Air compressor impeller multi-objective optimization method based on turbulence coherent regulation and control

Abstract

The invention provides an air compressor impeller multi-objective optimization method based on turbulent flow coherent regulation and control, and relates to the field of air compressors for fuel cell automobiles, the optimization objective of the optimization method is isentropic efficiency and maximum turbulent flow kinetic energy, and the optimization parameters are a blade inlet mounting angle beta1A, a blade top gap Cc, a blade outlet width bc and a blade outlet mounting angle beta2A. According to the optimization method, a Latin hypercube sampling method is used for obtaining data points of a plurality of groups of impeller structures, impeller parameters are adjusted according to the data points, and a three-dimensional model is established. According to the optimization method, CFD simulation analysis is conducted on a three-dimensional model, the isentropic efficiency corresponding to each air compressor and the maximum turbulence kinetic energy in an impeller are obtained, optimization parameter data points and the corresponding isentropic efficiency and maximum turbulence kinetic energy are set as a training sample, and a sample set containing a plurality of training samples is constructed. Using the sample set to construct an agent model through neural network fitting, using a multi-target genetic algorithm to perform multi-target optimization on the agent model, obtaining a Pareto frontier solution set, and selecting a required optimal solution according to actual requirements.

Description

Technical Field

[0001] The invention relates to the technical field of air compressors for fuel cell vehicles, and in particular to a multi-objective optimization method for an air compressor impeller based on turbulence coherent regulation. Background Art

[0002] In recent years, the country's attention and strong support for the hydrogen energy industry have promoted the vigorous development of fuel cell vehicles. The air compressor used in fuel cell vehicles is a key component of fuel cell vehicles. Its performance is directly related to key indicators such as efficiency and noise of the fuel cell system. The main component of the air compressor is the impeller. The performance of the impeller has a great influence on the overall performance of the air compressor. Optimizing the design of the impeller can significantly improve the overall performance of the air compressor. However, the current optimization design of the impeller focuses on the relationship between the impeller structural parameters and the performance of the air compressor, without considering that the impeller structural parameters will also affect the turbulent coherent structure in the flow field. Summary of the invention

[0003] In view of this, the purpose of the present invention is to provide a multi-objective optimization method for air compressor impellers based on turbulent co-order regulation, analyze the air compressor impeller structural parameters from the perspective of turbulent co-order regulation, construct a proxy model of impeller structural parameters, isentropic efficiency, and maximum turbulent kinetic energy through neural network fitting, and combine a multi-objective genetic algorithm to perform multi-objective optimization. This method can effectively optimize the air compressor impeller structural parameters, improve isentropic efficiency, and reduce turbulent kinetic energy.

[0004] To achieve the above object, the present invention adopts the following technical solution: a multi-objective optimization method for an air compressor impeller based on turbulence co-regulation, comprising the following steps:

[0005] Step S1, constructing an initial model of an air compressor, performing CFD simulation analysis on the initial model of the air compressor, and obtaining a flow field of the initial model;

[0006] Step S2, extracting the turbulent coherent structure in the flow field by Q criterion, analyzing the influence of the turbulent coherent structure on isentropic efficiency and maximum turbulent kinetic energy, and determining optimization parameters and value ranges;

[0007] Step S3, using the Latin hypercube sampling method to obtain a number of data points about each optimization parameter, and establishing a three-dimensional model based on all the data points; performing CFD simulation analysis on all the three-dimensional models to obtain the corresponding isentropic efficiency and maximum turbulent kinetic energy, setting the optimization parameter data point and the corresponding isentropic efficiency and maximum turbulent kinetic energy as a training sample, and obtaining a sample set containing multiple training samples;

[0008] Step S4, optimizing the impeller geometry model parameters, the optimization includes the following steps: S41, constructing a proxy model based on the sample set using a neural network fitting; S42, combining a multi-objective genetic algorithm to perform multi-objective optimization on the proxy model, obtaining a Pareto frontier solution set, and selecting the desired optimal solution according to actual needs.

[0009] In a preferred embodiment, in step S2, the Q criterion definition is as follows:

[0010]

[0011] Where Ω is the rotation tensor; S is the strain rate tensor; Q is the vorticity;

[0012] In step S2, the turbulent co-ordered structure refers to an ordered structure in the irregular flow in the flow field, and its manifestation in the air compressor is a separation vortex, a leakage vortex, and a wake vortex.

[0013] In a preferred embodiment, in step S2, the air compressor isentropic efficiency calculation formula is as follows:

[0014]

[0015] Where, T 1c * is the measured total air intake temperature of the air compressor, in K; T 2c * It is the measured air outlet of the air compressor, the unit is K; π c * is the air compressor pressure ratio; k is the ideal gas adiabatic index.

