A flow design method for axial compressor based on three-dimensional data scaling

Abstract

The present invention provides an axial flow compressor flow design method based on three-dimensional data scaling, comprising: (1) initial design; (2) solving an initial two-dimensional flow design; (3) parameter scaling based on the three-dimensional data; (4) two-dimensional flow design based on the scaled data; and (5) implementing iterations of steps (3)-(4) through the scaled design to obtain a final two-dimensional flow design scheme. The present invention uses a cross-dimensional scaling method to achieve refined modeling and performance prediction of the axial flow compressor end region, thereby shortening the design cycle of the axial flow compressor while ensuring design quality; transferring high-dimensional data to low-dimensional data can improve the refinement of the low-dimensional compressor design and thereby enhance the accuracy of the axial flow compressor design or the axial flow compressor performance prediction, thereby making the initial low-dimensional flow field calculation more consistent with the three-dimensional calculation results, and making the designed compressor have better aerodynamic performance.

Description

Technical Field

[0001] The present invention specifically relates to a flow design method for an axial flow compressor based on three-dimensional data scaling, which is applied in the field of aerodynamic design of an axial flow compressor of a gas turbine. Background Art

[0002] Since the concept of multi-stage axial compressors was proposed, researchers have conducted research on axial compressors and their design methods. Especially driven by the development of aircraft and aero engines in the past seventy years, the design technology of axial compressors, as core components, has achieved great success. As gas turbines continue to develop towards high power, large flow and high pressure ratio, axial compressors, as one of the important components, increasingly require more advanced and more refined design methods. However, as the requirements for axial compressor performance indicators continue to rise, traditional axial compressor design technology is also facing huge challenges. Axial compressor design is inseparable from numerical simulation technology. Studies have shown that the application of simulation technology can reduce research and development costs and reduce the development cycle by more than one third.

[0003] Numerical simulation techniques for modern axial compressor design primarily include one-dimensional, two-dimensional, quasi-three-dimensional, and three-dimensional designs. In theory, full three-dimensional design is the most accurate design approach with the fewest adjustments. However, due to limitations in computational resources, design cycles, and the level of numerical computation development, using three-dimensional computation from the initial design stage is difficult to implement, and design still requires starting from a low-dimensional perspective. Traditional design methods, which progress from one-dimensional, two-dimensional, quasi-three-dimensional, to three-dimensional modeling and three-dimensional flow field numerical simulation, not only require extensive design experience but also require significant design time for parameter adjustment and three-dimensional flow field numerical simulation. This is due to the fact that three-dimensional flow losses are not fully considered in the early stages of design, resulting in poor flow field matching at different design stages and the need for repeated blade adjustments. Furthermore, the three-dimensional flow field numerical simulation stage accounts for a significant portion of the design process, leading to a long design cycle. Therefore, new design methods are needed to account for three-dimensional flow losses during the design process, improve design efficiency, and shorten the design cycle. Summary of the Invention

[0004] The purpose of the present invention is to provide an axial flow compressor flow design method based on three-dimensional data scaling to solve the spanwise flow loss calculation problem in the two-dimensional flow design of an axial flow compressor.

[0005] The technical solutions of the present invention are as follows:

[0006] The present invention provides an axial flow compressor flow design method based on three-dimensional data scaling:

[0007] (1) Initial design: First, the spanwise twisting law in the two-dimensional flow design of a multi-stage axial compressor is designed based on the mid-diameter position parameters obtained from the one-dimensional inverse problem. The radial equilibrium equation is solved according to the selected twisting law to obtain the pre-swirl distribution at each stage of the inlet.

[0008] (2) Initial two-dimensional flow design solution: Based on the results of the one-dimensional inverse problem design, given the constant distribution of the moving blade efficiency and the total pressure recovery coefficient of the stator blade along the radial direction, the streamline curvature method is used to solve the inverse problem and calculate the aerodynamic parameter distribution of the multi-stage compressor moving and stator blade rows along the span direction, thereby obtaining the initial two-dimensional flow design scheme;

[0009] (3) Parameter scaling based on three-dimensional data: Based on the two-dimensional flow design scheme, three-dimensional numerical simulation of blade shaping and multi-stage axial flow compressor is carried out, and the three-dimensional flow field results are averaged circumferentially to obtain the radial distribution of the compressor aerodynamic parameters, mainly the moving blade efficiency and the total pressure recovery coefficient of the static blade, which represents the flow loss parameters. Through numerical dimension scaling, the three-dimensional data is returned to the spanwise loss parameters in the two-dimensional flow design;

[0010] (4) Two-dimensional flow design based on scaled data: Using three-dimensional scaled data to replace the initial spanwise loss design parameters, the streamline curvature method is re-adopted to solve the inverse problem, and the spanwise aerodynamic parameter distribution of the scaled multi-stage compressor moving and stator blade rows is calculated;

[0011] (5) The iteration of steps (3) to (4) is realized through scaling design until the input parameters of the two 2D flow design are basically consistent and the difference in the moving blade efficiency values ​​is within 0.2%, thus obtaining the final 2D flow design scheme.

