Method for identifying mixed layer thickness based on carrier coordinate system

Abstract

The application discloses a mixed layer thickness identification method based on a carrier coordinate system, and the identification method comprises the following steps: importing flow field data; uniformly processing the flow field data, and marking fluid domains and non-fluid domains; identifying active flow and passive flow, and respectively tracking characteristic flow lines of the active flow and the passive flow to determine the boundary of a mixed area; establishing a carrier coordinate system, and determining an integral path set of the mixed area; integrating all the integral paths to obtain the mixed layer thickness of each integral path, i.e., to obtain the mixed layer thickness along the flow direction of the mixed area; screening out integral path discrete points with abnormal growth or reduction of the mixed layer thickness before and after the mixed layer thickness, and removing the integral path discrete points; drawing a curve graph of the mixed layer thickness along the flow direction, and completing the identification of the mixed layer thickness. The application is applied to the field of flow field mixing, can meet the calculation requirements of the mixed layer thickness under more general flow field conditions, and provides more accurate data for flow field analysis, mixed supercharging evaluation and other applications.

Description

Technical Field

[0001] This invention relates to the field of flow field mixing technology in fluid mechanics, specifically a method for identifying the thickness of the mixing layer based on a body coordinate system. Background Technology

[0002] The mixing layer thickness of ejectors, combined engines, and similar devices is a typical indicator of mixing effectiveness. Therefore, identifying the mixing layer thickness plays a crucial role in evaluating mixing efficiency. This type of mixing process is a typical jet mixing process under conditions of non-parallel velocity and static pressure mismatch. Currently, most studies on mixing processes focus on mixing processes where the jet is under pressure matching. Mixing layer studies typically use definitions such as velocity mixing layer, momentum mixing layer, vorticity mixing layer, and pressure mixing layer to describe the mixing layer thickness. Since velocity-type mixing layers depend on the similarity of the velocity profiles along the mixing layer, under pressure matching, when the active flow (primary flow, or high-speed flow) and passive flow (secondary flow, or low-speed flow) initially mix, they flow parallel to each other without a velocity angle. This ensures that the velocity profiles of the mixing layer in the Cartesian coordinate system develop similarly along the mixing layer. Therefore, using the traditional Cartesian coordinate system to describe the mixing layer thickness of the above-mentioned definitions yields good results.

[0003] Currently, the main methods for defining the thickness of the hybrid layer are:

[0004] 1. Average velocity thickness δ b Given that the velocities of the high-velocity layer and the low-velocity layer are U1 and U2 respectively, and the velocity difference is ΔU = U1 - U2, the velocity thickness can be defined as δ. b =y U0.9 -y U0.1 , where y U0.9 and y U0.1 For the lateral position corresponding to the velocity, here U 0.9 =U1-0.1ΔU,U 0.1 =U1+0.1ΔU.

[0005] 2. Brown & Roshko proposed the vorticity thickness δ in 1974. ω ,for:

[0006]

[0007] Integrating the equation for the conservation of normal momentum yields:

[0008]

[0009] in, y max It corresponds to Location;

[0010] Substituting equation (2) into equation (1) yields:

[0011]

[0012] 3. For example Figure 1 The momentum thickness δ shown θ ,for:

[0013]

[0014] The above three definitions of velocity-type hybrid layer thickness are basically based on the velocity-type hybrid layer definition.

[0015] Using the traditional Cartesian coordinate system to describe the mixing layer thickness based on velocity profiles yields good results when describing pressure-matched mixing of primary and passive flows (i.e., parallel flow mixing), as the velocity profiles develop similarly along the flow path. However, when velocity inclination is involved, using the Cartesian coordinate definition to calculate the mixing layer thickness, especially when velocity profile-based mixing layer thickness is not suitable. Figure 2 As shown, a distinct velocity angle is observed when typical active and passive flows mix. Simultaneously, coupling effects such as expansion and compression exist during the mixing process, causing the velocity profile in the mixing region to differ from that of the passive flow. Figure 1 The velocity profile shown is not full, but chaotic and lacks a specific velocity pattern. Therefore, in non-parallel mixing regions, such a velocity profile no longer exhibits the self-similar characteristic of varying along the coordinates. Traditional assumptions for mixing layer thickness identification rely on the parallelism of velocities within the control volume, with velocity patterns developing similarly along the process. Figure 2 In the process, it is found that the velocity profile of the mixed layer along the development path is irregular and dissimilar to each other, and has a non-negligible velocity component along the coordinate normal. Therefore, the velocity profile of the mixed layer thickness is very inaccurate in this coordinate system.

[0016] In particular, for complex flow structures such as vortex structures, Figure 3 As shown, the velocity profiles along the path exhibit even less similarity, making it even more difficult to adapt the velocity-based mixing layer definition to the Cartesian coordinate system, which leads to unreliable calculation results. Therefore, the reliability of velocity-based mixing layer identification in Cartesian coordinates for mixing under mismatched conditions, confined spaces, and shock-induced expansion coupling is low.

[0017] refer to Figure 4-6 The figure shows the flow path development curves of the mixing layer thickness for three velocity types in the mixing zone of the four-plate ejector in Cartesian coordinates, and the pressure contour plot of the flow field is shown below. Figure 25As shown, this flow field is a typical confined space mixing process involving shock waves, expansion waves, and a mixing layer. It can be observed that the momentum and velocity thicknesses fluctuate rapidly after a rapid increase, which is inconsistent with the development pattern of the mixing layer. The vorticity thickness clearly does not capture the mixing layer near the wall region. In general, the Cartesian coordinate system cannot accurately reflect more complex mixing flow fields. Therefore, in applications such as ejectors and combined engine ejector mixing, the high pressure ratio and shock wave / expansion wave coupling result in poor accuracy in identifying the thickness of the mixing layer using traditional velocity-based methods, leading to inaccurate assessments of mixing layer development and mixing effects. Summary of the Invention

[0018] To address the shortcomings of the existing technologies, this invention provides a method for identifying the thickness of the hybrid layer based on a body coordinate system. This method solves the problem of inaccurate prediction of the hybrid layer during expansion and compression and velocity angle mixing processes in the traditional Cartesian coordinate system. At the same time, the hybrid layer description method is modified to adapt to the hybrid layer evaluation calculation in the presence of eddies, non-fluid domains, shock wave & expansion wave coupling, and confined spaces. This provides theoretical support for the overall performance evaluation of ejectors and subsequent optimization design.

