Turbulent flow and non-turbulent flow interface identification method of hypersonic flow field

Abstract

The invention discloses a turbulent flow and non-turbulent flow interface identification method for a hypersonic flow field. The method comprises the following steps: acquiring a velocity field under the hypersonic flow field; converting the velocity field into a vorticity field to obtain regional volume fractions corresponding to different vorticity ranges; analyzing a vorticity threshold platform area of the TNTI interface according to the area volume fractions corresponding to different vorticity ranges; selecting a target threshold in the vorticity threshold platform area, and dividing the vorticity field into a turbulence area and a non-turbulence area according to the target threshold; a fuzzy clustering algorithm is adopted to extract a boundary between the turbulence area and the non-turbulence area, and an isosurface corresponding to the target threshold value is obtained; according to the method, one or several threshold values are not manually selected to identify the TNTI, the vorticity threshold value platform area is analyzed, all interfaces in the vorticity threshold value range conforming to the TNTI characteristics are identified, and finally the complete TNTI with the thickness is integrated.

Description

Technical Field

[0001] The present invention relates to the field of aerospace technology, and more particularly to a method for identifying turbulent and non-turbulent interfaces in a hypersonic flow field. Background Art

[0002] With the development of hypersonic vehicles, higher requirements are put forward for the external aerodynamic thermal protection of the vehicle and the flow and combustion stability in the core engine. Among them, the more common phenomenon is the process of material and energy transfer between turbulent and non-turbulent areas, which directly affects the overall safety, stability and controllability of the vehicle. In the study of turbulent and non-turbulent problems, the first focus is on how to identify the turbulent and non-turbulent interface (TNTI). This interface is actually a material layer with a small thickness, and is divided into a viscous superlayer with viscosity and an inviscid turbulent sublayer according to different characteristics. The two thin layers are at the Kolmogorov scale and Taylor scale respectively, which makes it difficult to accurately extract such a small-scale TNTI in a complex flow field.

[0003] At present, there are two main disadvantages for the commonly used method of identifying TNTI based on the threshold of vorticity or pseudo-vortex energy. First, the threshold of vorticity or pseudo-vortex energy corresponding to TNTI is relatively low, and due to interference factors such as internal dissipation of turbulence or shock waves, local small vortex structures will be formed inside the turbulent area or in the non-turbulent area. When extracting the isosurface under a certain threshold, it is easy to misidentify these non-TNTI small vortex structures, resulting in certain errors. Especially when there is shock wave interference, the area near the shock wave and its interaction with the boundary layer is often accompanied by complex small vorticity areas, which are not part of the energy and material transfer interface between the turbulent and non-turbulent regions that are of concern in the study. They are regarded as error points. Direct application of the vorticity threshold method will identify these error points and affect the final quantitative analysis results; secondly, this method requires manual selection of a suitable threshold, and the common normalization method is relatively simple and is not applicable to hypersonic conditions, which ultimately causes more errors; and, the common vorticity threshold method mostly selects a certain threshold to identify TNTI, while the real TNTI is a material layer with thickness, not a simple surface. Simple visualization can present it in this way, but the physical laws presented by different stratified areas within the TNTI are different during quantitative analysis.

[0004] In addition, the fuzzy clustering method can obtain a clearer vortex structure and does not require manual selection of thresholds, reducing manual selection errors. However, as the name of this method suggests, the extracted interface is "fuzzy", and it is difficult to explain the magnitude of the vorticity or turbulent kinetic energy of the TNTI identified by this method, resulting in the lack of clear physical meaning of the result.

[0005] Therefore, how to accurately and completely identify TNTI is a problem that technical personnel in this field need to solve urgently. Summary of the invention

[0006] In view of this, the present invention provides a method for identifying turbulent and non-turbulent interfaces in hypersonic flow fields. It no longer artificially selects one or several thresholds to identify TNTI, but analyzes the vortex threshold platform area, identifies all interfaces within the vortex threshold range that meet the characteristics of TNTI, and finally integrates them into a complete TNTI with thickness.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A method for identifying turbulent and non-turbulent interfaces in a hypersonic flow field comprises the following steps:

[0009] Obtain the velocity field under the hypersonic flow field.

[0010] The velocity field is converted into a vorticity field, and the area volume fractions corresponding to different vorticity ranges are obtained.

