Search engine for discovering works of Art, research articles, and books related to Art and Culture
ShareThis
Javascript must be enabled to continue!

Probabilistic Short-Term Solar Driver Forecasting with Neural Network Ensembles

View through CrossRef
Space weather indices are used to drive forecasts of thermosphere density, which directly affects objects in low-Earth orbit (LEO) through atmospheric drag force. A set of proxies and indices (drivers), F, S, M, and Y are used as inputs by the JB2008 thermosphere density model. The United States Air Force (USAF) operational High Accuracy Satellite Drag Model (HASDM), relies on JB2008, and forecasts of solar drivers from a linear algorithm. We introduce methods using long-short term memory (LSTM) model ensembles to improve over the current prediction method as well as a previous univariate approach. We investigate the usage of principal component analysis (PCA) to enhance multivariate forecasting. A novel method, referred to as striped sampling, is created to produce statistically consistent machine learning data sets. We also investigate forecasting performance and uncertainty estimation by varying the training loss function and by investigating novel weighting methods. Results show that stacked neural network model ensembles make multivariate driver forecasts which outperform the operational linear method. When using MV-MLE (multivariate multi-lookback ensemble), we see an improvement of RMSE for F, S, M, and Y of 17.7%, 12.3%, 13.8%, 13.7% respectively, over the operational method. We provide the first probabilistic forecasting method for S, M, and Y . Ensemble approaches are leveraged to provide a distribution of predicted values, allowing an investigation into robustness and reliability (R&R) of uncertainty estimates. Uncertainty was also investigated through the use of calibration error score (CES), with the approach providing an average CES of 5.63%, across all drivers.
Title: Probabilistic Short-Term Solar Driver Forecasting with Neural Network Ensembles
Description:
Space weather indices are used to drive forecasts of thermosphere density, which directly affects objects in low-Earth orbit (LEO) through atmospheric drag force.
A set of proxies and indices (drivers), F, S, M, and Y are used as inputs by the JB2008 thermosphere density model.
The United States Air Force (USAF) operational High Accuracy Satellite Drag Model (HASDM), relies on JB2008, and forecasts of solar drivers from a linear algorithm.
We introduce methods using long-short term memory (LSTM) model ensembles to improve over the current prediction method as well as a previous univariate approach.
We investigate the usage of principal component analysis (PCA) to enhance multivariate forecasting.
A novel method, referred to as striped sampling, is created to produce statistically consistent machine learning data sets.
We also investigate forecasting performance and uncertainty estimation by varying the training loss function and by investigating novel weighting methods.
Results show that stacked neural network model ensembles make multivariate driver forecasts which outperform the operational linear method.
When using MV-MLE (multivariate multi-lookback ensemble), we see an improvement of RMSE for F, S, M, and Y of 17.
7%, 12.
3%, 13.
8%, 13.
7% respectively, over the operational method.
We provide the first probabilistic forecasting method for S, M, and Y .
Ensemble approaches are leveraged to provide a distribution of predicted values, allowing an investigation into robustness and reliability (R&R) of uncertainty estimates.
Uncertainty was also investigated through the use of calibration error score (CES), with the approach providing an average CES of 5.
63%, across all drivers.

Related Results

Probabilistic Solar Proxy Forecasting With Neural Network Ensembles
Probabilistic Solar Proxy Forecasting With Neural Network Ensembles
AbstractSpace weather indices are used commonly to drive forecasts of thermosphere density, which affects objects in low‐Earth orbit (LEO) through atmospheric drag. One commonly us...
Detection of whale calls in noise: Performance comparison between a beluga whale, human listeners, and a neural network
Detection of whale calls in noise: Performance comparison between a beluga whale, human listeners, and a neural network
This article examines the masking by anthropogenic noise of beluga whale calls. Results from human masking experiments and a software backpropagation neural network are compared to...
Bicomplex Projection Rule for Complex-Valued Hopfield Neural Networks
Bicomplex Projection Rule for Complex-Valued Hopfield Neural Networks
A complex-valued Hopfield neural network (CHNN) with a multistate activation function is a multistate model of neural associative memory. The weight parameters need a lot of memory...
On Exponential Convergence Conditions of an Extended Projection Neural Network
On Exponential Convergence Conditions of an Extended Projection Neural Network
Recently the extended projection neural network was proposed to solve constrained monotone variational inequality problems and a class of constrained nonmonotontic variational ineq...
Comparative Policy Analysis of PV Industry Development: Case Studies of China and the United States
Comparative Policy Analysis of PV Industry Development: Case Studies of China and the United States
With global climate change, energy issues have evolved into critical global concerns. The development and utilization of new energy sources have garnered widespread attention, maki...
On Convergence Conditions of an Extended Projection Neural Network
On Convergence Conditions of an Extended Projection Neural Network
The output trajectory convergence of an extended projection neural network was developed under the positive definiteness condition of the Jacobian matrix of nonlinear mapping. This...
A Neural Model of Olfactory Sensory Memory in the Honeybee's Antennal Lobe
A Neural Model of Olfactory Sensory Memory in the Honeybee's Antennal Lobe
We present a neural model for olfactory sensory memory in the honeybee's antennal lobe. To investigate the neural mechanisms underlying odor discrimination and memorization, we exp...

Recent Results

The Sources and Themes of The Plough and the Stars
The Sources and Themes of The Plough and the Stars
THOUGH SEAN O'CASEY did not fight in the Easter Rising of 1916, he helped to organise the Irish Citizen Army and was a shrewd and passionate observer of life in Dublin before, duri...
Portrait of the People of Silla According to the Persian Texts
Portrait of the People of Silla According to the Persian Texts
Summary Persian and Arabo-Persian texts frequently referred to Silla as a highlighted toponym. Obviously, descriptions used for Silla in Persian texts cannot be comp...
Multi-Modal, Non-Invasive Investigation of Modern Colorants on Three Early Modern Prints by Maria Sibylla Merian
Multi-Modal, Non-Invasive Investigation of Modern Colorants on Three Early Modern Prints by Maria Sibylla Merian
Northwestern University’s Charles Deering McCormick Library of Special Collections owns three hand-colored copperplate engravings that once belonged to an edition of Matamorphosis ...
The effect of retirement and age at retirement on self-perceived health after three years of follow-up in Dutch 55–64-year-olds
The effect of retirement and age at retirement on self-perceived health after three years of follow-up in Dutch 55–64-year-olds
ABSTRACTHealth consequences of retirement have not been included in the current public debate about increasing the age at retirement, which might be due to the fact that studies ai...

Back to Top