[0019] To achieve the above objectives, the present invention provides a method for identifying the thickness of a hybrid layer based on a body coordinate system, comprising the following steps:

[0020] Step 1: Import flow field data, boundary data, and inlet data;

[0021] Step 2: Homogenize the flow field data and mark the fluid and non-fluid domains in the flow field data based on the boundary data;

[0022] Step 3: Identify the active and passive flows in the flow field data based on the inlet data, and trace the characteristic streamlines of the active and passive flows respectively to determine the boundary of the mixing region;

[0023] Step 4: Establish a body coordinate system based on streamlines, and determine the integration path set of the mixing region according to the body coordinate system;

[0024] Step 5: Map the flow field data of each integration path in the integration path set to the data space, integrate all integration paths to obtain the mixing layer thickness of each integration path, that is, obtain the mixing layer thickness along the path of the mixing region.

[0025] Step 6: Based on the consistency of the calculated thickness of the mixing layer in the same mixing region, filter out and remove discrete points of the integral path where the thickness of the mixing layer increases or decreases abnormally.

[0026] Step 7: Draw a curve of the mixed layer thickness along the flow direction to complete the identification of the mixed layer thickness.

[0027] To achieve the above objectives, the present invention also provides a method for evaluating the hybrid pressurization of a multi-plate ejector, comprising the following steps:

[0028] Obtain flow field simulation data for a multi-plate ejector;

[0029] Based on the above method, a curve of the current mixing layer thickness along the flow direction of the multi-plate ejector is obtained, and the mixing length of the multi-plate ejector is further obtained.

[0030] The mixing and pressurization evaluation of the multi-plate ejector was completed based on the curves of mixing length and mixing layer thickness along the flow direction.

[0031] The present invention provides a method for identifying the thickness of a mixing layer based on a body coordinate system. The consistency of the identified mixing layer thickness is significantly improved compared with that based on a Cartesian coordinate system. It can meet the calculation requirements of mixing layer thickness under more general flow field conditions and provide more accurate data for applications such as flow field analysis and mixing pressurization evaluation. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0033] Figure 1 This is a schematic diagram of momentum-based thickness in the background art of this invention;

[0034] Figure 2 This is a schematic diagram of the mixing velocity profile in the Cartesian coordinate system under the background technology of this invention.

[0035] Figure 3 This is a schematic diagram of the eddy mixing velocity profile in the Cartesian coordinate system in the background art of this invention;

[0036] Figure 4 This is a schematic diagram of the momentum-thickness path curve in the Cartesian coordinate system in the background art of this invention;

[0037] Figure 5 This is a schematic diagram of the velocity-thickness travel curve in the Cartesian coordinate system in the background art of this invention;

[0038] Figure 6 This is a schematic diagram of the vorticity thickness profile in the Cartesian coordinate system in the background of this invention.

[0039] Figure 7 This is a schematic diagram of the development of a general flow field mixing layer in an embodiment of the present invention;

[0040] Figure 8 This is a schematic diagram of momentum thickness based on the body coordinate system in an embodiment of the present invention;

[0041] Figure 9 This is a schematic diagram of velocity-thickness based on the body coordinate system in an embodiment of the present invention;

[0042] Figure 10 This is a schematic diagram illustrating the solution of vorticity thickness based on the body coordinate system in an embodiment of the present invention;

[0043] Figure 11 This is a flowchart illustrating the identification of the hybrid layer thickness in an embodiment of the present invention;

[0044] Figure 12 This is a schematic diagram of the original data interpolation and homogenization process in an embodiment of the present invention, wherein: (a) is a schematic diagram of the original data node distribution, and (b) is a schematic diagram of the data node distribution after homogenization.

[0045] Figure 13 This is a schematic diagram of the distribution of non-fluid domain boundary files under a uniform grid in an embodiment of the present invention;

[0046] Figure 14 This is a schematic diagram of the boundary file search and closure process in an embodiment of the present invention;

[0047] Figure 15 This is a schematic diagram illustrating the non-fluid region marking principle in an embodiment of the present invention;

[0048] Figure 16 This is a schematic diagram of the streamline solution for standard points in an embodiment of the present invention;

[0049] Figure 17 This is a schematic diagram illustrating the principle of velocity angle determination in an embodiment of the present invention, wherein: (a) is V x >0 and V y A schematic diagram of the velocity angle when >0, (b) is V x <0 and V y A schematic diagram of the velocity angle when >0, (c) is V x >0 and V y A schematic diagram of the velocity angle when <0, (d) is V x <0 and V y Schematic diagram of velocity angle when <0;

[0050] Figure 18 This is a schematic diagram of the division of the mixed region unit in an embodiment of the present invention;

[0051] Figure 19 This is a schematic diagram of a streamline-based body coordinate system in an embodiment of the present invention;

[0052] Figure 20This is a schematic diagram of solving for discrete points on the integration path in an embodiment of the present invention;

[0053] Figure 21 This is a schematic diagram of the integral path solution process in the reverse q2 direction in an embodiment of the present invention;

[0054] Figure 22 This is a schematic diagram of the integral path solution process along the q2 direction in an embodiment of the present invention;

[0055] Figure 23 This is a schematic diagram of the integral path solution in an embodiment of the present invention;

[0056] Figure 24 This is a schematic diagram illustrating the calculation of integral values ​​in an embodiment of the present invention;

[0057] Figure 25 This is a pressure cloud diagram of the four ejector skin trays in an embodiment of the present invention;

[0058] Figure 26 This is a flow field velocity cloud map after homogenization in an embodiment of the present invention.

