A method, device, medium and product for optimizing an outlet measurement point arrangement of a compressor test

Abstract

The application discloses a kind of compressor test outlet measuring point arrangement scheme optimization method, equipment, medium, product, it is related to the technical field of aero-engine compressor test.The method comprises: obtaining compressor flow field reduced-order model based on intrinsic orthogonal decomposition and Kriging algorithm, and predicting outlet section flow field pressure distribution;Determine the circumferential gap of probe and the radial gap of measuring point as optimization parameters, and set the physical space constraint of compressor;With the difference between the average value of measuring point pressure and the average value of section grid node pressure as the objective function, wherein the average value of measuring point pressure is calculated using mixed interpolation strategy and circumferential radial weighted average based on flow field pressure distribution;Optimization parameters are iteratively updated using intelligent optimization algorithm, and the optimal measuring point arrangement scheme is output.The application overcomes the defects that traditional measuring point arrangement is highly dependent on experience and is prone to measurement blind area, maximizes the coverage of key aerodynamic region of flow field under limited measuring points, and significantly improves the accuracy and reliability of compressor test data.

Description

Technical Field

[0001] This invention relates to the field of aero-engine compressor testing technology, specifically to a method, equipment, medium, and product for optimizing the layout of compressor test outlet measuring points. The aim is to solve the problem of high dependence on experience and inaccurate capture of flow field information in the preparation stage of compressor test by combining field order reduction technology and optimization technology, thereby improving the efficiency and accuracy of compressor test. Background Technology

[0002] The statements in this section are provided only as background information in connection with this disclosure and may not constitute prior art.

[0003] In the entire process of aero-engine compressor performance testing, the rational arrangement of measurement points at the outlet section is not only a fundamental step in obtaining flow field parameters, but also a crucial step that decisively affects the accuracy and comprehensiveness of the test results. For a long time, the measurement point arrangement strategies commonly used in the industry have relied heavily on the practical experience and subjective judgment of individual engineers, lacking systematic and scientific theoretical guidance and quantitative analysis support. This experience-based approach often results in insufficient objectivity in the selection of measurement point locations, uneven and incomplete distribution, and difficulty in effectively covering key aerodynamic regions and areas with significant parameter gradient changes in the flow field, such as vortex generation zones, boundary layer transition zones, and unsteady separation zones—complex structural regions. Due to the complex flow structures and large physical quantity gradients in these regions, traditional methods easily create measurement blind spots, resulting in the collected test data failing to truly and completely reflect the actual distribution characteristics and dynamic behavior of the compressor outlet flow field, significantly reducing the accuracy, reliability, and usability of the data.

[0004] Furthermore, an unreasonable arrangement of measurement points can lead to a series of problems related to test efficiency and resource consumption. For example, during the test preparation phase, sensors may need to be adjusted and rearranged multiple times, which not only prolongs the preparation cycle but also increases the number of debugging attempts, resulting in a waste of human and material resources, increased test costs, and decreased overall efficiency. Especially against the backdrop of modern aero-engines continuously developing towards higher loads, higher efficiency, and higher stability, the internal flow phenomena of compressors are becoming increasingly complex, exhibiting significant three-dimensional unsteady and multi-scale turbulent characteristics. This further increases the practical requirements for the spatial resolution and dynamic response capabilities of measurement technology. Faced with such complex flow structures, traditional experience-based measurement point arrangement methods are proving inadequate and cannot meet the dual needs of high-precision aerodynamic measurement and the refined aerodynamic design of modern compressors.

[0005] Therefore, addressing the shortcomings of current measurement point layout methods in testing, conducting systematic research based on fluid mechanics theory, numerical simulation, and intelligent optimization algorithms to establish a scientific, efficient, and reusable method for optimizing the layout of compressor test outlet measurement points has become an urgent task with significant practical engineering implications. The establishment of this method will not only significantly improve the quality and reliability of test data but also provide solid data support and methodological assurance for further optimization of engine performance and technological innovation. Summary of the Invention

[0006] The purpose of this invention is to propose an innovative method for optimizing the layout of measuring points at the compressor outlet during testing. This method is based on a field order reduction model and combines numerical simulation results, field order reduction techniques, and intelligent optimization algorithms to achieve scientific and systematic optimization of the measuring point layout. Specifically, this invention aims to overcome the shortcomings of traditional measuring point layout methods, such as over-reliance on experience, uneven distribution of measuring points, and difficulty in capturing key flow field information. It provides a method that can accurately identify key aerodynamic regions and locations of significant parameter gradient changes in the flow field, thereby ensuring that measuring points can comprehensively and uniformly cover the entire outlet cross-section, effectively improving the accuracy, reliability, and usability of test data. Simultaneously, by optimizing the measuring point layout, the number of adjustments and debugging time during the test preparation phase is reduced, the consumption of manpower and material resources is decreased, and test efficiency is improved, providing a more solid data foundation and technical support for the performance evaluation and optimization design of aero-engine compressors.

[0007] This invention aims to optimize the radial and axial spacing of probes on the compressor outlet section, minimizing the error between the average pressure at each measuring point and the average value of all grid nodes on the section. In other words, the measuring point arrangement can represent the flow field characteristics at the outlet section. The process is mainly implemented in two stages: first, training a reduced-order compressor flow field model to quickly obtain flow field distribution information, especially pressure distribution, at the compressor outlet section under different operating conditions; second, implementing the optimization process, supporting iteration of multiple measuring point arrangement schemes and the final optimized recommendation.

[0008] In the stage of training the compressor flow field reduction model, a large amount of simulation data of the compressor under different operating conditions was first collected. This data covered various flow field parameters at the outlet section, especially pressure data. Using this data, the high-dimensional data was reduced in dimensionality through field reduction techniques to identify the main modal information of the flow field. Combined with machine learning algorithms, a reduced-order model that can accurately reflect the characteristics of the compressor outlet flow field was constructed. This model can significantly reduce computational complexity while maintaining accuracy, thereby quickly obtaining flow field distribution information under different operating conditions.

[0009] In the optimization process phase, based on a trained compressor flow field reduction model and combined with intelligent optimization algorithms, the measuring point layout scheme is iteratively optimized. Specifically, by setting a reasonable optimization objective function, such as minimizing the error between the average pressure at each measuring point and the average value of all grid nodes in the cross-section, the intelligent optimization algorithm automatically searches for the optimal measuring point layout scheme in the solution space. During the iteration process, the radial and axial spacing of the measuring points on the probe is continuously adjusted, and the average value of the measuring points and the average value of the flow field grid nodes are calculated and compared based on the flow field characteristics, so that the measuring point layout can better represent the flow field characteristics at the outlet cross-section. Finally, through multiple iterations and evaluations, the optimal measuring point layout scheme is obtained, providing scientific and reasonable measuring point layout guidance for compressor testing.

