Space-time non-contact measurement method and device for spanwise two-dimensional frictional resistance of wall turbulent flow

Abstract

The invention belongs to the technical field of fluid mechanics testing, and particularly relates to a space-time non-contact measuring method and device for spanwise two-dimensional friction resistance of wall turbulent flow. The measurement method comprises the following steps of: measuring a particle image pair sequence in a flow field test, and processing and calculating particles in a flow spanwise image by adopting a particle tracking velocity measurement algorithm to obtain particles in a near-wall flow spanwise thin layer with relatively high temporal-spatial resolution and velocity distribution conditions of the particles; the flow friction resistance on a two-dimensional plane is calculated through the normal velocity gradient according to the probability density function velocity layering principle, the flow normal statistical law and the Newton internal friction theorem. The measuring device using the method comprises a flow field, a laser, an image recording instrument, a synchronous controller and a computer. The flow friction resistance measured by the method has the characteristics of instantaneity, no interference and high temporal-spatial resolution, the evolution condition of the wall flow spanwise two-dimensional plane friction resistance along with time and space can be measured, and the method has the advantages of low requirement on equipment, insensitivity to external conditions such as temperature and humidity and the like.

Description

Technical Field

[0001] The present invention belongs to the field of fluid mechanics testing technology, and relates to particle image velocimetry (PIV) technology in the field of laser velocimetry technology, and in particular to a time-space non-contact measurement method and device for the spanwise two-dimensional friction resistance of a wall turbulent flow. Background Art

[0002] The spanwise frictional drag in wall turbulence refers to the viscous drag generated by the friction between the fluid and the wall during fluid flow. The distribution and evolution of this drag in the two-dimensional (streamwise and spanwise) plane space of the wall are one of the important contents of studying wall turbulence and drag reduction control. It plays an important role in optimizing the fluid drag reduction control methods in various engineering systems and improving flow efficiency. The spatiotemporal distribution and evolution of wall turbulence frictional drag are caused by the continuous generation and dissipation of vortex structures in turbulence. The time-averaged frictional drag can be obtained by time averaging the instantaneous value; its spatial variation is mainly reflected in the shear stress distribution at different positions on the wall. Generally, the frictional drag is larger in the area close to the wall, and the frictional drag gradually decreases with the increase of the distance from the wall. At the same time, the measurement of spanwise frictional drag of wall turbulence with high spatiotemporal resolution is crucial to understanding the development of high- and low-speed strips in wall turbulence and the internal connection of turbulent coherent structures. It is the theoretical premise and basis for optimizing flow drag reduction control strategies and carrying out innovations in flow drag reduction.

[0003] In the prior art, the methods for measuring wall friction resistance mainly include: hot wire anemometer, oil film interferometry, particle image velocimetry, SRCC (Spectral Relaxation and Conjugate Correlation) algorithm and SPEC (Spectral Element Poisson-Cauchy) algorithm. However, these methods all have some disadvantages: (1) The hot wire anemometer follows Newton's law of internal friction and estimates the friction resistance of the wall by determining the vertical gradient of the flow velocity near the wall. This technology is often implemented by measuring the velocity at two different heights in the viscous bottom layer and then obtaining the velocity gradient. This method is susceptible to interference from the velocimeter and resolution limitations. (2) The oil film interferometry method measures shear stress by observing the principle that shear stress causes changes in the thickness of the thin oil film. This method can effectively measure shear force, but its calibration process is relatively complicated and requires extremely high accuracy of the optical system. (3) The PIV (Particle Image Velocimetry) technology developed in recent years uses a non-contact optical measurement method to measure the flow field velocity near the wall. It can solve the problem of probe / coating interference. The PIV velocity measurement technology is applied to the flow field measurement of the viscous bottom layer of wall turbulence. Based on Newton's internal friction law, the wall friction resistance can be obtained. Since the calculation of wall friction resistance depends on the normal gradient of the flow velocity at the wall, the key to improving the measurement accuracy of friction resistance is to obtain a high-resolution velocity field in the viscous bottom layer. However, the existing PIV technology cannot guarantee that a high-resolution viscous bottom layer flow field can be obtained under a certain flow scale. (4) The SRCC algorithm technology is an improved algorithm based on the particle image cross-correlation algorithm based on the characteristics of the large velocity gradient near the wall. It obtains the friction resistance of the flow normal to the wall by sacrificing the spatial resolution of the flow direction. (5) The SPEC algorithm technology is also based on the particle image cross-correlation algorithm based on the characteristics of the large velocity gradient near the wall. By sacrificing the time resolution, a single pixel is used to obtain the time-averaged flow normal to the wall friction resistance. These two algorithm technologies have high requirements on the particle images near the wall and cannot reflect the changes in the friction resistance in the flow span direction.

