KR19990068430A - 3-dimensional particle imaging velocimeter; 3D-PIV, so-called Thinker's EYE - Google Patents

Abstract

The present invention relates to a three-dimensional velocity measurement device for a fluid particle, it is possible to perform the three-dimensional motion analysis of the fluid flow field and dynamic structure by analyzing the behavior of the fluid particles by tracking the three-dimensional velocity of the fast-moving flow particles The price is low and the dynamic range of the measurement is good. To this end, the three-dimensional velocity measurement device, the tracking particle is introduced into the flow field moving along the flow of the flow field; At least two cameras installed at different points to photograph the tracking particles at different angles and positions; A light emitting device for emitting incident light toward the flow field; It is installed between the flow field and the light emitting device to block and adjust the incident light from the light emitting device to code the start point and the end point of the tracking particle flowing at high speed in the image of the one-frame image of the camera. Acoustic optical equipment to enable; An image recording device for receiving a tracking image of the tracking particles from each camera and recording the image; Receiving the recorded image from the image recording apparatus and visualizing the flow of the flow field by analyzing each position of the tracer particles in the image and controlling the digital image processing (Thinker's EYE); And a display unit which receives and displays the visualization information of the tracking particles from the control unit.

Description

3-dimensional particle imaging velocimeter using digital imaging; 3D-PIV, so-called Thinker's EYE}

The present invention relates to a three-dimensional high-speed flow measurement device, and more particularly, it is possible to perform the three-dimensional motion analysis of the fluid flow field and the dynamic structure by tracking the three-dimensional velocity of a plurality of fast moving particles to analyze the behavior of the flow particles The present invention relates to a three-dimensional velocity measurement device for a flow field, which can be manufactured at low cost and has excellent dynamic characteristics of measurement.

In general, accurately estimating the kinetic response of a floating structure floating in waves is an important issue in terms of the structure's operability and safety. Therefore, theoretical and experimental studies have been continued by many researchers for estimating the response of floating structures. These days, floating structures of various shapes have emerged for the purpose of marine development, and various theoretical analysis methods are used for the estimation of their motion responses, but in order to verify the validity of the estimates by these analysis methods, water tank tests are generally performed. Is being reviewed. Therefore, the accuracy of the motion response measurement technique of the floating structure by the water tank test has a significant influence on the performance evaluation of the structure.

In addition, in order to understand the physical mechanisms of turbulent motion fields such as wall turbulence, fractionation, and separation flow, which are engineeringally important flow phenomena, it is necessary to simultaneously quantitatively measure three components of instantaneous velocity in a wide range of flow fields. In particular, information about the spatial structure of the flow field is needed to interpret the mechanism from the generation of large-scale tissue structures to their extinction.

The conventional turbulent flow measurement method, the hot wire flowmeter or the laser Doppler flowmeter, is, in principle, a method of measuring the velocity at a point of a flow field, and in principle, does not directly provide information on the spatial structure.

On the other hand, Particle Imaging Velocimetry or Particle Tracking Velocimetry (hereinafter referred to as PIV / PTV) inserts fine tracking particles into the flow field and visualizes the paths of the particles with cameras and illumination sources and then corrects the correct particles for the visible images. Determining the path by means of measuring the velocity allows the velocity to be measured simultaneously throughout the flow field in a non-contact manner with little disturbance to the flow field. In addition, PIV / PTV is a great help in understanding the organizational structure of the flow field because the continuous measurement in time allows the change in the flow field. However, until now, most of the velocity measurement methods of the flow field by PIV / PTV measurement are only two-dimensional, and when the conventional two-dimensional measurement method is applied to the velocity measurement of the three-dimensional flow where the flow field is relatively strong To lose information. The measurement method developed to overcome this is the three-dimensional measurement method based on stereo photography.

However, most PIV / PTV measurement systems measure speed using a CCD (charge coupled device) camera of the international standard NTSC (national television system committee). Since the CCD camera has a limited shooting speed of 1/60 second, there is a problem in that it is difficult to apply the CCD camera to the fast measurement of the local velocity of the measurement area of the flow field. Of course, using a high-speed camera solves this problem, but the high-speed camera is expensive and expensive, and at the same time there is a problem that the dynamic characteristics of the measurement is not good.

Accordingly, an object of the present invention is to solve the above problems, to perform the flow analysis of the high speed flow field and the three-dimensional motion analysis of the dynamic structure by tracking the three-dimensional velocity of the flow particles in the high-speed fluid flow field It is possible to provide a three-dimensional velocity measurement device for a flow field that is economical and has improved dynamic characteristics.

