Available online at www.sciencedirect.

com

Energy Procedia 17 (2012) 1201 – 1208

2012 International Conference on Future Electrical Power and Energy Systems

Simulation and Test of the Blade Models’ Output

Characteristics of Wind Turbine
Xu Baoqinga,Tian Deb
a
College of Information and Engineering,Inner Mongolia University of Technology,Huhhot, China
b
College of Renewable Energy,North China Electric Power University,Beijing, China

Abstract

This electronic document is a “live” template. Based on the momentum-element theory, mathematical model of
numerical simulation of the output characteristics is set up for resemble models of MW-scale wind turbine blades,
which is under a test system with three different setting angles of 10.5e, 30.5eand 50.5e. The initial values of
axial and tangential induction factors are determined by Newton-Raphson method, the powers P, power coefficients
CP and torque coefficients CM of the three model test systems are simulated by programming in Matlab and Visual
Basic. At the same time, characteristic output tests are carried out by in-truck test method of the three model test
systems. The results show that each relative error of simulation and testing results of the model test systems is less
than 10%; it proves that the simulation method is reliable and the testing results are reasonable.

©
© 2012
2011 Published
Publishedby
byElsevier
ElsevierLtd.
Ltd.Selection
Selectionand/or peer-review
and/or under
peer-review responsibility
under of Hainan
responsibility University.
of [name organizer]

Keywords:wind turbine generator; blade; numerical simulation; test

1.Introduction

As a major component of wind turbine generator system, blade has been pay more and more attention
to, and its design and manufacture have become a kind of hot technology problem [1~3] in wind power
industry. The shape, airfoil, materials, number of blades, structure and processing technology of the blade
have been deeply studied both at home and abroad [4~7]. Aerodynamic shape which affects the strengths
and weaknesses of the whole unit performance is an important factor.
On the basis of the previous study [8], by improved the current optimal design method, the author [9]
gives the blade synthesis optimal design method taking the aspect ratio as the key parameter. With this
synthesis optimal design method, a kind of MW scale wind turbine blade external shape has been designed.
To further verify the blade aerodynamic performance, and consider the cost, personnel, facilities,
equipment and other test conditions, in accordance with the similarity theory, similar model tests have been
done by in-truck test method. In these tests, the model blades MSJ which have been designed and

1876-6102 © 2012 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of Hainan University.
doi:10.1016/j.egypro.2012.02.227
1202 Xu Baoqing and Tian De / Energy Procedia 17 (2012) 1201 – 1208

manufactured in similarity theory, are installed in three sets of setting angles of 10.5 °, 30.5 ° and 50.5 ° of
the testing systems.
In addition, in order to further understand the relationship between the various parameters of blade
shape and the performance, the mathematical simulation model for the output performance of the similar
models in the testing systems has been established. Using actual performance parameters of the other parts
of the model testing system, the output performance of the testing system with the model blades have been
simulated.
Comparative analysis to the test results and simulation results has been done, so as to master the rules
of how the blade shape acts on the output performance in variable pitch wind generator systems; and the
rules of variation of performance with various pitch angles.

2.Performance Simulation Model

2.1.Calculation Process

In model testing system with certain setting angles, the system output performance should be calculated
in various wind speed. Security level isĉ, and reference velocity vref = 50m/s when designs the prototype
blade. According to "GB / T 19068.2-2003 off-grid wind turbine generator Part 2: Test methods" and the
output characteristics of the testing system itself, it is determined that cut in wind speed vin = 3m/s, cut out
wind speed vout = 16m/s in simulation. Each section has different aerodynamic parameters under different
wind speed, so the whole system’s output performance various with the speed. Calculation process is
composed of several sub-modules shown in Fig. 1.

Figure 1. Flow chart of characteristics simulation.

2.2.Implementation of Sub-module

In module 1, the first step is to calculate the actual setting angle in according to the hub arm’s setting
angle and the twist angle of each blade section, then by relationship n = g (v) , the tip speed ratio can be
calculated as below:
 Xu Baoqing and Tian De / Energy Procedia 17 (2012) 1201 – 1208 1203

n ˜ S ˜ ri
 Oi 
30v
Secondly, axial induction factor a and tangential induction factor b of each section should be calculated.
Calculations need the initial values of a and b, and then revise them. Here using the expression of relation
of a and b in Glauert, the initial values of a and b can be get in Newton-Raphson iterative method.
According to Newton-Raphson iteration formula:

f ( Pk 1 )
 Pk g ( Pk 1 ) Pk 1  
f c( Pk 1 )

Substituted into a expression:

