Abstract

Atmospheric turbulence ($C_n^2$) modeling has been proposed by physics-based models, but they are unable to capture the many cases. Recently, machine learning surrogate models have been used to learn the relationship between local meteorological conditions and turbulence strength. These models predict $C_n^2$ at time $t$ from weather at time $t$. This work expands modeling capabilities by proposing a technique to forecast 3 h of future turbulence conditions at 30 min intervals from prior environmental parameters using artificial neural networks. First, local weather and turbulence measurements are formatted to pairs of the input sequence and output forecast. Next, a grid search is used to find the best combination of model architecture, input variables, and training parameters. The architectures investigated are the multilayer perceptron and three variants of the recurrent neural network (RNN): the simple RNN, the long short-term memory RNN (LSTM-RNN), and the gated recurrent unit RNN (GRU-RNN). A GRU-RNN architecture that uses 12 h of prior inputs is found to have the best performance. Finally, this model is applied to the test dataset and analyzed. It is shown that the model has generally learned the relationship between prior environmental and future turbulence conditions.

© 2023 Optica Publishing Group

Adaptive niche-genetic algorithm based on backpropagation neural network for atmospheric turbulence forecasting

Changdong Su, Xiaoqing Wu, Tao Luo, Su Wu, and Chun Qing
Appl. Opt. 59(12) 3699-3705 (2020)

Forecasting the atmospheric refractive index structure constant profile with an altitude-time correlations-inspired deep learning model

Muyu Hou, Shuhong Gong, Xue Li, Donghai Xiao, Yanchun Zuo, and Yu Liu
Opt. Express 31(2) 2426-2444 (2023)

Generalizable turbulent flow forecasting for adaptive optics control

Benjamin D. Shaffer, Jeremy R. Vorenberg, Christopher C. Wilcox, and Austin J. McDaniel
Appl. Opt. 62(23) G1-G11 (2023)

Continuous daytime and nighttime forecast of atmospheric optical turbulence from numerical weather prediction models

Florian Quatresooz, Ryan Griffiths, Lisa Bardou, Richard Wilson, James Osborn, Danielle Vanhoenacker-Janvier, and Claude Oestges
Opt. Express 31(21) 33850-33872 (2023)

Joint object classification and turbulence strength estimation using convolutional neural networks

Daniel A. LeMaster, Steven Leung, and Olga L. Mendoza-Schrock
Appl. Opt. 60(25) G40-G48 (2021)