Electric Arc Furnaces (EAFs) are widely used in the steel manufacturing industry to melt scrap steel by employing a large number of electric arcs. EAFs play an important role in ensuring the efficient production of steel. However, their nonlinear and variable load characteristics have a significant impact on power quality. Because the active power of an electric arc depends on its length, a system for controlling the electrode positions is necessary. This paper presents a control system based on a fuzzy logic controller for the active power control of an electric arc furnace. Individual simulation scenarios were chosen with both reference values and the process taken into consideration. The reference, constant value, step variation, and the sequence of step variation were investigated, as well as step disturbances and the sequence of step disturbances from the viewpoint of the process. Furthermore, the procedure of changing the tap on a transformer was investigated. The proposed solution minimizes the time required for charge elaboration, but the main benefit is that there are no additional costs in the implementation process because the installation remains identical, with the only changes being improvements to soft control management.

This paper presents an analysis regarding the modeling of the behavior for a three-phase electric arc furnace installation. Therefore, a block diagram is implemented in Simulink that represents the modeling of the entire electric arc furnace installation. This block diagram contains also the modeling of the electric arc which is the element that makes the electric arc furnace behaving as a nonlinear load. The values for the model parameters of the electric arc furnace installation are like the ones from the real installation taken into consideration. Other model parameters are the electric arc model ones. In order to study the influence of the parameters of the electric arc models, it is developed a Matlab program that contains the graphical user interfaces. These interfaces make connection with the models of the electric arc implemented in Simulink. The interfaces allow the user to modify parameters for each of the electric arc model. Current and voltage of the electric arc are the variables taken into account to study the influence of the parameters on the electric arc models. Waveforms for voltage and current of the electric arc are illustrated when a parameter of the model is modified in order to analyze the importance of this parameter on the electric arc model. Also, for each of the models is presented the voltage-current characteristic of the electric arc because this characteristic gives information about the behavior of the electric arc furnace installation.

Accurate modeling and short-term forecasting of electric arc furnace (EAF) voltage are important for power quality analysis, process monitoring, and the development of data-driven compensation and control strategies. The strongly nonlinear and time-varying behavior of the electric arc makes its voltage difficult to represent using simplified analytical or low-order parametric models. This study compares several modeling approaches using industrial voltage and current measurements acquired on the secondary side of a three-phase furnace transformer. The investigated methods include a hybrid Fourier–polynomial model optimized by a genetic algorithm, a CNN–GRU model, LSTM and BiLSTM networks, and a Dilated CNN–BiLSTM architecture. The models are first evaluated for same-interval voltage estimation over a common measurement interval using RMSE, MAE, R², and a current-interval V–I RMSE. A separate direct multi-horizon forecasting experiment is then performed using a 40 ms historical voltage–current window and prediction horizons of 1, 2, 5, 10, 20, 40, and 60 ms. Persistence and one-period predictors are used as reference baselines. For same-interval estimation, the Dilated CNN–BiLSTM achieves the lowest mean errors across three independent runs, with an RMSE of 5.33 ± 0.59 V, an MAE of 4.13 ± 0.44 V, an R² of 0.99986 ± 0.00003, and a current-interval V–I RMSE of 1.70 ± 0.41 V. The BiLSTM obtains the next-lowest mean RMSE, 9.26 ± 4.25 V. In the forecasting task, no single neural architecture achieves the lowest RMSE at all prediction horizons. The BiLSTM provides the lowest neural model RMSE at 5 and 20 ms, the LSTM at 2, 10, and 60 ms, while the BiLSTM and Dilated CNN–BiLSTM yield nearly identical RMSE values at 1 ms. At 40 ms, the one-period baseline produces a lower RMSE than the neural models. The results show that forecasting performance is strongly horizon-dependent and that increased architectural complexity does not necessarily guarantee lower forecasting error. This study also demonstrates the importance of distinguishing same-interval voltage estimation from true future forecasting when evaluating EAF voltage models.

In this paper is presented a mathematical model of the electric arc for an alternative current welding machine of low power. The electric arc model is based on dividing the voltage-current characteristic of the electric arc in many functioning zones. For the model of the entire welding machine are used real parameters as the ones of the proper welding machine. The voltage and current harmonics spectrum that is obtained during the welding process is presented. Also, the waveforms for the current and voltage of the electric arc plotted against time and the voltage-current characteristic of the electric arc are illustrated. The electric arc is considered as being supplied by alternative voltage from the electrical power network using a single phase transformer which has the output voltage of 80 volts. The model of the welding machine is developed in Simulink and the variations of some parameters of the electric arc are obtained by modifying of them in a Matlab function. Also, in this paper is presented the total harmonic distortion for the voltage and current of the electric arc obtained during simulation of the welding machine.