-05-17 | 2023-05-17 | Impeller mechanical performance prediction model parameter optimization method and device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116595874A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117648780A (en) * | 2024-01-30 | 2024-03-05 | 陕西空天信息技术有限公司 | Parameter optimization method and device for impeller machinery and computer storage medium |
-
2023
- 2023-05-17 CN CN202310558377.8A patent/CN116595874A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117648780A (en) * | 2024-01-30 | 2024-03-05 | 陕西空天信息技术有限公司 | Parameter optimization method and device for impeller machinery and computer storage medium |
| CN117648780B (en) * | 2024-01-30 | 2024-05-07 | 陕西空天信息技术有限公司 | Parameter optimization method and device for impeller machinery and computer storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111027148B (en) | Automatic calibration and industrial axial flow compressor performance calculation method for loss lag angle model | |
| US20130282307A1 (en) | Turbo-machinery stage families tuning/calibration system and method | |
| CN107203364B (en) | Prediction and identification method for full-working-condition characteristics of gas compressor | |
| CN112417773B (en) | Multidisciplinary optimization design method, device and equipment of multistage axial flow expander | |
| CN111898212B (en) | Impeller mechanical profile design optimization method based on BezierGAN and Bayesian optimization | |
| CN114254460B (en) | Turbomachine pneumatic robustness optimization method based on graph convolution neural network | |
| Derakhshan et al. | Optimization of GAMM Francis turbine runner | |
| CN116595874A (en) | Impeller mechanical performance prediction model parameter optimization method and device and storage medium | |
| CN107076162A (en) | Reversible type pump turbine includes the Optimization Design of its self-generating system and reversible type pump turbine | |
| CN112270139B (en) | Pneumatic optimization design method for centrifugal compressor of fuel cell based on mother type library | |
| CN115017843A (en) | Pneumatic performance optimization design method for centrifugal compressor | |
| Bashiri et al. | Design optimization of a centrifugal pump using particle swarm optimization algorithm | |
| CN110705079B (en) | Centrifugal compressor structure optimization method based on simulated annealing algorithm | |
| CN115374576A (en) | Integrated stability expansion design method for treatment of compressor blade and casing | |
| Ji et al. | Computer 3D vision-aided full-3D optimization of a centrifugal impeller | |
| Safari et al. | A novel combination of adaptive tools for turbomachinery airfoil shape optimization using a real-coded genetic algorithm | |
| CN117195760A (en) | Radial blending-based axial flow fan or compressor meridian plane through flow calculation method | |
| Obrovský et al. | Development of high specific speed Francis turbine for low head HPP | |
| Wang et al. | Performance dispersion control of a multistage compressor based on precise identification of critical features | |
| Sun et al. | Optimization design of IGV profile in centrifugal compressor | |
| Verstraete et al. | Multidisciplinary design and off-design optimization of a radial compressor for industrial applications | |
| CN117648780B (en) | Parameter optimization method and device for impeller machinery and computer storage medium | |
| Cuciumita et al. | Genetic algorithm for gas turbine blading design | |
| Ma et al. | Two Dimensional Optimization of Centrifugal Compressor Impellers Using Online Quasi-3D Flow Solver and Genetic Algorithm | |
| Kulkarni | Development of a Methodology to Estimate Aero-Performance and Aero-Operability Limits of a Multistage Axial Flow Compressor for Use in Preliminary Design |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |