Flow field reconstruction method based on test point position optimization and multi-source data fusion and related device

Abstract

The invention belongs to the field of turbine tests, and discloses a flow field reconstruction method based on test measuring point position optimization and multi-source data fusion and a related device, and the method comprises the steps: building a CFD data set and a sparse measuring point test data set; carrying out POD decomposition on the CFD data set; according to the sparse measurement point test data set, a correction modal coefficient is obtained by using a Gappy POD algorithm; according to the POD mode and the correction mode coefficient, obtaining a Gappy POD reconstruction flow field parameter; taking the minimum error of the global low-fidelity flow field parameter and the Gappy POD reconstruction flow field parameter as a target to obtain an optimal measurement point position; based on the optimal measurement point position, high-fidelity discrete test data is obtained, and an optimized modal coefficient is obtained by using a Gappy POD algorithm; obtaining a flow field reconstruction result of the turbine blade according to the POD mode and the optimization mode coefficient; according to the method, fine reconstruction of the global flow field distribution of the turbine blade is realized.

Description

Technical Field

[0001] The present invention belongs to the field of turbine test technology, relates to the field of CFD multi-source data flow field reconstruction, and particularly relates to a flow field reconstruction method based on test measurement point position optimization and multi-source data fusion and related devices. Background Art

[0002] As a core component of an aeroengine, the turbine's operating performance directly impacts the system's overall efficiency, stability, and operating life. Therefore, predicting and analyzing the turbine's detailed flow field is of great significance. However, the complex three-dimensional flow within the turbine, characterized by eddies, turbulence, and aerodynamic separation, poses challenges to existing experimental measurement methods and flow field assessment techniques, such as CFD (Computational Fluid Dynamics). Traditionally, CFD methods can easily obtain global flow field information for complex turbine structures. However, due to limitations in computational resources and model simplification, simulation accuracy is often limited by computational resources and model simplification, making it difficult to fully and faithfully reflect the complex flow characteristics within the turbine. Wind tunnel-based experimental measurement methods, on the other hand, can realistically capture flow phenomena within the turbine, but are limited by the accuracy of the test equipment and the layout of measurement points. Furthermore, the test data contain varying degrees of noise, resulting in spatial distributions that do not fully reflect the global flow field.

[0003] At present, in the existing turbine flow field research, the experimental / CFD multi-source data are often in a shallow communication situation of mutual verification, which makes it impossible to explore the potential correlation characteristics between the two, and thus cannot achieve the refined reconstruction of the internal flow field of the turbine; at the same time, due to the strong potential correlation between the experimental data and the CFD data and the mismatch characteristics of the two in spatial distribution, it is difficult to use simple methods to fuse the flow field information such as velocity field and pressure field represented by high-dimensional data structures such as matrices and tensors; therefore, how to achieve the complementary advantages between different sources of data such as experimental / CFD to quickly obtain a refined turbine flow field with higher spatial resolution and higher data fidelity has become an urgent problem to be solved in the current turbine flow field measurement. Summary of the Invention

[0004] In response to the technical problems existing in the prior art, the present invention provides a flow field reconstruction method and related devices based on test measurement point position optimization and multi-source data fusion, so as to solve the technical problem that in the existing turbine flow field research, the test / CFD multi-source data are often in a shallow communication situation of mutual verification, and the potential correlation characteristics between the two cannot be explored, and thus the technical problem of being unable to achieve refined reconstruction of the internal flow field of the turbine.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is: The present invention provides a flow field reconstruction method based on the optimization of test measurement point positions and multi-source data fusion, including: Sampling the design space variables of the turbine blade to be studied to obtain a sampling data sample; Performing CFD simulation calculations on the sampling data sample to obtain global low-fidelity flow field parameters; based on the global low-fidelity flow field parameters, establishing a CFD data set; collecting test data under the wind tunnel test of the turbine blade to be studied based on the pre-determined uniform measurement point positions to obtain a sparse measurement point test data set; Performing POD decomposition on the CFD data set to obtain POD modes and modal coefficients; According to the sparse measurement point test data set, using the Gappy POD algorithm to correct the modal coefficients to obtain corrected modal coefficients; According to the POD modes and the corrected modal coefficients, obtaining Gappy POD reconstructed flow field parameters; Taking the measurement point positions as design variables, and aiming at minimizing the error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters to optimize the design variables and obtain the optimal measurement point positions; Based on the optimal measurement point positions, collecting test data under the wind tunnel test of the turbine blade to be studied to obtain high-fidelity discrete test data; Based on the high-fidelity discrete test data, using the Gappy POD algorithm to correct the modal coefficients to obtain optimized modal coefficients; According to the POD modes and the optimized modal coefficients, obtaining the flow field reconstruction result of the turbine blade to be studied.

[0006] Furthermore, the design space variables of the turbine blade to be studied include the exit isentropic Mach number, inlet attack angle, inlet turbulence intensity, and relative axial pitch of the turbine blade to be studied.

[0007] Furthermore, the process of collecting test data under the wind tunnel test of the turbine blade to be studied based on the pre-determined uniform measurement point positions to obtain a sparse measurement point test data set includes: Determining the wind tunnel test conditions of the turbine blade to be studied under the initial sampling; wherein, the wind tunnel test conditions of the turbine blade to be studied under the initial sampling include the pre-determined uniform measurement point positions; Based on the wind tunnel test conditions of the turbine blade to be studied under the initial sampling, collecting test data under the wind tunnel test of the turbine blade to be studied to obtain a sparse measurement point test data set.

