Simulation method for entropy generation decomposition of compressor flow field based on quadruple proper orthogonal decomposition

Abstract

The present invention discloses a simulation method for entropy production decomposition of a compressor flow field based on quartic proper orthogonal decomposition. By performing quartic proper orthogonal decomposition on the unsteady flow field inside the compressor under pulsating back pressure and classifying the flow field structures, four flow field structures are obtained. Then, the entropy production is decomposed according to different turbulent dissipation mechanisms in the decomposed flow field structures, and the corresponding data is output to realize the analysis of the compressor unsteady flow field from the overall to the components one by one. The method of the present invention can quantitatively locate the irreversible losses in the compressor unsteady flow, extract and identify the unsteady characteristics generated by different mechanisms in the compressor flow field, and provide a systematic method for analyzing the evolution law of the compressor unsteady flow field under pulsating back pressure.

Description

Technical Field

[0001] The present invention relates to the field of impeller mechanical supercharging technology simulation, and in particular to a centrifugal compressor flow field entropy generation decomposition simulation method based on quadruple intrinsic orthogonal modal decomposition. Background Art

[0002] Turbocharging technology can significantly improve the power density of internal combustion engines and is a key technology for the development of the internal combustion engine industry. During the operation of an internal combustion engine equipped with a turbocharger, the engine intake pipe is connected to the compressor outlet section of the turbocharger, and each cylinder of the internal combustion engine inhales air periodically, so that there is always low-frequency pressure pulsation in the compressor outlet pipe. The propagation of pressure pulsation in the compressor will cause the internal flow field stability to become alienated and thus affect the compressor performance. At the same time, due to the presence of the volute tongue, unsteady characteristics of dynamic and static interference will appear between the impeller and the diffuser in the compressor components. This is coupled with the unsteady characteristics caused by the pulsating back pressure in the compressor, making the flow law in the compressor highly complex. Revealing the unsteady evolution law of the flow field inside the compressor has always been a research topic in this field.

[0003] Since the amount of data on the unsteady flow field of the compressor under pulsating back pressure is very large and the pulsating part of the turbulent field cannot be accurately identified, it is very difficult to extract the unsteady features generated only by the pulsating back pressure from the complex flow field inside the compressor. At the same time, quantifying and locating the irreversible losses in the unsteady flow of the compressor, extracting and identifying the unsteady features generated by different mechanisms in the compressor are also urgent problems to be solved by researchers in this field. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a centrifugal compressor flow field entropy generation decomposition simulation method based on quadruple intrinsic orthogonal modal decomposition, aiming to reveal the influence mechanism of pulsating back pressure on irreversible losses and extract the unsteady characteristics caused by pulsating back pressure.

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

[0006] The present invention provides a compressor flow field entropy generation decomposition simulation method based on quadruple intrinsic orthogonal modal decomposition, comprising the following steps:

[0007] S10, constructing a three-dimensional numerical model of the compressor flow channel according to the compressor geometry model, setting boundary conditions, and performing unsteady numerical calculations using CFD software to output unsteady flow field data;

[0008] S20, performing intrinsic orthogonal modal decomposition on the unsteady flow field data to obtain unsteady flow field modes and corresponding time coefficients of mode pairs;

[0009] S30. Perform a quadruple decomposition on the unsteady flow field mode, divide the flow field structure, and obtain the mean flow field, large-scale coherent structures, small-scale random turbulent structures, and noise fields;

[0010] S40. According to the entropy production caused by the dissipation of different flow field structures in step S30, decompose the entropy production accordingly, and output the corresponding data to achieve a step-by-step analysis of the unsteady flow field of the compressor from the overall to the components.

[0011] Further, in step S30, performing a quadruple decomposition on the unsteady flow field mode and dividing the flow field structure includes:

[0012] Take the first mode as the mean flow field structure, and use a certain threshold of the reconstructed flow field correlation coefficient between each pair of modes as the classification criterion, and classify the reconstructed flow fields of the remaining modes into large-scale coherent structures, small-scale random turbulent structures, and noise fields.

[0013] Further, in step S40, according to the entropy production caused by the dissipation of different flow field structures in step S30, decomposing the entropy production accordingly includes:

[0014] For the three flow field structures of the large-scale coherent structure, small-scale random turbulent structure, and noise field after classification, according to the entropy production generated by different dissipation mechanisms, the turbulent dissipation entropy production is decomposed accordingly;

[0015] The entropy production caused by the total turbulent dissipation consists of the entropy production caused by direct turbulent dissipation, the entropy production caused by the perturbation dissipation of large-scale coherent structures, and the entropy production caused by the dissipation of small-scale random turbulence.