[0004] Therefore, due to the limited resolution and accuracy of the measurement equipment, the rapid capture capability of the measurement technology, and the complexity and variability of the wall turbulence flow environment, there is no effective and practical method and device for measuring the friction resistance of wall turbulence flow in the laboratory. Therefore, it is still an unsolved technical problem to measure the flow friction resistance signal of the wall surface in two-dimensional space with high temporal and spatial resolution based on non-contact methods. Summary of the invention

[0005] The purpose of the present invention is to overcome the deficiencies in the prior art and to provide a method for measuring the spanwise frictional resistance of a turbulent flow based on particle image velocimetry (PIV). The method can measure the instantaneous frictional resistance of the flow spanwise to the wall by a non-contact optical method in a flow field simulation device such as a water tank, a water tunnel, a wind tunnel, or a real flow field, and perform statistics on its two-dimensional distribution.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solution: a time-space non-contact measurement method of the spanwise two-dimensional friction resistance of a wall turbulent flow, comprising the following steps:

[0007] Step 1: Use particle image velocimetry (PIV) technology to measure the flow velocity v of each tracer particle. 动流 :(1) Spreading tracer particles in the flow field;(2) Stabilizing the flow field environment and mixing the tracer particles evenly;(3) Generating flow span and normal laser illumination, using a laser to generate a laser strip light source to illuminate a thin layer close to the wall and parallel to the wall flow span, the thickness of the thin layer is less than the height of the corresponding wall turbulence buffer layer; using a laser to generate a laser strip light source to vertically illuminate the wall and illuminate the flow normal plane;(4) Collecting tracer particle images in the flow span and flow normal of the near-wall region, wherein the sampling time and the number of collected time series particle images are selected, and the time interval between two adjacent images is Δt;(5) Performing image processing on the collected tracer particle images, and using the PCSS particle image cross-frame tracking algorithm to obtain the motion flow velocity v of each tracer particle in the thin layer below the near-wall buffer and with the viscous bottom layer in the linear region as the core 动流 .

[0008] Among them, the flow field is generated by flow field simulation devices such as water flumes, water tunnels, wind tunnels, or real flow fields.

[0009] Preferably, the laser band emitted by the laser has a thickness, covers the flow field area below the near-wall buffer zone and centered on the viscous bottom layer of the linear zone, and the irradiated flow field is uniform laser sheet light, and the irradiated laser is close to the wall surface.

[0010] Preferably, in the first step (4), the pixel size of the tracer particles moving within the cross-frame time is less than or equal to R / 2.

[0011] Preferably, the image processing method comprises the following steps: (1) grouping particle images of adjacent frames: grouping the collected tracer particle images frame by frame, wherein the first group of images are tracer particle images recorded at time t, denoted as set1; the second group of images are adjacent tracer particle images recorded at time t+Δt, denoted as set2. (2) matching each tracer particle in set1 with each tracer particle in set2 images: displaying each tracer particle in each image in set1 with grayscale intensity, and using the PCSS method, finding the tracer particles in the image corresponding to the image in set1 in set2 for matching, specifically in the following manner:

[0012] In the tracer particle adjacent image, let the tracer particle i in set1 be the center and R be the radius to make a query area I. Then select a candidate particle j in set2 and make a query area J with R as the radius. Use S ij represents the similarity coefficient between the two particles, and the calculation expression is shown in formula (5):

[0013]

[0014] In the formula, ε r and ε θ are the thresholds of distance deviation and angle deviation respectively; r im and θ im are the distance and connection angle between particle i and particles in query area I respectively; r jn and θ jn are the distance and connection angle between particle j in the candidate particles and particles in the query area J, respectively; M and N represent the number of particles in the query areas I and J, respectively.

[0015] Preferably, the distance between the position of the candidate particle j in set2 and the position of the tracer particle i is less than or equal to R, and the size of R is related to the maximum particle displacement of the flow field.

[0016] Preferably, the time interval Δt in the first step (4) and the selected distance deviation ε in formula (5) are r and angular deviation ε θ The threshold value, tracer particle diameter and other parameters are selected according to the experimental test requirements.

[0017] H(x,y) in formula (5) is a step function, and its expression is shown in (6):

[0018]

[0019] That is, for all candidate particles j of particle i in set2, a similarity coefficient S can be calculated according to formula (6): ij, select the candidate particle j with the largest similarity coefficient as the paired particle of particle i. Then perform the above pairing on all particles in set1 to complete the particle pairing process of the entire image, and then obtain the displacement of the tracer particle between set1 and set2, and finally obtain the movement speed v of each tracer particle at adjacent moments 动流 .

[0020] Step 2: Use velocity probability density to stratify the particle image, obtain the average flow velocity in different velocity intervals, and then use the time-space average flow normal statistical law curve to obtain the average normal height corresponding to the average flow velocity, which specifically includes the following steps:

[0021] (1) Probabilistic statistics and stratification of particles in rectangular regions: Each frame of the particle image in the time series is divided into multiple rectangular regions, the number and size of the rectangular regions are set, and the flow velocity v of the tracer particles in each rectangle in the time series particle image is calculated. 动流 The PDF probability density analysis is performed, and according to the PDF results, the particle flow velocity is divided into three different speed ranges according to the flow velocity, namely v 1 、v 2 、v 3 .