[0008] Furthermore, the process of obtaining corrected modal coefficients according to the sparse measurement point test data set and combining the POD modes using the Gappy POD algorithm includes: Based on the sparse measurement point test data set and the POD modes, construct an optimization formula for modal coefficients; Use the least squares method to solve the optimization formula for modal coefficients to obtain corrected modal coefficients.

[0009] Further, the optimization formula for modal coefficients is specifically:

[0010]

[0011] where, is the corrected modal coefficient; is the discrete test measurement point data; is the modal coefficient; is the POD mode; is the modal order; is the th test data of the measurement point; is the total number of measurement points.

[0012] Further, the process of optimizing the design variables with the measurement point position as the design variable and the minimum error between the global low-fidelity flow field parameters and the GappyPOD reconstructed flow field parameters as the objective includes: Determine the number of measurement point positions; According to the number of measurement point positions, determine the initial value of the measurement point position; wherein, the initial value of the measurement point position is several equally spaced measurement point positions from the leading edge to the trailing edge of the turbine blade to be studied; According to the initial value of the measurement point position, with the minimum error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters as the objective and the measurement point position as the design variable, establish an objective function; Based on the global optimization algorithm, optimize the objective function to obtain the optimal measurement point position.

[0013] The present invention also provides a flow field reconstruction system based on the optimization of test measurement point positions and the fusion of multi-source data, including: A data sampling module, configured to sample the design space variables of the turbine blade to be studied to obtain a sampling data sample; A data set establishment module, configured to perform CFD simulation calculations on the sampling data sample to obtain global low-fidelity flow field parameters; based on the global low-fidelity flow field parameters, establish a CFD data set; based on the pre-determined uniform measurement point positions, collect the test data of the turbine blade to be studied under wind tunnel tests to obtain a sparse measurement point test data set; The POD flow field feature extraction module is used to perform POD decomposition on the CFD dataset to obtain POD modes and modal coefficients; The initial sampling coefficient correction module is used to correct the modal coefficients according to the sparse measurement point test dataset by using the Gappy POD algorithm to obtain corrected modal coefficients; The initial sampling flow field reconstruction module is used to obtain GappyPOD reconstructed flow field parameters according to the POD modes and the corrected modal coefficients; The measurement point optimization module is used to optimize the design variables with the measurement point position as the design variable and the minimum error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters as the goal to obtain the optimal measurement point position; The test module is used to collect test data under the wind tunnel test of the turbine blade to be studied based on the optimal measurement point position to obtain high-fidelity discrete test data; The optimized sampling coefficient correction module is used to correct the modal coefficients by using the Gappy POD algorithm based on the high-fidelity discrete test data to obtain optimized modal coefficients; The optimized sampling flow field reconstruction module is used to obtain the flow field reconstruction result of the turbine blade to be studied according to the POD modes and the optimized modal coefficients.

[0014] The present invention also provides a flow field reconstruction device based on test measurement point position optimization and multi-source data fusion, including: A processor suitable for executing a computer program; A computer-readable storage medium, in which a computer program is stored, and when the computer program is executed by the processor, it executes the flow field reconstruction method based on test measurement point position optimization and multi-source data fusion according to any one of claims 1-6.

[0015] The present invention also provides a computer-readable storage medium, in which a computer program is stored, and when the computer program is executed by a processor, it implements the flow field reconstruction method based on test measurement point position optimization and multi-source data fusion.

[0016] The present invention also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the flow field reconstruction method based on test measurement point position optimization and multi-source data fusion.

[0017] Compared with the prior art, the beneficial effects of the present invention are: The flow field reconstruction method based on the optimization of test measurement point positions and multi-source data fusion provided by the present invention efficiently couples multi-source data of test data and CFD data based on the Gappy POD algorithm and introduces an optimization strategy for test measurement point positions to achieve refined reconstruction of the global flow field distribution of turbine blades, thereby achieving refined acquisition of global flow field information under test conditions. Specifically, through the POD decomposition method, flow field feature extraction is performed on the global low-fidelity flow field parameters in the CFD dataset, and key modes in the full-space physical field can be efficiently extracted. Secondly, the Gappy POD algorithm is used, and combined with the sparse measurement point test dataset, the modal coefficients are corrected. The corrected modal coefficients are combined with the POD modes to obtain the Gappy POD reconstructed flow field parameters, which can effectively explore the close connection existing between different source data of test data and CFD data, enabling the complementary advantages of multi-source data, and further obtaining a reconstructed flow field with high precision and high spatial resolution, laying a solid foundation for the accurate analysis of turbine aerodynamic performance. Secondly, with the goal of minimizing the error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters, the optimal spatial distribution scheme of test measurement points is determined, thereby achieving refined reconstruction of the flow field at a lower sampling cost. Further, the sparse measurement point layout optimization strategy is combined with the multi-source data fusion method, and by optimizing the test measurement point layout, higher-fidelity flow field features can be captured with fewer test measurement points.

[0018] The flow field reconstruction system, flow field reconstruction device, computer-readable storage medium, and computer program product based on the optimization of test measurement point positions and multi-source data fusion provided by the present invention have all the advantages of the above-mentioned flow field reconstruction method based on the optimization of test measurement point positions and multi-source data fusion. Brief Description of the Drawings

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for description in the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0020] Figure 1 It is a flowchart of the flow field reconstruction method based on the optimization of test measurement point positions and multi-source data fusion provided for Embodiment 1; Figure 2 It is a schematic diagram of the principle of multi-source data fusion for turbine blade flow field reconstruction in Embodiment 1; Figure 3 It is a schematic diagram of the layout positions before and after the optimization of the measurement point positions in Embodiment 1; Figure 4Schematic diagram of flow field reconstruction results after optimizing the measuring point positions in Example 1; wherein, Figure 4 a is a schematic diagram of the flow field reconstruction results at 10 measuring points. Figure 4 b is a schematic diagram of the flow field reconstruction results at 15 measuring points; Figure 5 A structural block diagram of a flow field reconstruction system based on test point position optimization and multi-source data fusion provided in Example 2; Figure 6 This is a structural block diagram of the flow field reconstruction device based on test measurement point position optimization and multi-source data fusion provided in Example 3. DETAILED DESCRIPTION

[0021] In order to make the technical problems, technical solutions, and beneficial effects solved by this application more clearly understood, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application; it is obvious that the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of this application.

