A Multi-Stage Axial Compressor Characteristic Prediction Method Based on CFD / S2 Hybrid Dimensions

Abstract

This invention discloses a method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension approach, belonging to the field of axial compressor technology. The method comprises the following steps: obtaining the total pressure ratio and efficiency of the multi-stage axial compressor, as well as the flow field parameter distribution between internal blade rows, through S2 flow surface analysis; determining the boundary conditions for CFD numerical simulation of a selected blade row; performing CFD numerical simulation on the selected blade row, and after convergence, extracting the aerodynamic parameters of the leading and trailing edges of the blade row after circumferential averaging of the three-dimensional CFD results; performing flow analysis calculations on the multi-stage compressor; treating steps 1 to 4 as an iteration of the hybrid dimension method, and stopping the iteration after multiple iterations in which both flow analysis and CFD converge in each iteration; performing another flow analysis calculation based on the CFD results of the last iteration, and using the result as the predicted characteristics of the multi-stage axial compressor. This invention reduces the problem of extensive flow field calculations in the optimization process of multi-stage compressors based on CFD technology.

Description

Technical Field

[0001] This invention belongs to the field of axial compressor technology, specifically relating to a method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension. Background Technology

[0002] With the continuous improvement of the efficiency, thrust-to-weight ratio, reliability, noise, and pollution emission levels of aero engines, the performance requirements for multi-stage axial compressors, one of the three core components of aero engines, are also constantly increasing in advanced aero engines. However, the development of multi-stage axial compressors requires a large amount of human, material, and financial resources.

[0003] To reduce the number of iterative iterations and corrections in the design process of multi-stage axial compressors, aerodynamic optimization design methods for compressors have been a focus of researchers in recent years. These methods rely on CFD flow field calculations to evaluate the performance of design schemes and combine appropriate optimization algorithms to search for the optimal design. Although current computing power has made significant progress compared to the past, the cost of CFD flow field calculations remains enormous, especially the time-sensitive costs in the design process. When optimizing multi-stage axial compressors, the optimization scale grows exponentially with the number of design variables; the multi-stage design variables result in a substantial increase in computation time.

[0004] The streamline curvature method offers fast computation speed for S2 flow surface flow analysis, but the accuracy of the results depends on a systematic experimental database comprised of numerous cascade blowing experiments and compressor whole-machine experiments. Furthermore, due to the axisymmetric assumption, flow analysis cannot accurately predict the effects of three-dimensional flow. CFD numerical simulation offers high computational accuracy and can provide flow field details, but it is slow, costly, and limited by current computer capabilities and turbulence models. Summary of the Invention

[0005] The technical problem to be solved:

[0006] To avoid the shortcomings of existing technologies, this invention provides a method for predicting the characteristics of multi-stage axial compressors based on a CFD / S2 hybrid dimension, which reduces the problem of large-scale flow field calculations in the optimization process of multi-stage compressors based on CFD technology.

[0007] The technical solution of this invention is: a method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension, the specific steps of which are as follows:

[0008] Step 1: Obtain the total pressure ratio and efficiency of the multi-stage axial compressor, as well as the flow field parameter distribution between the internal blade rows, through S2 flow surface analysis;

[0009] Step 2: Extract the flow field parameters distributed radially along the leading and trailing edges of the selected blade row from the flow analysis results and use them as the boundary conditions for the CFD numerical simulation of the selected blade row.

[0010] Step 3: Perform CFD numerical simulation on the selected blade row. After the calculation converges, extract the aerodynamic parameters of the leading and trailing edges of the blade row after circumferential averaging of the three-dimensional CFD results.

[0011] Step 4: Based on the aerodynamic parameters obtained in Step 3, calculate the pressure ratio and efficiency, or circulation and total pressure recovery coefficient, of the selected blade row trailing edge radially distributed. Update the pre-given parameters of the selected blade row trailing edge station in the flow analysis input file, and then perform flow analysis calculations on the multi-stage compressor.

[0012] Step 5: Take steps 1 to 4 as one iteration of the mixed-dimensional method. After multiple iterations, if the flow analysis and CFD converge in each iteration, the iteration stops. Perform another flow analysis calculation based on the CFD results of the last iteration. The result is used as the predictive characteristic of the multi-stage axial compressor.

