Centrifugal impeller design verification method applied to small-power civil aero-engine

Abstract

The invention discloses a centrifugal impeller design verification method applied to a small-power civil aero-engine, and the method comprises the steps: deconstructing a three-dimensional model of a centrifugal impeller into a two-dimensional airway boundary, blade thickness distribution, blade angle distribution and other elements, carrying out the dimension reduction of continuous data points, and extracting feature values; meanwhile, in combination with other overall performance characteristics such as rated rotating speed, inlet total temperature, inlet total pressure and outlet static pressure, a BP neural network is trained, and the pressure ratio, flow and efficiency are output. The method can be used for preposed rapid verification calculation of small-power civil aero-engine centrifugal impeller design, centrifugal impellers which cannot meet design requirements are rapidly screened out, the workload of three-dimensional flow field simulation is reduced, and the development efficiency of the centrifugal impellers is improved.

Description

Technical field:

[0001] The invention relates to a centrifugal impeller design verification method applied to a low-power civil aviation engine, and belongs to the field of impeller machinery design such as civil aviation engines, gas turbines and fans. Background technology:

[0002] Centrifugal impellers can provide a large air compression ratio and are widely used in civil aviation engines, gas turbines, etc. The design efficiency of centrifugal impellers is highly related to the designer's engineering design experience of civil aviation engine compressor flow path, blade thickness, blade angle and other elements.

[0003] The current main approaches to the design of centrifugal compressors are as follows: one-dimensional and meridional flow channel design, and three-dimensional blade design. After years of development, one-dimensional and meridional flow channel design has accumulated rich design experience. Based on test results and design experience, the airflow loss in the centrifugal compressor is subdivided into various forms such as surface friction loss, blade load loss, leakage loss, backflow loss, separation loss, shock wave loss, etc., and corresponding empirical formulas are given to make the design process as accurate as possible. Three-dimensional blade design refers to the initial blade shape design based on one-dimensional and meridional flow channel design by giving the blade angle distribution and thickness distribution of the root tip blade, and then performing iterative design based on the three-dimensional flow field simulation results. If the simulation results show that the one-dimensional design parameters are not selected reasonably, it is necessary to return to the one-dimensional simulation and re-given the basic design parameters. Otherwise, it is directly iteratively designed through the three-dimensional blade shape parameters until the aerodynamic performance of the centrifugal impeller meets the design requirements and the flow field distribution is reasonable. The current approach has the following main disadvantages: (1) The design and verification process is long, which affects the design efficiency; (2) It is highly dependent on the engineering experience of the designer, and the design method is difficult to expand and promote.

[0004] Therefore, it is necessary to improve the existing technology to solve the shortcomings of the existing technology. Summary of the invention:

[0005] The present invention is to solve the problems existing in the above-mentioned prior art and provide a centrifugal impeller design verification method applied to low-power civil aviation engines. The method deconstructs the three-dimensional model of the centrifugal impeller into two-dimensional airway boundaries, blade thickness distribution, blade angle distribution and other elements, reduces the dimension of continuous data points to extract characteristic values, and combines other overall performance characteristics such as rated speed, inlet total temperature, inlet total pressure, outlet static pressure, etc. to train the BP neural network and output pressure ratio, flow rate, and efficiency. The method can be used for the pre-fast verification calculation of the design of centrifugal impellers for low-power civil aviation engines, quickly screen out centrifugal impellers that cannot meet the design requirements, reduce the workload of three-dimensional flow field simulation, and improve the development efficiency of centrifugal impellers.

[0006] The present invention adopts the following technical solution: a centrifugal impeller design verification method applied to a low-power civil aviation engine, the specific steps are as follows:

[0007] Step 1: Generate a large number of three-dimensional design models of centrifugal impellers for low-power civil aviation engines, output the pressure ratio π, flow rate m, and efficiency η values ​​through three-dimensional flow field simulation, and generate training sample data;

[0008] Step 2: For each centrifugal impeller three-dimensional design model, deconstruct the centrifugal impeller two-dimensional airway boundary points, the continuous blade thickness distribution at 100% blade height, the continuous blade thickness distribution at 0% blade height, the continuous angle distribution at 100% blade height, and the continuous blade angle distribution at 0% blade height;

[0009] Step 3: Using the principal component analysis method, the continuous data points of the two-dimensional airway boundary points of the centrifugal impeller, the continuous blade thickness distribution at 100% blade height, the continuous blade thickness distribution at 0% blade height, the continuous angle distribution at 100% blade height, and the continuous blade angle distribution at 0% blade height are reduced to two dimensions, and the single variable characteristics of rated speed N, inlet total temperature T, inlet total pressure P0, and outlet static pressure P1 are added;

