A method for correcting the blade shape of an axial-flow cooling fan for a fuel cell vehicle

Abstract

This invention discloses a method for correcting the blade shape of an axial flow cooling fan for fuel cell vehicles, comprising the following steps: 1) constructing a thin blade design model for the cooling fan; 2) performing mesh discretization by equating the thin blade model to a curved surface; 3) performing static analysis to solve for the deformation based on the blade's operating conditions and steady-state pressure distribution; 4) calculating the mean square error of the error matrix after deformation and determining whether the conditions are met; 5) solving for the nodal coordinate correction and iterating; 6) finally obtaining the corrected blade shape parameters. This invention can solve the problem of steady-state deformation of fan blades caused by airflow and high-speed rotation during operation. Using the corrected blade shape parameters, the morphology of the blades during operation can be guaranteed to meet design requirements, thus improving its performance.

Description

Technical Field

[0001] This invention relates to the field of cooling fan design, and more particularly to a method for correcting the blade shape of an axial cooling fan for fuel cell vehicles. Background Technology

[0002] Axial flow fans are widely used in light industry, food processing, metallurgy, chemical industry, and civil construction for ventilation and heat dissipation. When the impeller rotates, gas enters axially from the inlet, its energy is increased by the pushing action of the impeller blades, and then it flows into the guide vanes. However, during operation, due to complex and variable operating conditions, the fan blades undergo a certain amount of steady-state deformation and vibration displacement. Many scholars have discussed vibration reduction measures for blades. Vibration dampers can be installed to reduce vibration and noise, but this method is complex and difficult to implement. Blade perforation can be used, but an excessively large perforation coefficient can cause the pressure difference to drop too quickly, failing to meet the design energy requirements.

[0003] Currently, there is limited research on the steady-state deformation of wind turbine blades under high-speed rotation. Since wind turbine blades are designed according to optimal design principles, during operation, they are subjected to airflow and high-speed rotation. This causes a certain degree of shift in the blade's coordinates between its operational equilibrium state and its initial state, resulting in a difference between the blade's working shape and its initial design shape. Consequently, the blade cannot operate in its optimal design form, thus affecting its performance. Most scholars focus on the optimal design of the wind turbine blade shape to achieve comprehensive optimal performance, neglecting the deformation problem under operational equilibrium. Therefore, a compensation and correction scheme for blade deformation has not yet been developed. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a method for correcting the blade shape of an axial flow cooling fan for fuel cell vehicles. This method aims to modify the blade shape while considering fluid pressure and steady-state deformation, ensuring that the blade's morphology meets design requirements during operation.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] A method for correcting the blade shape of an axial flow cooling fan for a fuel cell vehicle, characterized by comprising the following steps:

[0007] S1. Digitize the wind turbine blade shape to be corrected to obtain the coordinate matrix A representing the blade shape, which is used as the blade shape matrix to be corrected.

[0008] S2: Perform fluid steady-state analysis calculations on the blade shape coordinate matrix to be corrected under the working conditions of the blade to obtain the steady-state pressure distribution on the blade surface;

[0009] S3. Combining the working conditions of the blade surface, through static analysis, analyze the blade shape deformation data caused by the pressure in the pressure distribution area being applied to the blade, and obtain the coordinate matrix of the deformed blade shape.

[0010] S3. Obtain the error matrix by subtracting the coordinate matrix of the deformed blade shape from the coordinate matrix A.

[0011] S4. Calculate the mean square error value based on the error matrix. When the error is less than the threshold, use the coordinate information in the blade forming coordinate matrix in step S2 as the final corrected shape parameter of the blade; otherwise, proceed to step S5.

[0012] Step S5: Correct the coordinates in the deformed blade shape coordinate matrix obtained in step S3 to obtain the corrected coordinate matrix as the new blade shape coordinate matrix to be corrected, and then return to step S2 to execute sequentially.

[0013] Step S1 includes:

[0014] S11: Obtain the thin blade model of the cooling fan to be corrected;

[0015] S12. Construct a blade coordinate system, quantize the blade shape into coordinates, obtain a coordinate point representing the blade shape, and establish a coordinate matrix A of the blade shape based on the coordinate point.