[0016] In a preferred embodiment, in step S2, the turbulent kinetic energy k tke It is an indicator of the development or decay of turbulence, and its calculation formula is as follows:

[0017]

[0018] Where U is the average velocity in m / s; I is the turbulence intensity;

[0019] In step S2, the optimization parameter determined is the blade inlet installation angle β 1A , blade tip clearance C c , blade outlet width b c , Blade outlet installation angle β 2A .

[0020] In a preferred embodiment, in step S3, the Latin hypercube sampling method provides a value and randomly arranges each interval (1, 1 / n), (1 / n, 2 / n) ... (1-1 / n, 1) in n groups of data for x impeller structures.

[0021] In a preferred embodiment, in step S3, the generated sample set includes all combinations of optimization parameters and the corresponding isentropic efficiency and maximum turbulent kinetic energy, wherein the combination of optimization parameters is the blade inlet installation angle β 1A , blade tip clearance C c , Blade outlet width b c and blade outlet installation angle β 2A parameter combination.

[0022] In a preferred embodiment, in the step S41, the sample set is fitted and trained using a neural network toolbox, and the steps are as follows: S411, taking the optimization parameter combination as the independent variable, the isentropic efficiency and the maximum turbulent kinetic energy as the dependent variables, and importing the data into the neural network toolbox; S412, setting the proportion of training data and test data, and setting the number of hidden neurons; S413, training to obtain a proxy model of the air compressor optimization parameters, isentropic efficiency and maximum turbulent kinetic energy, and using the regression line slope R as the proxy model accuracy evaluation index.

[0023] In a preferred embodiment, in S413, the accuracy of the proxy model obtained by the neural network fitting training is determined by the slope R of the regression line. The closer R is to 1, the better the training effect.

[0024] In a preferred embodiment, in step S42, a multi-objective genetic algorithm is used for optimization, and the optimization steps are as follows: S421, the blade inlet installation angle β 1A , blade tip clearance C c , Blade outlet width b c and blade outlet installation angle β 2A As the optimization variable, the proxy model serves as the fitness function; S422, setting the constraints of the optimization variables; S423, setting the population size, the optimal front-end individual coefficient, the number of iterations, the fitness function value deviation, the probability of crossover and mutation; S424, based on the optimization parameters and the constraints of the fitness function, establishing a mathematical model of multi-objective optimization; S425, using a multi-objective genetic algorithm to optimize the mathematical model and select the optimal solution in the resulting Pareto frontier.

[0025] In a preferred embodiment, in S424, the mathematical model is as follows:

[0026]

[0027] Where f1(x) is the isentropic efficiency of the air compressor; f2(x) is the maximum turbulent kinetic energy of the air compressor; β 1A C is the blade inlet installation angle; c is the blade tip clearance; b c is the blade outlet width; β 2A is the blade outlet installation angle; β 1A1 and β 1A2 are the minimum and maximum values ​​of the blade inlet installation angle, C c1 and C c2 are the minimum and maximum values ​​of the blade tip clearance, b c1 and b c2 are the minimum and maximum values ​​of the blade outlet width, β 2A1 and β 2A2 These are the minimum and maximum values ​​for the blade outlet installation angle, respectively.

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] 1. The present invention adjusts and controls the coherent structure of turbulence, focuses on the main vortex structure in the disordered flow field, and uses turbulent kinetic energy to characterize the development or decay of turbulence. Compared with the optimization method that only focuses on isentropic efficiency and pressure ratio, the present invention focuses on isentropic efficiency and the energy of coherent structure in the flow field, and the optimization angle is more reasonable;

[0030] 2. The optimization objectives are isentropic efficiency and maximum turbulent kinetic energy, and the optimization parameters are blade inlet installation angle β1A, blade tip clearance Cc, blade outlet width bc, and blade outlet installation angle β2A. By constructing an agent model combined with a multi-objective genetic algorithm, the optimization efficiency is higher while targeting two optimization objectives. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a flow chart of optimization design of a preferred embodiment of the present invention;

[0032] Figure 2 The turbulent kinetic energy at 50% blade height under different blade tip clearances of the preferred embodiment of the present invention is as follows: (a) blade tip clearance = 0.2 mm; (b) blade tip clearance = 0.3 mm; (c) blade tip clearance = 0.4 mm;

[0033] Figure 3 Comparison of vorticity at 90% flow position under different blade tip clearances of the preferred embodiment of the present invention: (a) blade tip clearance = 0.2 mm; (b) blade tip clearance = 0.3 mm; (c) blade tip clearance = 0.4 mm;

[0034] Figure 4 Error analysis diagram of the neural network proxy model of the preferred embodiment of the present invention: (a) isentropic efficiency; (b) maximum turbulent kinetic energy;

[0035] Figure 5 The Pareto front optimized for isentropic efficiency and maximum turbulent kinetic energy of the preferred embodiment of the present invention. DETAILED DESCRIPTION

[0036] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0037] It should be noted that the following detailed descriptions are illustrative and are intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present application belongs.