[0013] A further technical solution of the present invention is: the S2 flow surface flow analysis in step 1 includes two loops. The inner loop uses the streamline curvature method to solve the S2 flow surface design problem and obtain the aerodynamic parameter distribution of the transonic flow field. The outer loop substitutes the results of the inner loop into the lag angle model and the loss model to obtain a new circulation and efficiency distribution, and then substitutes it into the inner loop for recalculation.

[0014] A further technical solution of the present invention is: the inner loop includes the following steps:

[0015] Step 1.1: First, assume the position of the streamline based on the principle of equal area, then set up calculation stations along the projection of the leading and trailing edges of the blade row onto the meridional plane, and perform calculations sequentially along the calculation stations;

[0016] Step 1.2: After completing the calculations at all computing stations, determine the pressure distribution, velocity distribution, and flow rate distribution at each computing station. Then, connect the isoflow points at each computing station with a smooth spline curve to obtain a new streamline.

[0017] Step 1.3: Compare the positions of the new streamlines with the initial streamlines used in the calculation. If the difference is within the error limit... Then convergence is considered complete, and the inner loop ends; in the formula, Let ε be the radial coordinate at the nth iteration. r As the error standard;

[0018] Step 1.4: If convergence is not achieved, recalculate the airflow parameters at each calculation station based on the streamlines in Step 1.2, and obtain new streamlines until the calculated streamlines substantially coincide with the given streamline positions as known streamlines.

[0019] A further technical solution of the present invention is: the outer loop includes the following steps:

[0020] Step 1.5: First, assume the pressure ratio π and efficiency η of the blade trailing edge calculation station, and the circulation V. θ The distribution of r and total pressure recovery coefficient σ along the spanwise direction enters the inner circulation, and the flow field distribution on the S2 surface is solved using the complete radial equilibrium equation and the continuity equation in the form of a design problem.

[0021] Step 1.6: Introduce the lag angle model and loss model, and substitute the obtained flow field parameters into the model to calculate the lag angle δ and loss coefficient.

[0022] Step 1.7 Based on the model results from Step 1.6, namely the lag angle δ and the loss coefficient... Calculate the pressure ratio π and circulation rV corresponding to the lag angle. θ The efficiency η and total pressure recovery coefficient corresponding to the loss coefficient;

[0023] Step 1.8 compares the calculation results in Step 1.7 with the circulation and efficiency in the S2 flow surface flow analysis input file. If the convergence criteria are met, the loop ends; otherwise, the inner loop restarts based on the new circulation and efficiency.

[0024] A further technical solution of the present invention is: the boundary conditions for selecting the blade row CFD numerical simulation in step 2 include:

[0025] The inlet boundary conditions are: radially distributed airflow direction and total temperature T. * and total pressure P * ;

[0026] The export boundary condition is the mass flow rate.

[0027] A further technical solution of the present invention is: the radially distributed airflow direction is determined by the ratio V of the radial velocity to the absolute velocity. r / |V|, the ratio of circumferential velocity to absolute velocity. t / |V| and the ratio of the axial velocity component to the absolute velocity V z / |V| is given.

[0028] A further technical solution of the present invention is: the extraction of the aerodynamic parameters of the leading and trailing edges of the blade row after circumferential averaging of the three-dimensional CFD results in step 3 includes the following steps:

[0029] When the blade row is selected as the rotor, the total temperature T distributed radially along the leading and trailing edges is extracted. rotor * and total pressure P rotor * ;

[0030] When the selected blade row is the stator, the total pressure P distributed radially along the leading edge is extracted. stator * The total pressure P distributed radially at the trailing edge stator * and the circumferential velocity component V of the airflow t .

[0031] A further technical solution of the present invention is: in step 4, in order to avoid interference from non-physical factors in the CFD results near the hub and casing, pre-given parameters for the selected blade row trailing edge station at a relative spanwise position of 5% to 95% are calculated.

[0032] A further technical solution of the present invention is: in step 4, the loss and lag angle models are not used to update the parameters on the trailing edge station of the selected blade row during the calculation.

[0033] Beneficial effects

[0034] The beneficial effects of this invention are as follows:

[0035] 1. This invention reduces the reliance on loss and lag angle models based on experimental data, and improves the prediction accuracy of compressor characteristics compared with existing S2 flow surface flow analysis methods.

