Experimental and Analytical Study of Helical Cross Flow Turbines for a Tidal Micropower Generation System

Transcription

1 Experimental and Analytical Study of Helical Cross Flow Turbines for a Tidal Micropower Generation System Adam Niblick University of Washington Northwest National Marine Renewable Energy Center MSME Thesis Defense March 7, 2012

2 Outline Tidal Micropower System Overview Turbine Selection/Design Parameters Experimental Testing Modeling Conclusions and Future Work

3 Motivation for Micropower Admiralty Inlet potential site of tidal turbine installation for commercial power Environmental site characterization monitoring Sea Spider bottom lander tripod Instrumentation requires power Currently battery powered Grid, solar and wind not feasible

4 Motivation for Micropower Unit Battery Capacity (W hr): 165 Cost ($/W hr): $ 1.15 Power Usage (W) Deployment Duration (days) Total W-hr # of batteries Required Battery Cost Doppler Current Profiler $ 2,280.00

5 Motivation for Micropower Unit Battery Capacity (W hr): 165 Cost ($/W hr): $ 1.15 Power Usage (W) Deployment Duration (days) Total W-hr # of batteries Required Battery Cost Doppler Current Profiler $ 2, Doppler Current Profiler $ 4,560.00

6 Motivation for Micropower Unit Battery Capacity (W hr): 165 Cost ($/W hr): $ 1.15 Power Usage (W) Deployment Duration (days) Total W-hr # of batteries Required Battery Cost Doppler Current Profiler $ 2, Doppler Current Profiler $ 4, Imaging Sonar (20% duty cycle) $ 99, u 0 : Inflow velocity ρ: Fluid density

7 Motivation for Micropower Unit Battery Capacity (W hr): 165 Cost ($/W hr): $ 1.15 Power Usage (W) Deployment Duration (days) Total W-hr # of batteries Required Battery Cost Doppler Current Profiler $ 2, Doppler Current Profiler $ 4, Imaging Sonar (20% duty cycle) $ 99, Imaging Sonar (20% duty cycle) $ 2,850.00

8 Micropower System Concept Drawing courtesy of Stelzenmuller et al Micropower System A tidal hydrokinetic power generation system to power remote instrumentation on the order of 20 W continuous in 1.5 m/s peak tidal currents.

9 Research Goals Provide initial concept for a micropower tidal generation system Provide detailed design, analysis and experimental testing one of the components: the hydrokinetic turbine

10 Micropower System Block Diagram u 0 (m/s) 2 0 Tidal Current Profile time (days) Tidal Turbine Turbine Battery + Bank + Loads Loads Monitoring Equipment Gearbox Generator Rectifier Converter Controller/processor Controller Data storage

11 Outline Micropower System Overview Turbine Selection/Design Parameters Experimental Testing Modeling Conclusions and Future Work

12 Turbine Performance Metrics Efficiency for power and torque production Starting torque magnitude and range for self start H: Height D: Diameter R: Radius T: Torque u 0 : Inflow velocity ρ: Fluid density θ: Azimuthal angle ω: Angular velocity Tip Speed Ratio and Rotation Rate for generator integration Performance with respect to inflow orientation for deployment flexibility

13 Turbine Design Concepts High efficiency (~35%) Good self start Savonius capability Rotor Accepts flow from any horizontal Axial Turbine direction Darrieus Turbine Hybrid Turbine Helical Turbine Sources: Polagye, 2011; engr.mun.ca/~tariq/jahangiralam.pdf; newenergycorp.ca;

14 Blade Element Theory Streamwise velocity (u S ) Tip velocity (ωr) Induced by turbine s own rotation Relative velocity (u REL ) Resultant vector of u S and ωr Angle of attack (α) Angle between u REL and chord line Lift Drag Forces (F L, F D ) Generate turbine torque Best lift drag ratio when 10< α<10 c: Blade length h: Blade height A P : Planform area ch

15 Axial vs. Cross Flow Turbines Axial Cross-Flow Tidal Flow is Parallel to Axis of Rotation Tidal Flow is Perpendicular to Axis of Rotation

16 Blade Element Theory Angle of attack (α) Constantly changing as turbine rotates Often exceeds stall angle

17 Helical Turbine Parameters Blade Profile Blade Pitch Helical Pitch Aspect Ratio Solidity Ratio/ Chord to Radius Ratio Number of Blades Blade Wrap Attachment Design