[0059] Figure 27 This is a homogenized velocity cloud map after wall data correction in an embodiment of the present invention;

[0060] Figure 28 This is the division of the mixed region unit in the embodiment of the present invention;

[0061] Figure 29 This is a schematic diagram of the integration path marked in the flow field in an embodiment of the present invention;

[0062] Figure 30 This is a schematic diagram of the calculation results without removing outliers in an embodiment of the present invention;

[0063] Figure 31 This is a momentum-thickness curve in the body coordinate system after removing outliers in an embodiment of the present invention.

[0064] Figure 32 This is a velocity-thickness curve in the body coordinate system after removing outliers in an embodiment of the present invention.

[0065] Figure 33 This is a vortex thickness curve in the body coordinate system after removing outliers in an embodiment of the present invention.

[0066] Figure 34 This is a diagram of the Pitot pressure thickness in Cartesian coordinates in an embodiment of the present invention.

[0067] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0068] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0069] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0070] Furthermore, in this invention, descriptions involving "first," "second," etc., are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0071] In this invention, unless otherwise explicitly specified and limited, the terms "connection," "fixed," etc., should be interpreted broadly. For example, "fixed" can mean a fixed connection, a detachable connection, or an integral part; it can mean a mechanical connection, an electrical connection, a physical connection, or a wireless communication connection; it can mean a direct connection or an indirect connection through an intermediate medium; it can mean the internal communication of two elements or the interaction between two elements, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0072] Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are feasible for those skilled in the art. If the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0073] Explanation of terms used in this embodiment:

[0074] Active flow (primary flow, high-speed flow): When two jets are mixed, the jet with higher energy is the one on the side with the higher energy.

[0075] Passive flow (secondary flow, low-speed flow); the jet with lower energy when two jets are mixed;

[0076] Mixing layer: The region where the mixing process occurs when two jets are mixed;

[0077] Hybrid layer thickness: Describes the thickness of the hybrid layer region.

[0078] This embodiment discloses a method for identifying the thickness of a hybrid layer based on a body coordinate system. It can be applied to scenarios in industries such as ejectors and combined engines where there is a single jet or multiple jet mixing process to evaluate the growth of the hybrid layer and obtain relevant performance indicators.

[0079] Given the self-similarity of the hybrid layer's development along its path, especially based on the velocity-dependent definition of hybrid layer thickness, the underlying assumption is that the velocity profiles of the hybrid layer are similar along its path, and that the velocities within the control volume enclosed by these velocity profiles are parallel. Figure 2 As mentioned, the Cartesian coordinate system can well satisfy the similarity of the mixing layer's development along the flow path for parallel flow mixing, but it cannot satisfy more common mixing situations such as velocity angles and entrainment between the primary and passive flows. Therefore, the coordinate system for defining the traditional mixing layer thickness is modified. To satisfy the similarity and continuity of the mixing layer's development along the flow path, and considering that the velocity profile within the mixing layer must satisfy the parallelism of velocities within the coordinate system, the body coordinate system based on streamline development ensures that the velocity is perpendicular to the q2 direction everywhere. This guarantees that the velocities within the velocity profile intercepted by this coordinate system are parallel, and the velocity profile's development along the flow path is similar, satisfying the self-similarity characteristic of the mixing layer. Therefore, the body coordinate system based on streamline development, also known as the self-similar coordinate system, is adopted. Figure 7 As shown, the q1oq2 coordinate system is a body coordinate system.

[0080] The control volume obtained based on the body coordinate system is as follows: Figure 8-9 As shown, according to the q1oq2 coordinate system, the q2 curve axis is orthogonal to the streamline. Therefore, the velocity within the control volume is orthogonal to the q2 axis everywhere. Thus, mapping the control volume in physical space to the computational space under the q1oq2 coordinate system can be transformed into... Figure 1 The transformation relationship between the q1oq2 coordinate system and the Cartesian xoy coordinate system is as follows:

[0081]

[0082] In the formula, q1 is the streamline direction, q2 is the direction orthogonal to q1 determined by the right-hand rule, and θ is the direction angle. Based on this, the thickness of the mixing layer can be calculated in the computational space, and the result can be mapped to the actual physical space. By following the baseline streamline, the development of the mixing layer thickness in the mixing region can be obtained, and other flow field data, such as the mixing layer growth rate, can be obtained.

[0083] Therefore, the main idea of ​​this embodiment is to establish a body coordinate system based on streamline tracing, identify the active and passive flows using K-means clustering analysis, trace their characteristic streamlines, divide the mixing region, and obtain the physical region for calculating the mixing layer. Then, based on the divided mixing region, a series of control volumes are established along the body coordinate system using path search and interpolation. This yields a series of velocity profiles with self-similarity along the flow path. The control volume information is stored using an integral path (computation path), where the integral path is the normal curve of the streamline cluster. Data on the integral path is read one by one to form a series of column vectors, which are then mapped from the body coordinate system to the computational space. Various definitions of the mixing layer thickness are discretized for convenient numerical calculation. The calculated thickness is then mapped back to the physical space where the mixing region is located, finally obtaining information such as the flow path development characteristics of the mixing layer. When calculating the mixing layer thickness in a general flow field, this scheme follows the following... Figure 11 Perform the steps shown. These include steps 1 through 7.

[0084] Step 1: Import flow field data, boundary data, and inlet data:

[0085] Import the flow field data from simulation calculations or data obtained through experimental measurements, including: the surface boundary condition file, the velocity file of the flow field (including component velocities and resultant velocities), the density distribution file of the flow field, and if it is necessary to calculate the thickness of the pressure-type mixing layer, the pressure distribution file, the Mach number distribution file, and the active / passive inlet identifier file, etc. The file content needs to be arranged in column vector form.