[0038] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application; as used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or their combinations.

[0039] like Figure 1-5 As shown in the optimization design flow chart, the present invention provides a multi-objective optimization method for an air compressor impeller based on turbulence co-regulation, which is characterized by comprising the following steps:

[0040] S1. Construct an initial model of an air compressor, perform CFD simulation analysis on the initial model, and obtain the flow field of the initial model;

[0041] S2. Extracting the turbulent coherent structure in the flow field by Q criterion, analyzing the influence of the turbulent coherent structure on isentropic efficiency and maximum turbulent kinetic energy, and determining the optimization parameters and value range;

[0042] S3. Use the Latin hypercube sampling method to obtain several data points about each optimization parameter, and establish a three-dimensional model based on all the data points; perform CFD simulation analysis on all the three-dimensional models to obtain the corresponding isentropic efficiency and maximum turbulent kinetic energy, and use the optimization parameter data points and the corresponding isentropic efficiency and maximum turbulent kinetic energy as a training sample to obtain a sample set containing multiple training samples;

[0043] S4. Optimize the parameters of the impeller geometry model, and the optimization includes the following steps: S41. Based on the sample set, use neural network fitting to build a proxy model; S42. Combine the multi-objective genetic algorithm to perform multi-objective optimization on the proxy model, obtain the Pareto frontier solution set, and select the required optimal solution according to actual needs.

[0044] This embodiment conducts a simulation study on a certain air compressor.

[0045] Firstly, an initial model of the air compressor is constructed, and CFD simulation analysis is performed on the initial model to obtain the flow field of the initial model. Then, the turbulent co-ordered structure in the flow field is extracted using the Q criterion, and the influence of the co-ordered structure on the isentropic efficiency and the maximum turbulent kinetic energy is analyzed.

[0046] like Figure 2 As shown in the figure, the turbulent kinetic energy in the interference area between the leakage flow and the wake vortex is the largest. When the airflow moves downstream from the trailing edge, the turbulent kinetic energy increases first and then decreases. This is because the co-ordered structure generated by the interaction between the end wall vortex and the leakage vortex increases the turbulent kinetic energy and reaches the maximum in the interference area between the leakage vortex and the wake vortex. The vortex fragmentation caused by the interference gradually reduces the turbulent kinetic energy and dissipates it into a disordered vortex structure. The interaction between the vortex systems aggravates the unsteady loss of the flow field, and the loss increases with the increase of the tip clearance. Therefore, changes in the tip leakage vortex, wake vortex, and separation vortex are important reasons for the decrease in isentropic efficiency and the increase in turbulent kinetic energy.

[0047] Depend on Figure 3 It can be seen that the vorticity is concentrated at the blade tip, and the blade tip leakage vortex mainly comes from the leakage vortex 1 formed by the gap leakage flow shearing at the blade tip, and the leakage vortex 2 formed by the mixing of the mainstream and the leakage flow. As the blade tip clearance increases, the leakage flow increases, the volume of leakage vortex 1 increases and squeezes leakage vortex 2, reducing the effective flow area and increasing the flow loss. In summary, considering that a reasonable blade tip clearance can realize the regulation of the coherent structure and reduce the turbulent dissipation loss, the optimization range of the blade tip clearance is narrowed to 0.25-0.35 mm.

[0048] Based on this, the blade inlet installation angle, blade tip clearance, blade outlet width, and blade outlet installation angle are determined as optimization parameters, and their respective optimization ranges are 30°~35°, 0.25mm~0.35mm, 3.8mm~5.4mm, and 45°~60°.

[0049] The Latin hypercube sampling method is used to obtain several groups of impeller structure data points, and the impeller parameters are adjusted according to the data points to establish a three-dimensional model. Then, CFD simulation analysis is performed to obtain the corresponding isentropic efficiency and maximum turbulent kinetic energy in the impeller of each air compressor. The optimized parameter data points and the corresponding isentropic efficiency and maximum turbulent kinetic energy are used as a training sample to obtain a sample set containing multiple training samples. The sample set is shown in Table 1.

