Particle Image Velocimetry

Transcription

1 Particle Image Velocimetry Fundamentals and Application TMR 7: Experimental Methods Fall 2015 Chittiappa Muthanna

2 Overview Basic Principle Components Correlation analysis Practical issues Laser Safety The lab exercise Some examples TMR 7 lab test 3: Particle Image Velocimetry 2

3 Basic principle Particle Image Velocimetry (PIV) is a non-intrusive, indirect, whole-field method to measure fluid motion. It is a digital processing technique in 2 steps: Acquisition + Analysis The basic principle is to take two snapshots of the flow field at a known time interval, and evaluating the velocity TMR 7 lab test 3: Particle Image Velocimetry 3

4 Basic Principle of the PIV Technique What is Particle Image Velocimetry? It is a non-intrusive, whole field optical measurement technique used to get velocity information of the flow field. The basic principle is to take two snapshots of the flowfield at a known time interval, and evaluating the velocity frame 1 frame 2 Δt V V s t

5 Why use PIV? Consider a pitot-probe or hot-wire probe. To get the pictures on the right, we would have to take data one point at a time and then average them to get the velocity plots. We thus lose information as to how the flow develops around that point, and for time evolving flows, we have no idea how the flow behaves Note : to be able to get velocity vectors, special hot-wire probes, or 2 component LDA systems have to be used thus increasing complexity

6 Why use PIV? With PIV, we can actually take snapshots of the entire flow field at each instance, and thus are able to measure the evolving flowfield. Averaging these pictures, we thus can get the same average flow field we get with a single point measurement system. Note : Due to the low sampling frequency, PIV is very poor for spectral results.

7 Basic principle & components The main components are: Powerful light source (laser) Shaping optics lightsheet Seeding material Camera(s) recording particle images Synchronization unit Software to control acquisition and post-processing TMR 7 lab test 3: Particle Image Velocimetry 7

8 Components: light source Requirements: bright!, monochromatic (single wavelength), ability to be bundled, redirected and formed into a sheet. LASER, either 1. continous wave, or 2. pulsed. To obtain short flashes of light ( freeze motion), CW-lasers need rotating mirrors/prisms. Nowadays usually NdYAG-lasers, 532nm wavelength (green). Pulses with 4-10 ns length, 20mJ@1kHz - 800mJ@10Hz per pulse. 2 cavities (twin laser) to generate 2 independent pulses. TMR 7 lab test 3: Particle Image Velocimetry 8

9 Components: light path & light sheet optics Requirements: turn the point-shaped laser into an evenly illuminated flat sheet inside the field of interest. TMR 7 lab test 3: Particle Image Velocimetry 9

10 Components: light path & light sheet optics Requirements: turn the point-shaped laser into an evenly illuminated flat sheet inside the field of interest. TMR 7 lab test 3: Particle Image Velocimetry 10

11 Components: light path & light sheet optics Laser periscope for submerged PIV

12 Components: light path & light sheet optics TMR 7 lab test 3: Particle Image Velocimetry 12

13 Components: light path & light sheet optics TMR 7 lab test 3: Particle Image Velocimetry 13

14 Components: Seeding The P in PIV Remember: We measure particle motion, not fluid motion! Requirements: - small: follow the flow neutrally - same density as fluid, neither buoyant nor sinking - big and reflective: give good scattered light image - cheaply available or easy to generate - homogenous in size and behaviour - non-toxic, chemically inactive TMR 7 lab test 3: Particle Image Velocimetry 14

15 Components: Seeding The P in PIV Common particles in air: - Smoke ( 1 μm) - DEHS olive oil (0,5-1,5 μm) - He filled soap bubbles (1-3 mm) Common particles in water: - polystyrene/polyamide (10-90 μm) - silver coated hol. glass spheres (10-90 μm) - air bubbles (5-500 μm two-phase flow!) TMR 7 lab test 3: Particle Image Velocimetry 15

16 Components: Seeding particles Common particles in water: - polystyrene/polyamide (10, 50,90 μm) - silver coated hollow glass spheres (15μm)

17 Components: Cameras Requirements: fast image rate, good resolution (typically 1k by 1k, up to 4k by 4k = 16Mpixels Two sensor techniques commonly available, CCD: fast in recording double-frames (2 pics, 1 ms apart) slower in overall acquisition rate: 1-20 Hz best s/n-ratio, highest sensitivity established technology CMOS: highest effective frame rates (500, 1000, even 5000 Hz) image subsections can be picked for even higher speeds fast developing technique TMR 7 lab test 3: Particle Image Velocimetry 17

18 Components: Cameras CCD cameras CMOS cameras

19 Components: Synchronization frame 1 Δt frame 2 #1 #2 #1 #2 #1 #2 #1 #2 Time Interframe Time, Pulse Distance, Δt : e.g ms Acquisition Rate, 1/f : e.g. 1/10Hz = 100ms TMR 7 lab test 3: Particle Image Velocimetry 19