[0022] (2) For three different speed intervals v 1 、v 2 、v 3 , perform time and space averaging: perform time averaging on the particle speeds in the rectangular area at the same position in the time series to obtain the time-averaged speed interval v t1 、v t2 、v t3 , and then v t1 、v t2 、v t3 Rectangular space averaging is performed to obtain three different speed intervals v after spatial and temporal averaging. ts1 、v ts2 、v ts3 , calculate the average flow velocity of these three speed intervals, and obtain the average flow velocity value v of three different speed intervals 均1 、v 均2 、v 均3 .

[0023] (3) Statistics of flow normal: Using the flow normal particle image collected in step (4), the particle image is calculated using the query window-based cross-correlation algorithm commonly used in PIV technology. The spatiotemporal average flow velocity u under this condition is + With normal height y +The statistical law curve L is the classic scale velocity curve in the turbulent boundary layer.

[0024] (4) The average velocity corresponds to the average height, and the average flow velocity v in different speed intervals is calculated 均1 、v 均2 、v 均3 The corresponding different heights y 均1 ,y 均2 ,y 均3 :v 均1 、v 均2 、v 均3 Corresponding flow velocity u + With normal height y + In the statistical law curve L, the flow velocity u + , normal height y + The calculation expressions of are shown in formula (1) and formula (2), respectively.

[0025]

[0026] In formula (1), v is substituted into v 均1 、v 均2 、v 均3 ; and u τ From the second step (3),

[0027] In formula (2), υ is the kinematic viscosity coefficient of the fluid, which is a known quantity and is related to the temperature in the experiment.

[0028] Using formula (1) to obtain v 均1 、v 均2 、v 均3 in u + After the corresponding value on the axis, according to the flow velocity u + With normal height y + The statistical law curve L finds its + After obtaining the corresponding value on the axis, formula (2) is transformed to obtain formula (3):

[0029]

[0030] Then we can calculate y 均1 ,y 均2 ,y 均3 , that is, the single-layer particle image is divided into y 均1 Height particle image, y 均2 Height particle image, y 均3 Highly particle image.

[0031] Step 3: Calculate the friction resistance τ using Newton's internal friction formula ω:Use v obtained in the second step (4) 均1 →y 均1 ,v 均2 →y 均2 ,v 均3 →y 均3 , find the velocity gradient, and apply Newton's internal friction formula (4),

[0032]

[0033] Where μ is the dynamic viscosity coefficient of the fluid, It is the normal gradient of the flow velocity after spanwise stratification of the near-wall flow for which friction resistance measurement is required, indicating the change in velocity per unit normal height.

[0034] In view of the above-mentioned measurement method, the present invention also proposes a measurement device, including a flow field, a laser, an image recording instrument, a synchronous controller and a computer, wherein the flow field provides an experimental environment and a number of tracer particles are sown in the flow field; one laser in the flow span direction and one laser in the flow normal direction are arranged below the flow field to provide the flow field with light sources in the flow span direction and the flow normal direction; one image recording instrument in the flow span direction and one image recording instrument in the flow normal direction are arranged directly in front of the flow field positions corresponding to the lasers in the flow span direction and the flow normal direction to facilitate the collection of tracer particle images; the lasers in the flow span direction and the flow normal direction and the image recording instrument in the flow span direction and the flow normal direction are all electrically connected to a synchronous controller, and the synchronous controller is electrically connected to the computer.

[0035] Preferably, a sheet light system is installed at the front end of the laser in the flow span direction and the flow normal direction, so that the laser beam emitted by the laser is diffused into a uniform sheet light, and the laser light sheet in the flow normal direction is perpendicular to the velocity direction of the flow field and parallel to the interface surface between the buffer layer and the viscous bottom layer of the boundary layer plate where friction resistance measurement is required; the light sheet emitted by the laser in the flow span direction is parallel to the velocity direction of the flow field and perpendicular to the boundary layer plate.

[0036] Preferably, the image recording instrument is a high-speed camera, and the optical axis of the high-speed camera is perpendicular to the plane of the uniform sheet light provided by the laser.

[0037] A high-speed camera is used to collect images of tracer particles in the flow direction in the near-wall area along the outer normal direction of the surface where friction resistance measurement is required along the boundary layer within a certain period of time; images of tracer particles in the flow normal direction along the boundary layer within a certain period of time are collected by the high-speed camera, converted into digital signals and stored in a control computer; a synchronous controller has three interfaces, which are respectively connected to the laser, the high-speed camera and the computer through connecting lines; the digital signal sent by the computer is received by the synchronous controller, and is used to trigger the high-speed camera and the laser to work, so that the high-speed camera can collect cross-frame image pairs of tracer particles in single exposure mode.