[0012] The present invention also includes:

[0013] In step (3), the blade shape is determined by giving the trailing angle of each blade attack angle, and the trailing angle is determined by a modified calculation of the trailing angle empirical model based on the two-dimensional design results.

[0014] In step (3), the method used is to achieve data dimensionality reduction by circumferentially averaging the three-dimensional results, and the loss parameters of each column of moving and static blades after circumferential averaging are reflected by the moving blade efficiency and the total pressure recovery coefficient of the static blades. In this way, the purpose of extracting the two-dimensional spanwise loss design parameters from the three-dimensional flow field can be achieved. The specific method is as follows:

[0015] Based on the total pressure, total temperature, and dense flow at the inlet and outlet of each blade row, the radial distribution of the moving blade efficiency and the total pressure recovery coefficient of the stationary blades can be calculated. However, the loss parameters obtained in the three-dimensional results include the sudden change in boundary layer losses caused by the viscous boundary layer at the end wall. Since the two-dimensional flow design is inviscid, the losses caused by the viscous boundary layer in the three-dimensional results need to be eliminated. The location of the viscous boundary layer extraction is mainly determined by the location of the maximum value of the flow loss parameters of each blade row in the end region.

[0016] After determining the location of the viscous end zone that needs to be eliminated, the parameters of the mainstream core area are interpolated proportionally. Twelve parameters are selected at equal intervals to represent the input parameters of the two-dimensional flow design. At this time, the spanwise loss parameters in the two-dimensional flow design are the actual parameters that include various flow losses.

[0017] The present invention has the following advantages and outstanding technical effects:

[0018] 1. The axial flow compressor flow design method based on three-dimensional data scaling proposed in this invention uses a cross-dimensional scaling method to achieve refined modeling and performance prediction of the axial flow compressor end area, shortening the design cycle of the axial flow compressor while ensuring design quality.

[0019] 2. The axial flow compressor flow design method based on three-dimensional data scaling proposed in the present invention transfers high-dimensional data to low dimensions, which can improve the refinement of low-dimensional compressor design and then improve the accuracy of axial flow compressor design or axial flow compressor performance prediction, so that the initial low-dimensional flow field calculation is more consistent with the three-dimensional calculation results, and the designed compressor has better aerodynamic performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a flow chart of the present invention;

[0021] Figure 2a is the compressor reaction degree distribution;

[0022] Figure 2b is the density distribution of compressor stator blades;

[0023] Figure 2c is the density distribution of the compressor rotor blades;

[0024] Figure 2d is the compressor blade flow coefficient distribution;

[0025] Figure 2e is the compressor stator blade flow coefficient distribution;

[0026] Figure 2f is the compressor circulation distribution;

[0027] Figure 2g is the Mach number distribution at the compressor blade inlet;

[0028] Figure 2h is the Mach number distribution at the compressor stator inlet;

[0029] Figure 2i is the distribution of compressor stator expansion factor;

[0030] Figure 2j is the compressor blade expansion factor distribution;

[0031] Figure 3a The change of the blade efficiency before and after scaling for the embodiment;

[0032] Figure 3b The change of the total pressure recovery coefficient of the static blade before and after scaling of the embodiment;

[0033] Figure 4a is the compressor reaction degree distribution;

[0034] Figure 4b is the density distribution of compressor stator blades;

[0035] Figure 4c is the density distribution of the compressor rotor blades;

[0036] Figure 4d is the compressor blade flow coefficient distribution;

[0037] Figure 4e is the compressor stator blade flow coefficient distribution;

[0038] Figure 4f is the compressor circulation distribution;

[0039] Figure 4g is the Mach number distribution at the compressor blade inlet;

[0040] Figure 4h is the Mach number distribution at the compressor stator inlet;

[0041] Figure 4i is the distribution of compressor stator expansion factor;

[0042] Figure 4j is the compressor blade expansion factor distribution; DETAILED DESCRIPTION

[0043] The principle of the present invention will be described below with reference to the accompanying drawings:

[0044] like Figure 1 A flow design method for an axial compressor based on three-dimensional data scaling is implemented by the following steps:

[0045] Step 1: Initial Design. First, based on the mid-diameter position parameters obtained from the one-dimensional inverse problem, the spanwise twisting law in the two-dimensional flow design of the multi-stage axial compressor is designed. The radial equilibrium equation is solved according to the selected twisting law to obtain the pre-swirl distribution at each stage of the inlet.