[0011] According to the regional volume fractions corresponding to different vorticity ranges, the vorticity threshold platform area of ​​the TNTI interface is analyzed.

[0012] A target threshold is selected in the vorticity threshold platform region, and the vorticity field is divided into a turbulent region and a non-turbulent region according to the target threshold.

[0013] A boundary line between the turbulent area and the non-turbulent area is extracted to obtain an isosurface corresponding to the target threshold.

[0014] Preferably, the step further includes: normalizing the data points in the vortex field.

[0015] Preferably, the normalization process comprises:

[0016]

[0017] Among them, ω * is the normalized vorticity, ω is the vorticity, v is the kinematic viscosity coefficient, δ is the local boundary layer thickness, u τ is the local friction velocity.

[0018] Preferably, the step of analyzing the vorticity threshold platform area of ​​the TNTI interface comprises:

[0019] Setting a vorticity threshold variable, and treating an area where the vorticity is greater than the threshold variable as a suspected turbulent area;

[0020] Based on the area volume fractions corresponding to different vorticity, the volume fraction curve of the suspected turbulent area under the change of vorticity threshold variable is generated;

[0021] A slope threshold is set, a portion of the volume fraction curve that is smaller than the slope threshold is extracted, and an interval formed by the end values ​​of the vorticity threshold of the portion is used as the vorticity threshold platform area.

[0022] Preferably, the steps further include:

[0023] Multiple target threshold selections are performed in the vorticity threshold platform area, and multiple isosurfaces are obtained accordingly as identification results of the turbulent and non-turbulent interfaces.

[0024] Preferably, a fuzzy clustering algorithm is used to extract the boundary line between the turbulent area and the non-turbulent area, and the steps include:

[0025] Initialization is performed based on the division of turbulent and non-turbulent areas;

[0026] According to the membership matrix, calculate the cluster center;

[0027] Based on the cluster center, calculate the distance from each data point to the cluster center and update the membership matrix;

[0028] Iterate until the objective function converges and output the final membership matrix and cluster centers.

[0029] Preferably, the objective function is:

[0030]

[0031] Among them, U is the membership matrix, V represents the cluster center, and u ij is the degree of membership, d ij is the distance from the data point to the cluster center, and m is the fuzzy index.

[0032] Preferably, a neural network algorithm is used to extract the boundary line between the turbulent area and the non-turbulent area.

[0033] A turbulent and non-turbulent interface identification system for a hypersonic flow field, comprising a data acquisition module, a preprocessing module, a data analysis module and an interface generation module;

[0034] The data acquisition module is used to acquire the velocity field under the high sonic flow field;

[0035] The preprocessing module is used to convert the velocity field into a vorticity field;

[0036] The data analysis module is used to analyze the volume fractions of different vorticity ranges in the vorticity field, and confirm the vorticity threshold platform area of ​​the TNTI interface according to the area volume fractions corresponding to the different vorticity ranges;

[0037] The interface generation module is used to select a target threshold in the vortex threshold platform area, and divide the vortex field into a turbulent area and a non-turbulent area according to the target threshold; and is used to extract the boundary line between the turbulent area and the non-turbulent area using a fuzzy clustering algorithm to obtain an isosurface corresponding to the target threshold.

[0038] Preferably, the preprocessing module is further used to perform normalization processing on the vorticity value in the vorticity field, and the normalization processing comprises the following steps:

[0039]

[0040] Among them, ω * is the normalized vorticity, ω is the vorticity, v is the kinematic viscosity coefficient, δ is the local boundary layer thickness, u τ is the local friction velocity.

[0041] It can be known from the above technical solutions that, compared with the prior art, the present invention discloses a method for identifying turbulent and non-turbulent interfaces in hypersonic flow fields. It no longer artificially selects one or several thresholds to identify TNTI, but analyzes the vortex threshold platform area, identifies all interfaces within the vortex threshold range that meet the characteristics of TNTI, and finally integrates them into a complete TNTI with thickness. Secondly, the present invention divides the original vortex field into two regions of "0" and "1" according to different thresholds in the vortex threshold platform area, extracts the interface, and can obtain multiple isosurfaces with clear vortex values. The present invention uses local parameters for normalization in high-speed flow fields to avoid the influence of the Reynolds number. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0043] Figure 1 A schematic diagram of a method for identifying turbulent and non-turbulent interfaces in a hypersonic flow field provided by an embodiment of the present invention;

[0044] Figure 2 A schematic diagram of the structure of a turbulent and non-turbulent interface identification system for a hypersonic flow field provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] Example 1

[0047] like Figure 1 The embodiment of the present invention discloses a method for identifying turbulent and non-turbulent interfaces in a hypersonic flow field, comprising the following steps:

[0048] S1: Obtain the velocity field under the hypersonic flow field.