[0010] Specifically, the technical solution of the present invention is as follows: An optimization method for the layout of measuring points at the compressor test outlet includes: A reduced-order compressor flow field model is obtained. This model is based on single-channel flow field simulation data of the compressor under different operating conditions. The flow field feature vectors are extracted using the intrinsic orthogonal decomposition (POD) method for order reduction, and then trained using the Kriging algorithm. The reduced-order compressor flow field model is used to predict and output the flow field pressure distribution at the compressor outlet section based on the input compressor speed and outlet static pressure. Determine the optimization parameters and constraints for the arrangement of measuring points. The optimization parameters include the circumferential clearance of the probes and the radial clearance of the measuring points. The optimization constraints include that the sum of the circumferential clearances of all the probes does not exceed the circumferential angle range of a single channel of the compressor, and the sum of the radial clearances of all the measuring points does not exceed the distance between the compressor casing and the hub. The objective function is set to minimize the difference between the average pressure at the measuring points and the average pressure at the cross section. The average pressure at the cross section is calculated based on the pressure of the grid nodes on the compressor outlet cross section. The average pressure at the measuring points is calculated based on the flow field pressure distribution, using a hybrid interpolation strategy that combines linear interpolation and nearest neighbor interpolation to obtain the pressure values ​​at each measuring point, and then using a circumferential and radial weighted average method. Using an intelligent optimization algorithm, the optimization parameters are iteratively updated under the premise of satisfying the optimization constraints. In each iteration, the objective function value under the current measurement point layout scheme is calculated until the preset convergence condition is met or the maximum number of iterations is reached, and the optimal measurement point layout scheme is output.

[0011] Furthermore, the step of reducing the order of the flow field feature vector extracted by the intrinsic orthogonal decomposition (POD) method includes: The single-channel flow field simulation data of the compressor under different operating conditions were used to construct a flow field matrix; Construct the covariance matrix of the flow field matrix, and solve for the eigenvalues ​​and eigenvectors of the covariance matrix. The eigenvectors represent the spatial structure of the flow field, and the eigenvalues ​​represent the energy contribution of the corresponding eigenvectors. Arrange the features in descending order of their eigenvalues, and select the top... The original flow field matrix is ​​projected onto the reduced-order subspace as an optimal orthogonal basis to complete the order reduction process. in, It is a positive integer, and the first selected... The ratio of the sum of the eigenvalues ​​corresponding to each eigenvector to the sum of all eigenvalues ​​is not less than 95%.

[0012] Furthermore, the reduced-order model of the compressor flow field obtained by training with the Kriging algorithm includes: Using compressor speed and outlet static pressure as input variables, the Kriging algorithm is used to estimate the flow field pressure distribution at unknown points by weighted sum of known sample points. The weighting coefficients that satisfy the unbiasedness condition and the minimum variance condition are solved by the Lagrange multiplier method. The weight coefficients are solved by using a Gaussian radial basis function as the covariance function. The training data is divided into multiple subsets by cross-validation. The combination of variance and scale parameters that minimizes the prediction error is selected as the optimal parameters of the Gaussian radial basis function.

[0013] Furthermore, before acquiring the pressure values ​​at each measuring point, a single-channel mapping step for the probe measuring point data is also included: Since the compressor outlet flow field is periodically distributed, the flow field data measured by multiple probes inserted in different blade channels are redistributed and mapped to the same simulated single channel for centralized calculation based on the relative position of each probe in each channel; the probe has multiple measuring points distributed radially.

[0014] Furthermore, the method of obtaining the pressure value at each measuring point using a hybrid interpolation strategy combining linear interpolation and nearest neighbor interpolation includes: For each measurement point to be interpolated, within the spatial range of the original position data formed by the grid nodes on the compressor outlet section, a triangle containing the measurement point is found; based on the known pressure values ​​of the vertices of the triangle, linear weights are assigned according to the relative position of the measurement point within the triangle, and the pressure value of the measurement point is calculated by weighted average. If the measuring point exceeds the convex hull range formed by the original position data, causing linear interpolation to fail and be marked as invalid, then the Euclidean distance between the measuring point marked as invalid and all original position points is calculated, and the pressure value of the nearest original position point is selected as the interpolation result for that measuring point.

[0015] Furthermore, the average pressure at each measuring point is calculated using the circumferential-radial weighted average method. First, the average circumferential pressure at each measuring point is calculated by weighting the area occupied by the measuring point. The calculation formula is as follows:

[0016] in, Indicates the number of probes on all probes. The average circumferential pressure at each measuring point Indicates the circumferential position as Radial position is Pressure value at the measuring point, The radial area between two adjacent probes. This refers to the number of probes arranged circumferentially. .

[0017] Furthermore, after calculating the average circumferential pressure at each measuring point, a wall-simulated measuring point is constructed using a boundary layer correction factor, a radial torus is divided, and the average pressure at the measuring point of the outlet section is calculated using the radial torus weighted method. The calculation formula is as follows:

[0018] in, This refers to the number of measuring points arranged on a single probe. For the first The radial coordinate values ​​of each measuring point.

[0019] The present invention also proposes an electronic device, comprising: At least one processor; and a memory communicatively connected to said at least one processor; The memory stores instructions that can be executed by the at least one processor, and the at least one processor executes the instructions stored in the memory to perform the method described above.

[0020] The present invention also proposes a computer-readable storage medium for storing instructions that, when executed, cause the method described above to be implemented.

[0021] The present invention also proposes a computer program product, which implements the above-described method when executed by a processor.

[0022] Compared with existing technologies, the advantages of this invention are: 1. Improved accuracy in flow field identification: By introducing a field reduction model, key aerodynamic regions and locations with significant parameter gradient changes in the compressor outlet flow field can be accurately identified. This guides the scientific arrangement of measurement points, overcomes the drawbacks of traditional methods that rely on experience and have uneven distribution, and ensures that measurement points fully and uniformly cover the outlet section, effectively improving the accuracy, reliability, and usability of experimental data.

[0023] 2. Automated Layout Optimization: By combining intelligent optimization algorithms, the layout of measurement points is automatically optimized, reducing the number of adjustments and debugging time in the test preparation stage, significantly reducing the consumption of manpower and material resources, and improving test efficiency.