[0050] Table 1

[0051]

[0052]

[0053] The sample set was fitted and trained using the neural network toolbox, with the optimized parameter combination as the independent variable, isentropic efficiency and maximum turbulent kinetic energy as the dependent variable, and the data was imported into the neural network toolbox. The training data and test data ratios were set to 85% and 15%, respectively, and the number of hidden neurons was set to 10. The training obtained the proxy model of the air compressor optimization parameters, isentropic efficiency and maximum turbulent kinetic energy, and the regression line slope R was used as the proxy model accuracy evaluation index, such as Figure 4 As shown, the training effect is good.

[0054] The blade inlet installation angle β 1A , blade tip clearance C c , Blade outlet width b c and blade outlet installation angle β 2A As optimization variables, the proxy model is used as fitness function; the constraints of optimization variables are set, and the optimization intervals of blade inlet installation angle, blade tip clearance, blade outlet width, and blade outlet installation angle are 30°~35°, 0.25mm~0.35mm, 3.8mm~5.4mm, and 45°~60° respectively; the population size is set to 100, the optimal front-end individual coefficient is 0.3, the number of iterations is 150, the fitness function value deviation is 1e-5, and the crossover and mutation probabilities are 0.8 and 0.1; based on the optimization parameters and the constraints of the fitness function, a mathematical model of multi-objective optimization is established; a multi-objective genetic algorithm is used to optimize the mathematical model and the Pareto frontier is obtained, such as Figure 5 As shown in the figure, the x and y axes are the isentropic efficiency and the maximum turbulent kinetic energy, respectively. It can be seen from the figure that, in general, the improvement of the isentropic efficiency will lead to the increase of the maximum turbulent kinetic energy.

[0055] In this embodiment, the mathematical model is as follows:

[0056]

[0057] Where f1(x) is the isentropic efficiency of the air compressor; f2(x) is the maximum turbulent kinetic energy of the air compressor; β 1A C is the blade inlet installation angle; c is the blade tip clearance; b c is the blade outlet width; β 2A is the blade outlet installation angle; β 1A1 and β 1A2 are the minimum and maximum values ​​of the blade inlet installation angle, C c1 and C c2 are the minimum and maximum values ​​of the blade tip clearance, b c1 and b c2 are the minimum and maximum values ​​of the blade outlet width, β 2A1 and β 2A2 are the minimum and maximum values ​​of the blade outlet installation angle.1A1 Take 30°, β 1A2 Take 35°, C c1 Take 0.25mm, C c2 Take 0.35mm, b c1 Take 3.8mm, b c2 Take 5.4mm, β 2A1 Take 45°, β 2A2 Take 60°.

[0058] CFD simulation is used to verify the isentropic efficiency and maximum turbulent kinetic energy of the optimized air compressor. The impeller structural parameters after the original model optimization are shown in Table 2. The optimized isentropic efficiency is 81.21%, and the maximum turbulent kinetic energy is 2587.44m 2 / s 2 Compared with before optimization, the isentropic efficiency increased by 1.61% and the maximum turbulent kinetic energy decreased by 418.03m 2 / s 2 .

[0059] Table 2 Parameter comparison before and after optimization

[0060]

[0061] It should also be noted that modifications or replacements of the aforementioned embodiments by those skilled in the art cannot deviate from the scope of the technical solutions of the embodiments of the present invention; similarly, meaningless changes or refinements made to the main design concept and spirit of the present invention, which solve technical problems consistent with the present invention, should all be included in the protection scope of the present invention.

Claims

1. A multi-objective optimization method for air compressor impeller based on turbulence coherent control, characterized in that: The following steps are involved: Step S1, constructing an initial model of an air compressor, performing CFD simulation analysis on the initial model of the air compressor, and obtaining a flow field of the initial model; Step S2, extracting the turbulent coherent structure in the flow field by Q criterion, analyzing the influence of the turbulent coherent structure on isentropic efficiency and maximum turbulent kinetic energy, and determining optimization parameters and value ranges; Step S3, using the Latin hypercube sampling method to obtain a number of data points about each optimization parameter, and establishing a three-dimensional model based on all the data points; performing CFD simulation analysis on all the three-dimensional models to obtain the corresponding isentropic efficiency and maximum turbulent kinetic energy, setting the optimization parameter data point and the corresponding isentropic efficiency and maximum turbulent kinetic energy as a training sample, and obtaining a sample set containing multiple training samples; Step S4, optimizing the impeller geometry model parameters, the optimization includes the following steps: S41, constructing a proxy model based on the sample set using a neural network fitting; S42, combining a multi-objective genetic algorithm to perform multi-objective optimization on the proxy model, obtaining a Pareto frontier solution set, and selecting the desired optimal solution according to actual needs.

