Neural Networks for Modelling and Control of Dynamic Systems

Nørgaard, Magnus; Ole, Ravn; Poulsen, Niels Kjølstad; Hansen, Leif

doi:10.1007/978-1-4471-0453-7

Cited by 769 publications

(262 citation statements)

References 0 publications

Supporting

Mentioning

244

Contrasting

Unclassified

Order By: Relevance

“…Therefore, also coefficient D has to be optimized by trial and error; we let it vary between 0.0001 and 1. The extension of the Levenberg-Marquardt algorithm to case of regularized training function is given in Norgaard et al (2000). In order to prevent overfitting as much as possible, we use at the same time both regularization and early stopping for training FFNNs.…”

Section: Ffnns Trainingmentioning

confidence: 99%

See 1 more Smart Citation

Air quality prediction in Milan: feed-forward neural networks, pruned neural networks and lazy learning

Corani

2005

Ecological Modelling

218

View full text Add to dashboard Cite

Ozone and PM 10 constitute the major concern for air quality of Milan. This paper addresses the problem of the prediction of such two pollutants, using to this end several statistical approaches. In particular, feed-forward neural networks (FFNNs), currently recognized as state-of-the-art approach for statistical prediction of air quality, are compared with two alternative approaches derived from machine learning: pruned neural networks (PNNs) and lazy learning (LL). PNNs constitute a parameterparsimonious approach, based on the removal of redundant parameters from fully connected neural networks; LL, on the other hand, is a local linear prediction algorithm, which performs a local learning procedure each time a prediction is required. All the three approaches are tested in the prediction of ozone and PM 10 ; predictors are trained to return at 9 a.m. the concentration estimated for the current day.No strong differences are found between the forecast accuracies of the different models; nevertheless, LL provides the best performances on indicators related to average goodness of the prediction (correlation, mean absolute error, etc.), while PNNs are superior to the other approaches in detecting of the exceedances of alarm and attention thresholds. In some cases, datadeseasonalization is found to improve the prediction accuracy of the models.Finally, some striking features of lazy learning deserve consideration: the LL predictor can be quickly designed, and, thanks to the simplicity of the local linear regressors, it both gets rid of overfitting problems and can be readily interpreted; moreover, it can be also easily kept up-to-date.

show abstract

Section: Ffnns Trainingmentioning

confidence: 99%

“…For the sake of simplicity it is described in the case of a square error training function. However, the description of the OBS algorithm for a regularized error function can be found in Norgaard et al (2000).…”

Section: Pruned Neural Network (Pnns)mentioning

confidence: 99%

Air quality prediction in Milan: feed-forward neural networks, pruned neural networks and lazy learning

Corani

2005

Ecological Modelling

218

View full text Add to dashboard Cite

show abstract

“…The application of feedforward networks to dynamic systems modelling requires the use of external delay lines involving both input and output signals (Norgaard et al, 2000).…”

Section: Fig 1 Feedforward Neural Network Architecturementioning

confidence: 99%

Neural and Genetic Control Approaches in Process Engineering

Cañete¹,

Saz²,

García-Cerezo³

et al. 2012

Frontiers in Advanced Control Systems

View full text Add to dashboard Cite

“…5-neurons, the second one has six tansig neurons and the third has one linear neuron. For background on multilayer perceptrons the reader is referred to [38].…”

Section: Servovalvesmentioning

confidence: 99%

A high-accuracy trajectory following controller for pneumatic devices

Carneiro

Almeida

2011

Int J Adv Manuf Technol

View full text Add to dashboard Cite

The use of pneumatic devices is widespread among different industrial fields, in tasks like handling or assembly. Pneumatic systems are low-cost, reliable, and compact solutions. However, its use is typically restricted to simple tasks due to the poor performance achieved in applications where accurate motion control is required. This paper presents a novel nonlinear controller, using neural network-based models, that allows the use of common industrial servopneumatic components in applications where fine trajectory following tasks is required. Furthermore, several experimental trials show that the system is highly robust to payload variation without any controller retuning. These results encourage the use of pneumatics actuators in a set of applications for which they have not been traditionally considered.

show abstract

Neural Networks for Modelling and Control of Dynamic Systems

Cited by 769 publications

References 0 publications

Air quality prediction in Milan: feed-forward neural networks, pruned neural networks and lazy learning

Air quality prediction in Milan: feed-forward neural networks, pruned neural networks and lazy learning

Neural and Genetic Control Approaches in Process Engineering

A high-accuracy trajectory following controller for pneumatic devices

Contact Info

Product

Resources

About