[0024] 3. Improved quality of measurement point layout: For specific scenarios at the compressor outlet section, the distribution density and spatial location of measurement points are scientifically optimized to ensure maximum coverage of key areas of the flow field with a limited number of measurement points. This reduces measurement errors caused by redundant or missing measurement points, provides higher quality raw data support for the reconstruction of the flow field at the outlet section, and directly improves the spatial resolution and aerodynamic parameter characterization accuracy of the compressor test data. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments recorded in the embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0026] Figure 1 This is a technical roadmap for optimizing the layout of compressor test outlet measuring points provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of a single-channel flow area of ​​a compressor provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the pressure distribution on the surface of a compressor blade provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the compressor flow direction pressure change provided in an embodiment of the present invention; Figure 5 This is a compressor meridional pressure cloud map provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the probe mounting position and single-channel simulation provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of the distribution of exit probe measurement points provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of the test layout optimization parameters provided in an embodiment of the present invention; Figure 9 This is a schematic diagram of the custom parameter interface provided in an embodiment of the present invention; Figure 10 This is a schematic diagram of the constraint addition interface provided in an embodiment of the present invention; Figure 11 This is a schematic diagram of the compressor inlet and outlet cross sections and coordinate axes provided in an embodiment of the present invention; Figure 12 This is a schematic diagram of the cut-out plane setting interface provided in an embodiment of the present invention; Figure 13 This is a schematic diagram of circumferential and radial weighted average calculation provided in an embodiment of the present invention; Figure 14 The optimized calculation flowchart provided for embodiments of the present invention; Figure 15 This is a diagram showing the interface for displaying the optimization results of the measurement point layout provided in an embodiment of the present invention; Figure 16 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0027] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0028] The features and performance of the present invention will be further described in detail below with reference to embodiments.

[0029] Example 1 See Figure 1 The technical approach for optimizing the arrangement of measuring points at the compressor outlet section is shown in this embodiment. This embodiment provides a method for optimizing the arrangement of measuring points at the compressor test outlet. Given the significant three-dimensional unsteady and multi-scale turbulent characteristics of the internal flow of aero-engine compressors, traditional arrangement methods relying on engineer experience are prone to causing measurement blind spots. Therefore, this embodiment deeply integrates fluid dynamics numerical simulation with intelligent optimization algorithms, automatically finding the optimal solution under physical constraints.

[0030] Specifically, the method of this embodiment is based on the steps described in the following claims. To fully disclose its technical details and working principle, detailed descriptions are provided after each step: Step 1: Obtain the compressor flow field reduced-order model. The compressor flow field reduced-order model is obtained based on the single-channel flow field simulation data of the compressor under different operating conditions. The flow field feature vectors are extracted by the intrinsic orthogonal decomposition (POD) method for order reduction processing, and then trained by the Kriging algorithm. The compressor flow field reduced-order model is used to predict and output the flow field pressure distribution at the compressor outlet section based on the input compressor speed and outlet static pressure. In this embodiment, it should be noted that the verification object is a single-stage compressor. In actual engineering, the compressor flow field has a periodic distribution characteristic. Therefore, to reduce computing power consumption, the simulation calculation uses single-channel calculation instead of whole-machine calculation (e.g., Figure 2 The single-channel watershed shown and Figure 3 (The blade surface pressure is shown). In practice, commercial fluid dynamics simulation software such as CFX can be used to calculate the compressor characteristic curves for different speeds. For example, when calculating a single speed characteristic curve, the total inlet pressure is fixed at 101 kPa and the speed remains constant, while the outlet static pressure is changed to complete the full-condition calculation from the surge point to the blockage point (flow-direction pressure changes and meridional pressure contour maps are shown). Figure 4 , Figure 5 (As shown).

[0031] Compressor flow field data contains rich flow field information, which is highly dimensional and difficult to process and analyze directly. To efficiently extract key flow field features and reduce computational complexity, this invention employs a field order reduction technique to reduce the order of compressor flow field data. Using the POD method, matrix decomposition is performed on flow field data at a specific outlet pressure and a given rotational speed. Feature vectors characterizing the data are extracted, and the first few vectors with larger eigenvalues ​​are selected as the optimal orthogonal basis. The original full-order data is then projected onto the reduced-order subspace composed of this set of orthogonal bases, thus completing the order reduction of the flow field data.

[0032] Compressor flow field data can be regarded as being composed of grid nodes Composed of physical quantities (such as pressure, temperature, velocity, etc.), it can be represented as a high-dimensional matrix. In this matrix, each column corresponds to the spatial distribution of a physical quantity. If directly used for subsequent iterative optimization, the computational cost would explode exponentially. Therefore, this embodiment introduces a reduced-order model.

[0033] To achieve the above-mentioned order reduction and prediction, the specific sub-steps are as follows: Step 1.1: Construct a flow field matrix from the single-channel flow field simulation data of the compressor under different operating conditions; construct the covariance matrix of the flow field matrix, and solve for the eigenvalues ​​and eigenvectors of the covariance matrix. The eigenvectors represent the spatial structure of the flow field, and the eigenvalues ​​represent the energy contribution of the corresponding eigenvectors; arrange the eigenvalues ​​in descending order of size, and select the top... Using eigenvectors as the optimal orthogonal basis, a reduced-order subspace is constructed. The original flow field matrix is ​​then projected onto this reduced-order subspace to complete the order reduction process. It is a positive integer, and the first one selected is... The ratio of the sum of the eigenvalues ​​corresponding to each eigenvector to the sum of all eigenvalues ​​is not less than 95%.

[0034] Specifically, constructing the covariance matrix :

[0035] Solving eigenvalue problems ,in Let X be the eigenvalue matrix of the flow field matrix X, arranged in descending order; The corresponding eigenvector matrix satisfies orthogonality. ( (This is the Kronecker function).

[0036] Feature vector These are called POD modes, representing the spatial structure of the flow field. The corresponding eigenvalues... This represents the energy contribution of that mode. The original flow field matrix can be expressed as a linear combination of modes. ,in, For the coefficient matrix, the k-th column... The time coefficient for the k-th mode (which is a constant for a steady-state flow field like this in this invention) is obtained through... get.