[0016] Step S12 includes: converting the constructed thin blade model into a curved surface, and establishing a coordinate system OXYZ with the center of the blade hub as the origin O, the X-axis direction parallel to the blade extension direction as the X-axis direction, the Y-axis direction perpendicular to the blade extension direction as the Y-axis direction, and the Z-axis direction parallel to the center axis of the blade hub as the Z-axis direction.

[0017] The equivalent surface of the blade is discretized into a mesh of N nodes:

[0018]

[0019] In the formula: A i Let x be the coordinate matrix of the i-th discrete node; i y is the value of node i on the horizontal axis; i z is the value of node i on the vertical axis; i Let i be the value of node i on the vertical axis;

[0020] And store the coordinates of N nodes in the desired blade shape coordinate matrix A:

[0021] A = [A1, A2, ..., A i ,…,A N ].

[0022] Step S1 also includes defining the initial pre-deformation node matrix as B. 0 =A;

[0023] Let k be the iteration number of the current node, and initialize k = 1. Let K be the total number of iterations. Then, the node matrix before the transformation in the (k-1)th iteration is B. k-1 .

[0024] Step S2 includes: based on the blade's operating conditions, performing fluid steady-state analysis on the blade shape model corresponding to the node matrix before the (k-1)th iteration deformation, obtaining the steady-state pressure distribution on the blade surface, and storing the pressure results in the pressure distribution matrix P:

[0025] P = [p1, p2, ..., p i ,…,p N ]

[0026] In the formula: P i This represents the pressure data corresponding to the i-th node.

[0027] Step S3 includes: combining the operating conditions of the blade, loading the pressure distribution matrix P onto each node before deformation, performing static analysis to obtain the node coordinates after deformation, and defining the node coordinate matrix after deformation as C. k .

[0028] Calculate the error matrix E of the k-th iteration node. k :

[0029] E k =[e1 k e2 k ,…,e i k ,…,e N k ] = C k -A

[0030] In the formula: e i k This represents the error corresponding to the i-th node in the k-th iteration.

[0031] Step S4 includes solving for the mean square error of the error matrix in the k-th iteration, and dividing each e in the error matrix... i k The mean square error is obtained by solving for the error, and the mean square error is defined as rms.

[0032] Step S5 includes:

[0033] Calculate the correction amount u of the coordinates of the i-th node in the k-th iteration. i k :

[0034]

[0035] Then the correction amount for the coordinates of N nodes in the k-th iteration is U. k :

[0036] U k =[u1 k u2 k ,…,u i k ,…,u N k ]

[0037] Correct the node coordinate matrix B before the k-th iteration deformation k :

[0038] B k =B k-1 +U k

[0039] The modified node coordinate matrix B before the k-th iteration deformation. k Return to step S2 and execute sequentially.

[0040] When the iteration number k is 1 or 2,

[0041] Correction amount u i k :u i k =-e i k .

[0042] The advantages of this invention are as follows: 1. This invention proposes a method for correcting the blade shape of an axial flow cooling fan for fuel cell vehicles. Based on the working conditions of the blade and considering the steady-state deformation caused by fluid pressure and high-speed rotation, the shape of the hot air fan blade is iteratively corrected based on the correction amount proposed in this invention. This ensures that the shape of the blade meets the design requirements during the working process and improves the working performance of the blade.

[0043] 2. In the iterative process, this invention updates the correction amount for the next iteration based on the correction amount of the previous step and takes into account the error trend in each iteration. This allows for faster calculation of the final blade shape parameters, reduces the number of calculations and the cycle, and improves the efficiency of wind turbine blade R&D. Attached Figure Description

[0044] The following is a brief explanation of the contents of each of the accompanying drawings and the markings in the drawings:

[0045] Figure 1 This is a schematic diagram of the axial flow cooling fan for fuel cell vehicles in this invention;

[0046] Figure 2This is the coordinate system of the blade model constructed in this invention;

[0047] Figure 3 This is a flowchart illustrating the principle of wind turbine blade correction in this invention;

[0048] Figure 4 This is a detailed flowchart of the wind turbine blade correction method in this invention; Detailed Implementation

[0049] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and the description of the preferred embodiments.