18 Helical Turbine Parameters Blade Profile Helical Pitch Aspect Ratio Solidity Ratio/ Chord to Radius Ratio Number of Blades Attachment Design y NACA x Source: answers.com

19 Helical Turbine Parameters Blade Profile Helical Pitch Aspect Ratio Solidity Ratio/ Chord to Radius Ratio Number of Blades Attachment Design

20 Helical Turbine Parameters Blade Profile Helical Pitch Aspect Ratio Solidity Ratio/ Chord to Radius Ratio Number of Blades Attachment Design Size limited by deployment constraints

21 Helical Turbine Parameters Blade Profile Helical Pitch Aspect Ratio Solidity Ratio/ Chord to Radius Ratio Number of Blades (B) Attachment Design 30% Solidity 15% Solidity

22 Helical Turbine Parameters Blade Profile Helical Pitch Aspect Ratio Solidity Ratio/ Chord to Radius Ratio Number of Blades Attachment Design

23 Parameter Optimization Higher efficiency, Improved C P, Manufacturability Improved C P, Allows higher pitch Helical Pitch Aspect Ratio Lower static torque variation Easier Deployment, Less blade stress, Less vibration Better starting torque, Manufacturability Solidity Ratio Higher tip speed ratio Allows increase in pitch and reduced aspect ratio Number of Blades Fewer parts, Less blade wake interference

24 Outline Micropower System Overview Turbine Selection/Design Parameters Experimental Testing Modeling Conclusions and Future Work

25 Turbine Prototypes 3 Bladed Turbine Baseline Configuration 4 Bladed Turbine Low Solidity Turbine 3 Blades D=20 cm H=20 cm 1.00 Aspect Ratio 43.7 Pitch Angle 30% Solidity 4 Blades D=17.2 cm H=23.4 cm 1.36 Aspect Ratio 60 Pitch Angle 30% Solidity 3 Blades D=20 cm H=20 cm 1.00 Aspect Ratio 43.7 Pitch Angle 15% Solidity

26 Tests Performed Baseline Configuration Load Performance Test Static Torque Test Tilted Turbine Test Design Comparison Tests Helical Pitch/No. of Blades Strut Attachment Design Shaft Design

27 Flume Test Channel 3 meters length 72 cm wide Maximum 0.8 m/s flow rate Turbine located in center of flume width at 1.7 m from test channel inlet Acoustic Doppler Velocimeter (ADV) to measure flume velocity

28 Test Instrumentation Module Reaction Torque Cell Particle Brake Flex Coupling Optical Encoder Locking Plate Bearing

29 Turbine Operation

30 Load Performance Results Baseline Configuration

31 Lift and Drag for different α Region Region of of but Turbine increasing λ Operation decreases 0.4 at 0.8 m/s m/saverage α

32 Static Torque Baseline Configuration Flume: 0.7 m/s 0 view, looking downstream

33 Turbine Design Comparison Load Performance Test Results 3-Bladed Turbine 4-Bladed Turbine

34 Strut Design Comparison 2 mm Spoke 4.8 mm Spoke Circular Plate

35 Shaft Diameter Comparison No Shaft 6.4 mm Shaft 12.7 mm Shaft

36 Tilted Turbine Test Downstream end of turbine frame inclined 0, 2.5, 5 and 10 degree inclinations Performance test Baseline spoke design and circular plate configurations

37 Tilted Turbine Test

38 Test Conclusions Four bladed turbine is best design for optimum aspect ratio and performance Good self start and performance with high solidity Maximize helical pitch angle to get better max C P No impact of shaft diameter on performance Circular plate best attachment design eliminates profile drag, better tilt angle performance Static torque is not uniform Higher inflow velocity gives better C P

39 Outline Micropower System Overview Turbine Selection/Design Parameters Experimental Testing Modeling Conclusions and Future Work

40 Vortex Model Free vortex flow model Blade divided into several elements Blade element performance from C L,C D lookup tables Blades elements simulated as vortex generators Sum of velocity components gives u REL

41 Vortex Model Comparison to other model types Less computational time than CFD (days vs. minutes) Compared to momentum models, the vortex model: Can resolve helical geometry without yaw correction factors Better resolves the wake Model Modifications Helical geometry Strut drag for circular plate and end spokes Flow curvature effects Higher σ C R than previously modeled

42 Flow Curvature Change in α from leading edge to trailing edge Important for high chord to radius ratio turbines Approximated as Virtual camber Source: Migliore, 1980