1a-a first schematic diagram of a three-dimensional high speed flow measurement apparatus according to the present invention;

1b-a second schematic diagram of a three-dimensional high speed flow measurement apparatus according to the present invention;

Figure 2-Flowchart schematically showing a three-dimensional high speed flow measurement procedure.

3 is a coordinate diagram for the principle of general photogrammetry.

4A-a perspective view of a test apparatus for observing tracer particles in a flow field.

4b-4a.

5-Coordinate diagrams for the three-dimensional positioning principle of particles.

Fig. 6 is a state diagram showing a relationship between a camera signal and an acoustic optical device.

7-a state diagram showing the flow state of the flow field rotating at 300 rpm.

8-a state diagram showing the flow state of the flow field rotating at 400 rpm.

9-First screen of a three-dimensional particle image flowmeter ("Thinker's EYE ").

10 is an exemplary view of measurement results of a velocity vector distribution for a flow field around an automobile measured by "Thinker's EYE".

Fig. 11-Exemplary three-dimensional velocity measurement results of a rotating disk measured by "Thinker's EYE".

Fig. 12-Exemplary three-dimensional velocity measurement results for the flow field in the wash bath measured by "Thinker's EYE".

13-Setting screen of "Thinker's EYE" for obtaining the measurement result of FIGS. 11 and 12.

14-Setting screen of "Thinker's EYE" for obtaining the measurement result of FIGS. 11 and 12.

15-9 is a three-dimensional representation of Turbulent Intensity for the flow field of FIG. 9 as "Thinker's EYE."

16-9 is an exemplary view showing the pressure distribution for the flow field of FIG. 9 as "Thinker's EYE".

In order to achieve the above object, the three-dimensional velocity measurement apparatus according to the present invention includes a tracking particle which is introduced into a flow field and flows along the flow of the flow field; At least two cameras installed at different points to track and photograph the tracking particles at different angles; A light emitting device for emitting incident light toward the flow field; It is installed between the flow field and the light emitting device to block and adjust the incident light from the light emitting device start and end point of the tracking particles flowing at high speed in the measurement area with only one (One-Frame) image of the camera Acoustic optical device that can code; An image recording device for receiving a tracking image of the tracking particles from each camera and recording the image; Receiving a recorded image from the image recording apparatus and analyzing each position of the tracking particles in the image to visualize and develop the flow phenomenon of the flow field; And a display unit which receives and displays the visualization information of the tracking particles from the control unit.

Accordingly, by analyzing the three-dimensional behavior of the moving particles by tracking the three-dimensional velocity of the fast-moving flow particles, it is possible to perform the three-dimensional motion analysis of the fluid flow field and structure, and the manufacturing cost is low and the dynamic characteristics of the measurement (dynamic It is possible to provide a three-dimensional velocity measurement device for a good range of flow particles.

Hereinafter, a 3D particle image flowmeter device according to the present invention will be described in detail with reference to the accompanying drawings.

1A is a first schematic diagram of a three-dimensional high speed flow measurement apparatus according to the present invention, and FIG. 1B is a second schematic diagram of a three-dimensional high speed flow measurement apparatus according to the present invention. As shown in these drawings, the three-dimensional high-speed flow measurement apparatus includes three CCD cameras capable of capturing the inside of the flow field, a light emitting device for emitting incident light toward the flow field side, and installed between the flow field and the light emitting device. Acousto-optic device which blocks and adjusts the incident light from the emitting device so that the start and end points of the fast-flowing tracer particles can be coded with only one-frame image of the camera, and the tracer particles from each camera. An image recording apparatus for receiving a tracking image of the image and recording the image; By receiving the recorded image from the image recording apparatus and analyzing each position of the tracer particles in the image, the flow phenomenon of the flow field is visualized and composed of a digital image processing control unit (software unit: also known as Thinker's EYE).

The fluid flows at a high speed in the flow field, and the tracer particles flow along the flow of the flow fluid. Three CCD cameras are installed at different points to capture specific tracking particles flowing in the flow field at different angles.

The light emitting device employs a laser emitting device emitting device to emit a laser beam into the flow field. Acousto-optical modulator (AOM) is installed between the light emitting device and the flow field to block and control the incident light from the light emitting device to track particles with one image of three CCD cameras. It is possible to code the start point and the end point of. The AOM is controlled by the AOM controller to appropriately block and adjust the incident light from the light emitting device so that the timing of the start and end points of the tracking particles can be freely adjusted, i.e., to improve the dynamic range. It is.

