Extracting the main patterns of natural time series for long-term neurofuzzy prediction

Gholipour, Ali; Lucas, Caro; Araabi, Babak Nadjar; Mirmomeni, Masoud; Shafiee, Masoud

doi:10.1007/s00521-006-0062-x

Cited by 43 publications

(27 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The method has also shown its effectiveness in the long term prediction of the sunspot number time series (Gholipour et al, 2005). The use of singular spectral analysis results in the identification and separation of the main components with different periodicities.…”

Section: Resultsmentioning

confidence: 99%

“…In fact, the utilization of SSA indirectly amounts to using external information about the behavior of the time series in the previous cycles as well. Therein lies the secret of the success in longer term prediction (Loskutov et al, 2001a, b;Gholipour et al, 2005). We seem to be doing the impossible by predicting multi-steps ahead by using only a few hours' past data; but we are, in fact, indirectly using other data on periodicities of the time series obtained through consideration of much longer term past values that could not have been directly used in prediction tools.…”

Section: Resultsmentioning

confidence: 99%

“…Powerful black box neural and neurofuzzy methods fail when the prediction the time horizon increases and the number of training data decreases, especially when training set contains some few number of storms and contains minor storms but test set contains some extreme storms. In this article a novel combination of nonlinear and linear techniques that has been successfully utilized for long term prediction of space weather (Gholipour et al, 2005) has been adopted for multi-step forecasting Dst values. The paper is organized as follows: In Section 2, singular spectral analysis (SSA) method is introduced and in Section 3, locally linear neurofuzzy (LLNF) model is introduced for Black-Box modeling.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multi-step prediction of Dst index using singular spectrum analysis and locally linear neurofuzzy modeling

2006

Self Cite

View full text Add to dashboard Cite

Even one-step prediction of natural time series without delay especially in main phase of storm is difficult for many complicated time series such as Dst index. In this study, with a new method based on singular spectrum analysis, we extract the main components of the time series, model each component with a locally linear neurofuzzy network, and utilize the trained networks for multi-step ahead prediction of a validation set of data, and finally combine the predicted patterns for construction of general prediction. Our methods are compared with several previous studies for Dst index prediction. Several solar geomagnetic extreme events are predicted well with our state-of-the-art method; such as extreme events in 14 March 1989 that led to power black-out in Quebec, as well as other extreme storms.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multi-step prediction of Dst index using singular spectrum analysis and locally linear neurofuzzy modeling

2006

Self Cite

View full text Add to dashboard Cite

show abstract

“…Consequently, modern space weather forecasters rely on a great variety of forecast systems, ranging from simple nonlinear models to complex informationbased (empirical approaches) physical, and hybrid models (Bothmer and Daglis, 2007). Over the last two decades, which was when real-time data began to be available online, data-driven approaches, such as artificial neural networks (Bothmer and Daglis, 2007), neurofuzzy modeling (Gholipour et al, 2005(Gholipour et al, , 2007, Kalman filtering (Bothmer and Daglis, 2007), among others, have been shown to perform well in space weather forecasting. Space weather forecasting can be classified into five classes according to the forecasting frame-time:…”

Section: Problem Statementmentioning

confidence: 99%

Long-term prediction of solar and geomagnetic activity daily time series using singular spectrum analysis and fuzzy descriptor models

et al. 2009

Self Cite

View full text Add to dashboard Cite

Of the various conditions that affect space weather, Sun-driven phenomena are the most dominant. Cyclic solar activity has a significant effect on the Earth, its climate, satellites, and space missions. In recent years, space weather hazards have become a major area of investigation, especially due to the advent of satellite technology. As such, the design of reliable alerting and warning systems is of utmost importance, and international collaboration is needed to develop accurate short-term and long-term prediction methodologies. Several methods have been proposed and implemented for the prediction of solar and geomagnetic activity indices, but problems in predicting the exact time and magnitude of such catastrophic events still remain. There are, however, descriptor systems that describe a wider class of systems, including physical models and non-dynamic constraints. It is well known that the descriptor system is much tighter than the state-space expression for representing real independent parametric perturbations. In addition, the fuzzy descriptor models as a generalization of the locally linear neurofuzzy models are general forms that can be trained by constructive intuitive learning algorithms. Here, we propose a combined model based on fuzzy descriptor models and singular spectrum analysis (SSA) (FD/SSA) to forecast a number of geomagnetic activity indices in a manner that optimizes a fuzzy descriptor model for each of the principal components obtained from singular spectrum analysis and recombines the predicted values so as to transform the geomagnetic activity time series into natural chaotic phenomena. The method has been applied to predict two solar and geomagnetic activity indices: geomagnetic aa and solar wind speed (SWS) of the solar wind index. The results demonstrate the higher power of the proposed method-compared to other methods-for predicting solar activity.

show abstract

“…It is shown that the cyclic solar activity has chaotic characteristics especially during storm time which depicts the difficulties in long-term prediction of solar activity indices (Stefanski, 2003;Gholipour et al, 2007;Mirmomeni et al, 2007). It has to be said that, deterministic chaos appears in different fields of science like physics, biomedicine, and engineering (Ruelle, 1978;Kocarev et al, 2006).…”

Section: Introductionmentioning

confidence: 99%

Analyzing the variation of embedding dimension of solar and geomagnetic activity indices during geomagnetic storm time

Mirmomeni

Lucas

2009

Earth Planet Sp

View full text Add to dashboard Cite

Cyclic solar activity as one of the natural chaotic phenomena has significant effects on Earth, climate, and satellites. Rapid changes in the near-Earth space environment can affect the performance and reliability of both spacecraft and ground-based systems. This can imply major problems due to communication and satellite operational anomalies. Therefore, it is meaningful to analyze solar activity and geomagnetic indices to elicit the behavior of sun as the origin of most of these chaotic phenomena. One of the most important tools for analyzing the chaotic trends is the "Embedding Dimension" (ED). In this paper, the variation of ED for solar activity indices especially during storm time for two well-known storms is considered. The first storm is the super-storm on 13 March 1989, which shuts down the power supply system in Québec, Canada and the second one is the storm caused by Coronal Mass Ejection on 11 January 1997 which causes the failure of Telstar 401 satellite. The method of this paper is based on the fact that the reconstructed dynamics of an attractor should be a smooth map, i.e. with no self intersection in the reconstructed attractor. It is shown that the Embedding Dimension (and other chaotic characteristics) of some solar and geomagnetic activity indices during these storms varies rapidly.

show abstract

Extracting the main patterns of natural time series for long-term neurofuzzy prediction

Cited by 43 publications

References 23 publications

Multi-step prediction of Dst index using singular spectrum analysis and locally linear neurofuzzy modeling

Multi-step prediction of Dst index using singular spectrum analysis and locally linear neurofuzzy modeling

Long-term prediction of solar and geomagnetic activity daily time series using singular spectrum analysis and fuzzy descriptor models

Analyzing the variation of embedding dimension of solar and geomagnetic activity indices during geomagnetic storm time

Contact Info

Product

Resources

About