Regression modeling method of space weather prediction

Parnowski, A. S.

doi:10.1007/s10509-009-0060-4

Cited by 8 publications

(9 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Parnowski, (, ) has used a simple linear regression approach, that yielded a prediction efficiency as high as PE ∼0.975 for 1‐hr‐ahead forecast and PE ∼0.9 for 3 hr ahead. They used a statistical method based on the Fisher statistical parameter to calculate the significance of candidate regressors (Fisher, ).…”

Section: Review Of Machine Learning In Space Weathermentioning

confidence: 99%

The Challenge of Machine Learning in Space Weather: Nowcasting and Forecasting

2019

View full text Add to dashboard Cite

The numerous recent breakthroughs in machine learning make imperative to carefully ponder how the scientific community can benefit from a technology that, although not necessarily new, is today living its golden age. This Grand Challenge review paper is focused on the present and future role of machine learning in Space Weather. The purpose is twofold. On one hand, we will discuss previous works that use machine learning for Space Weather forecasting, focusing in particular on the few areas that have seen most activity: the forecasting of geomagnetic indices, of relativistic electrons at geosynchronous orbits, of solar flares occurrence, of coronal mass ejection propagation time, and of solar wind speed. On the other hand, this paper serves as a gentle introduction to the field of machine learning tailored to the Space Weather community and as a pointer to a number of open challenges that we believe the community should undertake in the next decade. The recurring themes throughout the review are the need to shift our forecasting paradigm to a probabilistic approach focused on the reliable assessment of uncertainties, and the combination of physics-based and machine learning approaches, known as gray box. Plain Language SummaryIn the last decade, machine learning has achieved unforeseen results in industrial applications. In particular, the combination of massive data sets and computing with specialized processors (graphics processing units, or GPUs) can perform as well or better than humans in tasks like image classification and game playing. Space weather is a discipline that lives between academia and industry, given the relevant physical effects on satellites and power grids in a variety of applications, and the field therefore stands to benefit from the advances made in industrial applications. Today, machine learning poses both a challenge and an opportunity for the space weather community. The challenge is that the current data science revolution has not been fully embraced, possibly because space physicists remain skeptical of the gains achievable with machine learning. If the community can master the relevant technical skills, they should be able to appreciate what is possible within a few years time and what is possible within a decade. The clearest opportunity lies in creating space weather forecasting models that can respond in real time and that are built on both physics predictions and on observed data.

show abstract

Section: Review Of Machine Learning In Space Weathermentioning

confidence: 99%

The Challenge of Machine Learning in Space Weather: Nowcasting and Forecasting

2019

View full text Add to dashboard Cite

show abstract

“…Our predictions where also compared with results of other authors, [10,11,25,31], and [26]. Our method provides a much more precise guaranteed forecast than most empirical and typical neural network models.…”

Section: Resultsmentioning

confidence: 72%

“…So, for its prediction it is necessary to solve the problem of model structure and parameters identification. There are many different models including a neural network, a linear regression model, and others [4,7,[9][10][11][12][13] for Dst-index prediction. The most accurate and reliable is a nonlinear discrete dynamic model that describes a relation between spacecraft observation data and ground measurements from magnetic observatories [14].…”

Section: Introductionmentioning

confidence: 99%

“…It would be shown that the identification problem can be reduced to the solution of the corresponding mathematical programming problem with constraints. Such an approach substantially differs from the widely used statistical methods [10,11,25,26] where the great number of input variables is exploited. This paper focuses on a novel approach to reconstruction of the guaranteed prediction interval, the so called "prediction tube".…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The combined approach for space weather prediction with a guaranteed method and evolutionary algorithm

Semeniv¹

2015

J. Phys. Stud.

View full text Add to dashboard Cite

The problem of space weather prediction using guaranteed approach and solar wind parameters is considered. The original approach for space weather prediction based on a discrete dynamic model, evolutionary algorithm and a guaranteed interval estimation method is proposed. The problem of discrete model identification is formulated as a nonlinear programming problem with constraints. A new algorithm of Dst-index prediction using guaranteed interval estimation method is described. The model for the prediction of geomagnetic storms was constructed and the interaction between the solar wind parameters and the magnetosphere was discussed. The dependence between prediction accuracy and the forecasting horizon is analyzed.

show abstract

“…The latest, third group uses various black box‐type statistical models, relating the solar wind and index: artificial neural networks, nonlinear auto‐regression schemes, etc. [ Valdivia et al , 1996; Vassiliadis et al , 1999; Lundstedt et al , 2002; Temerin and Li , 2002, 2006; Wei et al , 2004; Pallocchia et al , 2006; Sharifie et al , 2006; Zhu et al , 2006, 2007; Amata et al , 2008; Parnowski , 2009; Boynton et al, 2012]. These models with rather complex structure are capable to reveal information about the process without any prior assumptions.…”

Section: Introductionmentioning

confidence: 99%

Extended geomagnetic storm forecast ahead of available solar wind measurements

Podladchikova¹,

Petrukovich²

2012

Space Weather

View full text Add to dashboard Cite

We develop a technique to predict geomagnetic storm magnitudes (peak Dst values) several hours in advance using the first indications of extreme solar wind conditions, as well as an assumption of constant driving function V Bs. For larger storms with clear V Bs jump we predict lower and upper limits of expected peak Dst. For smaller storms and storms with gradually increasing V Bs we predict Dst several hours ahead. The data from 1995–2010 were used to design the technique. The actual advance time of reliable forecasts (before the peak value is reached) is on average 5–6 hours, illustrating the realistic horizon of such “constant input” assumption. Larger storms are developing faster and thus are better predicted. False warnings (predictions more than 25% larger than the actual peak) occur in about 10% of events. The algorithm is implemented in real‐time in Space Research Institute, Moscow (www.spaceweather.ru).

show abstract

Regression modeling method of space weather prediction

Cited by 8 publications

References 24 publications

The Challenge of Machine Learning in Space Weather: Nowcasting and Forecasting

The Challenge of Machine Learning in Space Weather: Nowcasting and Forecasting

The combined approach for space weather prediction with a guaranteed method and evolutionary algorithm

Extended geomagnetic storm forecast ahead of available solar wind measurements

Contact Info

Product

Resources

About