[0016] Compared with the prior art, the present invention has the following beneficial effects:

[0017] A simulation method for decomposing the entropy production of a compressor flow field based on quadruple proper orthogonal mode decomposition provided by an embodiment of the present invention, through quadruple proper orthogonal mode decomposition of the unsteady flow field inside the compressor under pulsating back pressure and classifying the flow field structure, obtains four flow field structures; then, decomposes the entropy production according to different turbulent dissipation mechanisms in the decomposed flow field structure and outputs the corresponding data to achieve a step-by-step analysis of the unsteady flow field of the compressor from the overall to the components. The method of the present invention can quantitatively locate the irreversible losses in the unsteady flow of the compressor, extract and identify the unsteady characteristics generated by different mechanisms in the compressor flow field, and provide a systematic method for analyzing the evolution law of the unsteady flow field of a centrifugal compressor under pulsating back pressure. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flowchart of the simulation method for decomposing the entropy production of a compressor flow field based on quadruple proper orthogonal mode decomposition provided by an embodiment of the present invention;

[0019] Figure 2The three-dimensional numerical model of the compressor flow passage provided by the embodiment of the present invention;

[0020] Figure 3a The change curve of the correlation coefficient between mode pairs 2-7 provided by the embodiment of the present invention;

[0021] Figure 3b The change curve of the correlation coefficient between mode pairs 119-145 provided by the embodiment of the present invention;

[0022] Figure 4 The change curve of the difference in turbulent dissipation entropy production with and without pulsating back pressure provided by the embodiment of the present invention;

[0023] Figure 5 The entropy production proportion diagram of the compressor components at each moment provided by the embodiment of the present invention;

[0024] Figure 6 The distribution nephogram of turbulent dissipation entropy production of the impeller channel section provided by the embodiment of the present invention;

[0025] Figure 7 The distribution nephogram of turbulent dissipation entropy production of the diffuser section provided by the embodiment of the present invention. Detailed implementation manners

[0026] In order to make the technical means, creative features, achieved purposes and functions of the present invention easy to understand, the present invention will be further described below in conjunction with specific implementation manners.

[0027] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "upper", "lower", "inner", "outer", "front end", "rear end", "both ends", "one end", "the other end", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation to the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0028] In the description of the present invention, it should be noted that unless otherwise clearly specified and limited, the terms "installed", "provided with", "connected", etc. should be understood in a broad sense. For example, "connected" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0029] Refer to Figure 1As shown in the figure, a simulation method for decomposing the entropy production of a compressor flow field based on quartic proper orthogonal decomposition provided by the present invention includes the following steps:

[0030] S10. Construct a three-dimensional numerical model of the compressor flow passage according to the compressor geometric model, set boundary conditions, and perform unsteady numerical calculations using CFD software to output unsteady flow field data;

[0031] S20. Perform proper orthogonal decomposition on the unsteady flow field data to obtain unsteady flow field modes and corresponding time coefficients of the modes;

[0032] S30. Perform quartic decomposition on the unsteady flow field modes, divide the flow field structure, and obtain the mean flow field, large-scale coherent structures, small-scale random turbulent structures, and noise fields;

[0033] S40. Decompose the entropy production correspondingly according to the entropy production caused by the dissipation of different flow field structures in step S30, and output the corresponding data to realize the analysis of the compressor unsteady flow field from the overall to each component one by one.

[0034] The following will separately elaborate on each of the above steps in detail:

[0035] In the above step S10, the compressor studied in the embodiment of the present invention is a compressor matched with a free piston internal combustion generator, and its specific geometric model parameters are shown in Table 1, including: the number of blades, the impeller inlet diameter, the impeller outlet diameter, the impeller axial length, the tip clearance, the diffuser width, the cross-sectional area of the compressor housing 0-0, etc. Build a full three-dimensional grid model according to the three-dimensional model of the compressor flow passage, referring to Figure 2 , the number of grids of the vaneless diffuser a is 780,000, the number of grids of the compressor housing grid b is 2,190,000, the circular flow passage model c adopts a butterfly topology structure, and the total number of grids of the full-channel impeller d is 4,290,000. The impeller grid adopts an HHCHH type grid topology structure.

[0036] Table 1 Specific geometric parameters of the compressor

[0037]

[0038]

[0039] In the embodiment of the present invention, the Reynolds-averaged Navier-Stokes equations are used for unsteady numerical calculations, and the physical time step of the unsteady calculation iteration is 6.5×10 -6During the complete rotation period of the rotor, the time step is 103 steps, that is, the iterative calculation is performed every 3.5°, and the total calculation duration is 0.033 s. The harmonic function y = 10130sin720πt + 202600 Pa (where t is the calculation time and y is the compressor outlet pressure) is used to simulate the compressor outlet pressure pulsation. The inlet and outlet boundary conditions are set according to Table 2, including the compressor outlet pressure corresponding to three working conditions of steady state (steady), non-pulsating back pressure (unsteady), and pulsating back pressure (unsteady), as well as the unified compressor inlet parameters. To accelerate the calculation convergence speed, the steady-state calculation result is used as the initial field for the unsteady calculation. To extract the unsteady characteristics inside the compressor caused only by the outlet pulsating back pressure, the unsteady calculation under the condition of no outlet pulsating back pressure is also carried out. After the calculation is completed, the flow field data is output.

