Optimization of Wind Turbine Vortex Generator Based on Back Propagation Neural Network

doi:10.16183/j.cnki.jsjtu.2022.169

Abstract

Abstract:

The optimal Latin hypercube experimental design method is used to refine the vortex generator parameters, determine the test scheme, simulate and calculate the thrust and torque of the wind turbine, and obtain the experimental data. Based on the back propagation (BP) neural network, the aerodynamic performance model of the wind turbine vortex generator optimized by genetic algorithm is constructed. The reliability of the aerodynamic performance model is verified by calculating the error and root mean square of the predicted and simulated values of the aerodynamic performance model. Coupling the fish swarm algorithm and the aerodynamic performance model of the wind turbine vortex generator, an optimization method of the wind turbine vortex generator is established, and the height, length, and installation angle of the vortex generator are solved iteratively to realize the optimization of the vortex generator. The results show that compared with the original vortex generator scheme, the flow separation of the wind turbine blade section optimized by the vortex generator is effectively restrained and delayed, the surface fluid separation phenomenon is improved, the power of the wind turbine is increased by 1.711%, and the thrust of the wind turbine is decreased by 0.875%.

Key words: genetic algorithm, back propagation (BP) neural network, fish swarm algorithm, vortex generator, wind turbine power

CLC Number:

TM315

XIA Yunsong, TAN Jianfeng, HAN Shui, GAO Jin’e. Optimization of Wind Turbine Vortex Generator Based on Back Propagation Neural Network[J]. Journal of Shanghai Jiao Tong University, 2023, 57(11): 1492-1500.

