Diffuser blade optimization design method based on mean camber line curvature control
Through the optimization design method of diffuser blades based on medium arc curvature control, the problems of uneven air flow and matching difficulties at high pressure ratio are solved, and good performance and reduced total pressure are achieved within a wider range of attack angles. Loss effect.
Patent Information
- Application Number
- CN202411442331.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-16
- Publication Date
- 2025-02-14
AI Technical Summary
The high pressure ratio leads to uneven air flow at the outlet of the centrifugal compressor and an increase in the number of Mach inlets, resulting in difficulty in matching between the diffuser impeller and the diffuser, reducing efficiency and working range.
Using the diffuser blade optimization design method based on the curvature control of the medium arc, the curve is generated by using the DeBoor recurrence algorithm to generate the curve and the integral is generated. The geometric angle and flexural angle distribution of the middle arc are introduced to ensure the consistency of the angle and realize direct control of the curvature of the middle arc.
Maintain good performance within a wider range of angle of attack, and the total pressure loss at the optimized design point is reduced by about 6.45%, the range of angle of attack is expanded, and the loss changes relatively smoothly with the angle of attack.
Smart Images
Figure CN119442509A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of rotary machinery, and in particular to a diffuser blade optimization design method based on mid-camber curvature control. Background Art
[0002] Centrifugal compressors have been widely used in superchargers due to their advantages such as high single-stage pressure ratio, wide operating range, simple and compact structure, few parts, and high reliability. The demand for modern high-performance superchargers has led to a continuous increase in the pressure ratio of compressors. However, the high pressure ratio makes the airflow at the centrifugal impeller outlet very uneven. In addition, the distance between the diffuser inlet and the centrifugal impeller outlet is very small, and there is a strong unsteady interaction between the two. At the same time, the high pressure ratio increases the radial diffuser inlet Mach number, and even supersonic sound occurs, further deteriorating the matching between the impeller and the diffuser, reducing the efficiency and operating range of the centrifugal compressor, and making the design of a compact and efficient diffuser very challenging.
[0003] When the centrifugal compressor is running at a near-stall operating point, the diffuser inlet airflow angle is a positive angle of attack, causing flow separation on its suction surface, which leads to stall or even surge of the compressor. In order to further expand the stable operating range of the centrifugal compressor, the flow separation at the near-stall operating point of the diffuser needs to be suppressed. In order to effectively suppress the flow separation in its channel, improve the performance of the entire stage, and increase the stall margin of the compressor, based on this, the present invention provides a diffuser optimization design method based on the mid-camber curvature control. Summary of the invention
[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a diffuser blade optimization design method based on mid-camber curvature control, which can maintain good performance in a wider range of attack angles.
[0005] The object of the present invention is achieved in that:
[0006] A diffuser blade optimization design method based on mid-camber curvature control.
[0007] Firstly, a cubic non-uniform B-spline curve is used to give the distribution of the curvature of the mid-arc along the arc length, and the DeBoor recursive algorithm is used to generate a cubic non-uniform B-spline curve.
[0008]
[0009] In the formula,
[0010] is the coordinate of the control point, k=3 is the degree of the curve, N j,k(u) is the B-spline basis function, u is the knot vector, the knot vector is generated by the HJ method, and the first and last knots are triple knots. The formula for calculating the knot value using the HJ method is:
[0011]
[0012] in,
[0013]
[0014] Where l j =|d i -d i-1 |, (i=1,2...,n) is the side length of the control polygon. After generating the node vector using formula (2), it is substituted into the recursive formula of formula (1) to generate the curvature distribution of the mid-arc;
[0015] The geometric meaning of curvature is the rotation rate of the unit tangent vector of the curve to the arc length, that is:
[0016]
[0017] Integrate the curvature along the arc length s once, and you will get the distribution of the mid-arc geometric angle β along the arc length and the turning angle Δβ of the blade. The geometric inlet angle and outlet angle of the blade are given in advance. When integrating the curvature distribution to generate the β angle distribution, the correction coefficient m is introduced to correct the curvature distribution to ensure that the turning angle Δβ obtained by integration is consistent with the geometric inlet and outlet angles, that is:
[0018]
[0019] The value of m is obtained after integration. After the distribution of the β angle along the arc length is obtained, another integration is performed to obtain the parameterized representation of the median arc, thereby realizing direct control of the curvature of the median arc.
[0020] Preferably, a distribution diagram of the curvature of the mid-arc line and the turning angle along the arc length is generated, in which the arc length s of the horizontal axis is dimensionless, and the thickness distribution is generated using a Bezier curve with 5 control points. The thickness distribution is superimposed on the mid-arc line to generate the suction surface and pressure surface profile lines, and the leading edge and trailing edge of the blade are processed.
[0021] Preferably, the thickness distribution law of the original blade profile is extracted for fitting and used as the input of parametric modeling. During optimization, the shape of the middle arc line is changed by controlling the middle arc line curvature distribution function, and the thickness distribution of the original blade profile is superimposed on the generated middle arc line to obtain an optimized blade profile.