[0010] Step 4: Train the BP neural network model;

[0011] Step 5: For the subsequent centrifugal impeller three-dimensional design model data, the feature vectors selected from the previous data set are used for dimensionality reduction. The characteristic values ​​of the centrifugal impeller samples after dimensionality reduction and the rated speed, inlet total temperature, inlet total pressure, and outlet static pressure are input into the trained neural network function calculation results, and the pressure ratio, flow rate, and efficiency are output as a rapid design evaluation before the three-dimensional flow field simulation.

[0012] Further, step one is as follows:

[0013] (1.1) Obtain samples using conventional centrifugal impeller design methods;

[0014] (1.2) The output value array of the three-dimensional flow field simulation of sample 1 is [π1, m1, η1], denoted as Op (1) , where π1 is the pressure ratio of sample 1, m1 is the flow rate of sample 1, η1 is the efficiency of sample 1, and the three-dimensional flow field simulation output results of 100 samples form a matrix [Op (1) ;Op (2) ; ... ; Op (100) ] T , denoted as Op.

[0015] Furthermore, step 2 is as follows:

[0016] (2.1) The two-dimensional airway boundary of the centrifugal impeller design sample includes the airway boundary of the casing surface located at the top and the airway boundary of the hub surface located at the bottom. The two-dimensional curve of the airway boundary of the casing surface consists of 100 two-dimensional coordinates. In the i-th two-dimensional curve of the airway boundary of the casing surface, 100 point coordinate samples are uniformly taken as A i , the hub surface airway boundary two-dimensional curve is also composed of 100 two-dimensional coordinates. In the i-th hub surface airway boundary two-dimensional curve, 100 point coordinate samples are uniformly taken as Note: B i ;

[0017] (2.2) In the centrifugal impeller design, 100 points are uniformly selected from the continuous blade thickness distribution at 0% blade height of the i-th sample, and the array is used Indicated by Ha i ; At the 100% leaf height of the i-th sample, the continuous leaf thickness distribution is uniform, and 100 points are taken using an array Denotes Hb i ;

[0018] (2.3) In the centrifugal impeller design, 100 points are uniformly selected from the continuous blade angle distribution at 0% blade height of the i-th sample, and the array is used Indicated by Aa i ; uniformly select 100 points from the continuous leaf angle distribution at 100% leaf height of the i-th sample and use the array Denotes Ab i .

[0019] Furthermore, step three is as follows:

[0020] (3.1) Two-dimensional coordinate matrix A for airway boundary samples i and B i , reduced to one dimension using principal component analysis;

[0021] (3.2) Calculate the matrix A i The covariance matrix C of C is calculated, and the eigenvalues ​​λ1 and λ2 of C are calculated. The eigenvalue λ with a covariance contribution rate greater than 90% is selected. k ;

[0022] (3.3) Choose λ k The corresponding eigenvectors form the feature transformation matrix P k ∈R 1*2 ;

[0023] (3.4) Matrix A i Dimensionality reduction transformation: The result after dimensionality reduction is recorded as As i , for the matrix B iTake the same operation to obtain the feature transformation matrix P m ∈R 1*2 , B i The result after dimensionality reduction is recorded as Bs i .

[0024] (3.5) All A i (i=1,2……100) after dimensionality reduction, the matrix [As1,As2……As 100 ], denoted as As;

[0025] (3.6) Calculate the covariance matrix C1 of the matrix As and calculate the eigenvalue λ of C1 a1 , a2 ... a100 , select two eigenvalues ​​λ whose sum of covariance contribution is greater than 90% i and λ j ;

[0026] (3.7) Choose λ i and λ j The corresponding eigenvectors form the feature transformation matrix P k2 ∈R 2*100 ;

[0027] (3.8) Dimension reduction transformation of matrix As: Therefore, the 2D curve sample of the airway boundary of the i-th casing surface is reduced to two eigenvalues: To B i Take the same operation to get the feature transformation matrix P m2 , the two-dimensional curve samples of the hub surface airway boundary are reduced to two eigenvalues:

[0028] (3.9) The blade thickness and blade angle feature matrix Ha for the i-th sample i , Hb i , Aa i ,Ab i Repeat steps (3.6) to (3.8) to obtain the feature transformation matrix P x1 , P x2 , P x3 , P x4 , after dimensionality reduction, the leaf thickness distribution at 0% leaf height, the leaf thickness distribution at 100% leaf height, the leaf angle distribution at 0% leaf height, and the leaf angle distribution at 100% leaf height of the i-th sample are obtained as the reduced eigenvalues: (3.10) Therefore, the two-dimensional design continuous features of the i-th sample can be extracted into 12 eigenvalues, combined with the overall performance parameters of the engine, rated speed N (i) , inlet total temperature T (i) , total inlet pressure Outlet static pressure The design characteristics of centrifugal impeller sample i can be expressed by the following 16 characteristic parameter groups:

[0029]

[0030] Ps (i) ,i=1,2,……100, sample set matrix [Ps (1) ;Ps (2) ;……;Ps (100) ] T , denoted as Ps.

[0031] Furthermore, step four is as follows:

[0032] The reduced-dimensional data sample set Ps is used as the input matrix, the simulation result matrix Op is used as the output matrix, and a hidden layer network structure is adopted. The hidden layer nodes are calculated according to the empirical formula Determine, where H is the number of hidden layer nodes, I is the number of input layer nodes, O is the number of output layer nodes, select H = 6;

[0033] The weight matrix between the input layer and the hidden layer is recorded as

[0034] The weight matrix between the hidden layer and the output layer is recorded as

[0035] The bias vector between the input layer and the hidden layer is denoted by B = (B1, B2, ..., B6) T ;

[0036] The bias vector between the hidden layer and the output layer is denoted by C = (C1, C2, C3) T ;

[0037] For the weight matrix W (1) , W (2) And the bias vectors B and C are initialized and assigned, and the initialization value is a random number in the range of (0.5,1);

[0038] The transfer function is selected as the Sigmoid function, that is

[0039] Calculate the jth hidden layer value of the i-th sample

[0040]

[0041] in is the weight matrix W (1) The jth row of Ps (i) is the i-th sample in the sample set Ps, B j is the jth value in the bias vector B;

[0042] Similarly, calculate the jth output layer value of the i-th sample

[0043]

[0044] in is the weight matrix W (2) The jth row of Hid (i) is the hidden layer output value vector of the i-th sample in the sample set, C j is the jth value in the bias vector C;

[0045] Refer to the output sample Op in the training sample and calculate the i-th sample Op (i) and Output (i) The error between (i) :

[0046]

[0047] in is the i-th sample Op (i) The kth value in is the i-th sample Output (i) The kth value in ;

[0048] The total (average) cost E of all training samples total for:

[0049]

[0050] The weights and biases are updated using batch gradient descent until the system error is minimized:

[0051]

[0052] Where μ is the learning rate, and the selection range is (0.05, 0.8)

[0053] The neural network model after training is recorded as f BP (x),x∈R 16*1 .

[0054] Furthermore, step five is as follows:

[0055] For the new centrifugal impeller 3D model design, deconstruct the 2D geometric elements;

[0056] Select 100 points evenly and equidistantly on the curve, and record the airway boundary data of the casing surface as A0∈R 100*2 , the hub surface airway boundary data is recorded as B0∈R 100*2The blade angle of the centrifugal impeller at 0% blade height is recorded as Aa0∈R 100*1 , the blade angle of the centrifugal impeller at 100% blade height is recorded as Ab0∈R 100*1 , the blade thickness of the centrifugal impeller at 0% blade height is recorded as Ha0∈R 100*1 , the blade thickness of the centrifugal impeller at 100% blade height is recorded as Hb0∈R 100*1 ;

[0057] Perform data dimension reduction on the above data:

[0058]

[0059] Add overall performance parameters Rated speed N (0) , inlet total temperature T (0) , total inlet pressure Outlet static pressure Construct a new input sample Ps (0) ∈R 16*1 :

[0060] in, is the eigenvalue of the two-dimensional airway boundary of the new sample after dimensionality reduction, is the eigenvalue of the leaf thickness distribution at 0% leaf height and 100% leaf height of the new sample after dimensionality reduction, It is the eigenvalue of the leaf angle distribution at 0% leaf height and 100% leaf height of the new sample after dimensionality reduction;

[0061] Substitute into the neural network function calculation, f BP (Ps (0) )=[π0,m0,η0];

[0062] Evaluate whether the output pressure ratio π0, flow rate m0, and efficiency η0 values ​​meet the design requirements. If so, continue with the three-dimensional flow field simulation verification; if not, redesign.

[0063] The present invention has the following beneficial effects:.