3 3 O2 1
 f (a) a3  a2  (3  O2 )a  
2 16 16

3
 f c(a) 3a 2  3a  (3  O2 ) 
16
Get

f (a k 1 )
 a k a k 1  
f c(a k 1 )

That is:

16ak31  24ak21  3(3  O2 )ak 1  (O2 1)

ak ak 1  
16ak21 16ak 1  3  O2

Where Ȝ is tip ratio at a section; k is the number of iterations. The iterations terminate
when a k  a k 1  10
6
. Then

a(1  a)
 b(1  b) 
O2
For a and b obtained in the method above, revised by the relationship:
1204 Xu Baoqing and Tian De / Energy Procedia 17 (2012) 1201 – 1208

­
° g1
°a
° 1  g1
°
° Bc ( C L cos I  C d sin I )
®g1 ˜ ˜H 
° 2 Sr 4 F sin 2 I
° ­1 . 0 , a d 0 . 3539
° °
°H ® 4 a (1  a )
°¯ ° 0 . 6  0 . 61 a  0 . 79 a 2 , a ! 0 . 3539
¯

­ g2
°b 1  g
° 2
® 
°g Bc (C L sin I  C d cos I )
˜
°¯ 2 2Sr 4 F sin I cos I
Finaly, the ith section axial and tangential induction factors ai and bi can be get. So the other parameters
of section i can be get:

1  ai 1
 I i tan 1 ( ˜ ) 
1  bi Oi

ª  2BR( Rsin rIi ) º

2 cos «e 1 i
»
 «¬ »¼ 
Fti
S

ª  2Br( ri sin
rrub )
Ii
º
2 cos «e 1hub
»
 
Fhi ¬« ¼»
S

 Fi Fti ˜ Fhi 

 D i Ii  E sti 

 Xu Baoqing and Tian De / Energy Procedia 17 (2012) 1201 – 1208 1205

1 1 1 ai 1 ai
 'D1i (tg  tg1 ) 
4 Oi (1 2bi ) Oi

1 B ˜ ci ˜ ri
˜
4 Oi SR 2 t
'D 2i ˜ ˜ max i 
15 1 2 r ci
( )  ( i )2
Oi R

 'D i 'D 1i  'D 2i 

 D sti D i  'D i 

CLi 0.004D sti2  0.145D sti  0.328

CDi 0.006587  0.006807CLi


 0.005029CLi2  0.004404CLi3

For Module 2 is concerned, after find the parameters of each section presented above, and based on the
momentum – section element theory, Thrust Coefficient CT, Torque Coefficient CM and the Power
Coefficient CP by the following relations can be get:

B R
CT ³ c ˜ vw2 ˜ c f ˜ dr 
SR 2 rhub

B R
CM ³ c ˜ v w2 ˜ c g ˜ r ˜ dr 
SR 3 rhub

B R
CP ³ c ˜ v w2 ˜ c g ˜ O ˜ dr 
SR 2 rhub

Where

cf C L cos I  C D sin I 

1206 Xu Baoqing and Tian De / Energy Procedia 17 (2012) 1201 – 1208

cg C L sin I  C D cos I 

(1  a ) 2
 v
2
w 
sin 2 I

Substituting CP to the following formula:

1
 Pout C P ˜ U 0 ˜ S ˜K1 ˜K 2 ˜ R 2 ˜ v 3 
2
System output power Pout in wind speed v can be get.
In (27), ¨1 is generator efficiency; ¨2 is mechanical transmission efficiency. According to the test
method for testing the generator efficiency in [10], ¨1 can be obtained; get a set of blades for 100W small
scale wind energy generator which output characteristics is known, fix them in the test system, carry out
the in-truck test to get its output characteristics, compare and analysis the test results and its original
properties, then the mechanical transmission efficiency¨2 can be gained.

3.Comparative Analyses of Simulation Results and Test Results

3.1.Output Power

Application of the performance calculation method and in-truck test method, the simulation and test
results of the model testing system are obtained. Fig. 2 shows simulation and test output results of the
model testing system.



 06-VLPXODWHG 06-WHVWHG

 06-VLPXODWHG 06-WHVWHG

 06-VLPXODWHG 06-WHVWHG




SRZHU:




             
ZLQGVSHHGPgV 

Figure 2. Simulated and test output powers of the three model sets

In the effective wind speed range of 3m/s ~ 16m/s, the relative errors of simulation and test output of
the three model systems are 7.2%, 6.9% and 5.3%. Obviously, the relative error of MSJ305 and MSJ505 is
smaller, below 12m/s, relative error of MSJ505 reaches 2.3%.
 Xu Baoqing and Tian De / Energy Procedia 17 (2012) 1201 – 1208 1207

3.2.Power Coefficient

Fig. 3 shows the simulated values and test values of the power coefficient are generally consistent. As
the power coefficient CP is proportional to power P, so the variation rule of the relative error of CP is the
same in value.