The image recording device and the control unit (the software unit: Thinker's EYE) are embedded in a general personal computer (PC) and include two-dimensional and three-dimensional PIV / PTV measurement algorithms. A two-dimensional PIV / PTV measurement algorithm is inserted to compare the three-dimensional PIV / PTV measurement results as necessary to measure the two-dimensional velocity distribution over the measurement area of the cross section of the fluid flow field.

Each unit of the 3D high speed flow measurement device is connected to a PC and can operate. The PC is connected to a display unit that receives and displays visualization information of the tracer particles from a controller (software unit: Thinker's EYE). In this embodiment, a monitor, which is a display device which is generally used, is used.

In summary, the flow state flowing inside the flow field is represented by tracking particles, and the flow state of these tracking particles is captured by a CCD camera and sent to an image recording device for human observation by a control unit (software unit). Are displayed graphically. Accordingly, the three-dimensional flow state of the flow field can be visualized. The image of the flow state of the flow field is stored in a general file that can be variously applied as needed.

Hereinafter will be described the operation and effects of the three-dimensional high-speed flow measurement apparatus according to the present invention.

Since the images of the tracking particles taken by the three CCD cameras appear in a separate form, a three-dimensional stereoscopic image can be obtained by synthesizing three images taken by the three CCD cameras. In order for these image synthesis to be made correctly, first, as shown in Fig. 2, the calibration operation of these three CCD cameras must be preceded. The calibration of cameras is the process of finding out where each camera is located at what angle and with respect to the world coordinate system.

The principle of camera calibration is the same as photogrammetry.

The principle of the photogrammetry method will be described with reference to FIG. 3 through the relationship between the absolute coordinate system (X, Y, Z), the camera coordinate system ( X , Y , Z ), and the photo coordinate system (x, y). The tracking particle P corresponds to the point p in the camera coordinate system (photo coordinate system). Equation 1 is established from the collinear condition that the projection center, the photographic image, and the ground tracking particles are in a straight line.

Here, the rotation matrix a _ij is represented by Equation 2, and correction amounts Δx and Δy due to distortion of the camera lens are represented by Equation 3. (X, Y, Z) is the absolute coordinate of the object P, (X ₀ , Y ₀ , Z ₀ ) is the absolute coordinate of the projection center, c is the screen distance, (x, y) is the photo coordinate on the corresponding picture, (x ₀ , y ₀ ) represents the amount of movement of the camera tavern position.

Applying the photogrammetry described above, the operation of obtaining the three-dimensional position and posture of the camera by the above-described photogrammetry in order to calculate the three-dimensional position P (X, Y, Z) of the tracking particles is called a camera calibration operation. This camera calibration operation will be described with reference to FIGS. 4A and 4B.

Camera calibration is performed by assuming that the reference point is three-dimensionally arranged in the measurement area in the absolute coordinate system (ground coordinate system) as the tracking particle. The camera's external parameters (X ₀ , Y ₀ , Z ₀ ,) from a sequential approximation method combined with the least-squares equations for Equations 1, 2, and 3 using a reference point that knows the three-dimensional absolute coordinates and the photo coordinates. ω, φ, κ) and internal parameters (c, x ₀ , y ₀ , k ₁ , k ₂ ) are obtained.

Here, as shown in FIG. 4 for calibrating three cameras, three-dimensional coordinate values are already known and use a plurality of reference points, for example, 36 reference points. These reference points are marked on the apex of vertical rods of different heights fixed on the disk, and the disk rotates at a constant speed by the driving force of a driving motor (not shown) for the verification of an already constructed three-dimensional measurement system. Camera calibration is done with this disc stationary.

When the calibration process of the camera is completed through the above process, the position and inclination of each camera with respect to the absolute coordinate system are known. Therefore, as in the principle of triangulation, the intersection of straight lines drawn toward one point from the position of each camera is determined. A three-dimensional position is obtained with respect to the point. Then, the three-dimensional position of two different views is obtained and the time difference between the two views is divided to obtain the three-dimensional velocity of the point.

This process will be explained in detail again.

After the calibration of three CCD cameras is completed, 64 frame images are inputted through each camera into the PC's built-in image recording device for continuous continuous tracking of tracer particles in three-dimensional space. After interpolating and separating three sets of temporal frame images input at the same time into field images, the centers of the particles appearing in the images are obtained and used as the photo coordinates of each camera. The reason for the interpolation separation is to obtain images of two time points with time difference. In other words, when the frame image is interpolated, two feed images at two points of view are generated.