[0064] (1) The present invention expresses the three-dimensional entity of the centrifugal impeller through continuous two-dimensional geometric data, then uniformly and densely samples the continuous two-dimensional data, performs data dimension reduction through principal component analysis, deconstructs and reduces the dimension layer by layer, and finally realizes the parametric expression of the three-dimensional entity of the centrifugal impeller. Based on the aforementioned parametric expression, the BP neural network model is trained to fit the output results of the three-dimensional flow field simulation, thereby improving the design evaluation efficiency of the designers of the compressor end of the civil aviation engine;

[0065] (2) The centrifugal impeller design verification method applied to small-power civil aviation engines of the present invention has great reference significance for the future participation of artificial intelligence large models in the auxiliary design of civil aviation engine compressor components. Description of the drawings:

[0066] Figure 1 This is the two-dimensional airway boundary map of centrifugal impeller sample 1.

[0067] Figure 2 Graphs showing blade thickness distribution at 0% blade height and 100% blade height of centrifugal impeller sample 1.

[0068] Figure 3 Graph showing blade angle distribution at 0% blade height and 100% blade height of centrifugal impeller sample 1.

[0069] Figure 4 Iterative flow chart for BP neural network training.

[0070] Figure 5(a) , 5(b) , 5(c), and 5(d) are schematic diagrams of the three-dimensional model of the centrifugal impeller and the deconstruction of the two-dimensional geometric elements.

[0071] Figure 6 Flowchart for the implementation of the design verification method for centrifugal impellers of small-power civil aviation engines. Specific implementation method:

[0072] The present invention will be further described below in conjunction with the accompanying drawings.

[0073] The present invention is applied to the centrifugal impeller design verification method of a low-power civil aviation engine, and the specific steps are as follows:

[0074] Step 1: For small-power civil aviation engines, especially small-power civil aviation engines with single-stage centrifugal compressor configuration below 500kW, generate a large number of centrifugal impeller three-dimensional design models, output the pressure ratio π, flow rate m, and efficiency η values ​​through three-dimensional flow field simulation, and generate training sample data.

[0075] Step 1 is as follows:

[0076] (1.1) The conventional centrifugal impeller design method is used to obtain samples, including: a one-dimensional centrifugal impeller design that is artificially optimized based on the key design parameter selection criteria, a two-dimensional air duct profile that is artificially optimized based on the load distribution in the end wall area of ​​the centrifugal impeller, a three-dimensional blade aerodynamic shape that is artificially optimized based on the three-dimensional blade load distribution of the centrifugal impeller, and aerodynamic performance parameters of the centrifugal impeller obtained by means of three-dimensional flow field simulation.

[0077] (1.2) The output value array of the three-dimensional flow field simulation of sample 1 is [π1, m1, η1], denoted as Op (1), where π1 is the pressure ratio of sample 1, m1 is the flow rate of sample 1, and η1 is the efficiency of sample 1. The three-dimensional flow field simulation output results of 100 samples form a matrix [Op (1) ;Op (2) ; ... ; Op (100) ] T , denoted as Op.

[0078] Step 2: For each centrifugal impeller 3D design model, deconstruct the 2D airway boundary points of the centrifugal impeller, the continuous blade thickness distribution at 100% blade height, the continuous blade thickness distribution at 0% blade height, the continuous angle distribution at 100% blade height and the continuous blade angle distribution at 0% blade height. See the attached figure for details. Figure 1 , Attachment Figure 2 , Attachment Figure 3 For each deconstructed two-dimensional design element, 100 data points are evenly taken.

[0079] Step 2 is as follows:

[0080] (2.1) The two-dimensional airway boundary of the centrifugal impeller design sample is shown in the attached Figure 1 The above is the two-dimensional curve of the airway boundary of the casing surface, which consists of 100 two-dimensional coordinates. The sample of the two-dimensional curve of the airway boundary of the i-th casing surface (100 points are evenly taken) is A i The following is the two-dimensional curve of the hub surface airway boundary, which is also composed of 100 two-dimensional coordinates. The i-th hub surface airway boundary two-dimensional curve sample (100 points are evenly taken) is Note: B i .

[0081] (2.2) The blade thickness distribution of the centrifugal impeller design sample at 0% blade height and 100% blade height is shown in the attached figure. Figure 2 The upper curve is the continuous leaf thickness distribution at 0% leaf height (100 points are uniformly taken in the continuous distribution). The i-th sample can be used in the array Indicated by Ha i ; The lower curve is the continuous leaf thickness distribution at 100% leaf height (100 points are uniformly selected in the continuous distribution), and the i-th sample can be used in the array Denotes Hb i .