[0038] The beneficial effects of the present invention are:

[0039] (1) The innovative velocity stratification principle based on probability density function is used to stratify the thin linear regions of the wall turbulent viscosity bottom layer illuminated by the PIV light source. This stratification method is based on finite flow direction and span direction scales and instantaneous flow field. Therefore, the friction signal obtained by applying Newton's internal friction law to the corresponding stratified unit velocity and gradient information has a smaller spatial scale and time scale; (2) It solves the technical difficulties of measuring the spatiotemporal signal of flow friction resistance in two-dimensional wall space with high spatiotemporal resolution based on non-contact methods, and has low equipment requirements and is insensitive to external conditions such as temperature and humidity. This invention can promote the coupling research of turbulent friction resistance and turbulent structure on a self-consistent scale of time and space, and is conducive to providing a theoretical basis for further optimizing flow drag reduction strategies and innovating flow drag reduction methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a flow chart of the measuring method of the present invention;

[0041] Figure 2 This is the experimental picture of the PCSS particle image cross-frame tracking algorithm;

[0042] Figure 3 This is a schematic diagram of the layered particle image in the flow span direction;

[0043] Figure 4 The average velocity distribution u of the two-dimensional turbulent boundary layer is obtained by calculating the flow normal particle image. + (y + );

[0044] Figure 5 This is a schematic diagram of the principle of calculating the wall friction resistance based on Newton's law of internal friction;

[0045] Figure 6 This is a tracer particle image taken by a high-speed camera;

[0046] Figure 7(a) is the calculated flow span-wise wall friction velocity distribution image;

[0047] FIG7( b ) is a pre-multiplied spectrum distribution image calculated by obtaining the flow spanwise wall friction velocity map;

[0048] FIG8( a ) is a relative error distribution of the flow spanwise wall friction resistance calculated in the flow spanwise layering algorithm embodiment of the present invention compared with the DNS data;

[0049] FIG8( b ) is the distribution of autocorrelation and cross-correlation signals of the flow spanwise wall friction resistance and DNS data calculated in the embodiment of the flow spanwise layering algorithm of the present invention;

[0050] Fig. 9It is a schematic diagram of the structure of the measuring device in the present invention.

[0051] Reference numerals:

[0052] 1. Water tank; 2. High-speed camera; 3. Laser; 4. Synchronous controller; 5. Computer; 6. Tracer particles; 7. Light sheet system. DETAILED DESCRIPTION

[0053] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the structure of the present invention.

[0054] Embodiment 1: A non-contact time-space measurement method for the spanwise two-dimensional friction resistance of a wall turbulent flow according to the present invention, such as Figure 1 As shown, this is achieved through the following steps:

[0055] First, the first step is to use particle image velocimetry (PIV) technology to measure the flow velocity v of each tracer particle. 动流 :(1) Spreading tracer particles in the flow field;(2) Stabilizing the flow field environment and mixing the tracer particles evenly;(3) Generating flow span and normal laser illumination, using a laser to generate a laser strip light source to illuminate a thin layer close to the wall and parallel to the wall flow span, the thickness of the thin layer is less than the height of the corresponding wall turbulence buffer layer; using a laser to generate a laser strip light source to vertically illuminate the wall and illuminate the flow normal plane;(4) Collecting tracer particle images in the flow span and flow normal of the near-wall region, wherein the sampling time and the number of collected time series particle images are selected, and the time interval between two adjacent images is Δt;(5) Performing image processing on the collected tracer particle images, and using the PCSS particle image cross-frame tracking algorithm to obtain the motion flow velocity v of each tracer particle in the thin layer below the near-wall buffer and with the viscous bottom layer in the linear region as the core 动流 .

[0056] Among them, the flow field is generated by flow field simulation devices such as water flumes, water tunnels, wind tunnels, or real flow fields.

[0057] Specifically, in the PIV experiment, the selection of appropriate tracer particle size and concentration has a significant impact on the accuracy of the measurement. Therefore, before the experiment begins, it is necessary to determine the particle size and concentration according to the specific conditions and requirements of the experiment. The tracer particles are consistent with the density of the flowing medium in the flow field. Generally, the comprehensive performance of particles in a liquid environment is good at 10-20μm. The present invention has shown through verification experiments that when using a particle cross-frame matching algorithm, the optimal particle diameter is 4 to 6 pixels. The specific particle diameter selection also needs to consider factors such as the field of view of the experiment, the distance between the macro lens and the test plane, and the output light intensity of the laser to determine.

[0058] Specifically, in a conventional flume experiment, a high-speed camera collects 6,000 particle images at a time. Adjacent tracer particle images are captured by a high-speed camera at two consecutive times t and t+Δt, where Δt represents the time interval between frames. The exposure interval and the time interval between frames of these high-speed cameras should be selected based on the flow velocity and field of view of the flow field to ensure that the moving distance of the tracer particles between two frames is greater than the number of pixels in the image, but less than about 20 pixels.

[0059] like Figure 2 As shown, the image processing method includes the following steps:

[0060] (1) Grouping particle images of adjacent frames: The collected tracer particle images are grouped frame by frame, where the first group of images is the tracer particle images recorded at time t, denoted as set1; the second group of images is the adjacent tracer particle images recorded at time t+Δt, denoted as set2.