[0086] Step 2: In view of the non-uniformity of spatial distribution between simulation results and experimental results, in order to facilitate the reading, processing and storage of subsequent calculation data and improve calculation efficiency, the flow field data is homogenized and the fluid domain and non-fluid domain in the flow field data are marked based on the boundary data. Specifically, it includes the following steps 2.1-2.2.

[0087] Step 2.1, homogenization of flow field data: The flow field data is uniformly distributed across data nodes on the same uniform grid using interpolation methods. The implementation process is as follows:

[0088] Because calculations or experiments may focus on certain key flow field areas, the distribution of watershed data is not uniform. Figure 12As shown, the data recorded in reality is not uniform in the actual physical space, and the obtained flow field data cannot be directly calculated. Therefore, it is necessary to process the original data to obtain uniformly distributed flow field data. The method used in this embodiment is two-dimensional numerical interpolation. Linear interpolation, parabolic interpolation, spline interpolation, and other methods can be used to solve for the values ​​at the reset nodes. During the interpolation process, multiple nodes of the original data can be used for interpolation calculations to finally obtain uniformly distributed flow field data, which is then stored in an array.

[0089] The flow field data file generated after interpolation is a 2D array file, and the interpolation intervals dx and dy can be determined according to the required calculation accuracy and computing power.

[0090] It is important to note that in practice, the homogenization of the flow field data does not need to be strictly uniform, as long as it meets the basic computational requirements. For the specific homogenization method, one can use the interpolation methods mentioned above, or use techniques such as fitting to form local functions to supplement the flow field data, ensuring that all factors required for subsequent calculations are met.

[0091] Step 2.2, mark the fluid domain and non-fluid domain:

[0092] The initial flow field data, including the labeling of non-fluid domains such as walls, is affected by different meshing and experimental focus, and is not suitable for subsequent calculations. It is necessary to adapt it to a uniform mesh. At the same time, the obtained non-fluid domain boundaries are not closed, but only a few feature points of the boundary. In order to provide effective boundary constraints and labeling of non-fluid domains for subsequent calculations, they need to be closed.

[0093] like Figure 13 The image shows wall data from the imported data, but it only shows the wall surface and does not provide data on the non-fluid regions inside the wall. First, the original data needs to be interpolated to obtain an array file of wall data under a uniform grid. However, this cannot be done using the usual two-dimensional interpolation method, otherwise data spikes will occur, making the wall surface uneven. This solution uses a path search method to close the boundary, and its specific implementation is as follows:

[0094] First, the boundary data (column vector file) is converted into a two-dimensional array of all 1s with the same dimensions as the new flow field data after interpolation. This array represents a series of isolated scattered points, such as... Figure 14 The black dots shown are distributed in the same way. An initial search point (i) is set. start ,j start ), where i and j represent the indices of the point in the array, and the initial search point can be set to (i start ,j start)={min||(i k ,j k )-(1,1)|||(i k ,j k )∈raw data{(i1,j1),...,(i num ,j num The initial point is chosen as the closest point to (1,1) in the array; then, the search radius r is set. scan First, calculate the minimum index distance r between two points in the original data. min_scan and the maximum index distance r max_scan The exploration radius of the search point is r. scan ∈[r min_scan ,r max_scan During the search process, the search radius of each search point is incremented sequentially from smallest to largest until the next point is found. The direction angle between the two points is obtained, and then the search proceeds along the current search line, filling and marking the grids it passes through by changing the value of the node from 1 to 0, indicating that the node has changed from a fluid domain to a non-fluid domain. This process continues until the boundary file is updated, resulting in a continuous closed non-fluid domain boundary in the array.

[0095] For a closed domain, each time a ray crosses the boundary of the domain, it increments by 1. If the number of crossings is even and the point of interest is not on the boundary, it means the point of interest on the ray is outside the closed domain. If the number of crossings is odd and the point of interest is not on the boundary, it means the point is still inside the closed domain. Figure 15 As shown. Based on this principle, the non-fluid region surrounded by the non-fluid domain boundary can be marked in the flow field, obtaining a complete flow field boundary file marked with the non-fluid domain. It should be noted that the method of forming a continuous closed boundary is not limited to the path search method used. Numerical interpolation, function fitting, and other methods can also be used to reconstruct the surface or profile, thereby forming a continuous boundary in the flow field.

[0096] The generated, labeled non-fluid domain is used as a constraint to correct the homogenized flow field data, eliminating the "spurs" caused by interpolation in step 2.1, making the flow field smoother.

[0097] Step 3: Identify the active and passive flows in the flow field data based on the inlet data, and trace the characteristic streamlines of the active and passive flows respectively to determine the boundary of the mixing region. Specifically, this includes the following steps 3.1-3.2.

[0098] Step 3.1, Identification of active and passive flows:

[0099] For ejectors, combined engines, and other applications where multiple jets may mix, it is necessary to divide the mixing region into units to analyze each mixing layer and obtain relevant equipment performance data. This embodiment uses cluster analysis to determine the active and passive flows. The inlet data undergoes the same homogenization process as in step 2.2 to pinpoint the inlet location region corresponding to the homogenized flow field data. The active and passive flows are identified based on the total pressure at each inlet: higher total pressure indicates an active flow, and vice versa. The specific operation is as follows:

[0100] Acquire flow field data for the inlet region (if total pressure is used, this data represents the total pressure at the corresponding node). Starting with the node closest to (1,1) at the index position, use it as the standard point. Perform clustering using K-means clustering analysis, considering the distances between other nodes and the standard point, to obtain a data cluster centered on that point. Since the initially selected standard point may not reflect the core characteristics of a primary / passive flow, it is necessary to repeatedly select a standard point within the newly obtained data cluster and update the data cluster based on that standard point until the data cluster effectively represents all the characteristics of the primary / passive flow at the inlet. Then, remove the standard point from the current data cluster and move it to a pseudo-data cluster that can characterize the next primary / passive flow. Continue this process until all inlet data are divided into several data clusters. Based on the relative spatial positions between the data clusters and the total pressure of the corresponding standard point, it can be determined whether the flow corresponding to the data cluster is an active or passive flow. It is important to note that the distinction between primary and passive flows is relative, meaning it is based on a comparison of a certain characteristic (such as total pressure) of the two fluids, rather than being absolute.