[0022] Preferably, when selecting the optimization parameters, the control points P1, P6, and P7 are guaranteed to remain unchanged, and P2 (s2, κ2), P3 (s3, κ3), P4 (s4, κ4), and P5 (s5, κ5) are determined as the optimized geometric parameters of the blade. According to the blade modeling algorithm, the front and rear arc length positions cannot cross and overlap, and the front and rear threshold parameters M2, M3, and M4 are set for the arc length position; in order to avoid excessive curvature of the local mid-arc line, the curvature thresholds K2, K3, K4, and K5 are also set accordingly, and the constraints are:
[0023]
[0024] Preferably, the parameters in the constraint conditions are all empirical parameters of blade shaping, which are given based on experience.
[0025] Due to the adoption of the above technical solution, the present invention has the following beneficial effects:
[0026] The total pressure loss at the optimized design point (the angle of attack on the horizontal axis is 0) is reduced from 0.0186 to 0.0174, a decrease of about 6.45%; the angle of attack range is also greatly increased, and the loss of the optimized blade shape changes more slowly with the angle of attack. In other words, the optimized blade shape can maintain better performance in a wider range of angles of attack. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] Figure 1 The figure is a meridian view of the impeller and diffuser, including mainstream blades, splitter blades and diffusers, and the arrow direction in the figure is the airflow direction;
[0028] Figure 2 is the distribution diagram of curvature and turning angle;
[0029] Figure 3 To fit the original leaf shape diagram;
[0030] Figure 4 This is a comparison of blade profiles before and after optimization;
[0031] Figure 5 This is a comparison chart of total pressure loss. DETAILED DESCRIPTION
[0032] A diffuser blade optimization design method based on mid-arc curvature control is proposed. Taking into account that the changing law of the streamline curvature in the blade channel is mainly affected by the changing law of the curvature of the mid-arc line, the mid-arc line curvature distribution is directly given and then integrated to generate the mid-arc line. This can achieve direct control of the mid-arc line curvature, thereby achieving a more refined design of the blade profile.
[0033] In order to facilitate the flexible control of the curvature distribution of the middle arc, a cubic non-uniform B-spline curve is first used to give the distribution of the curvature of the middle arc along the arc length. The curvature distribution of the middle arc defined by the cubic non-uniform B-spline curve has the property of second-order differentiability, which ensures that the curvature of the middle arc of the blade is high-order differentiable, which is conducive to the refined control of the flow. The De Boor recursive algorithm is used to generate the cubic non-uniform B-spline curve.
[0034]
[0035] In the formula,
[0036] is the coordinate of the control point, k=3 is the degree of the curve, N j,k (u) is the B-spline basis function, and u is the knot vector. The knot vector is generated using the HJ method, and the first and last knots are triple knots. The formula for calculating the knot value using the HJ method is:
[0037]
[0038] in,
[0039]
[0040] Where l j =|d i -d i-1 |, (i=1,2...,n) is the side length of the control polygon. After generating the node vector using formula (2), it is substituted into the recursive formula of formula (1) to generate the curvature distribution of the mid-arc.
[0041] Considering that the geometric meaning of curvature is the rotation rate of the unit tangent vector of the curve to the arc length, that is,
[0042]
[0043] By integrating the curvature along the arc length s, we can get the distribution of the geometric angle β of the mid-arc along the arc length and the turning angle Δβ of the blade. In actual blade design, the geometric inlet angle and outlet angle of the blade are often given in advance. Therefore, when integrating the curvature distribution to generate the β angle distribution, the correction coefficient m is introduced to correct the curvature distribution to ensure that the turning angle Δβ obtained by integration is consistent with the geometric inlet and outlet angles, that is,
[0044]
[0045] The value of m can be obtained after integration. After obtaining the distribution of the β angle along the arc length, another integration can be performed to obtain the parameterized representation of the median arc. The distribution of the median arc curvature and the turning angle along the arc length generated by the method in this paper is as follows: Figure 2As shown in the figure, the arc length s of the horizontal axis is dimensionless. The thickness distribution is generated using a Bezier curve with 5 control points. The thickness distribution is superimposed on the mid-arc line to generate the suction surface and pressure surface profiles, and the leading edge and trailing edge of the blade are processed.
[0046] In order to verify the rationality of the parametric modeling method in this paper, the target blade profile is fitted using the patented method. The fitting results are shown in Figure 3 As shown in the figure, it can be seen that the difference between the blade thickness fitted by the method of this patent and the original blade thickness is only 0.2%, and the method is feasible.