[0046] Step 2: Solve the initial two-dimensional flow design. Again, based on the results of the one-dimensional inverse problem design, and given the constant radial distribution of the moving blade efficiency and the total pressure recovery coefficient of the stator blades, the inverse problem is solved using the streamline curvature method. The aerodynamic parameter distribution of the multi-stage compressor moving and stator blade rows along the span direction is calculated, thereby obtaining the initial two-dimensional flow design scheme.

[0047] Step 3: Parameter Scaling Based on 3D Data. Based on the 2D flow design, a 3D numerical simulation of blade shaping and a multi-stage axial flow compressor is performed. The 3D flow field results are circumferentially averaged to obtain the radial distribution of the compressor aerodynamic parameters, primarily the moving blade efficiency and the total pressure recovery coefficient of the stationary blades, which represent flow losses. Through numerical dimensional scaling, the 3D data is returned to the spanwise loss parameters used in the 2D flow design.

[0048] In this step, the blade shape is determined by giving the trailing angle of each blade attack angle, and the trailing angle is determined by modifying the trailing angle empirical model based on the two-dimensional design results.

[0049] The method used in this step is to achieve data dimensionality reduction by circumferentially averaging the three-dimensional results. The loss parameters of each column of moving and stationary blades after circumferential averaging are reflected by the moving blade efficiency and the total pressure recovery coefficient of the stationary blades. In this way, the purpose of extracting the two-dimensional spanwise loss design parameters from the three-dimensional flow field can be achieved. The specific method is as follows:

[0050] Based on the total pressure, total temperature, and dense flow at the inlet and outlet of each blade row, the radial distribution of the moving blade efficiency and the total pressure recovery coefficient of the stationary blades can be calculated. However, the loss parameters obtained in the three-dimensional results include the sudden change in boundary layer losses caused by the viscous boundary layer at the end wall. Since the two-dimensional flow design is inviscid, the losses caused by the viscous boundary layer in the three-dimensional results need to be eliminated. The location of the viscous boundary layer extraction is mainly determined by the location of the maximum value of the flow loss parameters of each blade row in the end region.

[0051] After determining the location of the viscous end zone that needs to be eliminated, the parameters of the mainstream core area are interpolated proportionally. Twelve parameters are selected at equal intervals to represent the input parameters of the two-dimensional flow design. At this time, the spanwise loss parameters in the two-dimensional flow design are the actual parameters that include various flow losses.

[0052] Step 4: 2D flow design based on scaled data. Using the 3D scaled data to replace the initial spanwise loss design parameters, the streamline curvature method is re-applied to solve the inverse problem, calculating the spanwise aerodynamic parameter distribution of the scaled multi-stage compressor moving and stator blade rows.

[0053] Step 5: Iterate through steps 3 and 4 through scaling design until the input parameters of the two 2D flow-through designs are substantially consistent and the rotor blade efficiency values ​​differ within 0.2%. This results in the final 2D flow-through design. The features and performance of the present invention are further described in detail below with reference to the following examples.

[0054] Example:

[0055] First, a two-dimensional inverse flow design is performed based on the one-dimensional design results, including spanwise twisting law design and spanwise loss parameter design. The twisting law design is obtained by specifying the inlet pre-swirl at each stage, while the loss parameters are given by the rotor efficiency and the total pressure recovery coefficient of the stationary blades.

[0056] The main parameters of the one-dimensional design results are shown in Table 1 below:

[0057] Table 1 Main parameters of one-dimensional design results

[0058]

[0059] The pre-swirl speed of the moving blade tip is set to -10 m / s, and the pre-swirl distribution of each blade height section is obtained based on the radial balance equation, as shown in Table 2. The distribution of the moving and stationary blade loss parameters is given based on one-dimensional data.

[0060] Table 2 Pre-rotation distribution of blade height section

[0061]

[0062] Then, based on the streamline curvature method, a two-dimensional inverse problem calculation was performed to obtain the two-dimensional flow design solution, such as Figures 2a to 2j shown.