[0040] Table 2 Boundary Condition Setting

[0041]

[0042] In the above step S20, in the embodiment of the present invention, the proper orthogonal decomposition is performed on the unsteady flow field data in step S10 to obtain the unsteady flow field modes and the corresponding time coefficients of the modes. The specific operation is as follows:

[0043] The pulsating part of the compressor unsteady flow field (s (k) ) can be decomposed into the product of a set of spatial eigenmode matrices ( orthogonal basis vectors) and a time coefficient matrix , that is, Equation (1).

[0044]

[0045] S = [s 1 , s 2 ..., s k-1 , s k (2)

[0046] where m is the number of snapshots of the compressor unsteady flow field, k is the kth snapshot of the compressor unsteady flow field. In the embodiment of the present invention, 145 flow field snapshots are selected, and m = 145. These flow field snapshot data are written into the matrix S in the form of stacked column vectors. represents an n-column matrix, n is the number of finite basis functions, and i is the ith basis function.

[0047] For any flow field snapshot, to reconstruct the flow field using the least number of basis functions (1) to capture more energy in the compressor unsteady flow field, Equation (3) needs to be satisfied.

[0048]

[0049] where ||·||F is the Frobenius norm of the matrix. The eigenmode matrix is huge, and singular value decomposition can be used to obtain an optimal low-order matrix approximation, which is introduced into the covariance matrix L:

[0050]

[0051] where S T represents the transpose of matrix S; λ i is the quadruple proper orthogonal decomposition eigenvalue, and ψ i is the eigenvector of the flow field snapshot. The eigenmode of the compressor unsteady flow field is obtained by transforming the eigenvector ψ i :

[0052]

[0053] The time coefficients corresponding to the compressor unsteady flow field modes can be obtained from the eigenmodes:

[0054]

[0055] represents the transpose of matrix ;

[0056] Through equations (1)-(7), the compressor unsteady flow field modes and the time coefficients corresponding to the modes are obtained.

[0057] In step S30 above, in the embodiment of the present invention, the quadruple decomposition is performed on the compressor unsteady flow field modes obtained in step S20 to divide the flow field structure. The specific operations are as follows:

[0058] The first mode has the most flow information, so it is used as the mean flow field structure. The correlation index (RI) is introduced to quantitatively evaluate the similarity of the reconstructed flow field structures between mode pairs.

[0059] RI = R[e i,j (n i :n j ),e i,j+1 (n i :n j+1 )] (8)

[0060]

[0061] Taking a certain threshold of the correlation index of the reconstructed flow fields (e i,j+1 (n i :n j+1 )) between each mode pair as the classification criterion, in the embodiment of the present invention, this threshold is taken as 0.98, and the reconstructed flow fields of the remaining modes are classified into large-scale coherent structures, small-scale random turbulent structures, and noise fields, referring toFigure 3a , the correlation coefficient between mode pairs 2 - 7 exceeds 0.98, and the reconstructed flow field is regarded as large-scale coherent structures. Refer to Figure 3b , the reconstructed flow field between mode pairs 119 - 145 is regarded as noise structures, and the remaining mode reconstructed flow fields are small-scale turbulent structures.

[0062] In the above step S40, in the embodiment of the present invention, according to the entropy production caused by the dissipation of different flow field structures in step S30, the entropy production is decomposed accordingly. The specific operation is as follows:

[0063] For the three classified flow field structures, the total turbulent dissipation entropy production (E gen，tur ) is decomposed accordingly. According to the entropy production generated by different dissipation mechanisms, the entropy production caused by turbulent dissipation consists of the entropy production generated by direct turbulent dissipation the entropy production generated by the perturbation of large-scale coherent structures and the entropy production generated by small-scale random turbulent dissipation (E′ gen，tur ).

[0064]

[0065]

[0066] In the formula, μ is the aerodynamic viscosity, is the average temperature, and x, y, z are the three directions in space. are the average velocities in the x, y, and z directions respectively, are the large-scale perturbation velocities in the x, y, and z directions respectively, and u′, v′, w′ are the small-scale random pulsation velocities in the x, y, and z directions respectively.