20 Components: Synchronization Ext. Triggering (e.g. propeller revolution) or fixed acquisition rate Camera Timing unit lamp 1 lamp 2 Q 1 Q 2 Laser Cavity 1 Cavity 2 TMR 7 lab test 3: Particle Image Velocimetry 20

21 Components: Software 1.: Image acquisition adjust measurement parameters control hardware devices create image database with acqu. information 2.: Data analysis image processing / enhancement (if necessary) masking of unwanted areas analysis: image pair > vector map visualization, statistics, export & 2. might come in one packet --- TMR 7 lab test 3: Particle Image Velocimetry 21

22 Components: Software FlowManager 4.70 by Dantec Dynamics TMR 7 lab test 3: Particle Image Velocimetry 22

23 Correlation analysis 1600 by 1200px full image U frame #1 TMR 7 lab test 3: Particle Image Velocimetry 23

24 Correlation analysis 1600 by 1200px full image U frame #2 TMR 7 lab test 3: Particle Image Velocimetry 24

25 Correlation analysis Zoomed in: 64 by 64px section Green grid: 24 by 24px interrogation areas frame #1 frame #2 TMR 7 lab test 3: Particle Image Velocimetry 25

26 Correlation analysis frame #1 frame #2 TMR 7 lab test 3: Particle Image Velocimetry 26

27 Correlation analysis TMR 7 lab test 3: Particle Image Velocimetry 27

28 Summary PIV recipe : FLOW seeding illumination imaging sampling registration quantization selection acquisition enhancement correlation validation pixelization estimation analysis interrogation RESULT (Poelma & Westerweel) TMR 7 lab test 3: Particle Image Velocimetry 28

29 Calibration: Relating Camera Images to Distances The images seen on the camera have no distance information in them. All that is registered is by how many pixels the particles have moved on the CCD chip. Thus, we have to be able to tell the software, what a movement on the CCD chip corresponds to in the measurement plane. This is done by placing a calibration grid in the measurement plane, thus giving us a relationship between pixels and physical space. pix Image size: (0,0), 8-bits (frame 2) Burst#; rec#: 1; 1 (1), Date: , Time: 17:11:47:812 Analog inputs: ; ; ; pix 1600 TMR 7 lab test 3: Particle Image Velocimetry 29

30 Stereoscopic PIV (SPIV) 3 component (U, V, and W) velocity measurement systems consist of 2 cameras. This is know as Stereoscopic PIV (SPIV). Sometimes referred to as 3D-PIV, but technically it is not 3D. SPIV combines the views from the two cameras to resolve the out-of-plane component. The two basic camera orientations are shown below. TMR 7 lab test 3: Particle Image Velocimetry 30

31 Stereoscopic PIV (SPIV) True 3D displacement ( X, Y, Z) is estimated from a pair of 2D displacements ( x, y) as seen from left and right camera respectively TMR 7 lab test 3: Particle Image Velocimetry 31

32 Stereoscopic PIV (SPIV): Scheimpflug condition Focussing problem and Scheimpflug condition TMR 7 lab test 3: Particle Image Velocimetry 32

33 Stereoscopic PIV (SPIV): Scheimpflug condition Focusing an off-axis camera requires tilting of the CCD-chip (Scheimpflug condition) 3D evaluation requires a numerical model, describing how objects in space are mapped onto the CCD-chip of each camera Object plane (Lightsheet plane) Object coordinates (X,Y,Z) Parameters for the numerical model are determined through camera calibration Left image coordinates (x,y) Lens plane left & right Right image coordinates (x,y) Image plane left & right TMR 7 lab test 3: Particle Image Velocimetry 33

34 Stereoscopic PIV (SPIV): Scheimpflug condition TMR 7 lab test 3: Particle Image Velocimetry 34

35 Stereo Camera calibration Images of a calibration target are recorded. The target contains calibration markers in known positions. Comparing known marker positions with corresponding marker positions on each camera image, model parameters are adjusted to give the best possible fit. TMR 7 lab test 3: Particle Image Velocimetry 35

36 Overlapping fields of view 3D evaluation is possible only within the area covered by both cameras. Due to perspective distortion each camera covers a trapezoidal region of the light sheet. Careful alignment is required to maximize the overlap area. Interrogation grid is chosen to match the spatial resolution Right camera's field of view Overlap area Left camera's field of view TMR 7 lab test 3: Particle Image Velocimetry 36

37 Laser Safety Laser Hazard Classification: Class 1 safe if not disassembled (CD-drives, players) Class 2/2a eye hazard if you stare into the beam (supermarket scanners) Class 3a eye hazard if collected or focused into eye (laser pointers) Class 3b eye hazard if direct or reflected beam is viewed (research) Class 4 eye hazard if direct, reflected, or diffusely-reflected beam is viewed; skin & fire hazard in direct beam (research, industry) For PIV, we use only class 4 lasers! TMR 7 lab test 3: Particle Image Velocimetry 37