[0022] Example 1 As attached Figure 1 As shown, this embodiment 1 provides a flow field reconstruction method based on test measurement point position optimization and multi-source data fusion, including the following steps: Step 1: Sample the design space variables of the turbine blade to be studied to obtain sampling data samples.

[0023] Specifically, the process of sampling the design space variables of the turbine blade to be studied and obtaining the sampled data samples includes: Step 1-1, determine the design space variables of the turbine blade to be studied, and determine the numerical variation range of each design space variable; wherein the design space variables of the turbine blade to be studied include the outlet isentropic Mach number, inlet angle of attack, inlet turbulence and relative axial pitch of the turbine blade to be studied.

[0024] Step 1-2: Use the Latin hypercube sampling method to sample the design space variables of the turbine blade to obtain sampling data samples.

[0025] Step 2: Perform CFD (Computational Fluid Dynamics) simulation calculations on the sampled data to obtain global low-fidelity flow field parameters; establish a CFD data set based on the global low-fidelity flow field parameters; and collect test data from a wind tunnel test of the turbine blade to be studied based on predetermined uniform measurement point positions to obtain a sparse measurement point test data set.

[0026] Among them, based on the pre-determined uniform measurement point positions, the process of collecting the test data under the wind tunnel test of the turbine blade to be studied and obtaining the sparse measurement point test data set is as follows: Determine the wind tunnel test conditions of the turbine blade to be studied under the initial sampling; among them, the wind tunnel test conditions of the turbine blade to be studied under the initial sampling include the pre-determined uniform measurement point positions; conduct the wind tunnel test on the turbine blade to be studied; then, according to the pre-determined uniform measurement point positions, collect the test data under the wind tunnel test of the turbine blade to be studied to obtain the sparse measurement point test data set.

[0027] Step 3: Perform POD decomposition (Proper Orthogonal Decomposition) on the CFD data set to obtain POD modes and modal coefficients. Specifically, performing POD decomposition on the CFD data set is to extract the POD flow field characteristics from the CFD data set to obtain POD modes and modal coefficients. The process is as follows: Use the POD reduced-order model, that is, the proper orthogonal decomposition method, to extract the characteristics of the CFD data set to decompose the full-space physical field contained in the CFD global flow field data into a set of orthogonal basis vectors .

[0028] Based on the singular value method, decompose the orthogonal basis vectors to complete the POD decomposition operation; among them, solve the singular values of the elements in the orthogonal basis vectors , truncate the first orthogonal vectors whose sum of the solved energy values is greater than 99.99%, and use the first orthogonal vectors to approximately express the flow field vectors in the space, complete the extraction of the global flow field characteristics, and obtain the POD modes and modal coefficients.

[0029] Step 4: According to the sparse measurement point test data set, use the Gappy POD algorithm to correct the modal coefficients to obtain the corrected modal coefficients. Among them, the process of obtaining the corrected modal coefficients includes: Step 4-1: Based on the sparse measurement point test data set and the POD modes, construct an optimization formula for the modal coefficients.

[0030] Step 4-2: Use the least squares method to solve the optimization formula for the modal coefficients to obtain the corrected modal coefficients.

[0031] Specifically, in this Embodiment 1, based on the flow field characteristics and the sparse flow field point data in the sparse measurement point test data set, the relationship between the modal sparsity and the sparse flow field point data is established according to the regression model to correct the low-fidelity modal coefficients, as shown in the appendixFigure 2 As shown; specifically, the process of correcting the modal coefficients of low-fidelity is as follows The spatial flow field obtained by POD decomposition, that is, the POD mode can be approximately expressed as a set of orthogonal basis vectors of high spatial resolution features The linear summation of, that is ; where is the spatial flow field, that is, the POD mode; is the modal coefficient.

[0032] Define the sparse measurement point test data set as a vector group ; where is the test data of the th measurement point; is the total number of measurement points.

[0033] Based on the Gappy POD algorithm, the process of correcting the modal coefficients of low-fidelity can be transformed into an optimization problem, that is, constructing a modal coefficient optimization formula; where, the modal coefficient optimization formula is specifically:

[0034] where is the corrected modal coefficient; is the discrete test measurement point data; is the modal coefficient; is the POD mode; is the modal order; is the th measurement point test data; is the total number of measurement points.

[0035] Finally, use the least squares method to solve the modal coefficient optimization formula to obtain the corrected modal coefficient.

[0036] Step 5: Obtain the Gappy POD reconstructed flow field parameters according to the POD mode and the corrected modal coefficient. Specifically, combine the orthogonal basis obtained by CFD simulation calculation, that is, the global low-fidelity flow field parameters, with the corrected modal coefficient to obtain the reconstructed flow field data with high precision and high spatial resolution for the whole field, that is, obtain the Gappy POD reconstructed flow field parameters.

[0037] Step 6: Take the measurement point position as the design variable, and optimize the design variable with the goal of minimizing the error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters to obtain the optimal measurement point position. Among them, take the measurement point position as the design variable, take the minimum of the flow field reconstruction error obtained by the reduced-order model and the regression model as the objective function, and combine the global optimization algorithm to carry out the optimization of the sparse test measurement point layout to obtain the optimal side point position.