[0037] The POD modes are arranged in descending order of energy contribution, with the first few modes typically containing the main energy of the flow field. For steady-state flow fields, this means that the dominant modes can capture the key spatial structures of the flow field (such as vortices, boundary layers, etc.). By selecting a few dominant modes, high-dimensional flow field data can be compressed into low-dimensional coefficient vectors, significantly reducing data storage and computation costs. For example, if the original data consists of 3 physical quantities with 10,000 grid nodes (30,000 dimensions), POD may only require 10 modes to reconstruct 99% of the energy, reducing the data volume to 300 dimensions. In this invention, during the flow field order reduction process, adaptive order reduction is performed to ensure that the energy proportion after order reduction is not less than 95%, i.e.:

[0038] in,m The total number of eigenvalues. n The first one selected after order reduction n Each modality represents and reconstructs the information of the entire field.

[0039] Step 1.2: Using compressor speed and outlet static pressure as input variables, the Kriging algorithm is used to estimate the flow field pressure distribution at unknown points through the weighted sum of known sample points; the weight coefficients satisfying the unbiasedness and minimum variance conditions are solved using the Lagrange multiplier method; the Gaussian radial basis function is selected as the covariance function to solve for the weight coefficients, and the training data is divided into multiple subsets using the cross-validation method. The combination of variance and scale parameters that minimizes the prediction error is selected as the optimal parameters of the Gaussian radial basis function.

[0040] Specifically, based on the reduced-order flow field data described above, the Kriging algorithm is used to train the model. Compressor speed and outlet static pressure are used as input parameters, while flow pressure, temperature, and velocity parameters are used as output parameters. As a statistical interpolation method, the Kriging algorithm fully utilizes the spatial correlation of known data points, estimating the attribute values ​​of unknown points through a weighted sum of known sample points.

[0041] in, As weighting coefficients, they must satisfy two conditions: unbiasedness and minimum variance. ① Unbiasedness condition: This ensures that the expected value of the estimated value equals the true value. ②Minimum variance condition: Solve for the weighting coefficients using the Lagrange multiplier method to minimize the estimated variance.

[0042] covariance function C ( h The magnitude of C(0) directly determines the weighting parameter for interpolation calculations of unknown points from adjacent points in the input variable space, and is key to describing the structural characteristics of the input variable space composed of compressor speed and flow field outlet pressure. By describing the correlation between the attribute values ​​(i.e., flow field variable parameters, such as pressure, temperature, and velocity) of two sample points at a distance h, the structural distribution of flow field attribute values ​​within the variable space is reflected. When h=0, C(0) is the variance of the sample point itself; as h increases, C(h) gradually decreases, reflecting the law that spatial correlation weakens with increasing distance.

[0043] The specific process of solving for the weight coefficients using the covariance function is as follows: Construct a covariance matrix K (where the elements of the matrix are the covariances between the sample points of the input variables) and introduce Lagrange multipliers. μ Establish a system of equations:

[0044]

[0045] in, λ Let C be the weight coefficient vector, C be the covariance vector between the unknown point and the sample point, and E be a unit vector of appropriate dimension (the dimension is the same as the dimension of matrix K). This equation, representing the transpose of a unit vector, is used to impose a normalization constraint on the weight coefficient vector λ, ensuring that the sum of all weight coefficients is 1. Solving the above system of equations yields the weight coefficients. λ .

[0046] The covariance function, as a core component, not only describes the correlation structure of spatial data but also provides crucial support for solving the weighting coefficients and quantifying estimation errors. Appropriate selection and construction of the covariance function is a key step in improving the accuracy of Kriging interpolation, requiring flexible adjustments based on the spatial characteristics and data distribution of the study area. In this invention, based on the linear variation characteristics of the flow field, a radial basis function is selected as the covariance function. Radial basis functions possess good locality and smoothness, and can better adapt to the complex characteristics of the compressor flow field. Specifically, this invention selects a Gaussian radial basis function as the covariance function, whose expression is:

[0047] in, and Let represent the spatial coordinate vectors of two different sample points in the input variable space of the flow field reduction model. The Euclidean distance between these two sample points. For variance, These are the scaling parameters. The appropriate values ​​of these two parameters have a significant impact on the performance of the covariance function and need to be determined through optimization methods. and In this invention, cross-validation is employed. The training data is divided into multiple subsets, with each subset serving as the validation set and the remaining subsets as the training set. This allows for the training and validation of the Kriging model under different parameter combinations. Based on the prediction error on the validation set, the parameter combination that minimizes the prediction error is selected as the optimal parameter. This method effectively avoids overfitting and improves the model's generalization ability.

[0048] This method of training a reduced-order compressor flow field model based on the Kriging algorithm and radial basis covariance function fully utilizes the spatial correlation of flow field data to improve the prediction accuracy of flow field parameters. Simultaneously, the establishment of the reduced-order model significantly reduces data storage and computation costs, making it possible to quickly obtain flow field distribution information at the compressor outlet section under different operating conditions. This provides a solid foundation for the iteration and optimization of the measurement point layout scheme in subsequent optimization processes, contributing to the scientific and systematic arrangement of measurement points and further improving the quality and reliability of compressor test data.

[0049] The optimized process development scheme is based on a trained compressor flow field reduced-order model, combined with optimization algorithms to construct a measuring point layout optimization framework. Specifically, after determining the optimization objective of this invention—minimizing the difference between the average pressure measurement value at the measuring points and the average pressure value at the outlet section—the optimization parameters are defined, namely the adjustable parameters in the measuring point layout. Within this optimization space, intelligent design of the optimization parameters is performed, and the combination of optimization parameters forms a layout scheme. Then, the optimization constraints are verified based on the layout scheme to determine whether a new round of iteration is needed. Finally, after multiple rounds of comparison, the solution that satisfies the optimization objective is found, which is the final recommended layout optimization scheme. The core lies in the configuration of the optimization space, optimization constraints, and optimization objective, as well as the optimization calculation and result display.

[0050] Step 2: Optimize the spatial design; Given the periodic distribution of the compressor outlet stator blades, the flow field is assumed to also be periodic. Therefore, only the flow field parameters of one blade channel at the compressor outlet need to be measured to obtain the outlet flow field. Since the spatial distribution of one blade channel at the test specimen outlet is relatively narrow, it is impossible to insert multiple probes simultaneously to measure its flow field. Therefore, generally 6-8 comb-shaped total temperature and total pressure composite probes are inserted into different blade channels, but the relative positions of each probe in each channel are different, which is equivalent to simulating the measurement of the outlet flow field of one channel. Figure 6 As shown, although the first and seventh probes are distributed in different blade channels, they are both located in the middle of the blade channel. The remaining probes simulate the positions from the blade channel to the rear of the blade trail. The measured data are redistributed to one channel for calculation, as shown below. Figure 7 As shown.