[0050] Example 1:

[0051] like Figure 1-4 As shown in this embodiment, a method for correcting the blade shape of an axial flow cooling fan for a fuel cell vehicle includes the following steps:

[0052] Step 1: As Figure 1 As shown, a thin blade model of a cooling fan is constructed. The blade shape in this model is the blade shape under a relatively ideal state obtained during the design.

[0053] Step 2: To reduce modeling difficulty and improve computational efficiency, the constructed thin blade model is equivalent to a curved surface, such as... Figure 2 As shown, a coordinate system OXYZ is established with the center of the blade hub as the origin O, the X-axis parallel to the blade extension direction as the X-axis, the Y-axis perpendicular to the blade extension direction as the Y-axis, and the Z-axis parallel to the center axis of the blade hub as the Z-axis.

[0054] Step 3: To achieve boundary fitting of the region, applicable to calculations involving fluid dynamics and surface stress concentration, the equivalent surface of the blade is discretized into a mesh of N nodes:

[0055]

[0056] In formula (1): A i Let x be the coordinate matrix of the i-th discrete node; i y is the value of node i on the horizontal axis; i z is the value of node i on the vertical axis; i Let i be the value of node i on the vertical axis; Discretizing into N points means dividing the leaf into multiple tiny squares of the same size. The center point of each square is taken as the coordinate point of the square, and the coordinate position is represented by this point. The area of ​​the square should be extremely small, and the smaller the square, the higher the accuracy.

[0057] Furthermore, the coordinates of N nodes are stored in the desired blade shape matrix A, which is also the design shape required under operating conditions.

[0058] A = [A1, A2, ..., A i ,…,A N (2)

[0059] Step 4: Define the initial pre-deformation node matrix as B. 0 =A, using the initial design form to simulate and solve for the deformation under steady-state conditions;

[0060] Let k be the iteration number of the current node, and initialize k = 1. Let K be the total number of iterations. Then, the node matrix before the transformation in the (k-1)th iteration is B. k-1 ;

[0061] Step 5: Based on the blade's operating conditions, perform a fluid steady-state analysis on the nodal model before deformation in the (k-1)th iteration to obtain the steady-state pressure distribution on the blade surface, and store the pressure results in the pressure distribution matrix P:

[0062] P = [p1, p2, ..., p i ,…,p N (3)

[0063] In equation (3): P i This represents the pressure data corresponding to the i-th node;

[0064] Step 6: Based on the operating conditions of the blades, and by loading the pressure distribution matrix P onto each node before deformation, the wind turbine blades have reached equilibrium under operating conditions. Therefore, static analysis can be performed to obtain the node coordinates after deformation, and the node coordinate matrix after deformation is defined as C. k ;

[0065] Step 7: Subtract the expected matrix from the node matrix after the k-th iteration transformation to obtain the error matrix of the k-th iteration node. Calculate the error matrix E of the k-th iteration node using equation (4). k :

[0066] E k =[e1 k e2 k ,…,e i k ,…,e N k ] = C k -A (4)

[0067] In equation (4): e i k This represents the error corresponding to the i-th node in the k-th iteration;

[0068] Step 8: Solve for the mean square error of the error matrix of the k-th iteration, and define the mean square error as rms;

[0069] Step 9: Compare the mean square error rms of the k-th iteration with the specified error ε. If rms < ε, jump to step 12; otherwise, continue to step 10.

[0070] Step 10: Calculate the correction amount u of the coordinates of the i-th node in the k-th iteration using equation (5). i k :

[0071]

[0072] For the first and second iterations, the error matrix is ​​directly used as the correction quantity according to equation (6):

[0073] u i k =-e i k (6)

[0074] However, for the third iteration, it is necessary to consider the error trend in the previous iteration and use the error matrix information of each iteration to correct the trend. That is, use equation (5) to update the correction amount of the next iteration, which can greatly improve the computational efficiency.

[0075] Then the correction amount for the coordinates of N nodes in the k-th iteration is U. k :

[0076] U k =[u1 k u2 k ,…,u i k ,…,u N k (7)

[0077] Step 11: Use equation (8) to correct the node coordinate matrix B before the k-th iteration deformation. k :

[0078] B k =B k-1 +U k (8)

[0079] Return the corrected node coordinate matrix before the kth iteration deformation to step 5 for sequential execution;

[0080] Step 12: Set the node coordinates B before the (k-1)th iteration deformation. k-1 The output is the final blade shape parameter. When the blade reaches the working equilibrium state under this parameter, the resulting steady-state working shape is exactly the same as the desired working shape.