SRZHUFRHIILFLHQW&S





 06-VLPXODWHG 06-WHVWHG

06-VLPXODWHG 06-WHVWHG

06-VLPXODWHG 06-WHVWHG

             
ZLQGVSHHGPgV 

Figure 3. Simulated and test power coefficients of the three model sets

3.3.Torque Coefficient

In the effective wind speed range of 3m/s ~ 16m/s, the relative errors of simulation and test torque
coefficient of the three model systems are 4.1%, 4.2% and 3.7%, as shown in Fig. 4. Below 12m/s, relative
error is 3.5%, 3.8% and 4.5% respectively; within 12m/s ~ 16m/s, relative error of MSJ505 reaches 2.8%.
And similar to P and CP, CM in a larger setting angle system has smaller errors.




WRUTXHFRHIILFLHQW& 0


06-VLPXODWHG 06-WHVWHG
06-VLPXODWHG 06-WHVWHG
06-VLPXODWHG 06-WHVWHG

             
ZLQGVSHHGPgV 

Figure 4. Simulated and test torque coefficients of the three model sets

4.Conclusion

x The relative errors of simulation and test results in three sets models are less than 10%; in system
with larger setting angles or in wind speed below 12m/s, the simulation and test relative errors are
less than 5%.
x Simulation and test results show that, for the same set of blades, among the setting angles of 10.5 °,
30.5 ° and 50.5 °, the system with smaller angle has better output performance. Figure 1 shows the
average test powers, MSJ105 is 17.33% greater than MSJ305; and is 62.16% greater than MSJ505.
x Compared the simulation and test relative errors with that of power and power coefficient, the
relative errors of torque coefficient are greater, it is because, apart from the system errors, the
expression wind speed and turning speed of rotor is used in calculation of torque coefficients. But
1208 Xu Baoqing and Tian De / Energy Procedia 17 (2012) 1201 – 1208

the expression which is fitted in Matlab also has some errors, therefore, results in greater relative
errors of torque coefficient.

References

[1] M Jureczkom, M Pawlak, “Optimization of Wind Turbine Blades”, Journal of Materials Processing Technology, vol. 167,
2005, pp. 463–471.
[2] Guanpeng Xu, Lakshmi NSankar, Computational Study of Horizontal Axis Wind Turbines, Journal of Solar Energy
Engineering, 2nd ed., vol. 122, 2000, pp. 35–39.
[3] Xin Zuoxia, Yao Xingjia, Chen Lei, The analysis of 1MW variable speed wind turbine parameters optimal design based
on cost modeling method, Acta Energiae Solaris Sinica, 5th ed., vol. 28, 2007, pp. 538–544.
[4] C Kong, J Bang, Y Sugiyama, Structural investigation of composite wind turbine blade considering various load cases
and fatigue life , Energy, vol. 30, 2005, pp. 2101–2114.
[5] Liu Xiong, Chen Yan, Ye Zhiquan, Application of genetic algorithms to HAWT rotor blades optimization, Acta
Energiae Solaris Sinica, 2nd ed., vol. 27, 2006, pp. 180–185.
[6] Tian De, Diao Mingguang, Wang Haikuan, Three & four blades impeller’s wind tunnel experiment study of concentrated
wind energy turbine, Acta Energiae Solaris Sinica, 1st ed., vol. 28, 2007, pp. 74–80.
[7] D.J. Malcolm, A.C. Hansen, Wind-PACT Turbine Rotor Design Study, USA: NREL Subcontractor Report, 2002, pp.
W17–W27.
[8] Xu Baoqing, Tian De, Wu Hua, et al, Methods for solving two-parameter wind velocity weibull distribution,
Transactions of the CSAE, 10th ed., vol. 23, 2007, pp. 180–185.
[9] Xu Baoqing, Tian De, Wang Haikuan, Zhao Danping, Optimal design of the wind turbine blade shape with VB,
Computer Technology and Development, 5th ed., vol. 20, 2010, pp. 211–214.
[10] Xu Baoqing, Tian De, Han Qiaoli, et al. Numerical simulation and experiment of the output power characteristic of the
concentrated wind energy turbine set, Agricultural Engineering Technology, 3rd ed., vol. 327, 2008, pp. 30–33.