IDENTIFICATION OF THE INFLUENCE OF THE SPATIAL ORIENTATION OF THE DEPOSITED LAYERS, AS WELL AS THEIR OVERLAP COEFFICIENT ON THE SURFACE SHAPE ERROR DURING ADDITIVE SHAPING BY AN ELECTRIC ARC

Статья посвящена изучению вопросов управления процессом аддитивного формообразования изделий. Представлены результаты исследования процесса аддитивного формообразования поверхности электрической дугой в среде защитного газа. Проведен анализ погрешности формы поверхностей, полученных с различным заполнением слоев. Подтверждено экспериментально, что такие параметры процесса, как ориентация слоев, коэффициент их перекрытия являются значимыми. Так, погрешность формообразования образцов наплавки слой на слой в вертикальном направлении выше по сравнению с другими способами наплавки, реализованными в эксперименте. Средние значения погрешности формы образцов составляют 0,75 мм, 0,88 мм, 1,15 мм, соответственно, для способов наплавки слой к слою на горизонтальную поверхность с коэффициентом перекрытия 0,3, слой к слою на горизонтальную поверхность с коэффициентом перекрытия 0,5, слой на слой в вертикальном направлении. Максимальные значения погрешности определены на уровне 0,85 мм, 1,2 мм, 1,5 мм для соответствующих способов наплавки, реализованных в эксперименте. Таким образом, пространственная ориентация слоев, а также коэффициент перекрытия слоев являются значимыми, оказывают влияние на численное значение погрешности формы получаемой поверхности, должны быть учтены при проектировании алгоритмов разделения на слои, их заполнения при аддитивном формообразовании электрической дугой в среде защитного газа The article is devoted to the study of the issues of managing the process of additive shaping of products. The paper presents the results of a study of the process of additive surface shaping by an electric arc in a protective gas medium. We analyzed the error of the shape of the surfaces obtained with different filling layers. We confirmed experimentally that such process parameters as the orientation of the layers, their overlap coefficient are significant. Thus, the error of forming samples of surfacing layer on layer in the vertical direction is higher compared to other methods of surfacing implemented in the experiment. The average values of the sample shape error are 0.75 mm, 0.88 mm, 1.15 mm, respectively, for methods of surfacing layer to layer on a horizontal surface with an overlap coefficient of 0.3, layer to layer on a horizontal surface with an overlap coefficient of 0.5, layer to layer in the vertical direction. The maximum error values are determined at the level of 0.85 mm, 1.2 mm, 1.5 mm for the corresponding surfacing methods implemented in the experiment. Thus, the spatial orientation of the layers, as well as the overlap coefficient of the layers, are significant, affect the numerical value of the shape error of the resulting surface, should be taken into account when designing algorithms for dividing into layers, filling them with additive shaping by an electric arc in a protective gas medium

Physico-Chemical Characterization of the Electric Arc Furnace Slag (EAFS) of the Sonasid-Jorf Steelworks - Morocco

The steelmaking process results in the by-product formation of electric arc furnace slag (EAFS). Slag is recovered at two different stages of the steelmaking process, the first recovery is black and the second is white. The present research focuses on the composition differences between the two types of slag from SONASID-Jorf steel in Morocco. A granular separation of the black and white slag was carried out to monitor the chemical and mineralogical composition. XRD and Fourier Transform Infrared Spectroscopy are performed on the samples in this paper. The slags suggest good hydraulic binder properties. It would be useful for research in the field of building materials to correlate the results of the characterization of EAFS with other types of slag with the aim of improving the potential for partial replacement of cement in the matrix. The slag can also be used as binders in mixtures of bio-based building materials. The electric arc furnace slag (EAFS), exhibiting appropriate cementitious activity, can be utilized as mineral admixture in cement and concrete. Black and white slags are studied in this paper in order to determine their characteristics according to their granularity.

Deep Learning Based Predictive Compensation of Flicker, Voltage Dips, Harmonics and Interharmonics in Electric Arc Furnaces

<div> <div> <div> <p>In this research work, deep machine learning based methods together with a novel data augmentation are developed for predicting flicker, voltage dip, harmonics and interharmonics originating from highly time-varying electric arc furnace (EAF) currents and voltage. The aim with the prediction is to counteract both the response and reaction time delays of active power filters (APFs) specifically designed for electric arc furnaces (EAF). Multiple synchronous Reference frame (MSRF) analysis is used to decompose the frequency components of the EAF current and voltage waveforms into dqo components. Then using low- pass filters and prediction of the future values of these dqo components, reference signals for APFs are generated. Three different methods have been developed. In two of them, a low- pass Butterworth filter is used together with a linear FIR based prediction or long short-term memory network (LSTM) for prediction. In the third method, a deep convolutional neural network (CNN) combined with a LSTM network is used to filter and predict at the same time. For a 40 ms prediction horizon, the proposed methods provide 2.06%, 0.31%, 0.99% prediction errors of the dqo components for the Butterworth and linear prediction, Butterworth and LSTM and CNN with LSTM, respectively. The error of the predicted reconstructed waveforms of flicker, harmonics, and interharmonics resulted in 8.5%, 1.90%, and 3.2% reconstruction errors for the above-mentioned methods. Finally, a Simulink and GPU based implementation of predictive APF using Butterworth filter + LSTM and a trivial APF resulted 96% and 60% efficiency on compensation of EAF current interharmonics. </p> </div> </div> </div>

Deep Learning Based Predictive Compensation of Flicker, Voltage Dips, Harmonics and Interharmonics in Electric Arc Furnaces

<div> <div> <div> <p>In this research work, deep machine learning based methods together with a novel data augmentation are developed for predicting flicker, voltage dip, harmonics and interharmonics originating from highly time-varying electric arc furnace (EAF) currents and voltage. The aim with the prediction is to counteract both the response and reaction time delays of active power filters (APFs) specifically designed for electric arc furnaces (EAF). Multiple synchronous Reference frame (MSRF) analysis is used to decompose the frequency components of the EAF current and voltage waveforms into dqo components. Then using low- pass filters and prediction of the future values of these dqo components, reference signals for APFs are generated. Three different methods have been developed. In two of them, a low- pass Butterworth filter is used together with a linear FIR based prediction or long short-term memory network (LSTM) for prediction. In the third method, a deep convolutional neural network (CNN) combined with a LSTM network is used to filter and predict at the same time. For a 40 ms prediction horizon, the proposed methods provide 2.06%, 0.31%, 0.99% prediction errors of the dqo components for the Butterworth and linear prediction, Butterworth and LSTM and CNN with LSTM, respectively. The error of the predicted reconstructed waveforms of flicker, harmonics, and interharmonics resulted in 8.5%, 1.90%, and 3.2% reconstruction errors for the above-mentioned methods. Finally, a Simulink and GPU based implementation of predictive APF using Butterworth filter + LSTM and a trivial APF resulted 96% and 60% efficiency on compensation of EAF current interharmonics. </p> </div> </div> </div>