Atmospheric Optical Turbulence Profile Model Fitting Based on Improved Particle Swarm Algorithm

Ketao Feng; Xiaoyi Li; Xuan Qian; Lehua Wu; He Zheng; Mou Chen; Mengru Li; Bo Liu

doi:10.3788/LOP202259.0501002

Journals >Laser & Optoelectronics Progress >Volume 59 >Issue 5 >Page 0501002 > Article

Laser & Optoelectronics Progress
Vol. 59, Issue 5, 0501002 (2022)

Atmospheric Optical Turbulence Profile Model Fitting Based on Improved Particle Swarm Algorithm

Ketao Feng¹, Xiaoyi Li^1、*, Xuan Qian², Lehua Wu¹, He Zheng¹, Mou Chen¹, Mengru Li¹, and Bo Liu³

Author Affiliations

¹Communication Sergeant School, Army Engineering University of PLA, Chongqing 400035, China

²National Astronomical Observatory, Chinese Academy of Sciences, Beijing 100101, China

³Unit 78092 of PLA, Chengdu , Sichuan 610036, China

show less

DOI: 10.3788/LOP202259.0501002 Cite this Article Set citation alerts

Ketao Feng, Xiaoyi Li, Xuan Qian, Lehua Wu, He Zheng, Mou Chen, Mengru Li, Bo Liu. Atmospheric Optical Turbulence Profile Model Fitting Based on Improved Particle Swarm Algorithm[J]. Laser & Optoelectronics Progress, 2022, 59(5): 0501002 Copy Citation Text

show less

Fig. 1. Particle position update method

Download full size

$Curves of fitting formula for refractive index structure constant$

Fig. 2. Curves of fitting formula for refractive index structure constant

Download full size

Fig. 3. Flowchart of the improved PSO algorithm

Download full size

Fig. 4. Comparison of the fitted profiles and the statistical average profiles of

C_{n}^{2}

for different periods in Ali. (a) In the morning; (b) in the evening

Download full size

Fig. 5. E_BIAS between the profile of

C_{n}^{2}

fitting model and the statistical average profiles in the morning and evening of Ali. (a) In the morning; (b) in the evening

Download full size

Fig. 6. [in Chinese]

Download full size

Fig. 7. E_BIAS between the profile of

C_{n}^{2}

fitting model and the statistical average profiles for different seasons of Ali. (a) In the spring; (b) in the summer; (c) in the autumn; (d) in the winter

Download full size

Fig. 8. Comparison of turbulence profile model optimization curves in autumn

Download full size

Method	Spring	Summer	Autumn	Winter
SPGD	0.9989	0.9895	0.9918	0.9989
Our method	0.9995	0.9977	0.9992	0.9996

Table 0. R² of four seasons turbulence profile models

Time	a₁	a₂	a₃	b₁	b₂	b₃	c₀	h_T
Morning	4.435×10^-20	1.302×10^-17	8.629×10^-16	2.064	3.061	0.032	3.881	5.3
Evening	7.453×10^-20	2.456×10^-18	1.162×10^-16	1.978	5.195	0.028	3.743	5.1

Table 1. Turbulence profile model constants in the morning and evening

Time	β₀	β₁	β₂	β₃	β₄	β₅	β₆	β₇
Morning	-16.66	-1.479	1.416	-0.569	0.1023	-0.00692	0	0
Evening	-18.46	2.167	-2.289	1.034	-0.1997	0.01376	0	0

Table 2. Turbulence profile model coefficients in the morning and evening

Method	Morning	Evening
SPGD	0.9983	0.9707
Our method	0.9997	0.9976

Table 3. R² of turbulence profile model in the morning and evening

Season	a₁	a₂	a₃	b₁	b₂	b₃	c₀	h_T
Spring	1.702×10^-20	4.013×10^-18	3.812×10^-16	1.376	5.964	0.030	5.385	10.4
Summer	9.587×10^-29	6.411×10^-18	1.772×10^-16	0.813	4.727	0.0207	15.399	10.3
Autumn	3.571×10^-21	5.081×10^-18	2.423×10^-16	1.402	5.102	0.0293	5.967	5.0
Winter	3.701×10^-21	7.590×10^-18	6.098×10^-16	1.391	4.994	0.0329	6.014	9.8

Table 4. Model constants of turbulent profiles in four seasons

Season	β₀	β₁	β₂	β₃	β₄	β₅	β₆	β₇
Spring	-17.3625	-0.7477	0.779	-0.30482	0.05486	-0.004568	0.0001425	0
Summer	-17.39	-0.2824	0.3231	-0.0981	0.011	-0.000421	0	0
Autumn	-15.607	-10.598	18.91	-15.95	7.145	-1.74	0.2176	-0.010954
Winter	-16.799	-1.47205	1.25	-0.4465	0.07668	-0.006254	0.0001944	0

Table 5. Model coefficients of turbulent profiles in four seasons

Time	Method in Ref.［36］	Method in Ref.［41］	Method in Ref.［42］	Our method
Morning	1.4808	1.4808	1.4796	1.4795
Evening	0.9471	0.9468	0.9469	0.9465
Spring	1.2292	1.2296	1.2286	1.2277
Summer	1.4931	1.4945	1.4921	1.4916
Autumn	1.6139	1.6143	1.6131	1.6122
Winter	1.7725	1.7725	1.7725	1.7725