[0101] In practical applications, clustering algorithms other than K-means clustering analysis can be used to identify and label the active and passive flows. At the same time, the evaluation criteria can be other indicators such as energy and enthalpy value, in addition to total pressure, to distinguish between the active and passive flows.

[0102] Step 3.2: Trace the characteristic streamlines to determine the mixing region and computational boundary:

[0103] Based on the active and passive flows identified in step 3.1, find the standard point corresponding to the center of the data group for each jet, and trace the streamlines emitted from the standard point:

[0104] Since the direction of streamlines is always the direction of velocity in the flow field, for a uniform discrete flow field data—a two-dimensional array—it is only necessary to find the velocity angle of the current point point(k), where point(k) is the kth discrete point on the streamline along the standard point. To ensure that the discrete points can be well adapted to the homogenized data and facilitate subsequent calculations, it is agreed that the step size along the i-th direction (the overall flow direction) is dx. Based on the velocity angle θ of point(k), the (k+1)th discrete point can be calculated until the streamline marking of the standard point is completed. The principle is as follows: Figure 16 As shown.

[0105] In specific calculations, the flow field data at point(k) is obtained by interpolating the data from neighboring nodes, such as the velocity component V at that point. x V y The calculation process is as follows:

[0106]

[0107] In the formula, dis x dis y These represent the distances from point(k) to the nodes with the smallest corresponding physical quantities between points (i,j) and (i,j+1), respectively. Flow field information for all discrete points of the standard point streamline can be obtained through calculations similar to those described above.

[0108] Based on the velocity component of point(k), the velocity angle of that point can be obtained using the velocity component. The result is obtained. However, since the solution process is a vector-based process, it is necessary to determine the magnitude of the velocity component during the specific calculation to ensure the correct velocity angle is obtained. (Refer to...) Figure 17 :

[0109] refer to Figure 17 (a), when the partial speed V x >0 and V y When >0, the velocity angle θ is:

[0110] θ = arctan(V) y / V x )

[0111] refer to Figure 17 (b), when the partial speed V x <0 and V y When >0, the velocity angle θ is:

[0112] θ = π - arctan(|V y / V| x )

[0113] refer to Figure 17 (c), when the partial speed V x>0 and V y When <0, the velocity angle θ is:

[0114] θ = -arctan(|V y / V x )

[0115] refer to Figure 17 (d), when the partial speed V x <0 and V y When <0, the velocity angle θ is:

[0116] θ=arctan(|V y / V| x )-π

[0117] The above calculations yielded the following results: Figure 18 The marking of the active / passive flow in the flow field shows the determination of each mixing region, and the streamlines of the standard points, i.e. the characteristic streamlines, serve as the upper and lower boundaries for calculating each mixing region.

[0118] In practical applications, when calculating discrete point (k) during the standard point streamline tracing of active and passive flows, multi-node interpolation can be used. Methods such as linear interpolation, spline interpolation, and parabolic interpolation can be used to solve the flow field data of the point. If the computing power allows, function fitting can also be performed to obtain the flow field data of the point.

[0119] Step 4: Establish a body coordinate system based on streamlines, and determine the integration path set of the mixing region according to the body coordinate system:

[0120] Considering the good self-similarity of the velocity profile along streamlines in the mixing layer, which satisfies the basic premise of the velocity-type mixing layer definition, this scheme determines the integration path set based on a body coordinate system established by streamlines. The integration paths generally coincide with the normal curves of the streamline clusters. Due to shock wave expansion coupling, eddies, and backflow, the integration paths may not laterally penetrate the mixing region. Therefore, the direction of the integration paths needs to be specifically declared to ensure that they reasonably penetrate the mixing region. The obtained integration path set constitutes the computational space, which can then be used to calculate the thickness of the mixing layer.

[0121] Based on equation (5), establish as follows Figure 19 The body coordinate system shown is q1oq2, where the q1 direction is the streamline direction and the q2 direction is orthogonal to the q1 direction and determined by the right-hand rule.

[0122] Based on the hybrid region units divided in step 3.2, and using the mainstream standard streamline, i.e., the boundary of the i-th hybrid region unit, as the starting boundary for computation, two different computation paths can be obtained. Specifically, the integration path along the hybrid layer can have both forward and reverse directions relative to the body coordinate q2, as shown below. Figure 21-22As shown. This computational path is called the integration path, and the boundaries of the mixed region are called the upper and lower bounds of the integration.

[0123] To obtain the integration path, we first need to obtain the velocity angle θ corresponding to the discrete points on the integration path. Based on θ and the initial flow direction, we obtain the integration path direction angle θ'. Then, based on the direction angle θ', we search forward with a step size dr until the integration path is determined. Therefore, the specific process of determining the integration path set is as follows:

[0124] Step 4.1, Discrete point interpolation calculation:

[0125] When calculating the flow field data corresponding to discrete points on the integration path, the discrete point (k) needs to be interpolated using the grid node data of the flow field. Here, the interpolation calculation uses data from four nodes. Figure 20 As shown, nodes (i-1,j), (i,j), (i,j+1), and (i-1,j+1) are numbered as node 1, node 2, node 3, and node 4 in a counter-clockwise order. First, the position information (x) of point(k) is obtained. point(k) ,y point(k) This allows us to obtain the spatial positions (x, y) of the virtual interpolation points interp1 and interp2. point(k) ,y4),(x point(k) ,y1), and then use nodes 3 and 4 and nodes 1 and 2 respectively to perform interpolation calculations on the virtual interpolation points, as follows:

[0126]

[0127]

[0128] Finally, the virtual interpolation point is used to interpolate point(k) to obtain the corresponding discrete flow field data:

[0129]

[0130] In the specific implementation process, the other flow field data of the discrete point point(k) can be calculated according to the method of equations (7)-(9).