[0049] S2: Convert the velocity field into the vorticity field to obtain the area volume fraction corresponding to different vorticity ranges;

[0050] S3: Analyze the vorticity threshold platform area of ​​the TNTI interface according to the area volume fraction corresponding to different vorticity ranges;

[0051] S4: selecting a target threshold in the vorticity threshold platform area, and dividing the vorticity field into a turbulent area and a non-turbulent area according to the target threshold;

[0052] S5: Extract the boundary line between the turbulent area and the non-turbulent area to obtain the isosurface corresponding to the target threshold.

[0053] In this embodiment, the present invention can extract a complete interface with thickness based on the vorticity field, that is, the vorticity threshold platform area, and by selecting the target threshold in the platform area for interface division, it can generate an isosurface with a clear vorticity level, and by extracting all the vorticity in the complete TNTI interface, all the isosurfaces are obtained, and all the isosurfaces together constitute the final TNTI interface.

[0054] Among them, regarding the acquisition of velocity field, in actual flight tests, various types of probes (such as pitot tubes, total pressure probes, etc.) can be installed to directly measure parameters such as static pressure, total pressure and temperature around the aircraft, and then calculate the velocity field information. Ground radar or other remote sensing equipment can also be used to detect the flow field characteristics around the aircraft from a long distance. Micro sensor arrays can also be installed on the aircraft to directly measure the pressure distribution on the aircraft surface and infer the flow field velocity through the pressure distribution. Modern sensor technology allows operation under hypersonic conditions and can withstand extremely high temperatures.

[0055] Regarding the transformation of the velocity field to the vorticity field, the vorticity is defined as the curl of the velocity field; in three-dimensional space, the data points in the velocity field are represented as u = (u, v, w), where u, v and w represent the velocity components in the x, y and z directions respectively, and the vorticity can be calculated by the following formula:

[0056]

[0057]

[0058] Among them, ω x ,ω y and ω z represent the vorticity components in the x, y and z directions respectively.

[0059] In order to further implement the above technical solution, the step also includes: normalizing the data points in the vortex field.

[0060] Further, the normalization process includes the following steps:

[0061]

[0062] Among them, ω * is the normalized vorticity, ω is the vorticity, v is the kinematic viscosity coefficient, δ is the local boundary layer thickness, u τ is the local friction velocity.

[0063] In this embodiment, the present invention uses the local boundary layer thickness and friction velocity to normalize the vortex field. Compared with the simple normalization method, it avoids the influence caused by changes in the Reynolds number and the like in the hypersonic flow field.

[0064] In order to further implement the above technical solution, S3 specifically includes:

[0065] S31: setting a vorticity threshold variable, and taking the area where the vorticity is greater than the threshold variable as a suspected turbulent area;

[0066] S32: Based on the volume fractions of the regions corresponding to different vorticities, a volume fraction curve of the suspected turbulent region under the change of the vorticity threshold variable is generated;

[0067] S33: setting a slope threshold, extracting a portion of the volume fraction curve that is smaller than the slope threshold, and taking the interval formed by the end values ​​of the vorticity threshold of the portion as the vorticity threshold platform area.

[0068] In this embodiment, due to the existence of TNTI, when the vorticity threshold variable changes from small to large or from large to small, there will be a slowly changing area in the volume fraction change curve of the suspected turbulent area. The present invention extracts the vorticity threshold corresponding to this slowly changing area as the vorticity range of the complete TNTI.

[0069] Regarding the calculation of volume fraction: the vorticity field is composed of the vorticity corresponding to multiple data points. By obtaining the number of data points that meet the threshold conditions and calculating the total proportion, the corresponding volume fraction can be obtained.

[0070] In order to further implement the above technical solution, S5 uses a fuzzy clustering algorithm to extract the boundary between the turbulent area and the non-turbulent area. The steps include:

[0071] S51: Initialization is performed based on the division of turbulent areas and non-turbulent areas.

[0072] S52: Calculate cluster centers according to the membership matrix.