[0036] 2. This invention only requires CFD simulation of the blade row to be optimized in a multi-stage axial compressor. Compared with the existing CFD simulation of the entire multi-stage axial compressor, the amount of computation is greatly reduced. Moreover, it can simultaneously obtain the detailed parameters of the blade row flow field and the characteristics of the entire compressor that researchers are interested in.

[0037] 3. This invention can effectively combine the advantages of flow analysis and CFD, especially when applied to the optimization problem of multi-stage axial compressors, it can significantly reduce optimization time and cost. Attached Figure Description

[0038] Figure 1 This is a flowchart of the present invention;

[0039] Figure 2 The following are CFD convergence curves during the iterative process in an embodiment of the present invention: (a) represents the global residual, and (b) represents the mass flow rate. Detailed Implementation

[0040] The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain the invention, and should not be construed as limiting the invention.

[0041] To address the issue of extensive flow field calculations in the optimization of multi-stage compressors based on CFD technology in existing technologies, this invention proposes a method for predicting the characteristics of multi-stage axial compressors based on a CFD / S2 hybrid dimension. The method steps are as follows:

[0042] Step 1: Obtain the total pressure ratio and efficiency of the multi-stage axial compressor, as well as the flow field parameter distribution between the internal blade rows, through S2 flow surface analysis;

[0043] Step 2: Extract the flow field parameters distributed radially along the leading and trailing edges of the selected blade row from the flow analysis results and use them as the boundary conditions for the CFD numerical simulation of the selected blade row.

[0044] Step 3: Perform CFD numerical simulation on the selected blade row. After the calculation converges, extract the aerodynamic parameters of the leading and trailing edges of the blade row after circumferential averaging of the three-dimensional CFD results.

[0045] Step 4: Based on the aerodynamic parameters obtained in Step 3, calculate the pressure ratio and efficiency (rotor) or circulation and total pressure recovery coefficient (stator) of the selected blade row trailing edge radially distributed. Update the pre-given parameters of the selected blade row trailing edge station in the flow analysis input file. Then perform flow analysis calculations on the multi-stage compressor. Do not use loss and lag angle models to update the parameters on the selected blade row trailing edge station during the calculation.

[0046] Step 5: Take steps 1 to 4 as one iteration of the mixed-dimensional method. After multiple iterations, if the flow analysis and CFD converge in each iteration, the iteration stops. Perform another flow analysis calculation based on the CFD results of the last iteration. The result is used as the predictive characteristic of the multi-stage axial compressor.

[0047] This invention reduces reliance on experimentally based loss and lag angle models, improving the prediction accuracy of compressor characteristics compared to existing S2 flow surface analysis methods. Furthermore, this invention only requires CFD simulation of the blade rows to be optimized in a multi-stage axial compressor, significantly reducing computational load compared to whole-machine CFD simulation. It also simultaneously obtains detailed parameters of the blade row flow field and the overall compressor characteristics of interest to researchers. This invention effectively combines the advantages of flow analysis and CFD, particularly when applied to multi-stage axial compressor optimization problems, significantly reducing optimization time and cost.

[0048] Example 1:

[0049] This embodiment presents a method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension. Taking a two-stage fan as the object, and using the first-stage stator as the selected blade row, the method predicts the characteristics of the compressor at its highest efficiency point (mass flow rate of 73.65 kg) at 90% of its design speed (9648 r.pm). The specific steps are as follows:

[0050] Step 1: Obtain the total pressure ratio and efficiency of the two-stage fan, as well as the flow field parameter distribution between the internal blade rows, through S2 flow surface analysis;

[0051] The flow analysis of the S2 flow surface is as follows:

[0052] First, to solve the analytical problems of subsonic and transonic flow fields using the complete radial equilibrium equations of the design problem, the distribution of the airflow angle must be transformed into the distribution of circulation, rV. θ Based on the velocity triangle, it can be represented as:

[0053] rV θ =r(U+W θ ) = r 2 ω+rV m tanβ

[0054] In the formula, r is the radial coordinate, V θ Let U be the circumferential component of the absolute velocity of the airflow, and W be the circumferential velocity. θ V is the circumferential component of the relative velocity of the airflow, ω is the angular velocity of rotation, and V m β is the airflow velocity along the meridional plane, and β is the relative airflow angle.

