Application of radial basis function networks to model electric arc furnaces

Sadeghian, Alireza; Lavers, J.D.

doi:10.1109/ijcnn.1999.830798

Cited by 28 publications

(36 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The load characteristics can be reflected by the signal energy in different sub-bands. Then, the radial basis function neural network (RBFNN) [10,11] is used to design a smart method to detect the serial arc faults. The line current waveforms of the system under normal operation or a serial arc fault are used to train the RBFNN.…”

Section: Introductionmentioning

confidence: 99%

Smart detection technology of serial arc fault on low-voltage indoor power lines

Liu

2015

International Journal of Electrical Power & Energy Systems

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Smart detection technology of serial arc fault on low-voltage indoor power lines

Liu

2015

International Journal of Electrical Power & Energy Systems

View full text Add to dashboard Cite

“…26,27 Predicting the EAF's reactive power by creating a neural network to enhance compensator performance is proposed in the literature. 28,29 Several papers use fuzzy logic 30 or adaptive fuzzy 31,32 in developing their models. A hybrid wavelet transform and a neural-network-based approach is used in Chang et al 33 to present a dynamic model for EAF.…”

Section: Introductionmentioning

confidence: 99%

Polynomial-dynamic electric arc furnace model combined with ANN

Golestani

Samet

2018

Int Trans Electr Energ Syst

View full text Add to dashboard Cite

Summary Two approaches exist for modeling electric arc furnace, the largest electrical load. Distinct mathematical equations are used as the first approach. However, the process is nonstationary and the equation is limited to 1 form and unable to extract all nonlinear components. Black‐box models such as artificial neural networks are used as the second approach. However, using a black‐box model for a process with partially known equations is not efficient, as it cannot benefit from known knowledge of the process. This paper proposes a combination of the 2 mentioned approaches. The first phase consists of nonlinear and dynamic terms. The appropriate model orders relate to dynamic and nonlinear terms are attained. In the second phase, artificial neural network is used to model the rest of the nonlinear components, which are not captured in the first phase. Numerous actual records are used to compare the proposed model with several electric arc furnace models.

show abstract

“…In , EAF is modeled using the chaotic dynamics. Radial basis function (RBF) networks are used in to model EAF v–i characteristic, and the results are compared with multi‐layer perceptron networks. In , the application of adaptive fuzzy logic systems (FLS) to model EAFs is presented, and the results are compared with FLS and RBF.…”

Section: Introductionmentioning

confidence: 99%

A dynamic, nonlinear and time-varying model for electric arc furnace

Samet

Farjah

Sharifi

2014

Int. Trans. Electr. Energ. Syst.

View full text Add to dashboard Cite

Summary Electric arc furnace (EAF) is a nonlinear and time‐varying load that causes power quality problems such as voltage flicker and harmonics. Field measurement of voltage and current from EAFs in Mobarakeh Steel Company in Isfahan‐Iran are used to introduce a comprehensive model for EAF. A new dynamic, nonlinear and time‐varying model is presented here. In other words, the function between EAF voltage and current is nonlinear, differential and with time‐varying parameters. By using the recorded data, the order of the dynamic part is determined by a new approach. The nonlinear characteristic is modeled with proper polynomial functions. All parameters related to the dynamic and nonlinear functions are considered time varying as time series. Autoregressive moving average (ARMA) models are implemented and the suitable ARMA order is attained for all EAF time‐varying parameters. Copyright © 2014 John Wiley & Sons, Ltd.

show abstract

Application of radial basis function networks to model electric arc furnaces

Cited by 28 publications

References 17 publications

Smart detection technology of serial arc fault on low-voltage indoor power lines

Smart detection technology of serial arc fault on low-voltage indoor power lines

Polynomial-dynamic electric arc furnace model combined with ANN

A dynamic, nonlinear and time-varying model for electric arc furnace

Contact Info

Product

Resources

About