performance studies on a direct drive turbine for wave power generation in a numerical wave tank

Performance Studies on a Direct Drive Turbine for Wave Power Generation in

a Numerical Wave Tank

Professor Young-Ho LEE Division of Mechanical & Energy System

Engineering,College of Engineering,

Korea Marit ime University, South Korea.

Deepak PrasadDiscipl ine of Mechanical Engineering,

University of the South Pacif ic, Fij i Is lands.

Dr. M. R. AhmedDiscipline of Mechanical Engineering,

University of the South Pacif ic, Fij i Islands.

Renewable Energy Sources

2

Wind Power Solar Power Biomass Hydro Power Ocean Energy

Thermal Energy

Mechanical Energy

WaveTide

Fig. 1 Renewable energy sources

3

Introduction

Wave energy: Power flux – 15 to 20 times more than wind and solar. Estimated wave energy is in order of 1 to 10 TW. Most consistent of all the intermittent sources and truly

renewable.

Wave tanks have been used over the years to provide: helpful information on wave characteristics and

employed to conduct prototype testing. However it is: Expensive, time consuming. Due to time constraint all design variables can not be tested.

4

Introduction

To overcome these problems much effort has been focused on the development of Numerical Wave Tank (NWT). NWT allows for rapid design changes and

improvements in short time.

With improving computer capabilities it is possible for these CFD packages to solve and give accurate solutions of real life problems.

5

Introduction

The current study employs a 3D NWT based on Reynolds Averaged Navier-Stokes Equation (RANSE) to generate waves using commercial CFD code ANSYS-CFX.

A cross flow turbine is employed to generate power from incoming waves.

Aim: To simulate waves using a Numerical Wave Tank. To validate the CFD code with the experimental data. To study flow characteristics.

6

Base Model

All Dimensions in mm

Fig. 2 Schematic of the entire model

Fig. 3 Schematic of the turbine and runner blade

Parameter Value

Blade entry angle, α 30º

Blade exit angle, β 90º

Outer Diameter, Do 260 mm

Inner Diameter, Di 165 mm

Number of Blades 30

Table 1 Turbine and runner parameters

Full model calculation. Numerical Wave Tank (NWT). Front Guide Nozzle. Augmentation Channel.

Front Nozzle. Rear Nozzle. Turbine.

Rear Chamber.

7

Test Section

Fig. 4 Schematic of the test section

CFX-Pre Computational Domain

8

Simulation Type Transient Wave Maker Piston-Type (Specified Motion)

Turbulence Model k-Epsilon Opening 1 atm

Phase Water & Air @ 25°C Walls No-Slip Condition

Surf. Tension Coeff.

0.075 N/m Density (kg/m3) Water = 997, Air 1.18

Opening

Moving mesh sect ion

Wall Motion: Asin(ωt)

Fig. 5 Computational domain.

NWT

Front GuideNozzle

Augmentat ion channel

Rear chamber

Table 2 ANSYS CFX-Pre condit ions

9

Wave Height & Velocity (NWT)

Point

Fig. 6 Water wave height in the numerical wave tank

Fig. 8 Velocity contour in the numerical wave tank

Fig. 7 Volume fraction showing the formation of waves in the wave tank

10

Front Guide Nozzle

Fig. 9 Velocity vector in front guide nozzle

There is a recirculation region observed near the top left corner, denoted by A when water is flowing in.

Due to this, the flow was directed towards the bottom and hence higher velocity recorded in region B.

When water was retreating, higher velocity was observed in region A.

11

Front Guide Nozzle

Fig. 10 Average velocity in the front guide nozzleat dif ferent rpm

It is observed that the velocity increases as the rpm increases, reaches a maximum and then decreases.

This peak at 35 rpm is due to better flow characteristic because of small re-circulating region in the front guide nozzle.

This also leads to better flow in the augmentation channel.

12

Turbine Power

Fig. 11 Comparison between experimental data and CFD results

Variable Unit Experiment CFD

H m 0.2 0.195

T s 2.0 2.0

ΔH m 0.071 0.065

Q m3/s 0.03 0.032

PWave W/m 86.74 82.46

PWP W 20.85 20.36

Table 3 Comparing experimental and CFD results

For CFD, the peak power is 6.71 W compared to 6.8 W obtained experimentally. The efficiency at 35 rpm is 44.73% and 45.33% respectively from CFD and experiments. The difference is within 3%.

13

Flow Characteristics

Fig. 12 Velocity vector in the augmentation channel at 35 rpm

The flow accelerates approaching stage 1 as expected. (Nozzle effect)

The water passes through the turbine passage at stage 1 while imparting energy to the runner.

From the exit at stage 1 to the entry of blades at stage 2, the flow again accelerates a little.

At stage 2, the passing water imparts energy to the runner once more before flowing into the rear nozzle.

Flow Characterist ics

14

15

Conclusion

Commercial CFD code ANSYS-CFX was successfully used to generate waves in a NWT using a piston type wave-maker.

The results of CFD simulation showed good agreement with the experimental data. The difference in result was within 3%.

The maximum turbine power was obtained at 35 rpm. For CFD, the maximum power was 6.71 W compared to 6.8 W obtained experimentally.

The efficiency at 35 rpm was 44.73% and 45.33% respectively for CFD and experiment.