[0081] Example 2:

[0082] A method for correcting the blade shape of an axial cooling fan for a fuel cell vehicle includes the following steps:

[0083] S1. Digitize the wind turbine blade shape to be corrected to obtain the coordinate matrix A representing the blade shape, which is used as the blade shape matrix to be corrected.

[0084] Step S1 includes:

[0085] S11: Obtain the thin blade model of the cooling fan to be corrected;

[0086] S12. Construct a blade coordinate system, quantize the blade shape into coordinates, obtain a coordinate point representing the blade shape, and establish a coordinate matrix A of the blade shape based on the coordinate point.

[0087] Step S12 includes: converting the constructed thin blade model into a curved surface, and establishing a coordinate system OXYZ with the center of the blade hub as the origin O, the X-axis direction parallel to the blade extension direction as the X-axis direction, the Y-axis direction perpendicular to the blade extension direction as the Y-axis direction, and the Z-axis direction parallel to the center axis of the blade hub as the Z-axis direction.

[0088] The equivalent surface of the blade is discretized into a mesh of N nodes:

[0089]

[0090] In the formula: A i Let x be the coordinate matrix of the i-th discrete node; i y is the value of node i on the horizontal axis; i z is the value of node i on the vertical axis; i Let i be the value of node i on the vertical axis;

[0091] And store the coordinates of N nodes in the desired blade shape coordinate matrix A:

[0092] A = [A1, A2, ..., A i ,…,A N ].

[0093] Iterate through steps S2-S5 until the corrected blade coordinate parameters are output. Where:

[0094] Step S2 includes:

[0095] S21. The blade shape coordinate matrix to be corrected is simulated iteratively to simulate the shape change of the blade under real working conditions. Matrix B represents the node coordinate matrix before deformation, and C represents the matrix after deformation. Define the iteration number of the current node as k, and initialize k = 1. The total number of iterations is K. Then, the node matrix before deformation in the (k-1)th iteration is B. k-1 .

[0096] Define the initial pre-deformation node matrix as B. 0 =A;B 0 The first is the coordinate matrix of the blade shape to be corrected. After multiple iterations, the node matrix before deformation in the (k-1)th iteration is B. k-1 , which serves as the coordinate matrix of the blade shape to be corrected;

[0097] S22. For the blade shape coordinate matrix to be corrected, perform fluid steady-state analysis calculations under the working conditions of the blade to obtain the steady-state pressure distribution on the blade surface.

[0098] Based on the blade's operating conditions, a fluid steady-state analysis is performed on the blade shape model corresponding to the node matrix before the (k-1)th iteration deformation to obtain the steady-state pressure distribution on the blade surface, and the pressure results are stored in the pressure distribution matrix P:

[0099] P = [p1, p2, ..., p i ,…,p N ]

[0100] In the formula: P i This represents the pressure data corresponding to the i-th node.

[0101] S3. Based on the operating conditions of the blade surface, static analysis is used to analyze the blade shape deformation data caused by the pressure distribution matrix P applied to the blade before deformation, obtaining the coordinate matrix of the deformed blade shape. Then, based on the operating conditions of the blade, the pressure distribution matrix P is applied to each node before deformation, and static analysis is performed to obtain the coordinates of the deformed nodes. The coordinate matrix of the deformed nodes is defined as C. k C k and B k-1 It is a pair of matrices with the same number of rows and columns before and after the transformation.

[0102] S3. Obtain the error matrix by subtracting the coordinate matrix of the deformed blade shape from the coordinate matrix A; calculate the error matrix E of the k-th iteration node. k :

[0103] E k =[e1 k e2 k ,…,e i k ,…,e N k ] = C k -A

[0104] In the formula: e i k Let C be the error corresponding to the i-th node in the k-th iteration. k and B k-1 Subtracting the values ​​at the same position in the matrix yields the error value e at that position.i k These error values ​​are then combined into an error matrix.

