RESEARCH ARTICLE Enhancement of SVC performance in electric arc furnace for flicker suppression using a Gray‐ANN based prediction method Haidar Samet 1 | Aslan Mojallal 2 | Teymoor Ghanbari 3 | Mohammad Reza Farhadi 1 1 School of Electrical and Computer Engineering, Shiraz University, Shiraz, Iran 2 School of Electrical Engineering and Computer Science, Washington State University, Pullman, WA, USA 3 School of Advanced Technologies, Shiraz University, Shiraz, Iran Correspondence Haidar Samet, School of Electrical and Computer Engineering, Shiraz University, Shiraz, Iran. Email: samet@shirazu.ac.ir Summary Delays in reactive power measurement and thyristor ignition limit SVC performance in flicker mitigation of electric arc furnaces (EAFs). To overcome this limitation, prediction methods can be employed to forecast the EAF reactive power for half cycle ahead, used as a reference signal of the SVC. The utilized prediction methods in this area can be divided into linear and black‐box approaches. However, the linear approaches cannot extract the nonlinear governed relations, and using a black‐box model is not efficient for linear relations. A Gray‐ANN method is proposed here to take the advantages of the two mentioned approaches. Results from indices based on actual records of Mobarakeh Steel Company confirm superiorities of the proposed method over previously utilized prediction methods in this application. Furthermore, SVC's flicker mitigation ability is evaluated using the actual data. The results confirm the significant reduction of flicker compared with the regular system. KEYWORDS artificial neural network, electric arc furnace, gray system theory, reactive power compensation 1 | INTRODUCTION Electric arc furnace (EAF) is considered as one of the largest reactive power loads in the power systems globally. 1 EAF is a harmonic‐rich nonlinear load from voltage and current points of view. 2-8 Moreover, an EAF requires large, time‐varying, and nondeterministic reactive power to operate. 9-11 This issue leads to low‐frequency (around 0.5‐25 Hz) oscillation in the reactive power of the system, known as flicker. 12-15 For instance, Figure 1 shows actual reactive power data for one EAF installed in Mobarakeh Steel Company (MSC), Isfahan, Iran. Regarding these two main power quality problems of EAFs, harmonics and flicker, the solutions are divided into filtering the harmonics 16,17 and mitigation of the flicker. The main focus of the present paper is the mitigation of the flicker caused by EAF plants. Although there are several approaches presented in the literature to tackle this issue, 18 the direct compensation of EAF reactive power is the well‐ known approach. 19,20 Static VAr compensator (SVC) is the most suitable candidate for this purpose, which uses power Abbreviations: AGO, Accumulated generating operation; ANN, Artificial neural network; ARMA, Autoregressive moving average; EAF, Electric arc furnace; FMF, Flicker mitigation factor; GM, Gray model; HMF, High‐frequency mitigation factor; MFA, Modified firefly algorithm; MSC, Mobarakeh steel company; ND, Nonlinear deterministic; NLMS, normalized least mean square; PSD, Power spectral density; RLS, recursive least square; STATCOM, Static synchronous compensator; SVC, Static VAr Compensator; SVR, Support vector regression; TCR, Thyristor‐controlled reactor; TLF, Time series load flow Received: 15 March 2018 Revised: 29 October 2018 Accepted: 30 November 2018 DOI: 10.1002/etep.2811 Int Trans Electr Energ Syst. 2018;e2811. https://doi.org/10.1002/etep.2811 © 2018 John Wiley & Sons, Ltd. wileyonlinelibrary.com/journal/etep 1 of 20