[0061] (2) Match each tracer particle in set1 and set2 images: Display each tracer particle in each image in set1 in grayscale intensity, and use the PCSS method to find the tracer particles in the image in set2 that corresponds to the image in set1 for matching. The specific method is as follows:

[0062] In the tracer particle adjacent image, let the tracer particle i in set1 be the center and R be the radius to make a query area I. Then select a candidate particle j in set2 and make a query area J with R as the radius. Use S ij represents the similarity coefficient between the two particles, and the calculation expression is shown in formula (5):

[0063]

[0064] In the formula, ε r and ε θ are the thresholds of distance deviation and angle deviation respectively; r im and θ im are the distance and connection angle between particle i and particles in query area I respectively; r jn and θ jn are the distance and connection angle between particle j in the candidate particles and particles in the query area J, respectively; M and N represent the number of particles in the query areas I and J, respectively.

[0065] The distance between the position of the candidate particle j in set2 and the position of the tracer particle i is less than or equal to R, and the size of R is related to the maximum particle displacement of the flow field.

[0066] In this experiment, the thresholds ε of the distance deviation and angle deviation selected in formula (5) are r =5,εθ =30.

[0067] H(x,y) in formula (5) is a step function, and its expression is shown in (6):

[0068]

[0069] That is, for all candidate particles j of particle i in set2, a similarity coefficient S can be calculated according to formula (6): ij , select the candidate particle j with the largest similarity coefficient as the paired particle of particle i. Then perform the above pairing on all particles in set1 to complete the particle pairing process of the entire image, and then obtain the displacement of the tracer particle between set1 and set2, and finally obtain the movement speed v of each tracer particle at adjacent moments 动流 .

[0070] Secondly, if Figure 3 As shown, the particle image is layered using the velocity probability density to obtain the average flow velocity in different velocity intervals, and then the average normal height corresponding to the average flow velocity is obtained using the time-space average flow normal statistical law curve. Specifically, the following steps are included:

[0071] (1) Probabilistic statistics and stratification of particles in rectangular regions: According to the characteristics of the wall turbulent flow in the plane, each frame of the particle image in the time series is divided into multiple rectangular regions, the number and size of the rectangular regions are set, and the flow velocity v of the tracer particles in each rectangle in the time series particle image is calculated. 动流 The PDF probability density analysis is performed, and according to the PDF results, the particle flow velocity is divided into three different speed ranges according to the flow velocity, namely v 1 、v 2 、v 3 .

[0072] (2) For three different speed intervals v 1 、v 2 、v 3 , perform time and space averaging: perform time averaging on the particle speeds in the rectangular area at the same position in the time series to obtain the time-averaged speed interval v t1 、v t2 、v t3 , and then v t1 、v t2 、v t3 Rectangular space averaging is performed to obtain three different speed intervals v after spatial and temporal averaging. ts1 、v ts2 、v ts3, calculate the average flow velocity of these three speed intervals, and obtain the average flow velocity value v of three different speed intervals 均1 、v 均2 、v 均3 .

[0073] (3) Statistics of flow normal: Using the flow normal particle image collected in step (4), the particle image is calculated using the query window-based cross-correlation algorithm commonly used in PIV technology. The spatiotemporal average flow velocity u under this condition is + With normal height y + The statistical law curve L is the classic scale velocity curve in the turbulent boundary layer, such as Figure 4 shown.

[0074] (4) The average velocity corresponds to the average height, and the average flow velocity v in different speed intervals is calculated 均1 、v 均2 、v 均3 The corresponding different heights y 均1 ,y 均2 ,y 均3 :v 均1 、v 均2 、v 均3 Corresponding flow velocity u + With normal height y + In the statistical law curve L, the flow velocity u + , normal height y + The calculation expressions of are shown in formula (1) and formula (2), respectively.

[0075]

[0076] In formula (1), v is substituted into v 均1 、v 均2 、v 均3 , and u τ is obtained in the second step (3);

[0077] In formula (2), υ is the kinematic viscosity coefficient of the fluid, which is a known quantity and is related to the temperature in the experiment.

[0078] Using formula (1) to obtain v 均1 、v 均2 、v 均3 in u + After the corresponding value on the axis, according to the flow velocity u + With normal height y + The statistical law curve L finds its + After obtaining the corresponding value on the axis, formula (2) is transformed to obtain formula (3):

[0079]

[0080] Then we can calculate y 均1 ,y 均2 ,y 均3 , that is, the single-layer particle image is divided into y 均1 Height particle image, y 均2 Height particle image, y 均3 Highly particle image.

[0081] Finally, the friction resistance τ is calculated using Newton's internal friction formula ω :like Figure 5 As shown, using the v obtained in the second step (4) 均1 →y 均1 ,v 均2 →y 均2 ,v 均3 →y 均3 , find the velocity gradient, and apply Newton's internal friction formula (4),

[0082]

[0083] Where μ is the dynamic viscosity coefficient of the fluid, It is the normal gradient of the flow velocity after spanwise stratification of the near-wall flow for which friction resistance measurement is required, indicating the change in velocity per unit normal height.