[0038] Specifically, the process of obtaining the optimal measurement point positions includes: Step 6-1: Select the measurement point positions for the test as design variables and determine the number of measurement point positions.

[0039] Step 6-2: Determine the initial values of the measurement point positions according to the number of measurement point positions; wherein, the initial values of the measurement point positions are several equally spaced measurement point positions from the leading edge to the trailing edge of the turbine blade to be studied; specifically, after determining the number of measurement point positions, select several equally spaced measurement point positions from the leading edge to the trailing edge of the turbine blade to be studied as the initial values to obtain the initial values of the measurement point positions.

[0040] Step 6-3: Based on the initial values of the measurement point positions, with the goal of minimizing the error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters, and taking the measurement point positions as design variables, establish an objective function.

[0041] Step 6-4: Based on the global optimization algorithm, conduct a rapid search on the objective function to obtain the optimal measurement point positions.

[0042] Step 7: Based on the optimal measurement point positions, collect the test data under the wind tunnel test of the turbine blade to be studied to obtain high-fidelity discrete test data.

[0043] Step 8: Based on the high-fidelity discrete test data, use the Gappy POD algorithm to correct the modal coefficients to obtain optimized modal coefficients. It should be noted that the process of using the Gappy POD algorithm to correct the modal coefficients based on the high-fidelity discrete test data to obtain optimized modal coefficients is basically the same as the process of obtaining the corrected modal coefficients in Step 4 above, and will not be elaborated here.

[0044] Step 9: According to the POD modes and the optimized modal coefficients, obtain the flow field reconstruction result of the turbine blade to be studied.

[0045] In this Embodiment 1, based on the Gappy POD method, the efficient coupling of experimental data and CFD multi-source data is realized, and the refined reconstruction of the global flow field distribution of the turbine is completed, so as to obtain the global flow field information under experimental conditions more precisely. Specifically, the POD decomposition is performed on the CFD data set with global low-fidelity flow field parameters to extract the flow field features and realize the efficient extraction of the key modes in the full-space physical field. At the same time, combined with the Gappy POD method, the correction mode coefficients under experimental conditions are obtained by using the discrete high-fidelity experimental measurement point data. Subsequently, the refined flow field distribution is obtained by combining the correction mode coefficients with the CFD flow field features. The present invention can effectively explore the close connection existing between different source data of experimental data and CFD data, make up for each other's advantages among multi-source data, and then obtain a reconstructed flow field with high precision and high spatial resolution, laying a solid foundation for the accurate analysis of the turbine aerodynamic performance and having important engineering practical value.

[0046] In this Embodiment 1, the measurement point position is optimized with the goal of minimizing the error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters to obtain the optimal measurement point position, so as to further combine the sparse measurement point layout optimization strategy with the multi-source data fusion method and capture higher-precision flow field features with fewer experimental measurement points by optimizing the experimental measurement point layout. Specifically, based on the global optimization algorithm and combined with the Gappy POD method, an objective function is constructed, and the goal is to minimize the global error of the reconstructed flow field after multi-source data fusion. Under this framework, the optimal spatial distribution scheme of the experimental measurement points is determined by the optimization algorithm, so as to realize the refined reconstruction of the flow field at a lower sampling cost.

[0047] The flow field reconstruction method based on the optimization of experimental measurement point positions and multi-source data fusion described in this Embodiment 1 effectively reduces the time cost and economic cost required in the process of arranging experimental measurement points, and significantly improves the reliability and spatial resolution of the reconstructed flow field. In addition, while ensuring high-fidelity reconstruction, it provides theoretical support and technical guarantee for the efficient analysis of complex turbine aerodynamic flow fields and has important engineering practical value.

[0048] Example illustration: Taking the second-stage rotor blade of a certain four-stage low-pressure turbine of PW as an example, the flow field reconstruction method based on the optimization of experimental measurement point positions and multi-source data fusion described above is explained as follows: (1) Determine the research object and working condition parameters and perform sampling Select the exit isentropic Mach number, inlet angle of attack, and inlet turbulence intensity of the second-stage rotor blade of a certain PW four-stage low-pressure turbine as the design space variables; select the corresponding change ranges for each design space variable within the design space; among them, the change range of the exit isentropic Mach number is 0.7 - 1.2, the change range of the angle of attack is -10° - 5°, and the change range of the inlet turbulence intensity is 0.01 - 0.1; use the Latin hypercube sampling method to collect 200 groups of data samples within the design space as the sampling data samples.

[0049] (2) Establish a CFD dataset and a sparse measurement point test dataset Based on the sampling data samples, use CFX (a commercial software for computational fluid dynamics analysis) for simulation to obtain the full-space low-fidelity flow field parameters of the relevant samples, that is, the global low-fidelity flow field parameters; based on the global low-fidelity flow field parameters, establish a CFD dataset.

[0050] Randomly select some samples from the CFD dataset, and add random noise to simulate and generate test data, and use the simulated and generated test data to construct a sparse measurement point test dataset; among them, the noise amplitude is 1% and 5% of the original data of the selected part of the samples; it should be noted that the measurement point positions are equally spaced from the leading edge to the trailing edge, and the number of measurement points is set to 20; after processing, high-fidelity measurement point data simulating real tests is obtained, and a sparse measurement point test dataset is established.