38 Laser Safety Laser injuries: 10Hz Nd:YAG TMR 7 lab test 3: Particle Image Velocimetry 38

39 Laser Safety Laser injuries TMR 7 lab test 3: Particle Image Velocimetry 39

40 Laser Safety TMR 7 lab test 3: Particle Image Velocimetry 40

41 Laser Safety: - styling eyewear TMR 7 lab test 3: Particle Image Velocimetry 41

42 Laser safety: Reflection hazards TMR 7 lab test 3: Particle Image Velocimetry 42

43 Lab test No 4: PIV of a cylinder wake Setup: Cylinder diameter: d = 50 mm Flow speed: m/s Acquisition rate: 1 15 Hz TMR 7 lab test 3: Particle Image Velocimetry 43

44 Lab test No 4: PIV of a cylinder wake Setup: Fixed uniform cylinder in a steady flow Objectives: Vortex formation, shedding and destruction Vortex street at different Reynolds numbers TMR 7 lab test 3: Particle Image Velocimetry 44

45 Lab test No 4: PIV of a cylinder wake Setup: Fixed uniform cylinder in a steady flow Objectives: Vortex formation, shedding and destruction Vortex street at different Reynolds numbers TMR 7 lab test 3: Particle Image Velocimetry 45

46 Lab test No 4: PIV of a cylinder wake Prepare yourselves: Determine what frequencies to expect. Be familiar with shedding regimes. In the report: Brief overview of objective, setup and measurement principle. Create meaningful results compare with literature. References! Wikipedia you lose. TMR 7 lab test 3: Particle Image Velocimetry 46

47 Lab test No 4: PIV of a cylinder wake Prepare yourselves, e.g.: Pettersen, B.(2008): Marin teknikk 3 Hydrodynamikk. Zdravkovich, M. (1997). Flow around Circular Cylinders; Volume 1. Fundamentals. Oxford Science Publications. Faltinsen, O. (1990). Sea Loads on Ships and Offshore Structures. Melbourne, Australia: Cambridge University Press Blevins, R. D. (1990). Flow-induced Vibration (2nd ed ed.). New York: Van Nostrand Reinhold. TMR 7 lab test 3: Particle Image Velocimetry 47

48 Recommended references Books: Raffel, Willert, Wereley, Kompenhans: Particle Image Velocimetry A Practical Guide, 2nd edition (2007) Adrian and Westerweel: Particle Image Velocimetry (2010) Tropea, Yarin, Foss (eds.): Springer Handbook of Experimental Fluid Mechanics System suppliers web pages: Users and Developers: Aero&Hydrodynamics TUDelft Dt. Zentrum f. Luft- und Raumfahrt, Department Experimental Methods (EV) Laboratoire de Mécanique de Lille TMR 7 lab test 3: Particle Image Velocimetry 48

49 Examples of PIV TMR 7 lab test 3: Particle Image Velocimetry 49

50 Turbulent wake behind tapered cylinders TMR 7 lab test 3: Particle Image Velocimetry 50

51 Turbulent wake behind tapered cylinders TMR 7 lab test 3: Particle Image Velocimetry 51

52 Flow behind intersecting cylinders (cross) camera 1 camera 2 Light sheet optics Frame and test model TMR 7 lab test 3: Particle Image Velocimetry 52

53 Flow behind intersecting cylinders (cross) TMR 7 lab test 3: Particle Image Velocimetry 53

54 Flow behind intersecting cylinders (cross) TMR 7 lab test 3: Particle Image Velocimetry 54

55 Flow around a moving propeller 2 comp. steady rps 3 comp. transient crashback/forward TMR 7 lab test 3: Particle Image Velocimetry 55

56 Flow around a moving propeller TMR 7 lab test 3: Particle Image Velocimetry 56

57 Flow around a moving propeller TMR 7 lab test 3: Particle Image Velocimetry 57

58 Flow around a moving propeller 3D results TMR 7 lab test 3: Particle Image Velocimetry 58

59 Example: Flow around straked cylinders 59

60 #1 #2 #3 #2 #3 #4 #5 #6 #5 #6 #7 #8 #9 #8 #9 ure 1: Mean, normalized velocity magnitude from the 9 measurement planes indicated in Figure 60

61 61

62 HTA flat plate benchmark 62

63 HTA flat plate benchmark 63

64 KMB ship-to-ship interaction 64

65 Warrick, D. R. et al: Aerodynamics of the hovering hummingbird, 2005, Nature 435, TMR 7 lab test 3: Particle Image Velocimetry 65

66 Source: Ecole Polytechnique Fédérale de Lausanne TMR 7 lab test 3: Particle Image Velocimetry 66

67 DLR-Institute Aerodynamics and flow technology 67

68 68

69 Thanks for listening! TMR 7 lab test 3: Particle Image Velocimetry 69