Figures/Tables 13

Fig.1

Fig.2

Tab.1

Fig.3

Fig.4

Fig.5

Fig.6

Tab.2

Fig.7

Tab.3

Fig.8

Fig.9

Fig.10

References 16

[1]	ZHAO Z Z, JIANG R F, FENG J X, et al. Researches on vortex generators applied to wind turbines: A review[J]. Ocean Engineering, 2022, 253: 111266. doi: 10.1016/j.oceaneng.2022.111266 URL
[2]	李爽. 风力机翼型动态失速的模型及流动控制机制研究[D]. 北京: 中国科学院大学(中国科学院工程热物理研究所), 2021.
	LI Shuang. Study of dynamic stall model and flow control mechanism of wind turbine airfoil[D]. Beijing: Institute of Physics, Chinese Academy of Sciences, 2021.
[3]	MANOLESOS M, VOUTSINAS S G. Experimental investigation of the flow past passive vortex generators on an airfoil experiencing three-dimensional separation[J]. Journal of Wind Engineering and Industrial Aerodynamics, 2015, 142: 130-148. doi: 10.1016/j.jweia.2015.03.020 URL
[4]	GAO L Y, ZHANG H, LIU Y, et al. Effects of vortex generators on a blunt trailing-edge airfoil for wind turbines[J]. Renewable Energy, 2015, 76: 303-311. doi: 10.1016/j.renene.2014.11.043 URL
[5]	TIMMER W A, VAN ROOIJ R P J O M. Summary of the delft university wind turbine dedicated airfoils[J]. Journal of Solar Energy Engineering, 2003, 125(4): 488-496. doi: 10.1115/1.1626129 URL
[6]	MUELLER-VAHL H, PECHLIVANOGLOU G, NAYERI C N, et al. Vortex generators for wind turbine blades: A combined wind tunnel and wind turbine parametric study[C]// Proceedings of ASME Turbo Expo 2012:Turbine Technical Conference and Exposition. Copenhagen, Denmark: ASME, 2012: 899-914.
[7]	DE TAVERNIER D, FERREIRA C, VIRÉ A, et al. Controlling dynamic stall using vortex generators on a wind turbine airfoil[J]. Renewable Energy, 2021, 172: 1194-1211. doi: 10.1016/j.renene.2021.03.019 URL
[8]	张惠, 赵宗德, 周广鑫, 等. 涡流发生器对风力机翼型气动性能影响的实验研究[J]. 太阳能学报, 2017, 38(4): 951-958.
	ZHANG Hui, ZHAO Zongde, ZHOU Guangxin, et al. Experimental investigation of effect of vortex generator on aerodynamic performance of wind turbine airfoil[J]. Acta Energiae Solaris Sinica, 2017, 38(4): 951-958.
[9]	张惠, 赵宗德, 周广鑫, 等. 涡发生器参数对风力机翼型性能影响实验研究[J]. 太阳能学报, 2017, 38(12): 3399-3405.
	ZHANG Hui, ZHAO Zongde, ZHOU Guangxin, et al. Experimental investigation of effect of vortex generator’s parameter on performance of wind turbine aerofoil[J]. Acta Energiae Solaris Sinica, 2017, 38(12): 3399-3405.
[10]	杨瑞, 马超善, 方亮, 等. 加装叶片涡流发生器对变桨距风力机功率的影响[J]. 兰州理工大学学报, 2019, 45(6): 64-68.
	YANG Rui, MA Chaoshan, FANG Liang, et al. Effect of installing vortex generator onto blade on power of variable-pitch wind turbine[J]. Journal of Lanzhou University of Technology, 2019, 45(6): 64-68.
[11]	FATEHI M, NILI-AHMADABADI M, NEMATOLLAHI O, et al. Aerodynamic performance improvement of wind turbine blade by cavity shape optimization[J]. Renewable Energy, 2019, 132: 773-785. doi: 10.1016/j.renene.2018.08.047 URL
[12]	赵清鑫, 张兰挺. 基于径向基神经网络的风力机叶片铺层优化[J]. 太阳能学报, 2020, 41(4): 229-234.
	ZHAO Qingxin, ZHANG Lanting. Ply parameter optimization of wind turbine blade based on radial basis function netural network[J]. Acta Energiae Solaris Sinica, 2020, 41(4): 229-234.
[13]	BI J X, FAN W Z, WANG Y, et al. A fault diagnosis algorithm for wind turbine blades based on BP neural network[J]. IOP Conference Series: Materials Science and Engineering, 2021, 1043(2): 022032. doi: 10.1088/1757-899X/1043/2/022032
[14]	HAND M M, SIMMS D A, FINGERSH L J, et al. Unsteady aerodynamics experiment phase VI: Wind tunnel test configurations and available data campaigns[R]. USA: Office of Scientific and Technical Information, 2001.
[15]	季宁, 张卫星, 于洋洋, 等. 基于Kriging代理模型和MOPSO算法的注塑成型质量多目标优化[J]. 塑料工业, 2020, 48(5): 67-71.
	JI Ning, ZHANG Weixing, YU Yangyang, et al. Multi-objective optimization of injection molding quality based on kriging agent model and MOPSO algorithm[J]. China Plastics Industry, 2020, 48(5): 67-71.
[16]	李晓磊. 一种新型的智能优化方法-人工鱼群算法[D]. 杭州: 浙江大学, 2003.
	LI Xiaolei. A new intelligent optimization method-artificial fish school algorithm[D]. Hangzhou: Zhejiang University, 2003.

序号	h/mm	l/mm	β/(°)	T/N	M/(N·m)
1	7.04	27.86	21.94	2129.69	1100.96
2	19.61	24.43	25.00	2051.26	872.01
3	17.37	15.86	6.63	2125.23	1032.05
?	?	?	?	?	?
46	22.76	13.29	10.20	2134.02	1083.02
47	24.10	24.86	12.76	2115.36	1054.24
48	20.51	10.71	21.43	2092.35	1016.70
49	9.29	17.57	7.65	2140.35	1077.87
50	16.47	25.29	15.31	2127.53	1027.86

序号	h/mm	l/mm	β/(°)	推力			转矩
序号	h/mm	l/mm	β/(°)	Y_T₁/N	Y_T₂/N	误差/%	Y_M₁/(N·m)	Y_M₂/(N·m)	误差/%
1	22.76	13.29	10.20	2134.016	2132.933	0.0510	1083.019	1083.484	0.0429
2	24.10	24.86	12.76	2115.363	2114.688	0.0320	1054.235	1053.252	0.0930
3	20.51	10.71	21.43	2092.350	2091.331	0.0490	1016.699	1016.375	0.0320
4	9.29	17.57	7.65	2140.347	2141.031	0.0319	1077.872	1078.278	0.0377
5	16.47	25.29	15.31	2127.533	2126.936	0.0280	1027.856	1024.256	0.0350

参数	VGs方案值		增幅/%
参数	原始	优化	增幅/%
T/kN	2.177	2.158	-0.875
P/kW	8.359	8.502	1.711