[0051] (3) Obtain Gappy POD reconstructed flow field parameters Based on the established CFD dataset, use a flow field reduced-order model, that is, proper orthogonal decomposition (POD), to process the CFD dataset, decompose the full-space physical field contained in the CFD dataset into a set of orthogonal basis vectors, and based on singular value decomposition, solve for the first r orthogonal vectors whose sum of energy values is greater than 99.99% for truncation, and finally approximately express the flow field vectors in the space with the first r orthogonal vectors to complete the extraction of global flow field characteristics.

[0052] Based on the fact that the spatial flow field obtained by the POD method can be approximately expressed as a linear sum of a set of orthogonal basis vectors with high spatial resolution characteristics, that is ; assume the vector group obtained through experiments as ; based on the GappyPOD method, the solution of the modified orthogonal basis coefficients can be transformed into an optimization problem, that is ; then use the least squares method to solve it to obtain the optimal orthogonal basis coefficients, that is, the correction mode coefficients; then, combine the orthogonal basis obtained from the CFD data with the correction mode coefficients to obtain the reconstructed flow field data with high accuracy and high spatial resolution for the entire field, that is, obtain the Gappy POD reconstructed flow field parameters.

[0053] (4) Optimization of measurement point positions Taking the measurement point positions as design variables, the design variable optimization is carried out with the goal of minimizing the error between the global low-fidelity flow field parameters and the flow field parameters reconstructed by the Gappy POD. Combining with the global optimization algorithm, the sparse test measurement point layout optimization is carried out to obtain the optimal side point positions. Among them, after determining the number of test measurement points, the equally spaced measurement point positions from the leading edge to the trailing edge are selected as the initial values.

[0054] (5)High-precision reconstruction of the flow field by multi-source data fusion Based on the optimal measurement point positions, the test data under the wind tunnel test of the turbine blade to be studied are collected to obtain high-fidelity discrete test data. Then, the Gappy POD method is used to couple the test / CFD data to complete the high-precision reconstruction of the turbine flow field. Among them, based on the high-fidelity discrete test data, the Gappy POD algorithm is used to correct the modal coefficients to obtain optimized modal coefficients. According to the POD modes and the optimized modal coefficients, the flow field reconstruction results of the turbine blade to be studied are obtained.

[0055] As shown in the appendix Figure 3 shown, in the appendix Figure 3 the schematic diagrams of the layout positions before and after the optimization of the measurement point positions in Example 1 are given; from the appendix Figure 3 it can be seen that when the number of test measurement points is 10, the average relative error after optimization is 1.019% and the maximum relative error is 3.159%. Compared with the flow field reconstruction results of the uniform layout before optimization, the average relative error and the maximum relative error are respectively reduced by 0.728% and 0.578%. When the number of test measurement points is 15, the average relative error and the maximum relative error after optimization are respectively 1.010% and 0.578%. Compared with the flow field reconstruction results of the uniform layout before optimization, the average relative error and the maximum relative error are respectively reduced by 0.192% and 0.070%. Thus, the correctness and effectiveness of the high-precision flow field reconstruction based on the sparse measurement point position optimization are verified.

[0056] As shown in the appendix Figure 4 shown, in the appendix Figure 4 the schematic diagrams of the flow field reconstruction results after the optimization of the measurement point positions in Example 1 are given; among them, Figure 4 a is the schematic diagram of the flow field reconstruction results at 10 measurement point positions, Figure 4 b is the schematic diagram of the flow field reconstruction results at 15 measurement point positions; from the appendix Figure 4It can be seen that after optimization, the number of test measurement points near the load peak (i.e., near the cascade throat) increases, and the accuracy of the load distribution reconstruction result near the load peak is improved. When adding measurement points to the leading edge part, the flow field information of the airfoil leading edge part can be increased, thereby improving the accuracy of the flow field reconstruction. Therefore, the proposed multi-source data flow field reconstruction method based on the optimization of test measurement point positions can improve the accuracy of load distribution reconstruction on the premise of not increasing the number of test measurement points.

[0057] Embodiment 2 As shown in the Figure 5 accompanying drawings, Embodiment 2 of the present invention provides a flow field reconstruction system based on the optimization of test measurement point positions and multi-source data fusion, including a data sampling module, a data set establishment module, a POD flow field feature extraction module, an initial sampling coefficient correction module, an initial sampling flow field reconstruction module, a measurement point optimization module, a test module, an optimized sampling coefficient correction module, and an optimized sampling flow field reconstruction module.

[0058] The data sampling module is used to sample the design space variables of the turbine blade to be studied to obtain a sampling data sample; the data set establishment module is used to perform CFD simulation calculations on the sampling data sample to obtain global low-fidelity flow field parameters; based on the global low-fidelity flow field parameters, a CFD data set is established; based on the pre-determined uniform measurement point positions, the test data under the wind tunnel test of the turbine blade to be studied is collected to obtain a sparse measurement point test data set; the POD flow field feature extraction module is used to perform POD decomposition on the CFD data set to obtain POD modes and mode coefficients; the initial sampling coefficient correction module is used to correct the mode coefficients according to the sparse measurement point test data set by using the Gappy POD algorithm to obtain corrected mode coefficients; the initial sampling flow field reconstruction module is used to obtain Gappy POD reconstructed flow field parameters according to the POD modes and the corrected mode coefficients; the measurement point optimization module is used to optimize the design variables with the measurement point positions as the design variables and the minimum error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters as the goal to obtain the optimal measurement point positions; the test module is used to collect the test data under the wind tunnel test of the turbine blade to be studied based on the optimal measurement point positions to obtain high-fidelity discrete test data; the optimized sampling coefficient correction module is used to correct the mode coefficients based on the high-fidelity discrete test data by using the Gappy POD algorithm to obtain optimized mode coefficients; the optimized sampling flow field reconstruction module is used to obtain the flow field reconstruction result of the turbine blade to be studied according to the POD modes and the optimized mode coefficients.