[0084] The particle image layered velocity measurement algorithm used in the present invention can obtain the temporal and spatial changes in the friction resistance of the near-wall flow. Obviously, the time interval Δt in the first step (4) and the selected distance deviation ε in formula (5) are r and angular deviation ε θ The threshold value, tracer particle diameter and other parameters are selected according to the experimental test requirements.

[0085] Example 2: Taking a typical two-dimensional turbulent boundary layer under zero pressure gradient as an example, the particle image layered velocimetry algorithm is used to measure the plane flow spanwise friction map and the DNS friction data calculation error to illustrate the accuracy and self-consistency of the particle image layered velocimetry algorithm.

[0086] This example simulates the free flow velocity U in a low-speed flume experiment. ∞ = 0.158m / s turbulent boundary layer. Figure 4 As shown, the average velocity distribution U measured by a single flow normal camera 2 is given. + (y + )’s experimental value.

[0087] like Figure 4 The following is a flow spanwise particle image obtained by measuring the flow spanwise camera in a low-speed water tank experiment. Figure 6 shown.

[0088] As shown in Figure 7(a), the instantaneous flow spanwise pulsating friction image is obtained by using the particle image layering algorithm. The white area is the low friction area, and the black area is the high friction area. Their scales in the flow spanwise direction are basically consistent with the strips. As shown in Figure 7(b), the pre-multiplied spectrum analysis of the flow spanwise friction can obtain that the point with the largest energy in the friction field corresponds to λ x + The scale is 10^3,λ z + The scale of the velocity is 10^2, which is comparable to the scale of high and low speed strips in classical boundary layer theory, indicating that the friction obtained by this method is accurate.

[0089] Figure 8(a) shows the relative error between the friction resistance calculated by the algorithm and the DNS friction resistance. 4 Five samples were randomly selected from the data, and the number of samples each time was 200. It can be seen that the relative error of each group of samples is basically within 20%, and most of them are concentrated within the relative error of 10%. As shown in Figure 8(b), the three curves are intercorrelated coefficients. The dotted line, solid line, and dotted line represent the friction signal autocorrelation obtained by the flow span stratification algorithm and the DNS friction signal autocorrelation. The cross-correlation of the two signals shows that the trends of the three cross-correlation signal graphs are basically consistent, which proves that the instantaneous wall friction resistance calculated by the flow span stratification algorithm proposed in the present invention is accurate and self-consistent.

[0090] The above embodiments illustrate that the present invention utilizes the flow-spanwise particle image layering algorithm to measure the instantaneous wall friction resistance of the flow-spanwise plane. Compared with other methods that can only obtain the average wall friction resistance in the flow-spanwise direction, the time and space resolution is improved while ensuring accuracy.

[0091] In view of the above method, the present invention proposes a time-space non-contact measurement device for the spanwise two-dimensional friction resistance of wall turbulent flow applicable to the measurement method, comprising a water tank 1, a high-speed camera 2, a laser 3, a synchronous controller 4, a computer 5 and tracer particles 6, such as Fig. 9 As shown. The water tank 1 is used to store and control water flow to form a flow field, and to spread a number of tracer particles 6 in the flow field; a laser 3 in the flow span direction and a laser 3 in the flow normal direction are arranged below the flow field formed by the water tank 1 to illuminate the flow field; a high-speed camera 2 in the flow span direction and a high-speed camera 2 in the flow normal direction are arranged in front of the flow field positions corresponding to the lasers 3 in the flow span direction and the flow normal direction, respectively, to facilitate the collection of tracer particle images; the lasers 3 in the flow span direction and the flow normal direction, and the high-speed cameras 2 in the flow span direction and the flow normal direction are all electrically connected to a synchronization controller 4, and the synchronization controller 4 is electrically connected to a computer 5.

[0092] In this embodiment, the flow field is generated by a water tank. Obviously, the flow field can also be generated in a flow field simulation device such as a water tunnel, a wind tunnel, or a real flow field.

[0093] A sheet light system 7 is installed at the front end of the laser 3 in the flow span direction and the flow normal direction, so that the laser beam emitted by the laser 3 is diffused into a uniform sheet light. The laser light sheet in the flow normal direction is perpendicular to the velocity direction of the flow field and parallel to the interface surface between the buffer layer and the viscous bottom layer of the boundary layer plate where the friction resistance measurement is required; the light sheet in the flow span direction is parallel to the velocity direction of the flow field and perpendicular to the boundary layer plate.