[0055] Then, by analyzing the distribution conditions of the known airflow angle β, and using the previously obtained V... m The distribution is calculated to obtain the circulation distribution required for the design problem, and then the flow field is solved using the equations of the design problem.

[0056] The flow analysis is divided into two loops. The inner loop uses the streamline curvature method to solve the S2 flow surface design problem and obtain the aerodynamic parameter distribution of the transonic flow field. The outer loop substitutes the results of the inner loop into the lag angle model and the loss model to obtain a new circulation and efficiency distribution, which is then substituted into the inner loop for recalculation.

[0057] The inner loop includes the following steps:

[0058] 1) First, assume the position of the streamline based on the principle of equal area, then set up calculation stations along the projection of the leading and trailing edges of the blade row onto the meridional plane, and perform calculations sequentially along the calculation stations;

[0059] 2) After all the calculations at all the stations are completed, the pressure distribution, velocity distribution and flow rate distribution at each station are obtained. Then, the isoflow points at each station are connected by a smooth spline curve to obtain a new streamline.

[0060] 3) Compare the positions of the new streamlines with the initial streamlines used in the calculation. If the difference is within the error limit... Then it can be considered converged, and the inner loop ends; in the formula, Let ε be the radial coordinate at the nth iteration. r This serves as the error standard.

[0061] If convergence is not achieved, recalculate the airflow parameters at each calculation station based on the streamlines in 2), and then obtain new streamlines until the calculated streamlines substantially coincide with the given streamline positions as known streamlines.

[0062] The outer loop includes the following steps:

[0063] 1) First, assume the pressure ratio π and efficiency η (rotor) of the blade trailing edge calculation station, and the circulation rV. θ The distribution of the total pressure recovery coefficient σ (stator) along the spanwise direction is used to enter the inner circulation. The flow field distribution on the S2 surface is solved using the complete radial equilibrium equation and the continuity equation in the form of a design problem.

[0064] 2) Introduce the lag angle model and loss model, and substitute the obtained flow field parameters into the model to calculate the lag angle δ and loss coefficient.

[0065] 3) Based on the model results in 2) (lag angle δ and loss coefficient) ), calculate the pressure ratio π and circulation rV corresponding to the lag angle. θ The efficiency η and total pressure recovery coefficient corresponding to the loss coefficient;

[0066] 4) Compare the calculation results in 3) with the circulation and efficiency in the S2 flow surface flow analysis input file. If the convergence criteria are met, end the loop; otherwise, restart the inner loop based on the new circulation and efficiency.

[0067] Step 2: Extract the radially distributed flow field parameters from the leading and trailing edges of the first-stage stator in the flow analysis results as the boundary conditions for the selected blade row CFD numerical simulation, as follows:

[0068] The inlet boundary condition is the airflow direction distributed radially (based on the ratio of radial velocity to absolute velocity, V). r / |V|, the ratio of circumferential velocity to absolute velocity. t / |V| and the ratio of the axial velocity component to the absolute velocity V z Given (|V|), total temperature T * and total pressure P * ;

[0069] To maintain consistency with the S2 flow surface analysis, the outlet boundary condition mass flow rate is... Set to 73.65kg.

[0070] Step 3: Perform CFD numerical simulation on the first-stage stator. After the calculation converges, extract the total pressure P distributed radially along the leading edge. stator * The total pressure P distributed radially at the trailing edge stator * and the circumferential velocity component V of the airflow t ;

[0071] Step 4: Based on the aerodynamic parameters obtained in Step 3, to avoid interference from non-physical factors in the CFD results near the hub and casing, calculate the radially distributed circulation V at the trailing edge of the first-stage stator at a position 5% to 95% of the spanwise direction. θ The parameters of the first-stage stator trailing edge station are updated using r and the total pressure recovery coefficient σ in the flow analysis input file. Then, the flow analysis is performed on the two-stage fan without using the loss and lag angle models to update the parameters on the first-stage stator trailing edge station.

[0072] Step 5: Steps 1 to 4 are taken as one iteration of the hybrid dimension method. After two iterations, and the flow analysis and CFD converge in each iteration, the iteration stops. Based on the CFD results in the second iteration, the flow analysis is performed again to obtain the predicted characteristics of the two-stage fan. The total pressure ratio is 2.448 and the efficiency is 87.98%. The entire hybrid dimension characteristic prediction process takes 211.75 seconds.