[0059] Embodiment 3 As shown in the Figure 3As shown in the figure, Embodiment 3 of the present invention provides a flow field reconstruction device based on the optimization of test measurement point positions and multi-source data fusion, including: a memory for storing a computer program; a processor for implementing the steps of the flow field reconstruction method based on the optimization of test measurement point positions and multi-source data fusion when executing the computer program.

[0060] When the processor executes the computer program, it implements the steps of the above-mentioned flow field reconstruction method based on the optimization of test measurement point positions and multi-source data fusion. For example: Sample the design space variables of the turbine blade to be studied to obtain a sample data set; perform CFD simulation calculations on the sample data set to obtain global low-fidelity flow field parameters; establish a CFD data set based on the global low-fidelity flow field parameters; collect the test data of the turbine blade to be studied under wind tunnel tests based on the pre-determined uniform measurement point positions to obtain a sparse measurement point test data set; perform POD decomposition on the CFD data set to obtain POD modes and mode coefficients; correct the mode coefficients using the Gappy POD algorithm according to the sparse measurement point test data set to obtain corrected mode coefficients; obtain the Gappy POD reconstructed flow field parameters according to the POD modes and the corrected mode coefficients; optimize the design variables with the measurement point positions as the design variables and the minimum error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters as the objective to obtain the optimal measurement point positions; collect the test data of the turbine blade to be studied under wind tunnel tests based on the optimal measurement point positions to obtain high-fidelity discrete test data; correct the mode coefficients using the Gappy POD algorithm based on the high-fidelity discrete test data to obtain optimized mode coefficients; obtain the flow field reconstruction result of the turbine blade to be studied according to the POD modes and the optimized mode coefficients.

[0061] Alternatively, when the processor executes the computer program, it implements the functions of each module in the above-mentioned flow field reconstruction system based on the optimization of test measurement point positions and multi-source data fusion. For example: A data sampling module is used to sample the design space variables of the turbine blade to be studied to obtain sampled data samples; a data set establishment module is used to perform CFD simulation calculations on the sampled data samples to obtain global low-fidelity flow field parameters; a CFD data set is established based on the global low-fidelity flow field parameters; based on the predetermined uniform measuring point positions, the test data under the wind tunnel test of the turbine blade to be studied is collected to obtain a sparse measuring point test data set; a POD flow field feature extraction module is used to perform POD decomposition on the CFD data set to obtain POD modes and modal coefficients; an initial sampling coefficient correction module is used to correct the modal coefficients using the Gappy POD algorithm based on the sparse measuring point test data set to obtain corrected modal coefficients; an initial sampling flow field reconstruction module is used to obtain Gappy POD reconstructed flow field parameters based on the POD modes and the corrected modal coefficients; a measuring point optimization module is used to use the measuring point positions as design variables, the global low-fidelity flow field parameters and the Gappy The design variables are optimized with the goal of minimizing the error between the POD reconstructed flow field parameters to obtain the optimal measurement point position; the test module is used to collect test data of the turbine blade to be studied under the wind tunnel test based on the optimal measurement point position to obtain high-fidelity discrete test data; the optimized sampling coefficient correction module is used to correct the modal coefficients using the Gappy POD algorithm based on the high-fidelity discrete test data to obtain optimized modal coefficients; the optimized sampling flow field reconstruction module is used to obtain the flow field reconstruction result of the turbine blade to be studied based on the POD mode and the optimized modal coefficients.

[0062] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing preset functions, and the instruction segments are used to describe the execution process of the computer program in the flow field reconstruction device based on test measurement point position optimization and multi-source data fusion.

[0063] For example, the computer program can be divided into a data sampling module, a data set establishment module, a POD flow field feature extraction module, an initial sampling coefficient correction module, an initial sampling flow field reconstruction module, a measurement point optimization module, a test module, an optimized sampling coefficient correction module and an optimized sampling flow field reconstruction module. The specific functions of each module are as follows: a data sampling module, for sampling the design space variables of the turbine blade to be studied to obtain sampling data samples; a data set establishment module, for performing CFD simulation calculations on the sampling data samples to obtain global low-fidelity flow field parameters; establishing a CFD data set based on the global low-fidelity flow field parameters; based on the predetermined uniform measurement point positions, collecting the test data under the wind tunnel test of the turbine blade to be studied to obtain a sparse measurement point test data set; a POD flow field feature extraction module, for performing POD decomposition on the CFD data set to obtain POD modes and modal coefficients; an initial sampling coefficient correction module, for using Gappy to calculate the POD decomposition of the CFD data set according to the sparse measurement point test data set. The POD algorithm corrects the modal coefficients to obtain corrected modal coefficients; an initial sampling flow field reconstruction module is used to obtain Gappy POD reconstructed flow field parameters based on the POD mode and the corrected modal coefficients; a measuring point optimization module is used to optimize the design variables with the measuring point position as the design variable and the goal of minimizing the error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters to obtain the optimal measuring point position; a test module is used to collect test data of the turbine blade to be studied under the wind tunnel test based on the optimal measuring point position to obtain high-fidelity discrete test data; an optimized sampling coefficient correction module is used to correct the modal coefficients using the Gappy POD algorithm based on the high-fidelity discrete test data to obtain optimized modal coefficients; an optimized sampling flow field reconstruction module is used to obtain the flow field reconstruction result of the turbine blade to be studied based on the POD mode and the optimized modal coefficients.