[0047] This patent focuses on the influence of the curvature distribution law of the arc on the performance. Therefore, the thickness distribution law of the original blade is extracted during fitting as the input of parameterized modeling. During optimization, the shape of the middle arc is changed by controlling the curvature distribution function of the middle arc, and the thickness distribution of the original blade is superimposed on the generated middle arc to obtain the optimized blade. When selecting the optimization parameters, the control points P1, P6, and P7 are kept unchanged, and P2 (s2, κ2), P3 (s3, κ3), P4 (s4, κ4), and P5 (s5, κ5) are determined as the optimized geometric parameters of the blade. According to the blade modeling algorithm, the front and rear arc length positions cannot cross and overlap, so the front and rear threshold parameters M2, M3, and M4 need to be set for the arc length position; in order to avoid excessive curvature of the local middle arc, the curvature thresholds K2, K3, K4, and K5 are also set accordingly; these values are empirical parameters for blade modeling to avoid "optimized irregularities". Its constraints are shown in formula (5).
[0048]
[0049] Figure 4 A comparison between the optimized blade shape and the original blade shape is given. Figure 5 A comparison of the angle of attack-total pressure loss characteristics of the optimized blade type and the original blade type is given. It can be seen from the figure that the total pressure loss at the optimized design point (the angle of attack on the horizontal axis is 0) is reduced from 0.0186 to 0.0174, a reduction of about 6.45%; the angle of attack range is also greatly increased, and the loss of the optimized blade type changes more slowly with the angle of attack. In other words, the optimized blade type can maintain better performance in a wider angle of attack range.
[0050] Figure 5 A comparison of the angle of attack-total pressure loss characteristics of the optimized blade type and the original blade type is given. It can be seen from the figure that the total pressure loss at the optimized design point (the angle of attack on the horizontal axis is 0) is reduced from 0.0186 to 0.0174, a reduction of about 6.45%; the angle of attack range is also greatly increased, and the loss of the optimized blade type changes more slowly with the angle of attack. In other words, the optimized blade type can maintain better performance in a wider angle of attack range.
[0051] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made in form and details without departing from the scope defined by the claims of the present invention.
Claims
1. A diffuser blade optimization design method based on mid-camber curvature control, characterized in that: Firstly, a cubic non-uniform B-spline curve is used to give the distribution of the curvature of the mid-arc along the arc length, and the De Boor recursive algorithm is used to generate the cubic non-uniform B-spline curve. In the formula, is the coordinate of the control point, k=3 is the degree of the curve, N j,k (u) is the B-spline basis function, u is the knot vector, the knot vector is generated by the HJ method, and the first and last knots are triple knots. The formula for calculating the knot value using the HJ method is: in, Where l j =|d i -d i-1 |, (i=1,2...,n) is the side length of the control polygon. After generating the node vector using formula (2), it is substituted into the recursive formula of formula (1) to generate the curvature distribution of the mid-arc; The geometric meaning of curvature is the rotation rate of the unit tangent vector of the curve to the arc length, that is: Integrate the curvature along the arc length s once, and you will get the distribution of the mid-arc geometric angle β along the arc length and the turning angle Δβ of the blade. The geometric inlet angle and outlet angle of the blade are given in advance. When integrating the curvature distribution to generate the β angle distribution, the correction coefficient m is introduced to correct the curvature distribution to ensure that the turning angle Δβ obtained by integration is consistent with the geometric inlet and outlet angles, that is: The value of m is obtained after integration. After the distribution of the β angle along the arc length is obtained, another integration is performed to obtain the parameterized representation of the median arc, thereby realizing direct control of the curvature of the median arc.
2. The diffuser blade optimization design method based on mid-camber curvature control according to claim 1, characterized in that: Generate the distribution diagram of the curvature of the mid-arc line and the turning angle along the arc length. The arc length s of the horizontal axis in the figure is dimensionless. The thickness distribution is generated using a Bezier curve with 5 control points. The thickness distribution is superimposed on the mid-arc line to generate the suction surface and pressure surface profile lines, and the leading edge and trailing edge of the blade are processed.
3. The diffuser blade optimization design method based on mid-camber curvature control according to claim 2, characterized in that: The thickness distribution law of the original blade profile is extracted for fitting and used as the input of parametric modeling. During optimization, the shape of the middle arc line is changed by controlling the curvature distribution function of the middle arc line. The thickness distribution of the original blade profile is superimposed on the generated middle arc line to obtain the optimized blade profile.
4. The diffuser blade optimization design method based on mid-camber curvature control according to claim 3 is characterized in that: When selecting the optimization parameters, the control points P1, P6, and P7 are guaranteed to remain unchanged, and P2 (s2, κ2), P3 (s3, κ3), P4 (s4, κ4), and P5 (s5, κ5) are determined as the optimized geometric parameters of the blade. According to the blade modeling algorithm, the front and rear arc length positions cannot cross and overlap, and the front and rear threshold parameters M2, M3, and M4 are set at the arc length position; in order to avoid excessive curvature of the local mid-arc, the curvature thresholds K2, K3, K4, and K5 are also set accordingly. The constraints are:
5. The method for optimizing diffuser blade design based on mid-camber curvature control according to claim 4, characterized in that: The parameters in the constraint conditions are all empirical parameters of blade modeling and are given based on experience.