[0051] The test layout scheme determines the location of the probes and the arrangement of the measurement points on the probes. Generally, the location and number of measurement points are the same on different probes. Therefore, the test layout optimization problem can be abstracted as optimizing the circumferential spacing of the probes and the radial spacing of the measurement points, such as... Figure 8 As shown. The number of probes determines the optimization parameters resulting from the circumferential arrangement. dθThe number of measurement points determines the number of optimization parameters dr generated by the radial layout, denoted as the circumferential gap; the number of measurement points determines the number of optimization parameters dr generated by the radial layout, denoted as the radial gap. Therefore, the optimization space can be defined as a parameter space composed of the circumferential gap and the radial gap. Furthermore, based on the above parameters, in order to more accurately describe the distribution of measurement points, the system automatically generates... The r0 parameter represents the initial position radian and initial radial coordinate of the probe distribution.

[0052] Optimizing the spatial design involves determining the range of values ​​for the optimization parameters. If this range is unclear, the optimization process will involve taking values ​​and iterating within the range of (-∞, +∞), significantly increasing the number of iterations and wasting computational resources. Generally, the circumferential gap... dθ The value range of θ is set to [0.01Θ, 0.15Θ] (Θ is the circumferential distance of a single compressor channel). This range ensures sufficient flexibility in the circumferential arrangement of the probes while avoiding mutual interference between probes due to excessively small spacing. Simultaneously, the value range of the radial clearance dr is set to [0.01H, 0.2H] (H is the distance between the compressor casing and the hub). This setting ensures that the measuring point can cover the entire outlet section radially and achieve a reasonable density distribution based on the flow field characteristics. After determining the value range of the optimization parameters, it is necessary to further consider the interaction between these parameters. For example, changes in the circumferential and radial clearances may jointly affect the coverage effect of the measuring point on key areas of the flow field. Therefore, the synergistic effect of these two parameters needs to be comprehensively considered in subsequent optimization processes.

[0053] Step 3: Optimize constraint configuration; that is, determine the optimization parameters and optimization constraints for the arrangement of measuring points. The optimization parameters include the circumferential clearance of the probes and the radial clearance of the measuring points. The optimization constraints include that the sum of the circumferential clearances of all the probes does not exceed the circumferential angle range of a single channel of the compressor, and the sum of the radial clearances of all the measuring points does not exceed the distance between the compressor casing and the hub. In the optimization process, configuring optimization constraints is a crucial step in ensuring the rationality and effectiveness of the measurement point layout. Considering the unique characteristics of the compressor test outlet measurement point layout, optimization constraints mainly cover the following aspects: First, it is necessary to ensure that the optimized radial and circumferential clearances meet the constraints of the actual channel dimensions. Specifically, the sum of all circumferential clearances should not exceed the circumferential angle range of a single channel, and the sum of all radial clearances should not exceed the distance between the compressor casing and the hub. Second, the measurement points should be able to cover the entire outlet cross-section to avoid measurement blind spots. Therefore, the location of some measurement points may need to be restricted, i.e., limiting the range of a certain clearance segment. Finally, considering structural constraints, the initial position of the probe layout may also need to be restricted, i.e., limiting the initial position parameters. These parameters can be added as constraints, and users can flexibly adjust and add constraints according to the actual compressor structural characteristics, test requirements, and flow field characteristics. This flexible configuration method ensures that the optimization process not only meets actual engineering needs but also effectively improves the scientific and rational nature of the measurement point layout.

[0054] The above optimization constraints need to be set according to specific circumstances, and multiple constraints may be required. Therefore, the optimization constraint configuration function fully considers the need for flexible settings, allowing users to add and customize parameters, and set limiting conditions for custom parameters. Adding and customizing parameters mainly involves defining expressions based on the initial position parameters and clearance parameters automatically generated by the system, generating new parameters with certain physical meaning. Defining optimization constraints allows users to define custom parameters as constraint parameters and set the value range of these parameters. For example, in this case, constraints on the total circumferential range and radial range are set to ensure the rationality of the final optimized values ​​of the circumferential and radial clearances.

[0055] For details, please refer to Figure 8 The test layout optimization parameters shown indicate that the number of probes determines the circumferential gap. The number of measuring points determines the radial clearance. Simultaneously, the system automatically generates initial position parameters (initial position in radians). and initial radial coordinates To avoid blindly searching for the optimal solution in an infinitely large space, this embodiment defines the circumferential clearance. The range of values ​​is ( (single-channel circumferential distance), radial clearance The range of values ​​is ( (This refers to the distance from the casing to the hub). Also, see [link / reference]. Figure 9 and Figure 10 The system provides a custom parameter and constraint addition interface, allowing users to flexibly and dynamically add constraints (such as the avoidance range of a certain gap) according to the specific structure.

[0056] Step 4: Optimize the objective calculation; set the objective function as minimizing the difference between the average pressure at the measuring points and the average pressure at the cross section; wherein, the average pressure at the cross section is calculated based on the pressure of the grid nodes on the compressor outlet cross section, and the average pressure at the measuring points is calculated based on the flow field pressure distribution, using a hybrid interpolation strategy combining linear interpolation and nearest neighbor interpolation to obtain the pressure values ​​at each measuring point, and then using a circumferential radial weighted average method. In this embodiment, it should be noted that for the specific problem of optimizing the test layout of the compressor outlet section, the optimization objective is definite, which is to ensure that the difference between the average pressure of the measuring point and the average pressure of the section is minimized. Therefore, the optimization objective in this invention does not need to be defined by the user; the core lies in the calculation of the optimization objective.

[0057] The average pressure at the outlet section is calculated by summing the pressure values ​​of all grid nodes at the outlet section. The geometric prototype of the field order reduction model in this invention is as follows: Figure 11 As shown, the direction from the inlet to the outlet coincides with the positive Z-axis, and the center of the outlet section coincides with the origin of the coordinate system. Therefore, the outlet section can be represented as Z = a (where a is the Z-coordinate of any point on the outlet section). Thus, specifying the Z-coordinate during optimization settings determines the location of the outlet section. During the background calculation, nodes meeting the given coordinate values ​​are searched, thereby completing the selection of outlet section nodes. Figure 12 As shown. Obtaining the outlet section nodal pressure values ​​requires first calculating the flow field pressure information based on the given test conditions, namely rotational speed and outlet static pressure, and then extracting the pressure values ​​on the desired optimized section according to the selected nodes. The grid nodes on the section are relatively uniformly distributed and have small intervals, so the average pressure of the section is calculated using the arithmetic mean method.