[0064] The flow field reconstruction device based on test point location optimization and multi-source data fusion can be a computing device such as a desktop computer, a notebook, a handheld computer and a cloud server. The flow field reconstruction device based on test point location optimization and multi-source data fusion can include, but is not limited to, a processor and a memory. Those skilled in the art will understand that the above is an example of a flow field reconstruction device based on test point location optimization and multi-source data fusion, and does not constitute a limitation on the flow field reconstruction device based on test point location optimization and multi-source data fusion, and can include more components than the above, or combine certain components, or different components. For example, the flow field reconstruction device based on test point location optimization and multi-source data fusion can also include input and output devices, network access devices, buses, etc.

[0065] The so-called processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the flow field reconstruction device based on the optimization of test point positions and multi-source data fusion, and connects various parts of the entire flow field reconstruction device based on the optimization of test point positions and multi-source data fusion through various interfaces and circuits.

[0066] The memory can be used to store the computer programs and / or modules. The processor realizes various functions of the flow field reconstruction device based on the optimization of test point positions and multi-source data fusion by running or executing the computer programs and / or modules stored in the memory, and by calling the data stored in the memory.

[0067] The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, phone book, etc.). In addition, the memory can include high-speed random access memory, and can also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.

[0068] Embodiment 4 Embodiment 4 of the present invention also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it realizes the steps of the flow field reconstruction method based on the optimization of test point positions and multi-source data fusion, for example: Sample the design space variables of the turbine blade to be studied to obtain a sample of sampling data; perform CFD simulation calculations on the sampling data sample to obtain global low-fidelity flow field parameters; establish a CFD data set based on the global low-fidelity flow field parameters; collect the test data under the wind tunnel test of the turbine blade to be studied based on the predetermined uniform measurement point positions to obtain a sparse measurement point test data set; perform POD decomposition on the CFD data set to obtain POD modes and mode coefficients; correct the mode coefficients using the Gappy POD algorithm according to the sparse measurement point test data set to obtain corrected mode coefficients; obtain the Gappy POD reconstructed flow field parameters according to the POD modes and the corrected mode coefficients; take the measurement point positions as design variables, and optimize the design variables with the goal of minimizing the error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters to obtain the optimal measurement point positions; collect the test data under the wind tunnel test of the turbine blade to be studied based on the optimal measurement point positions to obtain high-fidelity discrete test data; correct the mode coefficients using the Gappy POD algorithm according to the high-fidelity discrete test data to obtain optimized mode coefficients; obtain the flow field reconstruction result of the turbine blade to be studied according to the POD modes and the optimized mode coefficients.

[0069] If the module / unit integrated by the flow field reconstruction system based on test measurement point position optimization and multi-source data fusion is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0070] Based on such an understanding, all or part of the processes in the above-mentioned flow field reconstruction method based on test measurement point position optimization and multi-source data fusion of the present invention can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned flow field reconstruction method based on test measurement point position optimization and multi-source data fusion can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or preset intermediate form, etc.

[0071] The computer-readable storage medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0072] It should be noted that the content included in the computer-readable storage medium can be appropriately increased or decreased according to the requirements of legislation and patent practice within the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, the computer-readable storage medium does not include electrical carrier signals and telecommunication signals.

[0073] Embodiment 5 Embodiment 5 of the present invention provides a computer product. The computer program product includes a computer program, and the computer program is stored in a computer-readable storage medium; a processor of a flow field reconstruction device based on the optimization of test measurement point positions and multi-source data fusion reads the computer program from the computer-readable storage medium, and the processor executes the computer program, so that the flow field reconstruction device based on the optimization of test measurement point positions and multi-source data fusion can execute the flow field reconstruction method based on the optimization of test measurement point positions and multi-source data fusion described in Embodiment 1, which will not be elaborated here.

[0074] It should be noted that those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above methods.

[0075] For the flow field reconstruction method based on the optimization of test measurement point positions and multi-source data fusion according to the present invention, after determining the design space variables and their variation ranges, sample collection is carried out to establish initial sample data; the numerical calculation CFD method is used to obtain the global flow field distribution data set of low-fidelity samples and sparse the measured point test data set; the proper orthogonal decomposition (POD) reduced-order model is used to extract the key modal features of the CFD global flow field, that is, the orthogonal vector basis corresponding to the high-dimensional vector group, and the Gappy POD method is used to sparse the measured point test data set to obtain the correction modal coefficients, and then the high-fidelity global flow field distribution can be obtained; on this basis, taking the positions of the test measurement points as optimization variables and the minimum flow field reconstruction error obtained by Gappy POD as the objective function, sparse optimization of the test measurement point positions is carried out to obtain the optimal measurement point positions; based on the optimal measurement point positions, the optimal sparse test measurement point data in the high-dimensional flow field are collected, and a high-precision global flow field distribution is further obtained; the method can effectively utilize the characteristics of strong correlation of multi-source data such as test / CFD, combine the globality of CFD data with the discrete high-fidelity characteristics of test measurement point data, and reconstruct a high-precision and refined global flow field distribution through the Gappy POD method, effectively solving the problems of mismatched spatial distribution characteristics and different fidelity of test / CFD data.

[0076] The above embodiments are only one of the implementation manners capable of implementing the technical solution of the present invention. The scope of protection required by the present invention is not limited to this embodiment only, but also includes any changes, substitutions and other implementation manners that are easily conceivable by any person skilled in the art within the technical scope disclosed by the present invention.