5 is a coordinate diagram for the three-dimensional positioning principle of the particles. As shown in this figure, the target P whose ground coordinate system is (X, Y, Z) has a projection center of O ₁ (X ₁ , Y ₁ , Z ₀ ), and the three-axis tilt with respect to the absolute coordinate system of the camera coordinate system axis. The tilt angles of the three axes of the camera 1 with the angles (ω ₁ , φ ₁ , κ ₁ ), the projection center of O ₂ (X ₂ , Y ₂ , Z ₂ ), and the absolute coordinate system of the camera coordinate system are (ω ₂ , φ ₂ , κ ₂ ) and a camera with a projection center of O ₃ (X ₃ , Y ₃ , Z ₃ ) and a 3-axis tilt angle (ω ₃ , φ ₃ , κ ₃ ) with respect to the absolute coordinate system of the camera coordinate system axis. When shot at 3, the input image is projected on a plane away from the projection center by the screen distance. As described above, when the parameters of the camera are obtained, since the common line conditional expression is determined between the projection center, the photo coordinates, and the subject, the three-dimensional coordinates of the subject are determined as follows using this.

When the equation (1) is modified, the conversion from the corresponding set of photo coordinate systems (x, y) to the absolute coordinate system (X, Y, Z) is as shown in equation (4).

Photo coordinates p ₁ (x ₁ , y ₁ ), p ₂ (x ₂ , y ₂ ), and p ₃ (x ₃ , y ₃ ) corresponding to the particle images of camera 1, camera 2 and camera 3 are expressed in equation (4). The observation equation for the three-dimensional coordinates formed by these points and the cameras is represented by Equation 5, which represents a three-dimensional intersection where three straight lines formed by the three cameras intersect.

At this time, (X _pi , Y _pi , Z _pi ) (i = 1,2) is the image after the coordinate transformation of the first camera and the second camera, and the unknown parameters (t ₁ , t _2, t ₃ ) are It is obtained by the least-squares method to minimize the difference between the right side.

6A shows an outline for obtaining the three-dimensional velocity of the tracer particles. The three-dimensional velocity can be obtained by obtaining the temporal movement amount of the three-dimensional absolute coordinates. At this time, the same particle tracking applied the probability matching method to the three-dimensional position of the particles at different times. It separates one digital frame image into two field images and obtains all three-dimensional coordinates in the same manner as described above for the particles appearing in these two fed images. The three-dimensional velocity is obtained by dividing the two visual differences by obtaining the three-dimensional movement of the particles.

However, due to the large number of tracers present in the feed image at these two views, it is not easy to determine which one at the first time corresponds to which at the second time. Therefore, it is necessary to find pairs of identical points of tracking points for the feed image at two viewpoints. This is called pair matching. That is, the pair matched considers all possible displacement vectors of the particles between these three-dimensional position points at these two views, then defines a match-probability Pij for each displacement, and It is determined by iterating with the no-match probability Pi and determining the highest match probability density. Once the corresponding point is determined, the displacement vector is obtained by simply connecting these two points.

In the figure, i denotes the number of particle centers of the first picture, and j denotes the number of matchable particle centers of the second picture. Pij is the probability that particle j of the second image will match a particular particle (i) of the first image, and Pi is the match of the target particle (i) in the first image to any match in the second image. It is the probability that there is no point). Therefore, the sum of Pij and Pi for the target particle (i) is always one. It is assumed that during the time interval [Delta] t, the particles of the first image do not go beyond the maximum movement while moving to the second time. The maximum moving point is then generally matched to the maximum displacement in the flow. The concept of integral length scale also makes it possible to assume that all particles move close to solid motion within a region Tn (neighborhood threshold). That is, it can be assumed that the particles within the radius Tn centering on the particle i do not have a large quasi-rigidity threshold (Tq) indicating a difference in displacement. Using these assumptions, we can calculate the three-dimensional velocity vector by continuously calculating the coincident probability Pij and Pi , and then finding the displacement vector with the largest probability and dividing it by the time interval (Δt) between the two images.