[0073] In the mixed-dimensional characteristic prediction method, when performing CFD simulation on the first-stage stator, the number of grid points is 116,457. If the grid size of each blade row in the CFD simulation of the two-stage fan is consistent with that in the mixed-dimensional characteristic prediction method, the number of grid points for the whole machine is 459,588. Under the same operating conditions, the whole machine CFD simulation takes 1275.68 seconds, and the result is a total pressure ratio of 2.417 and an efficiency of 87.14%.

[0074] Using only the streamline curvature method for S2 flow surface analysis of the two-stage fan, the predicted characteristics of the two-stage fan are a total pressure ratio of 2.533 and an efficiency of 88.11%.

[0075] Example 2:

[0076] This embodiment presents a method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension. Taking a two-stage fan as the object, and the second-stage rotor as the selected blade row, the method predicts the characteristics of the compressor at its highest efficiency point (mass flow rate of 73.65 kg) at 90% of its design speed (9648 r.pm). The specific steps are as follows:

[0077] Step 1: Obtain the total pressure ratio and efficiency of the two-stage fan, as well as the flow field parameter distribution between the internal blade rows, through S2 flow surface analysis;

[0078] The flow analysis of the S2 flow surface is as follows:

[0079] First, to solve the analytical problems of subsonic and transonic flow fields using the complete radial equilibrium equations of the design problem, the distribution of the airflow angle must be transformed into the distribution of circulation, rV. θ Based on the velocity triangle, it can be represented as:

[0080] rV θ =r(U+W θ ) = r 2 ω+rV m tanβ

[0081] In the formula, r is the radial coordinate, V θ Let U be the circumferential component of the absolute velocity of the airflow, and W be the circumferential velocity. θ V is the circumferential component of the relative velocity of the airflow, ω is the angular velocity of rotation, and V m β is the airflow velocity along the meridional plane, and β is the relative airflow angle.

[0082] Then, by analyzing the distribution conditions of the known airflow angle β, and using the previously obtained V... m The distribution is calculated to obtain the circulation distribution required for the design problem, and then the flow field is solved using the equations of the design problem.

[0083] The flow analysis is divided into two loops. The inner loop uses the streamline curvature method to solve the S2 flow surface design problem and obtain the aerodynamic parameter distribution of the transonic flow field. The outer loop substitutes the results of the inner loop into the lag angle model and the loss model to obtain a new circulation and efficiency distribution, which is then substituted into the inner loop for recalculation.

[0084] The inner loop includes the following steps:

[0085] 1) First, assume the position of the streamline based on the principle of equal area, then set up calculation stations along the projection of the leading and trailing edges of the blade row onto the meridional plane, and perform calculations sequentially along the calculation stations;

[0086] 2) After all the calculations at all the stations are completed, the pressure distribution, velocity distribution and flow rate distribution at each station are obtained. Then, the isoflow points at each station are connected by a smooth spline curve to obtain a new streamline.

[0087] 3) Compare the positions of the new streamlines with the initial streamlines used in the calculation. If the difference is within the error limit... Then it can be considered converged, and the inner loop ends; in the formula, Let ε be the radial coordinate at the nth iteration. r This serves as the error standard.

[0088] If convergence is not achieved, recalculate the airflow parameters at each calculation station based on the streamlines in 2), and then obtain new streamlines until the calculated streamlines substantially coincide with the given streamline positions as known streamlines.

[0089] The outer loop includes the following steps:

[0090] 1) First, assume the pressure ratio π and efficiency η (rotor) of the blade trailing edge calculation station, and the circulation rV. θ The distribution of the total pressure recovery coefficient σ (stator) along the spanwise direction is used to enter the inner circulation. The flow field distribution on the S2 surface is solved using the complete radial equilibrium equation and the continuity equation in the form of a design problem.

[0091] 2) Introduce the lag angle model and loss model, and substitute the obtained flow field parameters into the model to calculate the lag angle δ and loss coefficient.