[0082] (2.3) The blade angle distribution of the centrifugal impeller design sample at 0% blade height and 100% blade height is shown in the attached figure. Figure 3 The upper curve is the continuous leaf angle distribution at 0% leaf height (100 points are uniformly selected in the continuous distribution). The i-th sample can be used in the array Indicated by Aa i; The lower curve is the continuous leaf angle distribution at 100% leaf height (100 points are uniformly selected in the continuous distribution), and the i-th sample can be used in the array Denotes Ab i .

[0083] Step 3: Using the principal component analysis method, the continuous data points of the centrifugal impeller airway boundary, the blade thickness distribution at 100% blade height and 0% blade height, and the blade angle distribution at 100% blade height and 0% blade height are reduced to two dimensions, and the single variable characteristics of rated speed N, inlet total temperature T, inlet total pressure P0, and outlet static pressure P1 are added. After dimensionality reduction, there are 16 eigenvalues ​​of a single centrifugal impeller sample.

[0084] Step 3 is as follows:

[0085] (3.1) Two-dimensional coordinate matrix A for airway boundary samples i and B i , reduced to one dimension using principal component analysis.

[0086] (3.2) Specifically, the matrix A is calculated i The covariance matrix C of the equation is used to calculate the eigenvalues ​​λ1 and λ2, and the eigenvalue λ with a covariance contribution greater than 90% is selected. k .

[0087] (3.3) Choose λ k The corresponding eigenvectors constitute the feature transformation matrix P k ∈R 1*2 .

[0088] (3.4) Matrix A i Dimensionality reduction transformation: The result after dimensionality reduction is recorded as As i . For the matrix B i By taking the same operation, we can obtain the feature transformation matrix P m ∈R 1*2 , the result after dimensionality reduction is recorded as Bs i .

[0089] (3.5) All A i (i=1,2……100) after dimensionality reduction, the matrix [As1,As2……As 100 ], denoted as As.

[0090] (3.6) Calculate the covariance matrix C1 of all sample matrices As and calculate the eigenvalue λ a1 , a2 ... a100 , select two eigenvalues ​​λ whose sum of covariance contribution is greater than 90% i and λ j .

[0091] (3.7) Choose λ i and λ j The corresponding eigenvectors constitute the feature transformation matrix P k2 ∈R 2*100 .

[0092] (3.8) Dimensionality reduction transformation of data set As: Therefore, the 2D curve sample of the airway boundary of the i-th casing surface can be reduced to two eigenvalues: To B i Take the same operation to get the feature transformation matrix P m2 , the two-dimensional curve samples of the hub surface airway boundary are reduced to two eigenvalues:

[0093] (3.9) The matrix Ha of blade thickness and blade angle for the i-th sample i , Hb i , Aa i ,Ab i Repeat steps (3.6) to (3.8) to obtain the feature transformation matrix P x1 , P x2 , P x3 , P x4 After dimensionality reduction, the leaf thickness distribution at 0% leaf height, the leaf thickness distribution at 100% leaf height, the leaf angle distribution at 0% leaf height, and the leaf angle distribution at 100% leaf height of the i-th sample are obtained as the reduced eigenvalues:

[0094] (3.10) Therefore, the two-dimensional design continuous features of the i-th sample can be extracted into 12 eigenvalues, combined with the overall performance parameters of the engine, rated speed N (i) , inlet total temperature T (i) , total inlet pressure Outlet static pressure The design characteristics of centrifugal impeller sample i can be expressed by the following 16 characteristic parameter groups:

[0095]

[0096] Ps (i) ,i=1,2,……100. Sample set matrix [Ps (1) ;Ps (2) ;……;Ps (100) ] T , denoted as Ps.

[0097] Step 4: Train the BP neural network model.

[0098] Step 4 is as follows:

[0099] The reduced-dimensional data sample set Ps is used as the input matrix, and the simulation result matrix Op is used as the output matrix. The neural network iterative training process is detailed in the attached Figure 4 .

[0100] Specifically, a hidden layer network structure is adopted, and the hidden layer nodes are based on the empirical formula Determine, where H is the number of hidden layer nodes, I is the number of input layer nodes, and O is the number of output layer nodes. In this example, H=6 is selected.

[0101] The weight matrix between the input layer and the hidden layer is recorded as

[0102] The weight matrix between the hidden layer and the output layer is recorded as

[0103] The bias vector between the input layer and the hidden layer is denoted by B = (B1, B2, ..., B6) T .