On the other hand, in the case of the measurement of high-speed flow with a fast particle moving speed in the measurement region, as shown in FIG. 6B, the shooting time interval of the CCD camera of the National Television System Committee (NTSC) is 1/60 second. Therefore, speed measurement is impossible at this time resolution. Therefore, the present invention employs an acoustic optical device, AOM (Acousto-Optic Modulator), is installed between the flow field and the light emitting device to block and control the incident light from the light emitting device to track in the flow field in which the CCD camera flows at high speed. You can shoot particles.

In other words, intermittent intermittent operation is possible at a time when a laser light source, which is a light emitting device, is desired for a fixed shooting time of 1/60 second of a CCD camera of a standard standard, so that the moving speed of particles passing through the measurement area is relatively high. In this case, the starting and ending points of the velocity vector can be condensed. Therefore, when coding as shown in FIG. 6B, it is easy to determine the start point and the end point simply by separating two odd and even fields. As a result, if only the three-dimensional absolute coordinates described above for the starting point and the end point are determined, the velocity measurement of the high speed three-dimensional flow field becomes possible. This allows three-dimensional velocity measurement using only one image (one-frame or two-field) image (hereinafter referred to as 3D-PIV: 3-Dimensional Particle Imaging Velocimeter).

Data of the three-dimensional velocity of the flow field measured through the above process is sent to the PC, the processing result is output to the screen through a monitor, a display device.

FIG. 7 shows the results of the measurement of the particles on the disc by using the three-dimensional particle image flowmeter (3-D PIV) constructed in the present invention when FIG. 4 rotates at 300 rpm, and FIG. 8 shows the result when FIG. 4 rotates at 400 rpm. Is a state diagram showing the measurement result. As shown in these figures, the image output through the monitor has a shape as shown in Figs. Thus, the observer can intuitively grasp the three-dimensional high-speed flow in the flow field through these images.

Therefore, as described above, it is possible to measure the three-dimensional high-speed flow, it is possible not only to perform accurate fluid flow analysis by applying to applications related to the three-dimensional measurement of the fluid, but also to analyze the three-dimensional motion of the dynamic structure Will be. Further, in the present invention, since the three-dimensional measurement of the high-speed flow particles is performed by using a relatively inexpensive CCD camera and AOM device, the economical efficiency of the device and the dynamic range of the measurement are improved.

The figure below shows a screen of a control unit (software unit) which automatically processes all the above processes on a PC.

9 shows the first screen of "Thinker's EYE", which is the name of the software.

Fig. 10 shows an example of the velocity vector distribution as a result of measuring the flow field around the vehicle with "Thinker's EYE".

Fig. 11 shows the results of the velocity vector measurement of the rotating disc measured by "Thinker's EYE".

Fig. 12 shows the distribution of the three-dimensional velocity vector for the flow field in the wash bath measured by "Thinker's EYE".

13 and 14 show setting screens for obtaining the measurement results of FIGS. 11 and 12 with "Thinker's EYE".

FIG. 15 shows the results of three-dimensional processing of Turbulent Intensity obtained with respect to the flow field of FIG. 9 by "Thinker's EYE" to make it easier for a measurer to understand.

FIG. 16 shows the results of three-dimensional processing of the pressure distribution obtained with respect to the flow field of FIG. 9 by "Thinker's EYE" to make it easier for a measurer to understand.

Above, "Thinker's EYE" can not only measure spatial velocity for quantitatively necessary quantitative data such as velocity distribution, pressure distribution, streamline, vorticity, turbulent intensity, etc. Also it can be displayed in three dimensions so that the user can easily understand.

Therefore, according to the three-dimensional high-speed flow measurement apparatus of the flow particles according to the present invention, it is possible to perform the analysis of the flow field by measuring the three-dimensional velocity of the plurality of fast flow particles in the fluid flow field, as well as the three-dimensional motion of the dynamic structure By measuring the speed, three-dimensional motion analysis can be performed, reducing the construction cost of the measurement system and improving the dynamic characteristics of the measurement.

Claims (1)

A tracer particle introduced into the flow field and flowing along the flow of the flow field; At least two cameras installed at different points to track and photograph the tracking particles at different angles; A light emitting device for emitting incident light toward the flow field; An acoustic optical device installed between the flow field and the light emitting device to block and adjust incident light from the light emitting device so that the camera can encode a start point and an end point of the tracking particle flowing at high speed; An image recording device for receiving a tracking image of the tracking particles from each camera and recording the image; Receiving the recorded image from the image recording apparatus and visualizing the flow of the flow field by analyzing each position of the tracer particles in the image and controlling the digital image processing (Thinker's EYE); And And a display unit for receiving and displaying the visualization information of the tracking particles from the control unit.