[0073] S53: Based on the cluster center, calculate the distance from each data point to the cluster center and update the membership matrix.

[0074] S54: Iterate until the objective function converges and output the final membership matrix and cluster centers.

[0075] Furthermore, the objective function is:

[0076]

[0077] Among them, U is the membership matrix, V represents the cluster center, d ij is the distance from the data point to the cluster center, m is the fuzzy index; n is the total number of data points contained in the data set, k is the total number of cluster centers, i represents the i-th data point, and j represents the j-th cluster center. ij Indicates the membership of the i-th data point to the j-th cluster center.

[0078] In this embodiment, after obtaining the complete TNTI threshold range, the present invention needs to identify each threshold value within the range separately. Before identification, certain processing needs to be performed, that is, the original vortex field is divided into "0" and "1" label array results according to the threshold value, and the vortex field greater than the threshold value is divided into "0", and less than or equal to the threshold value is divided into "1". This array result is input into the fuzzy clustering recognition part to identify all faces within the threshold range. The core of this part is still the fuzzy clustering algorithm, which initializes the center and membership matrix of the cluster to determine the cluster cluster to which each data point belongs. Then the distance from the data point to the cluster center is calculated to analyze the minimum value of the objective function under the constraint conditions, and finally the convergence is judged to give the clustering result. Compared with conventional fuzzy clustering, the input of the present invention is no longer a velocity field, but an array result of "0" and "1" with clear physical meaning, and the interface under a certain vortex threshold value is finally obtained, which is no longer ambiguous. Repeating the above clustering process will obtain TNTI under different thresholds until all faces within the threshold range in the first part are identified, and combined together to obtain a complete TNTI result.

[0079] In another embodiment, a neural network algorithm may be used to extract the boundary between the turbulent area and the non-turbulent area. The specific steps include:

[0080] Step 0: Build a 3-layer neural network with an input layer, a hidden layer, and an output layer.

[0081] Step 1: The number of nodes in the input layer is determined by the number of classes in the data, which is 2. Similarly, the number of nodes in the output layer is determined by the number of classes we have, which is also 2. (Since we only have 2 classes, we could actually just use one output node to predict 0 or 1, but having 2 makes it easier to scale the network to more classes).

[0082] Step 2: You can choose the dimensionality of the hidden layer. The more nodes you put in the hidden layer, the more complex the functions you can accommodate.

[0083] Step 3: Choose an activation function for the hidden layer. The activation function transforms the input of a layer into its output. Non-linear activation functions allow us to fit non-linear assumptions. Common choices for activation functions are tanh, sigmoidfunction, or ReLUs.

[0084] Step 4: The activation function of the output layer is softmax, which is a way to convert raw scores into probabilities. You can think of softmax as its generalization to multiple classes.

[0085] Step 5: Train the model by learning the fitting parameters to minimize the error of the training data in an iterative process.

[0086] Step 6: Predict the data based on the trained model, divide the original data points into two parts, and obtain the interface between the two parts.

[0087] Step 7: Use this model to analyze the “0” and “1” label arrays under different thresholds, obtain the interface, and integrate it into a complete TNTI.

[0088] Example 2

[0089] like Figure 2 , based on the same inventive concept, an embodiment of the present invention discloses a turbulent and non-turbulent interface identification system for a hypersonic flow field, including a data acquisition module, a preprocessing module, a data analysis module and an interface generation module;

[0090] The data acquisition module is used to obtain the velocity field under the high sonic flow field;

[0091] The preprocessing module is used to transform the velocity field into the vorticity field;

[0092] The data analysis module is used to analyze the volume fractions of different vorticity ranges in the vorticity field, and confirm the vorticity threshold platform area of ​​the TNTI interface according to the regional volume fractions corresponding to different vorticity ranges;

[0093] The interface generation module is used to select the target threshold in the vorticity threshold platform area, and divide the vorticity field into turbulent area and non-turbulent area according to the target threshold; it is used to extract the boundary line between the turbulent area and the non-turbulent area using the fuzzy clustering algorithm to obtain the isosurface corresponding to the target threshold.

[0094] In order to further implement the above technical solution, the preprocessing module is also used to normalize the vorticity value in the vorticity field. The normalization process includes the following steps:

[0095]

[0096] Among them, ω * is the normalized vorticity, ω is the vorticity, v is the kinematic viscosity coefficient, δ is the local boundary layer thickness, u τ is the local friction velocity.