[0058] The calculation of the average pressure at the cross-section measuring points requires first interpolating the pressure at the measuring points based on the cross-section mesh nodes, and then calculating the corresponding average value based on the interpolation results.

[0059] ①Measuring point pressure interpolation method The interpolation logic employs a hybrid interpolation strategy, primarily linear interpolation supplemented by nearest-neighbor interpolation. First, in N-dimensional space, for each new location point to be interpolated, a triangle containing the new point is found within the spatial range formed by the original location data. Based on the known physical quantities of all vertices of this triangle, linear weights are assigned according to the new point's relative position within the triangle (the closer to a vertex, the higher the weight), and the physical quantity of the new point is calculated through a weighted average. If the new point exceeds the convex hull range formed by the original location data (i.e., not within any triangle), linear interpolation cannot be calculated, and it is marked as an invalid value. After one round of interpolation, the points marked as invalid are identified. Nearest-neighbor interpolation is used to calculate the Euclidean distance between each invalid point and all original location points, finding the nearest original location point. The physical quantity of this nearest point is directly reused as the interpolation result for the invalid point, thus filling in all invalid values. This organic combination of linear interpolation and nearest-neighbor interpolation ensures high-precision interpolation within the effective interpolation region, while the supplementation by nearest-neighbor interpolation ensures the integrity and robustness of the entire outlet section pressure interpolation. This hybrid interpolation strategy can fully utilize the high accuracy of linear interpolation in areas with uniform data distribution, while leveraging nearest neighbor interpolation to effectively supplement data in sparse or boundary areas. This avoids the large errors that may be caused by a single interpolation method, providing a reliable data foundation for subsequent calculation of the average pressure value at cross-sectional measuring points.

[0060] ② Calculation of average pressure at measuring points Based on the characteristic that the radial distribution of measuring points on each probe is consistent, the average value of the measuring points is calculated using the circumferential radial weighted average method. That is, the radial distribution of the total pressure is obtained according to the weight of the circumferential measuring point position, and then the cross-sectional average total pressure is calculated using the radial toroidal weighted method.

[0061] Please see Figure 13 The circumferential average value of each measuring point is calculated by weighting the area occupied by the measuring point (N probes are arranged circumferentially, and M measuring points are arranged on each probe), and is calculated according to the following formula.

[0062]

[0063] in, Indicates the number of probes on all probes. The average circumferential pressure at each measuring point Indicates the circumferential position as Radial position is Pressure value at the measuring point, The radial area between two adjacent probes can be calculated using the coordinates of the measuring points. This refers to the number of probes arranged circumferentially. .

[0064] After obtaining the radial distribution of the outlet pressure, when considering the boundary layer correction, the wall simulation measuring points can be constructed according to the boundary layer correction factor of 0.98, that is, the casing and further divided into annular surfaces according to the reconstructed measuring point distribution. The area and characterization measurement value of each annular surface are calculated respectively, and the average total pressure of the outlet section is calculated according to the following formula.

[0065]

[0066] in, This refers to the number of measuring points arranged on a single probe. For the first The radial coordinate values ​​of each measuring point.

[0067] The key to optimizing the target lies in comparing the difference between the average pressure at the measuring points and the average pressure at the cross-section. This difference can be quantified by calculating the absolute or relative error between the two. During the optimization process, the algorithm continuously adjusts the arrangement of the measuring points to gradually reduce this error until the preset optimization target is reached or the convergence condition is met. In this way, it can be ensured that the final measuring point arrangement makes the average pressure at the measuring points as close as possible to the average pressure at the cross-section, thus more accurately representing the flow field characteristics at the outlet cross-section.

[0068] Step 5: Using an intelligent optimization algorithm, the optimization parameters are iteratively updated under the premise of satisfying the optimization constraints. In each iteration, the objective function value under the current measurement point layout scheme is calculated until the preset convergence condition is met or the maximum number of iterations is reached, and the optimal measurement point layout scheme is output.

[0069] In this embodiment, the specific optimization calculation process is as follows: Figure 14As shown in the diagram. In the optimization calculation process, the key information, such as the previously determined optimization space, optimization constraints, and optimization objective calculation method, must first be input into the optimization algorithm module. This module automatically initiates the iterative calculation process based on the input parameter range, constraints, and objective function. In each iteration, the optimization algorithm calculates the objective function value based on the current parameter combination (i.e., the current measuring point layout scheme), which is to calculate the difference between the average pressure at the measuring points and the average pressure across the cross-section under the current measuring point layout. Subsequently, the optimization algorithm generates a new parameter combination based on the calculated objective function value and a preset optimization strategy (such as gradient descent, genetic algorithm, etc.), thus updating and adjusting the measuring point layout scheme. This process repeats continuously; each iteration generates a new measuring point layout scheme and calculates the corresponding objective function value, comparing it with the result of the previous iteration. As the iteration progresses, the optimization calculation process stops when the change in the objective function value is less than a preset convergence threshold or when the preset maximum number of iterations is reached. At this point, the obtained measuring point layout scheme is the solution that satisfies the optimization objective, which is the final recommended layout optimization scheme. Finally, the final measurement point layout scheme obtained from the optimization calculation is presented to the user in an intuitive way. For example, the installation position of the probe and the distribution of the measurement points can be displayed through a graphical interface. At the same time, relevant data reports can be provided, such as the pressure value of each measurement point, the average pressure of the measurement point, the average pressure of the cross section, and the error between them, so that the user can comprehensively evaluate and analyze the optimization results.

[0070] The optimization results are circumferential and radial clearance values. To facilitate display and further application by users, the optimization results will be presented in a visual format, illustrating the distribution of measurement points, for example... Figure 15 As shown in the diagram. Simultaneously, a comparison of the measurement point layout before and after optimization will be displayed. "Before optimization" refers to the distribution of all grid nodes on the cross-section without measurement point layout. Through intuitive graphical comparison, users can quickly determine the optimization effect and clearly understand the improvements in the rationality of the measurement point distribution and the ability to cover key areas of the flow field after optimization. In addition to the visualization, the final circumferential and radial clearance values ​​will also be numerically displayed. These values ​​will be presented in a clear table format, detailing the specific values ​​of the optimized circumferential and radial clearances, and the results will be appropriately transformed into coordinates to facilitate user application of the results according to actual conditions.