[0104] The bias vector between the hidden layer and the output layer is denoted by C = (C1, C2, C3) T .

[0105] For the weight matrix W (1) , W (2) And the bias vectors B and C are initialized and assigned values, and the initialization values ​​are random numbers in the range of (0.5,1).

[0106] The transfer function is selected as the Sigmoid function, that is

[0107] Calculate the jth hidden layer value of the i-th sample

[0108]

[0109] in is the weight matrix W (1) The jth row of Ps (i) is the i-th sample in the sample set Ps, B j is the jth value in the bias vector B;

[0110] Similarly, calculate the jth output layer value of the i-th sample

[0111]

[0112] in is the weight matrix W (2) The jth row of Hid (i) is the hidden layer output value vector of the i-th sample in the sample set, C j is the jth value in the bias vector C;

[0113] Refer to the output sample Op in the training sample and calculate the i-th sample Op (i) and Output (i) The error between (i) :

[0114]

[0115] in is the i-th sample Op (i) The kth value in is the i-th sample Output (i) The kth value in ;

[0116] The total (average) cost E of all training samples total for:

[0117]

[0118] The weights and biases are updated using batch gradient descent until the system error is minimized:

[0119]

[0120]

[0121] Where μ is the learning rate, and the selection range is (0.05, 0.8).

[0122] The neural network model after training is recorded as f BP (x),x∈R 16*1 .

[0123] Step 5: For the subsequent centrifugal impeller three-dimensional design model data, the feature vectors selected from the previous data set are used for dimensionality reduction. The characteristic values ​​of the centrifugal impeller samples after dimensionality reduction and the rated speed, inlet total temperature, inlet total pressure, and outlet static pressure are input into the trained neural network function calculation results, and the pressure ratio, flow rate, and efficiency are output as a rapid design evaluation before the three-dimensional flow field simulation.

[0124] Step 5 is as follows:

[0125] For the new centrifugal impeller three-dimensional model design (Figure 5(a)), the two-dimensional geometric elements are deconstructed: airway (Figure 5(b)), blade angle (Figure 5(c)), and blade thickness (Figure 5(d)).

[0126] Select 100 points evenly and equidistantly on the curve, and record the airway boundary data of the casing surface as A0∈R 100*2 , the hub surface airway boundary data is recorded as B0∈R 100*2The blade angle of the centrifugal impeller at 0% blade height is recorded as Aa0∈R 100*1 , the blade angle of the centrifugal impeller at 100% blade height is recorded as Ab0∈R 100*1 , the blade thickness of the centrifugal impeller at 0% blade height is recorded as Ha0∈R 100*1 , the blade thickness of the centrifugal impeller at 100% blade height is recorded as Hb0∈R 100*1 .

[0127] Perform data dimensionality reduction on the above data:

[0128]

[0129] Add overall performance parameters Rated speed N (0) , inlet total temperature T (0) , total inlet pressure Outlet static pressure Construct input sample Ps (0) ∈R 16*1 :

[0130] in, is the eigenvalue of the two-dimensional airway boundary of the new sample after dimensionality reduction, is the eigenvalue of the leaf thickness distribution at 0% leaf height and 100% leaf height of the new sample after dimensionality reduction, It is the eigenvalue of the leaf angle distribution at 0% leaf height and 100% leaf height of the new sample after dimensionality reduction;

[0131] Substitute into the neural network function calculation, f BP (Ps (0) )=[π0,m0,η0].

[0132] Evaluate whether the output pressure ratio π0, flow rate m0, and efficiency η0 values ​​meet the design requirements. If so, continue with the three-dimensional flow field simulation verification; if not, redesign.

[0133] Please see the attached for details of the process Figure 6 .

[0134] The above description is only a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements without departing from the principle of the present invention, and these improvements should also be regarded as within the protection scope of the present invention.