[0094] Among them, water tank 1 is used to control the water flow rate so that the flow rate reaches the conditions required by the experiment. Water tank 1 plays a key role in the particle image velocimetry (PIV) experiment and provides an ideal environment for shooting and analyzing fluid flow. High-speed camera 2 is PhantomVEO E-340L, with a pixel depth of 12Bit and a 100mm fixed-focus lens, which converts optical signals into 0-4096 grayscale information and stores it in the computer. Laser 3 uses a dual-cavity high-frequency laser, Beamtech Vlite-Hi-527-30. The front ends of the two lasers 3 are equipped with a sheet light system. The sheet light system 7 is used to diffuse the laser beam emitted by laser 3 into a uniform sheet light with a thickness of about 1mm; one beam of light is perpendicular to the velocity direction of the flow field and parallel to the boundary layer plate, where the friction resistance measurement is required, and the interface surface between the buffer layer and the viscous bottom layer; one beam of light is parallel to the velocity direction of the flow field and perpendicular to the boundary layer plate. The high-speed camera 2 collects the images of the tracer particles 6 in the flow span direction and the images of the tracer particles 6 in the flow normal direction for a certain period of time, and converts them into grayscale information and stores them in the computer 5. The synchronization controller 4 is responsible for connecting the camera 2, the laser 3 and the computer 5 to ensure that the three are triggered at the same time. The computer 5 is used to save the particle images taken by the high-speed camera 2.

[0095] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for non-contact measurement of two-dimensional friction resistance of wall turbulent flow in time and space, characterized by: The following steps are involved: Step 1: Use particle image velocimetry (PIV) technology to measure the flow velocity v of each tracer particle in the wall layer. 动流 , (1) Spreading tracer particles in the flow field; (2) Make the flow field environment stable and the tracer particles mixed evenly; (3) Generate flow span and normal laser illumination, use a laser to generate a laser strip light source to illuminate a thin layer close to the wall and parallel to the wall flow span, the thickness of the thin layer is less than the height of the corresponding wall turbulence buffer layer; use a laser to generate a laser strip light source to illuminate the wall perpendicularly to illuminate the flow normal plane; (4) collecting tracer particle images in the flow span direction near the wall and in the normal direction of the wall turbulent flow, where the sampling time, the collected time series particle images and the time interval Δt between two adjacent images are included; (5) The collected tracer particle images are processed, and the PCSS particle image cross-frame tracking algorithm is used to obtain the motion velocity v of each tracer particle in the thin layer below the near-wall buffer zone and with the linear region viscous bottom layer as the core. 动流 ; Step 2: Use velocity probability density to stratify the particle image, obtain the average stream velocity in different velocity intervals, and then use the spatiotemporal average flow normal statistical law curve to obtain the average normal height corresponding to the average stream velocity; (1) Probabilistic statistics and stratification of particles in rectangular regions: Each frame of the particle image in the time series is divided into multiple rectangular regions, the number and size of the rectangular regions are set, and the flow velocity v of the tracer particles in each rectangle in the time series particle image is calculated. 动流 Perform PDF probability density analysis and statistics, and according to the PDF results, divide the particle flow velocity into three different speed intervals, namely v1, v2, and v3; (2) Perform time and space averaging on the three different speed intervals v1, v2, and v3: Perform time averaging on the particle speeds in the rectangular area at the same position in the time series to obtain the time-averaged speed interval v t1 、v t2 、v t3 , and then v t1 、v t2 、v t3 Rectangular space averaging is performed to obtain three different speed intervals v after spatial and temporal averaging. ts1 、v ts2 、v ts3 , calculate the average flow velocity of these three speed intervals, and obtain the average flow velocity value v of three different speed intervals 均1 、v 均2 、v 均3 ; (3) Statistics of flow normal: Using the flow normal particle image collected in step (4), the particle image is calculated using the query window-based cross-correlation algorithm commonly used in PIV technology to obtain the spatiotemporal average flow velocity u under this condition. + With normal height y + The statistical law curve L is the classic scale velocity curve in the turbulent boundary layer; (4) The average velocity corresponds to the average height, and the average flow velocity v in different speed intervals is calculated 均1 、v 均2 、v 均3 The corresponding different heights y 均1 ,y 均2 ,y 均3 :v 均1 、v 均2 、v 均3 Corresponding flow velocity u + With normal height y + In the statistical law curve L, the flow velocity u + , normal height y + The calculation expressions of are shown in formula (1) and formula (2), respectively. In formula (1), v is substituted into v 均1 、v 均2 、v 均3 ; and u τ From the second step (3), In formula (2), υ is the kinematic viscosity coefficient of the fluid, which is a known quantity and is related to the temperature in the experiment; Using formula (1) to obtain v 均1 、v 均2 、v 均3 in u + After the corresponding value on the axis, according to the flow velocity u + With normal height y + The statistical law curve L finds its + After obtaining the corresponding value on the axis, formula (2) is transformed to obtain formula (3): Then we can calculate y 均1 ,y 均2 ,y 均3 , that is, the single-layer particle image is divided into y 均1 Height particle image, y 均2 Height particle image, y 均3 Highly particle images; Step 3: Calculate the friction resistance τ using Newton's internal friction formula ω :Use v obtained in the second step (4) 均1 →y 均1 , v 均2 →y 均2 , v 均3 →y 均3 , find the velocity gradient, and apply Newton's internal friction formula (4), Where μ is the dynamic viscosity coefficient of the fluid, It is the normal gradient of the flow velocity after spanwise stratification of the near-wall flow for which friction resistance measurement is required, indicating the change in velocity per unit normal height.