[0071] The method proposed in this invention has significant advantages, mainly reflected in the following aspects: First, by optimizing the measurement point layout, the accuracy and reliability of the compressor test data were significantly improved. The optimized measurement point distribution can more comprehensively cover the key areas of the flow field and reduce measurement blind spots, thus providing more accurate raw data support for the reconstruction of the flow field at the outlet section.

[0072] Secondly, the field order reduction model training scheme adopted in this invention effectively reduces computational complexity and improves data processing efficiency. Through order reduction, key features of the flow field data are extracted, making model training more efficient while preserving the main features of the flow field information, providing strong support for subsequent optimization calculations.

[0073] In addition, the optimized process development solution is highly flexible and scalable, and can be customized according to different compressor models and test requirements to meet the optimization needs of measurement point layout in different scenarios.

[0074] Finally, the implementation of this invention helps to improve the overall level of compressor testing, providing more accurate and reliable test data for compressor design, optimization and performance evaluation, and promoting the continuous progress and development of compressor technology.

[0075] Example 2 The compressor test measurement point layout optimization method developed based on Example 1, using the previously trained compressor flow field reduced-order model as a foundation, can complete the test layout optimization under specified operating conditions. The specific steps are as follows: First, using the physical field order reduction model training function, the compressor single-channel flow field simulation data is used as training data to extract effective pressure field information and train the compressor flow field order reduction model. This type of model supports the prediction of flow field information. Then, the field model used for training is selected as the base model for optimization, the operating conditions of the compressor are set, the model calculation is executed, and flow field data is generated. Next, set the location of the exit section to be optimized and the location of the origin of the coordinate system in the long model, and specify the number of probes to be arranged and the number of measuring points on a single probe. Next, the optimization space and constraints are set to limit the optimization range; Finally, the optimization calculation is initiated. Based on the preset optimization algorithm and objective function, the system will automatically perform iterative calculations, continuously adjusting the arrangement of the measuring points. When the optimization calculation reaches the preset convergence condition or the maximum number of iterations, the system will automatically stop the calculation and output the final optimization results. These results include the optimized circumferential and radial clearance values, the specific distribution locations of the measuring points, and comparison graphs before and after optimization. Users can intuitively evaluate the optimization effect based on these results and understand the advantages of the optimized measuring point arrangement in improving measurement accuracy and covering key areas of the flow field.

[0076] Example 3 Based on the same technical concept, embodiments of the present invention provide an electronic device that can implement the method flow for optimizing the layout of compressor test outlet measuring points provided in the above embodiments of the present invention. In one embodiment, the electronic device can be a server, a terminal device, or other electronic equipment. Figure 16 As shown, the electronic device may include: At least one processor and a memory connected to the at least one processor. In this embodiment of the invention, the specific connection medium between the processor and the memory is not limited. Figure 16 The example used is the connection between the processor and memory via a bus. The bus... Figure 16 The connections between other components are indicated by thick lines and are for illustrative purposes only, not as limiting information. Buses can be categorized into address buses, data buses, control buses, etc., but for ease of representation, [the specific bus type is not shown here]. Figure 16 The processor is represented by a single thick line, but this does not mean that there is only one bus or one type of bus. Alternatively, the processor can also be called a controller; there is no restriction on the name.

[0077] In this embodiment of the invention, the memory stores instructions executable by at least one processor. By executing the instructions stored in the memory, the at least one processor can perform the aforementioned method for optimizing the arrangement of compressor test outlet measuring points. The processor can implement... Figure 16 The functions of each module in the device shown.

[0078] The processor is the control center of the device. It can connect to various parts of the control equipment through various interfaces and lines. By running or executing instructions stored in memory and calling data stored in memory, it can monitor the device's various functions and process data, thereby enabling overall monitoring of the device.

[0079] In an alternative design, the processor may include one or more processing units. The processor may integrate an application processor and a modem processor, wherein the application processor primarily handles the operating system, user interface, and applications, while the modem processor primarily handles wireless communication. It is understood that the modem processor may also not be integrated into the processor. In some embodiments, the processor and memory may be implemented on the same chip; in some embodiments, they may also be implemented separately on separate chips.

[0080] The processor can be a general-purpose processor, such as a CPU, digital signal processor, application-specific integrated circuit, field-programmable gate array or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method for optimizing the layout of compressor test outlet measuring points disclosed in the embodiments of this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0081] Memory, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules. Memory can include at least one type of storage medium, such as flash memory, hard disk, multimedia cards, card-type memory, random access memory (RAM), static random access memory (SRAM), programmable read-only memory (PROM), read-only memory (ROM), and electrically erasable programmable read-only memory (EPROM). Only memory (EEPROM), magnetic storage, magnetic disks, optical disks, etc. A memory is any other medium capable of carrying or storing desired program code in the form of instructions or data structures, and accessible by a computer, but is not limited thereto. The memory in embodiments of this invention can also be a circuit or any other device capable of performing storage functions for storing program instructions and / or data.

[0082] By designing and programming the processor, the code corresponding to the method for optimizing the layout of compressor test outlet measuring points described in the foregoing embodiments can be embedded into the chip, enabling the chip to execute the steps of the method described in the foregoing embodiments during operation. How to design and program the processor is a technique well-known to those skilled in the art and will not be elaborated upon here.

[0083] Based on the same inventive concept, embodiments of the present invention also provide a storage medium storing computer instructions, which, when executed on a computer, cause the computer to execute the aforementioned method for optimizing the arrangement of compressor test outlet measuring points.

[0084] In some alternative embodiments, various aspects of the method for optimizing the layout of compressor test outlet measuring points provided by the present invention can also be implemented in the form of a program product, which includes program code. When the program product is run on the device, the program code is used to cause the control device to perform the steps in the method for optimizing the layout of compressor test outlet measuring points according to various exemplary embodiments of the present invention as described above.

[0085] It should be noted that although several units or sub-units of the apparatus have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of the invention, the features and functions of two or more units described above can be embodied in one unit. Conversely, the features and functions of one unit described above can be further divided and embodied by multiple units. Furthermore, although the operation of the method of the invention is described in a specific order in the drawings, this does not require or imply that these operations must be performed in that specific order, or that all the operations shown must be performed to achieve the desired result. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0086] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can be implemented in one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs) containing computer-usable program code. The form of a computer program product implemented on ROM, optical memory, etc.