Claims

1. A centrifugal impeller design verification method for a low-power civil aviation engine, characterized in that: The specific steps are as follows: Step 1: Generate a large number of three-dimensional design models of centrifugal impellers for low-power civil aviation engines, output the pressure ratio π, flow rate m, and efficiency η values ​​through three-dimensional flow field simulation, and generate training sample data; Step 2: For each centrifugal impeller three-dimensional design model, deconstruct the centrifugal impeller two-dimensional airway boundary points, the continuous blade thickness distribution at 100% blade height, the continuous blade thickness distribution at 0% blade height, the continuous angle distribution at 100% blade height, and the continuous blade angle distribution at 0% blade height; Step 3: Using the principal component analysis method, the continuous data points of the two-dimensional airway boundary points of the centrifugal impeller, the continuous blade thickness distribution at 100% blade height, the continuous blade thickness distribution at 0% blade height, the continuous angle distribution at 100% blade height, and the continuous blade angle distribution at 0% blade height are reduced to two dimensions, and the single variable characteristics of rated speed N, inlet total temperature T, inlet total pressure P0, and outlet static pressure P1 are added; Step 4: Train the BP neural network model; Step 5: For the subsequent centrifugal impeller three-dimensional design model data, the feature vectors selected from the previous data set are used for dimensionality reduction. The characteristic values ​​of the centrifugal impeller samples after dimensionality reduction and the rated speed, inlet total temperature, inlet total pressure, and outlet static pressure are input into the trained neural network function calculation results, and the pressure ratio, flow rate, and efficiency are output as a rapid design evaluation before the three-dimensional flow field simulation.

2. The centrifugal impeller design verification method for a low-power civil aviation engine according to claim 1, characterized in that: Step 1 is as follows: (1.1) Obtain samples using conventional centrifugal impeller design methods; (1.2) The output value array of the three-dimensional flow field simulation of sample 1 is [π1, m1, η1], denoted as Op (1) , where π1 is the pressure ratio of sample 1, m1 is the flow rate of sample 1, η1 is the efficiency of sample 1, and the three-dimensional flow field simulation output results of 100 samples form a matrix [Op (1) ;Op (2) ; ...; Op (100) ] T , denoted as Op.

3. The centrifugal impeller design verification method for a low-power civil aviation engine as claimed in claim 2, characterized in that: Step 2 is as follows: (2.1) The two-dimensional airway boundary of the centrifugal impeller design sample includes the airway boundary of the casing surface located at the top and the airway boundary of the hub surface located at the bottom. The two-dimensional curve of the airway boundary of the casing surface consists of 100 two-dimensional coordinates. In the i-th two-dimensional curve of the airway boundary of the casing surface, 100 point coordinate samples are uniformly taken as A i , the hub surface airway boundary two-dimensional curve is also composed of 100 two-dimensional coordinates. In the i-th hub surface airway boundary two-dimensional curve, 100 point coordinate samples are uniformly taken as Note: B i ; (2.2) In the centrifugal impeller design, 100 points are uniformly selected from the continuous blade thickness distribution at 0% blade height of the i-th sample, and the array is used Indicated by Ha i ; At the 100% leaf height of the i-th sample, the continuous leaf thickness distribution is uniform, and 100 points are taken using an array Denotes Hb i ; (2.3) In the centrifugal impeller design, 100 points are uniformly selected from the continuous blade angle distribution at 0% blade height of the i-th sample, and the array is used Indicated by Aa i ; uniformly select 100 points from the continuous leaf angle distribution at 100% leaf height of the i-th sample and use the array Denotes Ab i .

4. The centrifugal impeller design verification method for a low-power civil aviation engine as claimed in claim 3, characterized in that: Step 3 is as follows: (3.1) Two-dimensional coordinate matrix A for airway boundary samples i and B i , reduced to one dimension using principal component analysis; (3.2) Calculate the matrix A i The covariance matrix C of C is calculated, and the eigenvalues ​​λ1 and λ2 of C are calculated. The eigenvalue λ with a covariance contribution rate greater than 90% is selected. k ; (3.3) Choose λ k The corresponding eigenvectors form the feature transformation matrix P k ∈R 1*2 ; (3.4) Matrix A i Dimensionality reduction transformation: The result after dimensionality reduction is recorded as As i , for the matrix B i Take the same operation to obtain the feature transformation matrix P m ∈R 1*2 , B i The result after dimensionality reduction is recorded as Bs i . (3.5) All A i (i=1,2……100) after dimensionality reduction, the matrix [As1,As2……As 100 ], denoted as As; (3.6) Calculate the covariance matrix C1 of the matrix As and calculate the eigenvalue λ of C1 a1 , a2 ... a100 , select two eigenvalues ​​λ whose sum of covariance contribution is greater than 90% i and λ j ; (3.7) Choose λ i and λ j The corresponding eigenvectors form the feature transformation matrix P k2 ∈R 2*100 ; (3.8) Dimension reduction transformation of matrix As: Therefore, the 2D curve sample of the airway boundary of the i-th casing surface is reduced to two eigenvalues: To B i Take the same operation to get the feature transformation matrix P m2 , the two-dimensional curve samples of the hub surface airway boundary are reduced to two eigenvalues: (3.9) The blade thickness and blade angle feature matrix Ha for the i-th sample i , Hb i , Aa i ,Ab i Repeat steps (3.6) to (3.8) to obtain the feature transformation matrix P x1 , P x2 , P x3 , P x4 , after dimensionality reduction, the leaf thickness distribution at 0% leaf height, the leaf thickness distribution at 100% leaf height, the leaf angle distribution at 0% leaf height, and the leaf angle distribution at 100% leaf height of the i-th sample are obtained as the reduced eigenvalues: (3.10) Therefore, the two-dimensional design continuous features of the i-th sample can be extracted into 12 eigenvalues, combined with the overall performance parameters of the engine, rated speed N (i) , inlet total temperature T (i) , total inlet pressure Outlet static pressure The design characteristics of centrifugal impeller sample i can be expressed by the following 16 characteristic parameter groups: Ps (i) ,i=1,2,……100, sample set matrix [Ps (1) ;Ps (2) ;……;Ps (100) ] T , denoted as Ps.