[0097] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0098] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying turbulent and non-turbulent interfaces in a hypersonic flow field, characterized in that: The following steps are involved: Obtain the velocity field under the hypersonic flow field; The velocity field is converted into a vorticity field to obtain the area volume fractions corresponding to different vorticity ranges; According to the area volume fraction corresponding to different vorticity ranges, the vorticity threshold platform area of ​​the TNTI interface is analyzed; Selecting a target threshold in the vorticity threshold platform region, and dividing the vorticity field into a turbulent region and a non-turbulent region according to the target threshold; A boundary line between the turbulent area and the non-turbulent area is extracted to obtain an isosurface corresponding to the target threshold.

2. The method for identifying turbulent and non-turbulent interfaces in a hypersonic flow field according to claim 1, characterized in that: The step also includes: normalizing the data points in the vortex field.

3. The method for identifying turbulent and non-turbulent interfaces in a hypersonic flow field according to claim 2, characterized in that: The normalization process comprises the following steps: Among them, ω * is the normalized vorticity, ω is the vorticity, v is the kinematic viscosity coefficient, δ is the local boundary layer thickness, u τ is the local friction velocity.

4. The method for identifying turbulent and non-turbulent interfaces in a hypersonic flow field according to claim 1, characterized in that: The steps of analyzing the vorticity threshold platform area of ​​the TNTI interface include: Setting a vorticity threshold variable, and treating an area where the vorticity is greater than the threshold variable as a suspected turbulent area; Based on the area volume fractions corresponding to different vorticity, the volume fraction curve of the suspected turbulent area under the change of vorticity threshold variable is generated; A slope threshold is set, a portion of the volume fraction curve that is smaller than the slope threshold is extracted, and an interval formed by the end values ​​of the vorticity threshold of the portion is used as the vorticity threshold platform area.

5. The method for identifying turbulent and non-turbulent interfaces in a hypersonic flow field according to claim 1, characterized in that: The steps also include: Multiple target threshold selections are performed in the vorticity threshold platform area, and multiple isosurfaces are obtained accordingly as identification results of the turbulent and non-turbulent interfaces.

6. The method for identifying turbulent and non-turbulent interfaces in a hypersonic flow field according to claim 1, characterized in that: The fuzzy clustering algorithm is used to extract the boundary line between the turbulent area and the non-turbulent area, and the steps include: Initialization is performed based on the division of turbulent and non-turbulent areas; According to the membership matrix, calculate the cluster center; Based on the cluster center, calculate the distance from each data point to the cluster center and update the membership matrix; Iterate until the objective function converges and output the final membership matrix and cluster centers.

7. The method for identifying turbulent and non-turbulent interfaces in a hypersonic flow field according to claim 6, characterized in that: The objective function is: Among them, U is the membership matrix, V represents the cluster center, and u ij is the degree of membership, d ij is the distance from the data point to the cluster center, and m is the fuzzy index.

8. The method for identifying turbulent and non-turbulent interfaces in a hypersonic flow field according to claim 1, characterized in that: A neural network algorithm is used to extract the boundary line between the turbulent area and the non-turbulent area.

9. A turbulent and non-turbulent interface identification system for a hypersonic flow field, characterized in that: It includes a data acquisition module, a preprocessing module, a data analysis module and an interface generation module; The data acquisition module is used to acquire the velocity field under the high sonic flow field; The preprocessing module is used to convert the velocity field into a vorticity field; The data analysis module is used to analyze the volume fractions of different vorticity ranges in the vorticity field, and confirm the vorticity threshold platform area of ​​the TNTI interface according to the area volume fractions corresponding to the different vorticity ranges; The interface generation module is used to select a target threshold in the vortex threshold platform area, and divide the vortex field into a turbulent area and a non-turbulent area according to the target threshold; and is used to extract the boundary line between the turbulent area and the non-turbulent area using a fuzzy clustering algorithm to obtain an isosurface corresponding to the target threshold.

10. A turbulent and non-turbulent interface identification system for a hypersonic flow field according to claim 9, characterized in that: The preprocessing module is also used to perform normalization processing on the vorticity value in the vorticity field, and the normalization processing comprises the following steps: Among them, ω * is the normalized vorticity, ω is the vorticity, v is the kinematic viscosity coefficient, δ is the local boundary layer thickness, u τ is the local friction velocity.

Images