[0105] S4. Calculate the mean square error based on the error matrix. If the error is less than the threshold, use the coordinate information in the blade forming coordinate matrix from step S2 as the final corrected shape parameters of the blade; otherwise, proceed to step S5. Solve for the mean square error corresponding to the error matrix from step S3, based on each e in the error matrix. i k The error is calculated to obtain the mean square error (RMS), which is defined as RMS. The RMS error of the (k-1)th iteration, RMS, is compared with a specified error ε. If RMS < ε, the coordinate information in the blade forming coordinate matrix from step S2 is used as the final corrected shape parameter of the blade, which is B. k-1 The blade coordinate parameters are used as the final shape parameters of the blade; otherwise, proceed to step S5.

[0106] Step S5: Correct the coordinates in the deformed blade shape coordinate matrix obtained in step S3 to obtain the corrected coordinate matrix as the new blade shape coordinate matrix to be corrected, and then return to step S2 to execute sequentially.

[0107] After proceeding to step S5, explain coordinate matrix B. k-1 The blade shape, as represented, deforms under downward pressure during actual operation, resulting in a value Ck that far exceeds the set range and does not meet the requirements. Therefore, a further iteration is needed to adjust the coordinate matrix B. k-1 Each coordinate value in the matrix is ​​adjusted by adding a correction factor to obtain the node coordinate matrix B before the next (Kth) deformation. k B k The coordinate matrix of the blade shape to be corrected is fed into step 2. Steps 2, 3, 4, and 5 are executed sequentially until the mean square error requirement is met. Then, the coordinate matrix before deformation in this iteration is output. The coordinate information of each coordinate node summarized in the coordinate matrix is ​​used as the blade shape parameters to output the final blade shape parameters.

[0108] coordinate matrix B k-1 Each coordinate value in the equation is supplemented with a correction factor. The method for calculating the correction factor is as follows:

[0109] When K is greater than 2 in the iteration, calculate the correction amount u of the coordinates of the i-th node in the k-th iteration. i k :

[0110]

[0111] Then the correction amount for the coordinates of N nodes in the k-th iteration is U. k :

[0112] U k =[u1 k u2 k ,…,u i k ,…,u N k ]

[0113] Correct the node coordinate matrix B before the k-th iteration deformation k :

[0114] B k =B k-1 +U k

[0115] The modified node coordinate matrix B before the k-th iteration deformation. k Return to step S2 and execute sequentially.

[0116] When the number of iterations k is 1 or 2

[0117] Correction amount u i k :u i k =-e i k .

[0118] The correction matrix U can be obtained using the above method. k :

[0119] U k =[u1 k u2 k ,…,u i k ,…,u N k ]

[0120] Then the coordinate matrix B before this deformation k-1 Add the correction matrix U k This will give us the coordinate matrix B before the deformation in the next iteration. k The blade shape coordinate matrix to be corrected is fed into step 2 and executed sequentially until the final coordinate matrix representing the blade shape is output, thus completing the blade shape correction. The coordinate information of each coordinate point in the output coordinate matrix can represent the blade shape.

[0121] Obviously, the specific implementation of this invention is not limited to the above-described methods. Any non-substantial improvements made using the inventive concept and technical solution of this invention are within the protection scope of this invention.

Claims

1. A method for correcting the blade shape of an axial flow cooling fan for a fuel cell vehicle, characterized in that: Includes the following steps: S1. Digitize the wind turbine blade shape to be corrected to obtain the coordinate matrix A representing the blade shape, which is used as the blade shape matrix to be corrected. S2: Perform fluid steady-state analysis calculations on the blade shape coordinate matrix to be corrected under the working conditions of the blade to obtain the steady-state pressure distribution on the blade surface; S3. Combining the working conditions of the blade surface, through static analysis, analyze the blade shape deformation data caused by the pressure in the pressure distribution area being applied to the blade, and obtain the coordinate matrix of the deformed blade shape. S3. Obtain the error matrix by subtracting the coordinate matrix of the deformed blade shape from the coordinate matrix A. S4. Calculate the mean square error value based on the error matrix. When the error is less than the threshold, use the coordinate information in the coordinate matrix of the blade to be corrected in step S2 as the final corrected shape parameter of the blade. Otherwise, proceed to step S5; Step S5: Correct the coordinates in the deformed blade shape coordinate matrix obtained in step S3 to obtain the corrected coordinate matrix as the new blade shape coordinate matrix to be corrected, and then return to step S2 to execute sequentially.