[0131] Step 4.2, Calculation of the velocity angle of discrete points:

[0132] In the specific implementation process, the velocity angle of the discrete point point(k) is the same as that disclosed in step 3.2, so it will not be described again in this embodiment.

[0133] Step 4.3, Calculation of the direction angle of the integration path:

[0134] In the body coordinate system, there are two cases in the mixed region: the overall direction of the integration path from the characteristic streamline of the active flow to the characteristic streamline of the passive flow is consistent with the positive direction of q2 (forward q2), otherwise it is in the reverse q2 direction. There are six typical discrete point orientation angle calculation cases in the specific integration path solution process. The following is the calculation process for the integration path in the reverse q2 direction:

[0135] 1. When the velocity angle At this point, the velocity direction makes an acute angle with the mainstream direction, and the overall direction remains consistent, indicating ordinary flow. When the q2 direction in the q1oq2 coordinate system at this point is inconsistent with the integration direction, the direction angle is:

[0136]

[0137] 2. When the velocity angle At this point, the velocity direction makes an acute angle with the mainstream direction, and the overall direction remains consistent, indicating ordinary flow. When the q2 direction in the q1oq2 coordinate system at this point is inconsistent with the integration direction, the direction angle is:

[0138]

[0139] 3. When the velocity angle At this point, the velocity direction makes an obtuse angle with the mainstream direction, and the directions generally remain opposite, meaning a backflow occurs at this point, and eddies appear near this point. To avoid the integration path "backflowing" with the backflow, it is necessary to declare this. Therefore, when the q2 direction in the q1oq2 coordinate system at this point is inconsistent with the integration direction, the direction angle is:

[0140]

[0141] 4. When the velocity angle At this point, the velocity direction makes an obtuse angle with the mainstream direction, and the directions generally remain opposite, meaning a backflow occurs at this point, and eddies appear near this point. To avoid the integration path "backflowing" with the backflow, it is necessary to declare this. Therefore, when the q2 direction in the q1oq2 coordinate system at this point is inconsistent with the integration direction, the direction angle is:

[0142]

[0143] 5. When the velocity angle At that point, the velocity direction is perpendicular to the mainstream direction, and the q1oq2 coordinate system and the xoy coordinate system at that point are rotating clockwise. The direction of q2 points inward into the vortex. Without a declaration, the integration path will either terminate inward into the vortex or, at that point, outward toward the main flow boundary. This would prevent the integration path from reaching the passive flow region, leading to errors in the mixing layer integration calculation. Therefore, it is necessary to declare that the basic direction of the integration path remains unchanged; that is, the direction of the integration path here is the velocity direction at that point, with the direction angle being:

[0144] θ'=θ

[0145] 6. When the velocity angle At that time, similar to case 5, the direction angle is:

[0146] θ'=-θ

[0147] In practical applications, the method for calculating the direction angle along the q2 direction is the same as that against the q2 direction, only the direction is changed. Here, θ' represents the angle between the integration path direction and the X-axis, and its calculation is mainly based on the local velocity angle θ. A vortex structure exists here, and the integration path is tangent to this point. To ensure the integration path can smoothly transition from the active flow region to the passive flow region, the integration path direction at this point is defined as the q1 direction, i.e., the velocity direction, to prevent backtracking and ambiguity. The search for the mixing layer path along the q2 direction is similar. Based on the solved path direction angle, iterate forward dr to obtain the next discrete point point(k+1), and repeat steps 4.1-4.3 until all integration paths for the mixing unit are obtained, thus determining the integration path set for mixing layer thickness identification. It is important to note that the new boundary file generated in step 4.2 will also be the termination constraint condition for the integration path, ultimately resulting in... Figure 23 The integral path is shown as a marker in the flow field.

[0148] In practical applications, based on the body coordinate system, the flow field information of the current search starting point is first solved, the direction of the integration path is defined, and the flow field data of the searched point is obtained by searching forward in the flow field according to the direction and a fixed search step size. Multi-node interpolation can be used to calculate the flow field data of the point, and linear interpolation, spline interpolation, parabolic interpolation and other methods can be used to solve the flow field data of the point. If the computing power allows, function fitting can also be performed to obtain the flow field data of the point.

[0149] Step 5: Map the flow field data of each integration path in the integration path set to the data space, and integrate over all integration paths to obtain the mixing layer thickness of each integration path, thus obtaining the friction-crossing mixing layer thickness of the mixing region. The specific implementation process is as follows:

[0150] The thicknesses of various hybrid layers are numerically calculated within the computational space formed by the integration path sets determined in step 4. Based on the meanings of velocity thickness, momentum thickness, and vorticity thickness, the following general formula can be written:

[0151] δ=C·∫F(x,y,z)dy(10)

[0152] In the formula, δ is the thickness of the hybrid layer, C is the integral coordinate-independent term, and F(x,y,z) is the main term;

[0153] Mapping the thickness of the hybrid layer from Cartesian coordinates to the q1oq2 coordinate system, we get:

[0154] δ=C(q1,q2)·∫G(q1,q2)dq2(11)

[0155] In the formula, G(q1,q2)=G(F(x,y,z)), which means that F(x,y,z) is mapped from Cartesian coordinates to the q1oq2 coordinate system;

[0156] Discretizing the above equation yields:

[0157]

[0158] In the formula, point(k) represents the kth discrete point on the corresponding integration path, num represents the number of discrete points contained in the corresponding integration path, and dr represents the spacing between discrete points on the corresponding integration path. The integration calculation of the thickness of the mixed layer can be completed by converting the integral into a series form. In numerical calculation, the trapezoidal rule, Simpson's rule, Newton-Cotes rule, composite trapezoidal rule, composite Simpson's rule, Romberg quadrature rule, etc. can also be used to discretize equation (11) to obtain a new equation (12).