2. The method for non-contact time-space measurement of spanwise two-dimensional friction resistance of wall turbulent flow according to claim 1, characterized in that: The image processing method described in the first step (5) comprises the following steps: (1) Group particle images of adjacent frames: The collected tracer particle images are grouped frame by frame, where the first group of images are the tracer particle images recorded at time t, denoted as set1; the second group of images are the adjacent tracer particle images recorded at time t+Δt, denoted as set2; (2) Match each tracer particle in the set1 and set2 images: Display each tracer particle in each image in set1 in grayscale intensity, and use the PCSS method to find the tracer particles in the image in set2 that corresponds to the image in set1 for matching, as follows: In the tracer particle adjacent image, let the tracer particle i in set1 be the center and R be the radius to make a query area I. Then select a candidate particle j in set2 and make a query area J with R as the radius. Use S ij represents the similarity coefficient between the two particles, and the calculation expression is shown in formula (5): In the formula, ε r and ε θ are the thresholds of distance deviation and angle deviation respectively; r im and θ im are the distance and connection angle between particle i and particles in query area I respectively; r jn and θ jn are the distance and the connecting angle between particle j in the candidate particles and the particles in the query area J, respectively; M and N represent the number of particles in the query area I and J, respectively; H(x,y) in formula (5) is a step function, and its expression is shown in (6): That is, for all candidate particles j of particle i in set2, a similarity coefficient S can be calculated according to formula (6): ij , select the candidate particle j with the largest similarity coefficient as the paired particle of particle i. Then perform the above pairing on all particles in set1 to complete the particle pairing process of the entire image, and then obtain the displacement of the tracer particle between set1 and set2, and finally obtain the movement speed v of each tracer particle at adjacent moments 动流 .

3. The method for non-contact time-space measurement of spanwise two-dimensional friction resistance of wall turbulent flow according to claim 2, characterized in that: The distance between the position of the candidate particle j in set2 and the position of the tracer particle i is less than or equal to R, and the size of R is related to the maximum particle displacement of the flow field.

4. The method for non-contact time-space measurement of spanwise two-dimensional friction resistance of wall turbulent flow according to claim 1, characterized in that: The density of the tracer particles is consistent with that of the flowing medium in the flow field, and the diameter of the tracer particles is statistically 3-6 pixels.

5. The method for non-contact time-space measurement of spanwise two-dimensional friction resistance of wall turbulent flow according to claim 1, characterized in that: The laser band emitted by the laser has a thickness, covers the flow field area below the near-wall buffer zone and with the linear zone viscous bottom layer as the core, and the irradiated flow field is uniform laser sheet light, and the irradiated laser is close to the wall surface.

6. The method for non-contact time-space measurement of spanwise two-dimensional friction resistance of wall turbulent flow according to claim 1, characterized in that: In the first step (4), the pixel size of the tracer particles moving within the cross-frame time is less than or equal to R / 2.

7. The method for non-contact time-space measurement of spanwise two-dimensional friction resistance of wall turbulent flow according to claim 1, characterized in that: The time interval Δt in step (4) and the selected distance deviation ε in formula (5) are r and angular deviation ε θ The threshold value, tracer particle diameter and other parameters are selected according to the experimental test requirements.

8. A device for non-contact measurement of the spanwise two-dimensional frictional resistance of a wall turbulent flow using the method for non-contact measurement of the spanwise two-dimensional frictional resistance of a wall turbulent flow in space and time according to any of the above claims, characterized in that: It includes a flow field, a laser, an image recording instrument, a synchronous controller and a computer. The flow field provides an experimental environment and a number of tracer particles are spread in the flow field. A laser in the flow span direction and a laser in the flow normal direction are arranged below the flow field to provide light sources in the flow span direction and the flow normal direction for the flow field; One image recording instrument in the flow span direction and one image recording instrument in the flow normal direction are arranged in front of the flow field positions corresponding to the lasers in the flow span direction and the flow normal direction, so as to facilitate the collection of tracer particle images; The lasers in the flow span direction and the flow normal direction, and the image recording instruments in the flow span direction and the flow normal direction are all electrically connected to a synchronous controller, and the synchronous controller is electrically connected to a computer.

9. The device for non-contact time-space measurement of spanwise two-dimensional friction resistance of wall turbulent flow according to claim 8, characterized in that: A sheet light system is installed at the front end of the laser in the flow span direction and the flow normal direction, so that the laser beam emitted by the laser is diffused into a uniform sheet light. The laser light sheet in the flow normal direction is perpendicular to the velocity direction of the flow field and parallel to the interface surface between the buffer layer and the viscous bottom layer of the boundary layer plate where friction resistance measurement is required; the light sheet emitted by the laser in the flow span direction is parallel to the velocity direction of the flow field and perpendicular to the boundary layer plate.

10. The device for non-contact time-space measurement of spanwise two-dimensional friction resistance of wall turbulent flow according to claim 8, characterized in that: The image recording instrument is a high-speed camera, and the optical axis of the high-speed camera is perpendicular to the plane of the uniform sheet light provided by the laser.

Images