43 Pitching and Dynamic Stall Lift, drag affected by pitch rate Virtual incidence Approximated by C L calculated from α at ¾ chord C D calculated at α at ½ chord Biases α positive Dynamic stall y x Vortex forms at the leading edge of hydrofoil and tracks back along the chord Lift is increased during tracking Lift drops as vortex fully sheds NACA0018 ¼ ½ ¾ Source: Ekaterinaris & Platzer, 1997

44 Dynamic Stall Lift Drag Source: NASA, 2009

45 Implementation of Secondary Effects Experimental Data at 0.8 m/s

46 Implementation of Secondary Effects Experimental Data at 0.8 m/s Model Prediction with no Secondary Effects

47 Implementation of Secondary Effects Experimental Data at 0.8 m/s Dynamic Stall added Model Prediction with no Secondary Effects

48 Implementation of Secondary Effects Experimental Data at 0.8 m/s Dynamic Stall & Pitch Rate Effects added Dynamic Stall added Model Prediction with no Secondary Effects

49 Implementation of Secondary Effects Dynamic Stall, Pitch Rate & Flow Curvature Effects added Dynamic Stall & Pitch Rate Effects added Dynamic Stall added Model Prediction with no Secondary Effects

50 Model Prediction of Inflow Variation

51 Extrapolation to 1.5 m/s

52 Analysis of High Tip Speed Ratio C Q =2.6, 1.5 m/s (a) Flow curvature active C P =0.46 C Q (degrees) Normalized Time (t*u 0 /R) Flow curvature disabled C P =0.43

53 Analysis of High Tip Speed Ratio α 3/4 10 α 1/2 6 Due to high chord to radius ratio: Flow curvature adds instability to model output Pitch rate effects and dynamic stall over amplifying performance This effect has been documented in previous literature (Dai et al., 2011), (Cardona, 1984)

54 Analysis of High Tip Speed Ratio α 3/4 10 α 1/2 6 Due to high chord to radius ratio: Flow curvature adds instability to model output Pitch rate effects and dynamic stall over amplifying performance This effect has been documented in previous literature (Dai et al., 2011), (Cardona, 1984)

55 Project Conclusions Demonstrated need for and viability of a tidal micropower system Designed a turbine for a micropower system 24% efficient at 0.8 m/s (and increasing with flume speed) Good self start and flow orientation capabilities Optimized select helical turbine design parameters Modified a vortex model to accept helical turbine design Demonstrated importance of secondary effects Demonstrated general trends in performance Need for further study of secondary effects for high chord toradius ratio turbines

56 Future Work Testing of the full scale turbine Design of system drive train and control Turbine Generator matching Selection of battery type Improving turbine efficiency Improved turbine modeling Examination and refinement of secondary effects Other model types (CFD, etc.)

57 Acknowledgements Dr. Brian Polagye Dr. Alberto Aliseda Dr. Brian Fabien Dr. Roy Martin Dr. Philip Malte Bill Kuykendall Kevin Soderlund, Eamon McQuaide Dr. Jim Thomson & APL Capstone Design Team: Nick Stelzenmuller, Bronwyn Hughes, Josh Anderson, Celest Johnson, Brett Taylor, Leo Sutanto Organization Sandia National Labs Marie

58 Questions?

59 Backup Slides

60 System Performance

61 System Performance Turbine generator matching Desired generator characteristics Low speed operation rpm for direct drive High efficiency Flat efficiency curve Fits within Sea Spider size constraints Low starting torque, low cogging

62 Battery Information Characteristic Sealed Lead Acid NiMH NiCd Lithium Ion Energy Density (Wh/L) [1] Low High Low High Cycle Life (cycles) [1] Low Medium High High Memory Effect [4] High Low Medium Low Periodic discharge No effect Less than NiCd No effect required Charging Time [2] High Medium Low Medium 8 16 hours 2 4 hours about 1 hour 1 4 hours Cost [3] ($/kwh) Low Medium Low Medium High $8.50 $18.50 $11.00 $24 Toxicity [2] Very high Low Very high Low Transportation Limitations [4] No No No Yes Comments [1,4] Slow charging time; low energy density Generates heat during high rate charge or discharge; tolerant of overcharge oroverdischarge Rugged; lowest cost per cycle; some limitations on use due to toxicity Data Source: [1] Reddy & Linden (2011) [2] Buchmann (2010d) [3] Buchmann (2011b) [4] Buchmann (2010e) Transportation limitations; requires complex circuit to operate safely; limited availability for high power applications

