The NNSYSID toolbox-a MATLAB(R) toolbox for system identification with neural networks

Nørgaard, Magnus; Ravn, Ole; Hansen, Lars Kai; Poulsen, Niels Kjølstad

doi:10.1109/cacsd.1996.555321

Cited by 25 publications

(24 citation statements)

References 11 publications

(5 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To verify whether the application of the FIS gives better results than traditional regression models, the results of the FNN model were compared to the results of an ARMA and an ARMAX model. The ARMA and ARMAX models are built on the NNARMAX2 algorithm of the NNSYSID Toolbox (Nørgaard et al, 2002). The regressors of the ARMA model are five past discharge values and two past residuals (ARMA(5,2)), as this is a common choice in hydrological analysis.…”

Section: Resultsmentioning

confidence: 99%

Long-range forecasts of River Po discharges based on predictable solar activity and a fuzzy neural network model / Prévisions à long terme des débits du Fleuve Pô basées sur l’activité solaire prévisible et sur un modèle de réseau de neurones flou

Tomasino

Zanchettin

Traverso

2004

Hydrological Sciences Journal

View full text Add to dashboard Cite

Is it possible to make seasonal and interannual forecasts of hydrological variables if one cannot predict next week's rainfall? Contrary to common view, some scientists support the hypothesis that variations in mean global temperature and precipitation are controlled more by external forcing (solar variability and volcanic eruptions) than by increasing atmospheric concentration of greenhouse gases. Temperature and precipitation are connected with special phases of the 11-year sunspot cycle, which coincide with significant accumulation of energetic solar eruptions. Because of the possibility of identifying years with many solar eruptions, the attractive prospect emerges of the long-term hydrological forecasting based on cycles of solar activity. Starting from this assumption, an expert system was built based on a fuzzy neural network model for seasonal and interannual forecasting of the Po River discharge. It was found that indices of solar activity and of global circulation are sufficient to yield useful forecasts of hydrological variables.Key words river discharge, fuzzy logic; artificial neural networks; seasonal forecasting; solar activity; Po River; Italy Prévisions à long terme des débits du Fleuve Pô basées sur l'activité solaire prévisible et sur un modèle de réseau de neurones flou Résumé Est-il possible de réaliser des prévisions saisonnières et interannuelles de variables hydrologiques, alors qu'on ne peut pas prévoir la pluie de la semaine prochaine? Contrairement à l'opinion courante, certains scientifiques considèrent que les variations de température globale moyenne et de précipitations sont plus gouvernées par des forçages externes (variabilité solaire et éruptions volcaniques) que par l'augmentation de la concentration atmosphérique des gaz à effet de serre. La température et les précipitations sont en relation avec les phases spéciales du cycle solaire de 11 ans, qui correspondent à une augmentation significative des éruptions énergétiques solaires. Dans la mesure où il est possible d'identifier les années à éruptions solaires nombreuses, on peut ouvrir la perspective de baser des prévisions hydrologiques à long terme sur les cycles d'activité solaire. Sur la base de cette hypothèse, un système expert a été conçu à partir d'un modèle de réseau de neurones flou pour réaliser des prévisions saisonnières et interannuelles du débit du Fleuve Pô. Il apparaît que des indices d'activité solaire et de circulation globale suffisent pour produire des prévisions utiles de variables hydrologiques.Mots clefs débit de rivière; logique floue; réseaux de neurones artificiels; prévision saisonnière; activité solaire; Fleuve Pô; Italie

show abstract

Section: Resultsmentioning

confidence: 99%

Long-range forecasts of River Po discharges based on predictable solar activity and a fuzzy neural network model / Prévisions à long terme des débits du Fleuve Pô basées sur l’activité solaire prévisible et sur un modèle de réseau de neurones flou

Tomasino

Zanchettin

Traverso

2004

Hydrological Sciences Journal

View full text Add to dashboard Cite

show abstract

“…By some kind of iterative minimization scheme: A improved Levenberg-Marquardt method is introduced for minimization of mean-square error criteria, due to its rapid convergence properties and robustness, the details is in literature [10] .…”

Section: Training Algorithmmentioning

confidence: 99%

A real-time fault diagnosis method of SACS based on combination of offline identification and online observing

et al. 2009

View full text Add to dashboard Cite

A novel fault diagnosis method based on the combination of offline identification and online observing is proposed in this paper, which can meet the requirement of both model complexity and real-time need for the satellite attitude control system. Accurate neural network models, both normal mode and faulty mode, can be obtained by off-line identification based on the data of fault simulation in different fault modes. With a parallel estimator derived from all models With all, fault determination based on threshold logic is designed for online fault detecting and isolating. Real-time simulation results, on the embedded fault simulation platform of satellite attitude control system, illustrate the effectiveness and superiority.

show abstract

“…On the other hand, an excessive number of neurons leads to overtraining problems and its computational cost is higher. The training and pruning of the neuronal network was carried out using the toolbox named ''Neural Based Network Identification System'' developed by Helsinki Technical University (Norgaard et al, 2000(Norgaard et al, , 2002. Fig.…”

Section: Model Selectionmentioning

confidence: 99%

Simulation of a coke wastewater nitrification process using a feed-forward neuronal net

Machón-González

Garcı́a

Rodríguez-Iglesias

et al. 2007

Environmental Modelling & Software

View full text Add to dashboard Cite

A laboratory-scale Activated Sludge System (ASS) was employed for the biodegradation of coke wastewater, which contains high concentrations of ammonium, thiocyanate, phenols and other organic compounds. The well-known kinetics models of Monod or Haldane are not very useful due to inhibition phenomena amongst the pollutants and also, they need the determination of a wide range of parameters to be introduced in the models. In this paper, a feed-forward neural network is outlined to obtain a satisfactory approach for estimating the effluent ammonium concentration of the treatment plant. The methodology consists in performing a group of different sizes of the hidden layer and different subsets of input variables.The developed model is useful to obtain simulations under different conditions of the influent stream, thus enabling the effluent ammonium concentration to be estimated. This neural network achieves better results than classical mathematical models for biological wastewater treatment as a result of the complex composition of the coke wastewater.

show abstract

The NNSYSID toolbox-a MATLAB(R) toolbox for system identification with neural networks

Cited by 25 publications

References 11 publications

Long-range forecasts of River Po discharges based on predictable solar activity and a fuzzy neural network model / Prévisions à long terme des débits du Fleuve Pô basées sur l’activité solaire prévisible et sur un modèle de réseau de neurones flou

Long-range forecasts of River Po discharges based on predictable solar activity and a fuzzy neural network model / Prévisions à long terme des débits du Fleuve Pô basées sur l’activité solaire prévisible et sur un modèle de réseau de neurones flou

A real-time fault diagnosis method of SACS based on combination of offline identification and online observing

Simulation of a coke wastewater nitrification process using a feed-forward neuronal net

Contact Info

Product

Resources

About