[0159] Taking momentum mixing layer calculation as an example, the thickness of the mixing layer in general mixing is calculated. Based on equation (4) and... Figure 8 The basic principle allows us to derive the formula for calculating momentum thickness in body coordinates:

[0160]

[0161] The above equation is processed using infinitesimal elements according to the definition of integral. Based on the integration path obtained in step 4, where the integration path itself stores a series of discrete points, the numerical calculation is as follows:

[0162]

[0163] The data U1(q1), U2(q1), ρ1(q1), and ΔU(q1) represent the active flow velocity, passive flow velocity, active flow side mixing boundary density, and active-passive velocity difference at q1 in the body coordinate system q1oq2 corresponding to the integration path. These values ​​can be obtained by interpolation calculation in Cartesian coordinates using equation (5), or by reading the flow field data corresponding to the upper and lower limits of the integration path. The principle diagram of numerical integration calculation is shown below. Figure 24 As shown. By performing the above integration calculation on all integration paths in the mixing region, the momentum thickness of the mixing layer along the path in the mixing region can be obtained.

[0164] Step 6: Based on the consistency of the calculated thickness of the mixing layer in the same mixing region, filter out and remove discrete points on the integral path where the thickness of the mixing layer increases or decreases abnormally.

[0165] Because errors exist during interpolation and integration along the integration path, the final result may contain obvious errors. Based on the consistency of the calculation results before and after the same mixed region, data with abnormal growth or decline is filtered out. In this embodiment, a pre-set upper and lower threshold for error is used; any error significantly exceeding the threshold is identified as an anomaly. Then, by comparing the growth rate or decay rate before and after, any data showing significantly abnormal growth or decay is identified as an anomaly. Specifically:

[0166] δ point(k) / δ point(k-1) ≥2orδ point(k-1) / δ point(k) ≤2 (15)

[0167] In the formula, δ point(k) The thickness of the mixed layer is the value corresponding to the kth discrete point along the main feature line, which is the result obtained by integrating along the kth integration path, δ. point(k-1) Similarly, this is the result obtained from the (k-1)th integration path. A point is considered an outlier if the value at each subsequent point differs by more than a factor of two, meaning the result obtained from the integration path at that point is an anomaly. For discrete points identified as outliers (on the main feature lines), their obtained hybrid layer thickness values ​​are removed, and interpolation is performed using the preceding and following points to fill in the gaps, thus eliminating ambiguous solutions. Alternatively, the 3σ principle can be used for outlier identification.

[0168] Step 7: Draw a curve of the mixed layer thickness along the flow direction to complete the identification of the mixed layer thickness.

[0169] The following examples further illustrate the hybrid layer thickness identification method based on the body coordinate system in this embodiment.

[0170] According to the method in this embodiment, the simulation data of a four-plate ejector is imported, and its flow field structure is as follows: Figure 25 As shown, this is a typical non-parallel jet mixing process. The flow field simulation data, including the surface boundary file, inlet condition file, and flow field parameter file (resultant velocity, component velocities, pressure, Mach number, density, etc.), is converted into a .txt file and stored in the program's running directory. The program is then run, and the flow field data is imported into the MATLAB workspace, completing step 1 of importing the flow field data.

[0171] Step 2 involves interpolating the column vectors in the source file into a spatially uniform two-dimensional array. In this case, dx = 0.0001 and dy = 0.0001 are used. The resultant velocity, after homogenization, is as follows: Figure 26 The flow field distribution shown is not smooth because the interpolation could not determine the termination region; data is present even where it should be on the wall. Therefore, subsequent trimming is required, and the "burrs" in the data are corrected by processing the wall file to obtain the desired smooth flow field. Figure 27 That is, complete step 2;

[0172] Step 3 involves determining the mixing region units, identifying the active and passive flows using a clustering algorithm, determining standard points, and tracing their streamlines to complete the division of the flow field mixing region units. Figure 28 As shown;

[0173] Step 4 involves determining the integration path for a given unit based on the defined hybrid region cells. Starting from the standard streamline of the mainstream flow within that cell, the process moves towards the standard streamline of the passive flow, thus completing the integration path determination for that cell. The flow field data along this path is then saved. Figure 29 The integral path distribution of the entire flow field is shown.

[0174] Step 5 involves discretizing the formula for the hybrid layer thickness and calculating the thickness along the integration path in each direction. Figure 30 As shown, the data contains ambiguous solutions due to accumulated numerical errors from interpolation and other processes, requiring further correction.

[0175] Step 6 involves using pre-set upper and lower thresholds. Any point that significantly exceeds the threshold is identified as an anomaly. The points before and after the threshold are then interpolated to fill the gap. Next, by comparing the growth rate or decay rate before and after the threshold, any point with significantly abnormal growth or decay is identified as an anomaly. This process clears the ambiguous solutions in the calculation results and finally completes the calculation and outputs the results.

[0176] After calculation, the following results were obtained respectively: Figures 31-33 The shown values ​​are momentum thickness, velocity thickness, and vorticity thickness. (Reference) Figure 4-6 and Figure 34 The calculation results are in the Cartesian coordinate system, from Figure 4-6 and Figure 34 It can be seen that, in the Cartesian coordinate system, besides Figure 34 The definition based on Pitot pressure in the Chinese model best reflects the mixing development of the flow field because it is not dependent on the velocity profile of the mixing layer and is a scalar thickness. Therefore, it is less affected by the choice of coordinate system. The other three definitions show significant improvements, particularly in that the curves for velocity thickness and vorticity thickness are largely consistent with the trends of the Pitot pressure thickness curve. Figure 31The momentum thickness shown is significantly affected by shock waves and expansion waves in the flow field, which also influences density, another important factor in the definition of momentum thickness. Therefore, momentum thickness reflects the coupling effect of shock waves and expansion waves along the flow path, and is also relatively... Figure 4 It was found that the momentum thickness based on the body coordinate system increases along the path, rather than developing rapidly and then fluctuating due to shock waves and expansion waves as in the Cartesian coordinate system. Therefore, the method for calculating the mixed layer thickness based on the body coordinate system is effective and the results are reasonable.