2. The method for correcting the blade shape of an axial flow cooling fan for a fuel cell vehicle as described in claim 1, characterized in that: Step S1 includes: S11: Obtain the thin blade model of the cooling fan to be corrected; S12. Construct a blade coordinate system, quantize the blade shape into coordinates, obtain a coordinate point representing the blade shape, and establish a coordinate matrix A of the blade shape based on the coordinate point.

3. The method for correcting the blade shape of an axial flow cooling fan for a fuel cell vehicle as described in claim 2, characterized in that: Step S12 includes: converting the constructed thin blade model into a curved surface, and establishing a coordinate system OXYZ with the center of the blade hub as the origin O, the X-axis direction parallel to the blade extension direction as the X-axis direction, the Y-axis direction perpendicular to the blade extension direction as the Y-axis direction, and the Z-axis direction parallel to the center axis of the blade hub as the Z-axis direction. The equivalent surface of the blade is discretized into a mesh of N nodes: In the formula: A i Let x be the coordinate matrix of the i-th discrete node; i y is the value of node i on the horizontal axis; i z is the value of node i on the vertical axis; i Let i be the value of node i on the vertical axis; And store the coordinates of N nodes in the desired blade shape coordinate matrix A: A=[A1,A2,…,A i ,…,A N ]。 4. The method for correcting the blade shape of an axial flow cooling fan for a fuel cell vehicle as described in claim 1, characterized in that: Step S1 also includes defining the initial pre-deformation node matrix as B. 0 =A; Let k be the iteration number of the current node, and initialize k = 1. Let K be the total number of iterations. Then, the node matrix before the transformation in the (k-1)th iteration is B. k-1 .

5. The method for correcting the blade shape of an axial flow cooling fan for a fuel cell vehicle as described in claim 4, characterized in that: Step S2 includes: based on the blade's operating conditions, performing fluid steady-state analysis on the blade shape model corresponding to the node matrix before the (k-1)th iteration deformation, obtaining the steady-state pressure distribution on the blade surface, and storing the pressure results in the pressure distribution matrix P: P=[p1,p2,…,p i ,…,p N ] In the formula: P i This represents the pressure data corresponding to the i-th node.

6. The method for correcting the blade shape of an axial cooling fan for a fuel cell vehicle as described in claim 4, characterized in that: Step S3 includes: combining the operating conditions of the blade, loading the pressure distribution matrix P onto each node before deformation, performing static analysis to obtain the node coordinates after deformation, and defining the node coordinate matrix after deformation as C. k .

7. The method for correcting the blade shape of an axial flow cooling fan for a fuel cell vehicle as described in claim 6, characterized in that: Calculate the error matrix E of the k-th iteration node. k : THE k =[e1 k ,e2 k ,…,the i k ,…,the N k ]=C k -A In the formula: e i k This represents the error corresponding to the i-th node in the k-th iteration.

8. The method for correcting the blade shape of an axial flow cooling fan for a fuel cell vehicle as described in claim 7, characterized in that: Step S4 includes solving for the mean square error of the error matrix in the k-th iteration, and dividing each e in the error matrix... i k The mean square error is obtained by solving for the error, and the mean square error is defined as rms.

9. The method for correcting the blade shape of an axial flow cooling fan for a fuel cell vehicle as described in claim 7, characterized in that: Step S5 includes: Calculate the correction amount u of the coordinates of the i-th node in the k-th iteration. i k : Then the correction amount for the coordinates of N nodes in the k-th iteration is U. k : U k =[u1 k ,u2 k ,…,u i k ,…,u N k ] Correct the node coordinate matrix B before the k-th iteration deformation k : B k =B k-1 +U k The modified node coordinate matrix B before the k-th iteration deformation. k Return to step S2 and execute sequentially.

10. The method for correcting the blade shape of an axial flow cooling fan for a fuel cell vehicle as described in claim 9, characterized in that: When the iteration number k is 1 or 2, Correction amount u i k :u i k =-e i k .

Images