2. The multi-objective optimization method for air compressor impeller based on turbulence coherent control according to claim 1 is characterized in that: In step S2, the Q criterion definition is as follows: Where Ω is the rotation tensor; S is the strain rate tensor; Q is the vorticity; In step S2, the turbulent co-ordered structure refers to an ordered structure in the irregular flow in the flow field, and its manifestation in the air compressor is a separation vortex, a leakage vortex, and a wake vortex.

3. The multi-objective optimization method for air compressor impeller based on turbulence coherent control according to claim 1 is characterized in that: In step S2, the air compressor isentropic efficiency calculation formula is as follows: Where, T 1c * is the measured total air intake temperature of the air compressor, in K; T 2c * It is the measured air outlet of the air compressor, the unit is K; π c * is the air compressor pressure ratio; k is the ideal gas adiabatic index.

4. The multi-objective optimization method for air compressor impeller based on turbulence coherent control according to claim 1, characterized in that; In step S2, the turbulent kinetic energy k tke It is an indicator of the development or decay of turbulence, and its calculation formula is as follows: Where U is the average speed in m / s; I is the turbulence intensity; In step S2, the optimization parameter determined is the blade inlet installation angle β 1A , blade tip clearance C c , blade outlet width b c , Blade outlet installation angle β 2A .

5. The multi-objective optimization method for air compressor impeller based on turbulence coherent control according to claim 1 is characterized in that: In step S3, the Latin hypercube sampling method provides a value for each interval (1, 1 / n), (1 / n, 2 / n) ... (1-1 / n, 1) in n groups of data for x impeller structures and arranges them randomly.

6. The multi-objective optimization method for air compressor impeller based on turbulence coherent control according to claim 1 is characterized in that: In step S3, the generated sample set includes all combinations of optimization parameters and the corresponding isentropic efficiency and maximum turbulent kinetic energy, where the combination of optimization parameters is the blade inlet installation angle β 1A , blade tip clearance C c , blade outlet width b c and blade outlet installation angle β 2A parameter combination.

7. The multi-objective optimization method for air compressor impeller based on turbulence coherent control according to claim 1 is characterized in that: In the step S41, the sample set is fitted and trained using a neural network toolbox, and the steps are as follows: S411, taking the optimization parameter combination as the independent variable, the isentropic efficiency and the maximum turbulent kinetic energy as the dependent variables, and importing the data into the neural network toolbox; S412, setting the proportion of training data and test data, and setting the number of hidden neurons; S413, training to obtain a proxy model of the air compressor optimization parameters, isentropic efficiency and maximum turbulent kinetic energy, and using the regression line slope R as the proxy model accuracy evaluation index.

8. The multi-objective optimization method for air compressor impeller based on turbulence coherent control according to claim 7, characterized in that: In S413, the accuracy of the proxy model obtained by the neural network fitting training is determined by the slope R of the regression line. The closer R is to 1, the better the training effect.

9. The multi-objective optimization method for air compressor impeller based on turbulence coherent control according to claim 1, characterized in that: In step S42, a multi-objective genetic algorithm is used for optimization, and the optimization steps are as follows: S421, the blade inlet installation angle β 1A , blade tip clearance C c , blade outlet width b c and blade outlet installation angle β 2A As the optimization variable, the proxy model serves as the fitness function; S422, setting the constraints of the optimization variables; S423, setting the population size, the optimal front-end individual coefficient, the number of iterations, the fitness function value deviation, the probability of crossover and mutation; S424, based on the optimization parameters and the constraints of the fitness function, establishing a mathematical model of multi-objective optimization; S425, using a multi-objective genetic algorithm to optimize the mathematical model and select the optimal solution in the resulting Pareto frontier.

10. The multi-objective optimization method for air compressor impeller based on turbulence coherent control according to claim 9, characterized in that: In the S424, the mathematical model is as follows: Where f1(x) is the isentropic efficiency of the air compressor; f2(x) is the maximum turbulent kinetic energy of the air compressor; β 1A C is the blade inlet installation angle; c is the blade tip clearance; b c is the blade outlet width; β 2A is the blade outlet installation angle; β 1A1 and β 1A2 are the minimum and maximum values ​​of the blade inlet installation angle, C c1 and C c2 are the minimum and maximum values ​​of the blade tip clearance, b c1 and b c2 are the minimum and maximum values ​​of the blade outlet width, β 2A1 and β 2A2 These are the minimum and maximum values ​​for the blade outlet installation angle, respectively.

Images