Claims

1. A flow field reconstruction method based on the optimization of test point positions and multi-source data fusion, characterized in that, Including: Sampling the design space variables of the turbine blade to be studied to obtain a sampling data sample; Performing CFD simulation calculations on the sampling data sample to obtain global low-fidelity flow field parameters; Based on the global low-fidelity flow field parameters, establishing a CFD data set; based on predetermined uniform measurement point positions, collecting test data under the wind tunnel test of the turbine blade to be studied to obtain a sparse measurement point test data set; Performing POD decomposition on the CFD data set to obtain POD modes and modal coefficients; According to the sparse measurement point test data set, using the Gappy POD algorithm to correct the modal coefficients to obtain corrected modal coefficients; According to the POD modes and the corrected modal coefficients, obtaining Gappy POD reconstructed flow field parameters; Taking the measurement point position as the design variable and aiming at minimizing the error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters, performing design variable optimization to obtain the optimal measurement point position; Based on the optimal measurement point position, collecting test data under the wind tunnel test of the turbine blade to be studied to obtain high-fidelity discrete test data; Based on the high-fidelity discrete test data, using the Gappy POD algorithm to correct the modal coefficients to obtain optimized modal coefficients; According to the POD modes and the optimized modal coefficients, obtaining the flow field reconstruction result of the turbine blade to be studied.

2. The flow field reconstruction method based on the optimization of test point positions and multi-source data fusion according to claim 1, characterized in that The design space variables of the turbine blade to be studied include the exit isentropic Mach number, inlet attack angle, inlet turbulence intensity and relative axial pitch of the turbine blade to be studied.

3. A flow field reconstruction method based on the optimization of test point positions and multi-source data fusion according to claim 1, characterized in that, The process of collecting test data under the wind tunnel test of the turbine blade to be studied based on predetermined uniform measurement point positions to obtain a sparse measurement point test data set includes: Determining the wind tunnel test conditions of the turbine blade to be studied under the initial sampling; wherein, the wind tunnel test conditions of the turbine blade to be studied under the initial sampling include predetermined uniform measurement point positions; Based on the wind tunnel test conditions of the turbine blade to be studied under the initial sampling, collecting test data under the wind tunnel test of the turbine blade to be studied to obtain a sparse measurement point test data set.

4. A flow field reconstruction method based on the optimization of test point positions and multi-source data fusion according to claim 1, characterized in that The process of obtaining corrected modal coefficients according to the sparse measurement point test data set and combining with the POD modes using the Gappy POD algorithm includes: Based on the sparse measurement point test data set and the POD modes, constructing a modal coefficient optimization formula; Using the least squares method to solve the modal coefficient optimization formula to obtain corrected modal coefficients.

5. A flow field reconstruction method based on the optimization of test point positions and multi-source data fusion according to claim 4, characterized in that The modal coefficient optimization formula is specifically: Among them, is the correction modal coefficient; is the discrete test measurement point data; is the modal coefficient; is the POD mode; is the modal order; is the th test data of the measurement point; is the total number of measurement points.

6. A flow field reconstruction method based on the optimization of test point positions and multi-source data fusion according to claim 1, characterized in that, The process of taking the measurement point position as the design variable and aiming at minimizing the error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters, performing design variable optimization to obtain the optimal measurement point position includes: Determining the number of measurement point positions; According to the number of measurement point positions, determining the initial value of the measurement point position; wherein, the initial value of the measurement point position is several equally spaced measurement point positions from the leading edge to the trailing edge of the turbine blade to be studied; Based on the initial values of the measurement point positions, with the goal of minimizing the error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters, and using the measurement point positions as design variables, an objective function is established. Based on a global optimization algorithm, the objective function is optimized to obtain the optimal measurement point positions.

7. A flow field reconstruction system based on the optimization of test point positions and the fusion of multi-source data, characterized in that, It includes: A data sampling module for sampling the design space variables of the turbine blade to be studied to obtain sampling data samples. A data set establishment module for performing CFD simulation calculations on the sampling data samples to obtain global low-fidelity flow field parameters; based on the global low-fidelity flow field parameters, a CFD data set is established; based on pre-determined uniform measurement point positions, the test data under the wind tunnel test of the turbine blade to be studied is collected to obtain a sparse measurement point test data set. A POD flow field feature extraction module for performing POD decomposition on the CFD data set to obtain POD modes and mode coefficients. An initial sampling coefficient correction module for correcting the mode coefficients using the Gappy POD algorithm according to the sparse measurement point test data set to obtain corrected mode coefficients. An initial sampling flow field reconstruction module for obtaining Gappy POD reconstructed flow field parameters according to the POD modes and the corrected mode coefficients. A measurement point optimization module for optimizing the design variables with the goal of minimizing the error between the global low-fidelity flow field parameters and the Gappy POD reconstructed flow field parameters using the measurement point positions as design variables to obtain the optimal measurement point positions. A test module for collecting the test data under the wind tunnel test of the turbine blade to be studied based on the optimal measurement point positions to obtain high-fidelity discrete test data. An optimized sampling coefficient correction module for correcting the mode coefficients using the Gappy POD algorithm based on the high-fidelity discrete test data to obtain optimized mode coefficients. An optimized sampling flow field reconstruction module for obtaining the flow field reconstruction result of the turbine blade to be studied according to the POD modes and the optimized mode coefficients.

8. A flow field reconstruction device based on the optimization of test point positions and multi-source data fusion, characterized in that, It includes: A processor suitable for executing a computer program. A computer-readable storage medium storing a computer program, and when the computer program is executed by the processor, it executes the flow field reconstruction method based on the optimization of test measurement point positions and multi-source data fusion as described in any one of claims 1-6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the flow field reconstruction method based on the optimization of test measurement point positions and multi-source data fusion as described in any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program, and when the computer program is executed by the processor, it implements the flow field reconstruction method based on the optimization of test measurement point positions and multi-source data fusion as described in any one of claims 1-6.

Images