[0177] After identifying the mixing layer thickness, it can be used for evaluation during the mixing and pressurization process of a four-plate ejector. The process is as follows: Based on the mixing layer thickness, the mixing length of the ejector (the distance from the start of mixing, i.e., the nozzle exit face, to the position where the mixing layer is fully developed) and the mixing efficiency of the mixing layer can be determined. The shorter the mixing length, the higher the mixing efficiency. The fluctuations of the mixing layer thickness development curve can determine the coupling effect between the wave system structure and the mixing layer development within the mixing section of the ejector. Combined with the operating conditions, this helps to evaluate the pressurization effect of the ejector's main flow on the secondary flow, thereby providing theoretical support for the overall performance evaluation of the ejector and subsequent optimization design.

[0178] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for identifying a mixed layer thickness based on a carrier coordinate system, characterized in that, The method comprises the following steps: Step 1, importing flow field data, boundary data and inlet data; Step 2, homogenizing the flow field data and marking the fluid domain and the non-fluid domain in the flow field data based on the boundary data, specifically: Firstly, the boundary data is converted into a two-dimensional full 1 array with the same specification as the new flow field data after interpolation, and the flow field boundary is closed in a path search manner, the process being: Setting initial search point ; Compute the minimum index distance between two points in the boundary data and the maximum index distance , resulting in a search radius for the search point of ; In the search process, the search radius of the search point is added from small to large each time until the next point position is searched to obtain the direction angle between the two points; Then, the search line is advanced according to the current search line, and the grid passed is filled with data to mark, that is, the value on the search point is changed from 1 to 0, indicating that the fluid domain is changed to the non-fluid domain, until the boundary data is updated to obtain a continuous closed non-fluid domain boundary in the array; Secondly, the fluid domain and the non-fluid domain are marked with the non-fluid domain boundary as the limit, and the generated non-fluid domain after marking is used as a constraint to modify the homogenized flow field data, so that the flow field is smoother; Step 3, identifying the active flow and the passive flow in the flow field data based on the inlet data, and tracking the characteristic streamlines of the active flow and the passive flow respectively to determine the boundary of the mixing region; Step 4, establishing a material coordinate system based on the streamlines, and determining the integral path set of the mixing region according to the material coordinate system; Step 5, mapping the flow field data of each integral path in the integral path set to the data space, and integrating all the integral paths to obtain the mixing layer thickness of each integral path, that is, the along-path mixing layer thickness of the mixing region, specifically: The mixing layer thickness in the Cartesian coordinate is: wherein is the mixed layer thickness, C is the integral coordinate independent term, is the bulk term; Mapping the mixing layer thickness from the Cartesian coordinate to the q1oq2 coordinate system is: In the formula, , represents from the Cartesian coordinate mapping to coordinate system; Discrete processing of the above formula is: wherein point k represents the i-th discrete point on the corresponding integration path, k num represents the number of discrete points comprised by the corresponding integration path, dr represents the distance between the discrete points on the corresponding integration path;​​ Step 6, screening out the integral path discrete points with abnormal growth or reduction of the mixing layer thickness before and after, and removing them; Step 7, drawing a curve graph of the mixing layer thickness along the flow direction to complete the identification of the mixing layer thickness. 2.The method of claim 1, wherein, In step 2, the homogenization processing of the flow field data is specifically: The flow field data is uniformly distributed on the data nodes of the same set of uniform grids through an interpolation method. 3.The method of claim 1, wherein, In step 3, the identification of the active flow and the passive flow in the flow field data based on the inlet data is specifically: The inlet data is uniformly distributed and processed, and the inlet position area corresponding to the homogenized flow field data is locked; The active flow and the passive flow are identified according to the total pressure of each inlet of the flow field: the active flow is the one with high total pressure, and vice versa.

4. The method according to claim 1 or 2 or 3, characterized in that, In step 3, the tracking of the characteristic streamlines of the active flow and the passive flow to determine the boundary of the mixing region is specifically: According to the identified active flow and passive flow, the standard points of the data group centers of each jet are found, and the streamlines emitted by the standard points are tracked: According to the direction of streamline is always the velocity direction of flow field, for a uniform discrete flow field data-two dimensional array, according to the data of adjacent grid nodes, the kth discrete point of the standard point along the development streamline is obtained by interpolation The area between the characteristic streamlines of the standard points of the active flow and the passive flow is the mixing region, and the characteristic streamlines of the standard points of the active flow and the passive flow are the upper and lower boundaries of the corresponding mixing region. ( k ) flow field data and the velocity angle, the flow field data of the k+1th discrete point Step 4 specifically includes: ( k +1) can be calculated, until the characteristic streamline marking of the standard point is completed. ​ 5. The method of claim 4, wherein the mixed layer thickness is identified based on the coordinates of the marker. ​ Establish a body coordinate system q1oq2, where the q1 direction is the streamline direction and the q2 direction is orthogonal to the q1 direction and the direction is determined by the right-hand rule; The direction of integration is determined by using either the direction along or against q2 as the integration path, and interpolation is performed using the grid node data of the flow field to obtain the flow field data and direction angle of each discrete point on the current integration path. iterating forward according to the direction angle of each discrete point on the current integral path obtaining the discrete points on the next integral path and again interpolating the grid node data of the flow field to obtain the flow field data and the direction angle of each discrete point on the next integral path; repeating this process until the integral paths of all the mixing regions are obtained, i.e. the integral path set for identifying the mixing layer thickness is determined.

Images