[0092] 3) Based on the model results in 2) (lag angle δ and loss coefficient) ), calculate the pressure ratio π and circulation rV corresponding to the lag angle. θ The efficiency η and total pressure recovery coefficient corresponding to the loss coefficient;

[0093] 4) Compare the calculation results in 3) with the circulation and efficiency in the S2 flow surface flow analysis input file. If the convergence criteria are met, end the loop; otherwise, restart the inner loop based on the new circulation and efficiency.

[0094] Step 2: Extract the radially distributed flow field parameters of the leading and trailing edges of the second-stage rotor from the flow path analysis results as the boundary conditions for the selected blade row CFD numerical simulation, as follows:

[0095] The inlet boundary condition is the airflow direction distributed radially (based on the ratio of radial velocity to absolute velocity, V). r / |V|, the ratio of circumferential velocity to absolute velocity. t / |V| and the ratio of the axial velocity component to the absolute velocity V z Given / |V|, total temperature T * and total pressure P * ;

[0096] To maintain consistency with the S2 flow surface analysis, the outlet boundary condition mass flow rate is... Set to 73.65kg.

[0097] Step 3: Perform CFD numerical simulation on the second-stage rotor. After the calculation converges, extract the total temperature T distributed radially along the leading and trailing edges. rotor * and total pressure P rotor * ;

[0098] Step 4: Based on the aerodynamic parameters obtained in Step 3, to avoid interference from non-physical factors in the CFD results near the hub and casing, calculate the radially distributed pressure ratio π and efficiency η at the 5% to 95% spanwise relative span position of the trailing edge of the second-stage rotor. Update the pre-given parameters of the trailing edge station of the second-stage rotor in the flow analysis input file. Then, perform flow analysis calculations on the two-stage fan. In the calculation, do not use loss and lag angle models to update the parameters on the trailing edge station of the second-stage rotor.

[0099] Step 5: Steps 1 to 4 are taken as one iteration of the hybrid dimension method. After two iterations, and the flow analysis and CFD converge in each iteration, the iteration stops. Based on the CFD results in the second iteration, the flow analysis is performed again to obtain the predicted characteristics of the two-stage fan. The total pressure ratio is 2.458 and the efficiency is 86.75%. The entire hybrid dimension characteristic prediction process takes 653.27 seconds.

[0100] In the mixed-dimensional characteristic prediction method, when performing CFD simulation on the second-stage rotor, the number of grid points is 136,863. If the grid size of each blade row in the whole-machine CFD simulation of the two-stage fan is consistent with that in the mixed-dimensional characteristic prediction method, the number of grid points in the whole machine is 459,588. Under the same operating conditions, the whole-machine CFD simulation takes 1275.68s, and the result is a total pressure ratio of 2.417 and an efficiency of 87.14%.

[0101] Using only the streamline curvature method for S2 flow surface analysis of the two-stage fan, the predicted characteristics of the two-stage fan are a total pressure ratio of 2.533 and an efficiency of 88.11%.

[0102] Using the hybrid dimension method for characteristic prediction significantly reduces computation time compared to CFD simulation of a multi-stage axial compressor, and the prediction accuracy is also higher than that of S2 flow surface flow analysis using the streamline curvature method. When applied to the optimization problem of multi-stage axial compressors, it can greatly reduce optimization time and optimization cost.

[0103] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention without departing from the principles and spirit of the present invention.

Claims

1. A method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension, characterized in that... The specific steps are as follows: Step 1: Obtain the total pressure ratio and efficiency of the multi-stage axial compressor, as well as the flow field parameter distribution between the internal blade rows, through S2 flow surface analysis; Step 2: Extract the flow field parameters distributed radially along the leading and trailing edges of the selected blade row from the flow analysis results and use them as the boundary conditions for the CFD numerical simulation of the selected blade row. Step 3: Perform CFD numerical simulation on the selected blade row. After the calculation converges, extract the aerodynamic parameters of the leading and trailing edges of the blade row after circumferential averaging of the three-dimensional CFD results. Step 4: Based on the aerodynamic parameters obtained in Step 3, calculate the pressure ratio and efficiency, or circulation and total pressure recovery coefficient, of the selected blade row trailing edge radially distributed. Update the pre-given parameters of the selected blade row trailing edge station in the flow analysis input file, and then perform flow analysis calculations on the multi-stage compressor. Step 5: Take steps 1 to 4 as one iteration of the mixed-dimensional method. After multiple iterations, if the flow analysis and CFD converge in each iteration, the iteration stops. Perform another flow analysis calculation based on the CFD results of the last iteration. The result is used as the predictive characteristic of the multi-stage axial compressor.