[0067] In the embodiment of the present invention, 5 calculation moments are selected to analyze the unsteady flow field of the compressor from the overall to the components one by one. First, in order to study the influence of pulsating back pressure on the internal irreversible loss of the compressor, the differences in turbulent dissipation entropy production in the compressor impeller, diffuser, and compressor housing under the conditions of with and without pulsating back pressure are compared. Refer to Figure 4 , it can be seen that the influence of pulsating back pressure is mainly reflected in the impeller and diffuser. The amplitude of the entropy production difference in the compressor housing is significantly reduced (about 1%), and the impeller and diffuser need to be focused on. After the triple decomposition of turbulent dissipation entropy production, refer to Figure 5 , it can be seen that the dissipation entropy production caused by the perturbation of large-scale coherent structures in the impeller is dominant, and the amplitude fluctuates greatly. Under the condition of pulsating back pressure, the increase in the dissipation of large-scale coherent structure perturbation and small-scale random turbulent dissipation in the impeller is Figure 4Reasons for the large difference in entropy production. After the high-speed airflow enters the vaneless diffuser from the impeller, it gradually stabilizes, and the flow field structure is mainly the mean flow. The entropy production due to direct turbulent dissipation accounts for more than 92%. Compared with the impeller, the dissipation of large-scale coherent structure disturbances in the diffuser is greatly reduced, but its volatility is still affected by the pulsating back pressure.

[0068] In the embodiments of the present invention, the specific locations of irreversible losses in each component of the compressor are further analyzed one by one. First is the impeller. Six axial planes are intercepted in the impeller passage along the flow direction, and the cloud diagrams of turbulent dissipation entropy production are drawn on them. The instantaneous flow fields at times t1 and t2 are selected for analysis, as Figure 6 shown. It can be seen that the unsteady characteristics generated by the pulsating back pressure in the impeller are that the entropy production caused by the disturbance dissipation of large-scale coherent structures is concentrated in the impeller tip clearance and the pressure side of the blade, and the entropy production caused by the disturbance dissipation of large-scale coherent structures and the entropy production caused by small-scale random turbulent dissipation are consistent with the volatility of the pulsating back pressure. The distribution of the cloud diagram of turbulent dissipation entropy production in the diffuser is as Figure 7 shown. It can be seen that the unsteady characteristics of the stator-rotor interaction in the unsteady flow field of the compressor are the phenomenon of alternating high and low entropy production areas near the diffuser inlet. The specific manifestation form of the unsteady characteristics under the action of the pulsating back pressure is that the high entropy production areas caused by the disturbance of large-scale coherent structures develop circumferentially in the diffuser annular passage.

[0069] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these changes and modifications.

Claims

1. A simulation method for decomposing the entropy generation of a compressor flow field based on quadruple proper orthogonal decomposition, characterized in that, It includes the following steps: S10. Construct a three-dimensional numerical model of the compressor flow passage according to the compressor geometric model, set boundary conditions, and perform unsteady numerical calculations using CFD software to output unsteady flow field data; S20. Perform proper orthogonal decomposition on the unsteady flow field data to obtain unsteady flow field modes and the corresponding time coefficients of the modes; S30. Perform quadruple decomposition on the unsteady flow field modes, divide the flow field structure, and obtain the mean flow field, large-scale coherent structures, small-scale random turbulent structures, and noise fields; S40. According to the entropy production caused by the dissipation of different flow field structures in step S30, decompose the entropy production accordingly and output the corresponding data to realize the analysis of the compressor unsteady flow field from the overall to each component one by one.

2. The simulation method for decomposing the entropy generation of a compressor flow field based on quartic proper orthogonal decomposition according to claim 1, wherein In step S30, performing quadruple decomposition on the unsteady flow field modes and dividing the flow field structure includes: Taking the first mode as the mean flow field structure, and using a certain threshold of the reconstructed flow field correlation coefficient between each pair of modes as the classification criterion, classifying the reconstructed flow fields of the remaining modes into large-scale coherent structures, small-scale random turbulent structures, and noise fields.

3. The simulation method for compressor flow field entropy generation decomposition based on quartic proper orthogonal decomposition according to claim 2, characterized in that In step S40, according to the entropy production caused by the dissipation of different flow field structures in step S30, decomposing the entropy production accordingly includes: For the three flow field structures of large-scale coherent structures, small-scale random turbulent structures, and noise fields after classification, the turbulent dissipation entropy production is decomposed accordingly according to the entropy production generated by different dissipation mechanisms; The entropy production caused by total turbulent dissipation consists of the entropy production caused by direct turbulent dissipation, the entropy production caused by the perturbation dissipation of large-scale coherent structures, and the entropy production caused by the dissipation of small-scale random turbulence.

Images