63 Helical Turbines Lift style device Advantages: Easily self starts Fairly high efficiency (~35%) No torque oscillation Disadvantages: Difficult to manufacture Less efficient than straight blade turbine

64 Blade Element Theory Lift Drag Force Normal Tangential Force

65 Helical Turbine Parameters Blade Profile

66 Helical Turbine Parameters Blade Profile Blade Pitch Helical Pitch Aspect Ratio Solidity Ratio/ Chord to Radius Ratio Number of Blades Blade Wrap Strut/shaft Design

67 Helical Turbine Parameters Blade Profile Blade Pitch Helical Pitch Aspect Ratio Solidity Ratio/ Chord to Radius Ratio Number of Blades Blade Wrap Strut/shaft Design 100% Wrap 50% Wrap

68 Test Matrix Turbine Geometry Flume Velocity (m/s): Turbine Design Strut Design Single Blade, from 4 bladed turbine 4.8 mm, 4 leg spokes Circular Plates 6.4 Performance test Static Torque Dynamic Torque Startup Shutdown Tilted Turbine Wake Measurement Shaft Diameter (mm) 6.4 Single blade length X X X X X X Full Revolution Load Performance Static Torque 4 bladed turbine 2.0 mm, 4 leg spokes No shaft 6.4 X X X X X X X X X X X X X X 12.7 X X X 3 bladed turbine Low Solidity Turbine X: Test Completed Circular Plates mm, 3 leg spokes 4.8 mm, 3 leg spokes X X X X X X X X X X X X X X X X

69 Test Calculations Parameter Symbol Data Range Flume Speed (m/s) u Static Water Height (cm) h sw Dynamic Water Height (cm) h Water Column Cross Sectional Area (m 2 ) A F Turbine Cross Sectional Area (m 2 ) A C Blockage Ratio ε Froude Number Fr Max Blade Reynold's Number (4 Bladed Turbine) Re B 4.0E E+04 Min Blade Reynold's Number (4 Bladed Turbine) Re B 3.6E E+04 Machine Reynold's Number (4 Bladed Turbine) Re M 7.7E E+05

70 Typical Test Results Load Performance Test Static Torque Test

71 Typical Test Results Load Performance Test Static Torque Test

72 Streamwise View θ=0 θ=20 θ=40 θ=70 Static torque near maximum Static torque drops

73 Single Blade Static Torque

74 Single Blade Torque Superposition

75 Streamwise View θ=0 θ=20 θ=40 θ=70 Static torque near maximum Unobstructed (-T) blade Partial obstruction of (+T) blade at Static torque location of drops best drag

76 Turbine Design Comparison 3-Bladed Turbine Static Torque Test Results 4-Bladed Turbine

77 Tilted Turbine Test Predicted C P Loss Strut Design Tilt Angle (degrees) Tilt Velocity from cos(tilt) Actual C P Loss 2 mm, 4 Leg Spoke 2 mm Circular Plate % 0.0% % 5.6% % 8.8% % 23.0% % 0.0% % 4.5% % 4.4% % 8.9%

78 Wake Measurement Test Baseline Configuration Flume: 0.78 m/s Wake measurements along turbine centerline

79 Start up Shutdown Test Profile 0 Turbine Torque, Nm 12 x (a) Starting Angle Load Torque (Nm) Cut In Speed (m/s) Cut Out 200 Speed Average cut in/cut out speed at Nm load torque: Turbine Angle, Degrees U Velocity, m/s Time, s (b) (c)

80 Vortex Model Blade is divided into segments, revolution into time steps 8 elements per blade 15 revolutions to establish wake Vortex influences the flow of surrounding blades

81 Model Solution Sum of velocity components gives u REL Total change in circulation over time is zero (Kelvin s theorem): Vortex induced velocity at a point is found by Biot Savart Law: Circulation strength is found from Kutta Joukowski theorem: Predictor corrector to solve system of equations

82 Boeing Vertol Dynamic Stall Model

83 Flow Curvature Change in α from leading edge to trailing edge Important for high chord to radius ratio turbines Approximated as Virtual camber

84 Full Scale Turbine Prediction All Secondary Effects Active

85 Full Scale Turbine Prediction Dynamic Stall only

86 Model Prediction of Strut Drag u 0 =0.8 m/s