[0087] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a server, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0088] Program code for performing the operations of this invention can be written using any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0089] In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0090] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0091] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0092] In addition, in some embodiments, a computer program product is also proposed, which, when executed by a processor, implements the above-described method for optimizing the layout of compressor test outlet measuring points.

[0093] The embodiments described above merely illustrate specific implementation methods of this application, and while the descriptions are detailed and specific, they should not be construed as limiting the scope of protection of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the technical solution of this application, and these modifications and improvements all fall within the scope of protection of this application.

[0094] This background section is provided to generally present the context of the invention. The work of the currently named inventors, the work to the extent described in this background section, and aspects of this section that did not constitute prior art at the time of application are neither expressly nor impliedly acknowledged as prior art to the invention.

Claims

1. A method for optimizing the layout of measuring points at the outlet of a compressor test, characterized in that, include: A reduced-order model of the compressor flow field is obtained. The reduced-order model of the compressor flow field is based on the single-channel flow field simulation data of the compressor under different operating conditions. The flow field feature vectors are extracted by the intrinsic orthogonal decomposition (POD) method for order reduction and then trained by the Kriging algorithm. The compressor flow field reduced-order model is used to predict and output the flow field pressure distribution at the compressor outlet section based on the input compressor speed and outlet static pressure. Determine the optimization parameters and constraints for the arrangement of measuring points. The optimization parameters include the circumferential clearance of the probes and the radial clearance of the measuring points. The optimization constraints include that the sum of the circumferential clearances of all the probes does not exceed the circumferential angle range of a single channel of the compressor, and the sum of the radial clearances of all the measuring points does not exceed the distance between the compressor casing and the hub. The objective function is set to minimize the difference between the average pressure at the measuring points and the average pressure at the cross section. The average pressure at the cross section is calculated based on the pressure of the grid nodes on the compressor outlet cross section. The average pressure at the measuring points is calculated based on the flow field pressure distribution, using a hybrid interpolation strategy that combines linear interpolation and nearest neighbor interpolation to obtain the pressure values ​​at each measuring point, and then using a circumferential and radial weighted average method. Using an intelligent optimization algorithm, the optimization parameters are iteratively updated under the premise of satisfying the optimization constraints. In each iteration, the objective function value under the current measurement point layout scheme is calculated until the preset convergence condition is met or the maximum number of iterations is reached, and the optimal measurement point layout scheme is output.

2. The method for optimizing the layout of compressor test outlet measuring points according to claim 1, characterized in that, The process of reducing the order of flow field feature vectors extracted using the intrinsic orthogonal decomposition (POD) method includes: The single-channel flow field simulation data of the compressor under different operating conditions were used to construct a flow field matrix; Construct the covariance matrix of the flow field matrix, and solve for the eigenvalues ​​and eigenvectors of the covariance matrix. The eigenvectors represent the spatial structure of the flow field, and the eigenvalues ​​represent the energy contribution of the corresponding eigenvectors. Arrange the features in descending order of their eigenvalues, and select the top... The original flow field matrix is ​​projected onto the reduced-order subspace as an optimal orthogonal basis to complete the order reduction process. in, It is a positive integer, and the first selected... The ratio of the sum of the eigenvalues ​​corresponding to each eigenvector to the sum of all eigenvalues ​​is not less than 95%.

3. The method for optimizing the layout of compressor test outlet measuring points according to claim 2, characterized in that, The compressor flow field reduction model obtained by training with the Kriging algorithm includes: Using compressor speed and outlet static pressure as input variables, the Kriging algorithm is used to estimate the flow field pressure distribution at unknown points by weighted sum of known sample points. The weighting coefficients that satisfy the unbiasedness condition and the minimum variance condition are solved by the Lagrange multiplier method. The weight coefficients are solved by using a Gaussian radial basis function as the covariance function. The training data is divided into multiple subsets by cross-validation. The combination of variance and scale parameters that minimizes the prediction error is selected as the optimal parameters of the Gaussian radial basis function.

4. The method for optimizing the layout of compressor test outlet measuring points according to claim 1, characterized in that, Before acquiring the pressure values ​​at each measuring point, a single-channel mapping step for the probe measuring point data is also included: Since the compressor outlet flow field is periodically distributed, the flow field data measured by multiple probes inserted in different blade channels are redistributed and mapped to the same simulated single channel for centralized calculation based on the relative position of each probe in each channel; the probe has multiple measuring points distributed radially.

5. The method for optimizing the layout of compressor test outlet measuring points according to claim 1, characterized in that, The method of obtaining pressure values ​​at each measuring point using a hybrid interpolation strategy combining linear interpolation and nearest neighbor interpolation includes: For each measurement point to be interpolated, within the spatial range of the original position data formed by the grid nodes on the compressor outlet section, a triangle containing the measurement point is found; based on the known pressure values ​​of the vertices of the triangle, linear weights are assigned according to the relative position of the measurement point within the triangle, and the pressure value of the measurement point is calculated by weighted average. If the measuring point exceeds the convex hull range formed by the original position data, causing linear interpolation to fail and be marked as invalid, then the Euclidean distance between the measuring point marked as invalid and all original position points is calculated, and the pressure value of the nearest original position point is selected as the interpolation result for that measuring point.

6. The method for optimizing the layout of compressor test outlet measuring points according to claim 1, characterized in that, The average pressure at each measuring point is calculated using a circumferential radial weighted average method. First, the average circumferential pressure at each measuring point is calculated by weighting the area occupied by the measuring point. The calculation formula is as follows: in, Indicates the number of probes on all probes. The average circumferential pressure at each measuring point Indicates the circumferential position as Radial position is Pressure value at the measuring point, The radial area between two adjacent probes. This refers to the number of probes arranged circumferentially. .

7. The method for optimizing the layout of compressor test outlet measuring points according to claim 6, characterized in that, After calculating the average circumferential pressure at each measuring point, a wall-simulated measuring point is constructed using the boundary layer correction factor. A radial annulus is then defined, and the average pressure at the measuring point of the outlet section is calculated using the radial annulus weighted method. The calculation formula is as follows: in, This refers to the number of measuring points arranged on a single probe. For the first The radial coordinate values ​​of each measuring point.

8. An electronic device, characterized in that, include: At least one processor; and a memory communicatively connected to the at least one processor; The memory stores instructions executable by the at least one processor, which executes the instructions stored in the memory to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store instructions that, when executed, cause the method as described in any one of claims 1-7 to be implemented.

10. A computer program product, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1-7.

Images