[0063] Based on the 2D design results, blade shaping is performed and 3D calculations are performed. The processed 3D radial parameters are scaled back to the 2D flow design input parameters, replacing the original spanwise loss parameters and performing 2D inverse problem calculations again. Figure 3a Figure 3b Represents the change of radial loss parameters before and after scaling, Figures 4a to 4j Compute results for the scaled 2D inverse problem.

[0064] The results are then subjected to a second 3D calculation and circumferential averaging. If the radial parameters of the flow field at this point match the 2D input parameter distribution, the scaling design is complete. Otherwise, the above steps are repeated until the 3D radial parameters match the 2D input parameter distribution.

[0065] The above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention without departing from the principles and purpose of the present invention.

[0066] The axial flow compressor flow design method based on three-dimensional data scaling proposed in the present invention is universal and is not only limited to multi-stage axial flow compressors of ship gas turbines, but is also applicable to the aerodynamic design process of various industrial gas turbine axial flow compressors and aircraft engine axial flow compressors.

Claims

1. A flow design method for an axial compressor based on three-dimensional data scaling, characterized by: The following steps are involved: (1) Initial design: First, the spanwise twisting law in the two-dimensional flow design of a multi-stage axial compressor is designed based on the mid-diameter position parameters obtained from the one-dimensional inverse problem. The radial equilibrium equation is solved according to the selected twisting law to obtain the pre-swirl distribution at each stage of the inlet. (2) Initial two-dimensional flow design solution: Based on the results of the one-dimensional inverse problem design, given the constant distribution of the moving blade efficiency and the total pressure recovery coefficient of the stator blade along the radial direction, the streamline curvature method is used to solve the inverse problem and calculate the aerodynamic parameter distribution of the multi-stage compressor moving and stator blade rows along the span direction, thereby obtaining the initial two-dimensional flow design scheme; (3) Parameter scaling based on three-dimensional data: Based on the two-dimensional flow design scheme, three-dimensional numerical simulation of blade shaping and multi-stage axial flow compressor is performed, and the three-dimensional flow field results are averaged circumferentially to obtain the circumferentially averaged compressor aerodynamic parameters, which are the radial distribution of the moving blade efficiency and the total pressure recovery coefficient of the static blade, that is, the parameters representing the flow loss. Through numerical dimension scaling, the three-dimensional data is returned to the spanwise loss parameters in the two-dimensional flow design; (4) Two-dimensional flow design based on scaled data: Using three-dimensional scaled data to replace the initial spanwise loss design parameters, the streamline curvature method is re-adopted to solve the inverse problem, and the spanwise aerodynamic parameter distribution of the scaled multi-stage compressor moving and stator blade rows is calculated; (5) The iteration of steps (3) to (4) is realized through scaling design until the input parameters of the two 2D flow design are basically consistent and the difference in the moving blade efficiency values ​​is within 0.2%, thus obtaining the final 2D flow design scheme.

2. The method for designing axial compressor flow path based on three-dimensional data scaling according to claim 1, wherein: In step (3), the blade shape is determined by giving the trailing angle of each blade attack angle, and the trailing angle is determined by a modified calculation of the trailing angle empirical model based on the two-dimensional design results.

3. The method for designing axial compressor flow path based on three-dimensional data scaling according to claim 1, wherein: Step (3) uses a method to achieve data dimensionality reduction by circumferentially averaging the three-dimensional results. The loss parameters of each column of moving and static blades after circumferential averaging are reflected by the moving blade efficiency and the total pressure recovery coefficient of the static blades, thereby achieving the purpose of extracting the two-dimensional spanwise loss design parameters from the three-dimensional flow field. The specific method is as follows: The radial distribution of the moving blade efficiency and the total pressure recovery coefficient of the stationary blades is calculated based on the total pressure, total temperature, and density flow at the inlet and outlet of each blade row. However, the losses caused by the viscous boundary layer in the three-dimensional results are eliminated. The extraction position of the viscous boundary layer is determined by the maximum value of the flow loss parameter of each blade row in the end region. After determining the location of the viscous end zone that needs to be eliminated, the parameters of the mainstream core area are interpolated proportionally. Twelve parameters are selected at equal intervals to represent the input parameters of the two-dimensional flow design. At this time, the spanwise loss parameters in the two-dimensional flow design are the actual parameters that include various flow losses.

Images