2. The method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension as described in claim 1, characterized in that: The S2 flow surface flow analysis in step 1 includes two loops. The inner loop uses the streamline curvature method to solve the S2 flow surface design problem and obtain the aerodynamic parameter distribution of the transonic flow field. The outer loop substitutes the results of the inner loop into the lag angle model and the loss model to obtain a new circulation and efficiency distribution, which is then substituted into the inner loop for recalculation.

3. The method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension as described in claim 2, characterized in that: The inner loop includes the following steps: Step 1.1: First, assume the position of the streamline based on the principle of equal area, then set up calculation stations along the projection of the leading and trailing edges of the blade row onto the meridional plane, and perform calculations sequentially along the calculation stations; Step 1.2: After completing the calculations at all computing stations, determine the pressure distribution, velocity distribution, and flow rate distribution at each computing station. Then, connect the isoflow points at each computing station with a smooth spline curve to obtain a new streamline. Step 1.3: Compare the positions of the new streamlines with the initial streamlines used in the calculation. If the difference is within the error limit... Then convergence is considered complete, and the inner loop ends; in the formula, Let ε be the radial coordinate at the nth iteration. r As the error standard; Step 1.4: If convergence is not achieved, recalculate the airflow parameters at each calculation station based on the streamlines in Step 1.2, and obtain new streamlines until the calculated streamlines substantially coincide with the given streamline positions as known streamlines.

4. The method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension as described in claim 3, characterized in that: The outer loop includes the following steps: Step 1.5: First, assume the pressure ratio π and efficiency η of the blade trailing edge calculation station, and the circulation V. θ The distribution of r and total pressure recovery coefficient σ along the spanwise direction enters the inner circulation, and the flow field distribution on the S2 surface is solved using the complete radial equilibrium equation and the continuity equation in the form of a design problem. Step 1.6: Introduce the lag angle model and loss model, and substitute the obtained flow field parameters into the model to calculate the lag angle δ and loss coefficient. Step 1.7 Based on the model results from Step 1.6, namely the lag angle δ and the loss coefficient... Calculate the pressure ratio π and circulation rV corresponding to the lag angle. θ The efficiency η and total pressure recovery coefficient corresponding to the loss coefficient; Step 1.8 compares the calculation results in Step 1.7 with the circulation and efficiency in the S2 flow surface flow analysis input file. If the convergence criteria are met, the loop ends; otherwise, the inner loop restarts based on the new circulation and efficiency.

5. The method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension as described in claim 4, characterized in that: The boundary conditions for selecting the blade row CFD numerical simulation in step 2 include: The inlet boundary conditions are: radially distributed airflow direction and total temperature T. * and total pressure P * ; The outlet boundary condition is the mass flow rate m.

6. The method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension as described in claim 5, characterized in that: The radially distributed airflow direction is determined by the ratio V of the radial velocity component to the absolute velocity component. r / |V|, the ratio of circumferential velocity to absolute velocity. t / |V| and the ratio of the axial velocity component to the absolute velocity V z / |V| is given.

7. The method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension as described in claim 6, characterized in that: Step 3, which involves extracting the aerodynamic parameters of the blade leading and trailing edges after circumferential averaging of the three-dimensional CFD results, includes the following steps: When the blade row is selected as the rotor, the total temperature T distributed radially along the leading and trailing edges is extracted. rotor * and total pressure P rotor * ; When the selected blade row is the stator, the total pressure P distributed radially along the leading edge is extracted. stator * The total pressure P distributed radially at the trailing edge stator * and the circumferential velocity component V of the airflow t .

8. The method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension as described in claim 7, characterized in that: In step 4, to avoid interference from non-physical factors in the CFD results near the hub and casing, pre-given parameters for the trailing edge of the selected blade row at 5% to 95% of the relative spanwise position are calculated.

9. The method for predicting the characteristics of a multi-stage axial compressor based on a CFD / S2 hybrid dimension as described in claim 8, characterized in that: In step 4, the loss and lag angle models are not used to update the parameters at the trailing edge station of the selected blade row.

Images