5. The centrifugal impeller design verification method for a low-power civil aviation engine as claimed in claim 4, characterized in that: Step 4 is as follows: The reduced-dimensional data sample set Ps is used as the input matrix, the simulation result matrix Op is used as the output matrix, and a hidden layer network structure is adopted. The hidden layer nodes are calculated according to the empirical formula Determine, where H is the number of hidden layer nodes, I is the number of input layer nodes, O is the number of output layer nodes, select H = 6; The weight matrix between the input layer and the hidden layer is recorded as The weight matrix between the hidden layer and the output layer is recorded as The bias vector from the input layer to the hidden layer is denoted by B = (B1, B2, ..., B6) T ; The bias vector between the hidden layer and the output layer is denoted by C = (C1, C2, C3) T ; For the weight matrix W (1) , W (2) And the bias vectors B and C are initialized and assigned, and the initialization value is a random number in the range of (0.5,1); The transfer function is selected as the Sigmoid function, that is Calculate the jth hidden layer value of the i-th sample Where W (1) is the weight matrix W (1) The jth row of Ps (i) is the i-th sample in the sample set Ps, B j is the jth value in the bias vector B; Similarly, calculate the j-th output layer value of the i-th sample Where W j (2) is the weight matrix W (2) The jth row of Hid (i) is the hidden layer output value vector of the i-th sample in the sample set, C j is the jth value in the bias vector C; Refer to the output sample Op in the training sample and calculate the i-th sample Op (i) and Output (i) The error between (i) : in is the i-th sample Op (i) The kth value in is the i-th sample Output (i) The kth value in ; The total (average) cost E of all training samples total for: The weights and biases are updated using batch gradient descent until the system error is minimized: Where μ is the learning rate, and the selection range is (0.05, 0.8) The neural network model after training is recorded as f BP (x),x∈R 16*1 .

6. The centrifugal impeller design verification method for a low-power civil aircraft engine as claimed in claim 5, characterized in that: Step 5 is as follows: For the new centrifugal impeller 3D model design, deconstruct the 2D geometric elements; Select 100 points evenly and equidistantly on the curve, and record the airway boundary data of the casing surface as A0∈R 100*2 , the hub surface airway boundary data is recorded as B0∈R 100*2 The blade angle of the centrifugal impeller at 0% blade height is recorded as Aa0∈R 100*1 , the blade angle of the centrifugal impeller at 100% blade height is recorded as Ab0∈R 100*1 , the blade thickness of the centrifugal impeller at 0% blade height is recorded as Ha0∈R 100*1 , the blade thickness of the centrifugal impeller at 100% blade height is recorded as Hb0∈R 100*1 ; Perform data dimensionality reduction on the above data: Add overall performance parameters Rated speed N (0) , inlet total temperature T (0) , total inlet pressure Outlet static pressure Construct a new input sample Ps (0) ∈R 16*1 : in, is the eigenvalue of the two-dimensional airway boundary of the new sample after dimensionality reduction, is the eigenvalue of the leaf thickness distribution at 0% leaf height and 100% leaf height of the new sample after dimensionality reduction, It is the eigenvalue of the leaf angle distribution at 0% leaf height and 100% leaf height of the new sample after dimensionality reduction; Substitute into the neural network function calculation, f BP (Ps (0) )=[π0,m0,η0]; Evaluate whether the output pressure ratio π0, flow rate m0, and efficiency η0 values ​​meet the design requirements. If so, continue with the three-dimensional flow field simulation verification; if not, redesign.

Images