A Novel Hybrid Framework for Precise Electric Energy Consumption Prediction in Steel Production via Electric Arc Furnace: Coupling Mechanistic Models with Advanced Data-Driven Algorithms

Low-carbon production of iron and steel: Technology options, economic assessment, and policy

Steel is an alloy of iron and one or more element(s), namely carbon, nickel, chromium, manganese, vanadium, molybdenum, tungsten, and so on. Chemically, steels may be classified in two groups: plain carbon steels and alloy steels. The former comprises the alloys of iron and carbon, whereas the latter contains one or more elements in addition to carbon. The alloying elements improve the mechanical, magnetic and electrical properties, as well as the corrosion resistance of steels. Impurities like Si, Mn, S, P, Al, and O are invariably present in steels due to their association in pig iron obtained by reduction smelting of iron ore with coke and lime in the blast furnace. Essentially, steelmaking is the conversion of molten pig iron (hot metal) containing variable amounts of 4.0–4.5% carbon, 0.4–1.5% silicon, 0.15–1.5% manganese, 0.05–2.5% phosphorus (normally between 0.06 and 0.25%), and 0.15% sulfur (normally between 0.05 and 0.08%) to steel containing about 1% of controlled amount of impurities by preferential oxidation. Alternatively, steel can be produced from solid sponge iron obtained by solid-state reduction of iron ore in the shaft furnace or retort. Thus, basically two routes are adopted in the production of steels. The first one employs the basic oxygen furnace (BOF – LD/Q-BOP/Hybrid converters) for treatment of hot metal, and the second route uses the electric arc furnace (EAF) to treat steel scrap/sponge iron or direct reduced iron (DRI). Electric arc or induction furnaces are generally used in the production of alloy steels. Pig iron contains a total of about 10% of C, Si, Mn, P, S, and so on as impurities, whereas sponge iron contains gangue oxides of iron ore, such as Al2O3, SiO2, CaO, and MgO. The amount and number of impurities depend on the quality of the iron ore, coke, and lime stone used in smelting. The molten pig iron is refined to molten steel under oxidizing conditions using iron ore and/or oxygen. On the other hand, scrap and sponge iron are melted in electric furnaces and refined for steel production.

The given chapter is devoted to the development of the mathematical support, in particular, mathematical models and algorithms, which can be successfully used for solving prediction tasks in various areas of human activity, including energetic and ecological management. The development peculiarities and the use of models and algorithms as elements of green technology to predict electric energy consumption based on mathematical apparatus of canonical expansions of random sequences are currently being discussed. Developed calculation method doesn’t impose any limitations on the qualities of random sequences of the change of electric energy consumption (requirement of linearity, Markovian behavior, monotony, stationarity etc.) and has maximal accuracy characteristics in this connection. Block diagrams of algorithms and results of the applied realization of the developed models and algorithms, for example the prediction of electric energy consumption by one of the local neighborhoods in Mykolaiv, Ukraine are introduced in the work. Comparative analysis of the results of a numerical experiment with the use of a Kalman filter and the linear prediction method confirms the high efficiency of the developed models and algorithms (relative error of prediction of electric energy consumption is 2–3%).

In the steel production, electrical arc furnace EAF is mainly used for melting the scrap iron. In steel production is increased by using EAF. The EAF is an extremely non-linear and massive power load, but due to EAF is stochastic functioning of an EAF, in the research of the power quality enhancement is motivated. The EAF operation power quality is introduced, the power quality negative consequences decides are characteristics and non-characteristics harmonics, voltage flicker effect, low power factor that arises in the electrical network in operation of the EAF. The quality disturbance are harmonics and current flicker. In the dynamic analysis of the power quality to voltage flicker is studied. The dynamic and static analysis of the EAF we taken examined to study the survival of the voltage flicker in the power system, The analysis of the harmonic is conveyed out to current harmonic is clarify under different types of EAF operation in the power system, so that controller, filter and PWM are used. The controller is hysteresis current and vector current are used to reduce the harmonic through the filter. The pulse width modulation generates the sinusoidal waveform. The EAF operation is beyond the permitted levels as per IEEE references, so as to mitigate the harmonic problems due to an EAF , a rapid dynamic reactive power compensation devices like series passive filter with vector current and hysteresis current controller is being proposed as a solution, The power system has been simulated using MATLAB and the results were tabulated.

Steel production in the modern steelmaking shops is accomplished by powerful electric arc furnaces (EAF) followed by ladle treatment. This scheme of production requires considerable capital investments and is widely used at large steel-works. For small steelmaking shops of machine-building plants, mini- and micro-mills an electric steelmaking aggregate ladle-furnace (ESA-LF) was developed, combining electric arc melting and ladle treatment. Description of a design, equipment and steelmaking technology in ESA-LF presented. Melting of steel in it is accomplished out of DRI-based raw materials, which are supplied into high temperature zone of electric arcs action through hollow graphitized electrodes. It enables to increase productivity of the aggregate, to decrease the electric energy consumption and to increase steel yield. The supply of bulk slag-forming materials and deoxydizers is made through axis channels of the electrodes. For homogenization of the melt chemical composition and temperature a blowoff by argon or nitrogen from the bottom is envisaged. The proposed scheme of combustible off gases afterburning during both the bath blow by oxygen and during its supply through additional tuyere, installed in the gas-outlet, enables to get additional heat for smelting running. Calculations of heat balance by various variants of carbon oxide afterburning presented. Simulation of smelting process accomplished, depending on carbon content in the metal and methods of combustible gases afterburning. It was determined, that application of the proposed technologies enables to decrease electric energy consumption and heat duration.

Continuous increase of the annual amount of steel scrap generated presents a problem to the sustainability of human society. The effective utilization of scrap is a serious problem to be solved in the near future, and the optimization of the global system of steel production should be considered from various viewpoints, particularly those of environmental load and material efficiency. For the consideration of a global system that contains many different kinds of flow, a criterion for evaluating the efficiency of the system is necessary for optimizing the utilization of materials in the system. The concept of exergy was adopted in this study and the application of exergy analysis to the evaluation of a complicated system was considered. A simulation model was developed for the steel industry in Japan and the exergy analysis of the Blast Furnace-LD converter (BF-LD) process and the Electric Arc Furnace (EF) process was conducted. The exergy loss was expressed as a function of parameters, such as the mixing rate of pig iron in the EF process and the total exergy loss in the system was calculated. The applicability of exergy as a criterion for the analysis of a production process and for evaluating material efficiency was discussed. Environment and recycling become important keywords in the consideration of a new industry structure for optimizing material efficiency and environmental load for production and utilization of materials. In the present study, a methodology for evaluating the degree of optimization in the system includ- ing material recycling was investigated. The Blast Furnace- LD converter (BF-LD) process for the production of high- grade steel and the Electric Arc Furnace (EF) process for the production of low-grade steel are the main steel produc- tion processes used in Japan. This division occurs because of contamination by tramp elements as impurities, which is inevitable in processes using low-grade scrap as feed. The percentage of EF steel in the total production of crude steel in Japan is about 33%, 1) and the total amount of the accumula- tive steel stock in Japan is increasing continuously. Generally, processes that consume large amounts of electric energy for heating and smelting are wasteful from the viewpoints of ef- fective energy use and environmental load. However, the op- timum steel production ratio in BF-LD and EF processes has not been investigated clearly. For the further improvement of steel productivity, it is necessary to optimize the steel pro- duction system from various viewpoints such as the suitable selection of feed materials, the total demand for steel products in society, and the impurities limit in the steel product. The recent state of the steel industry in Japan is that the amount of scrap is increasing annually and its quality is de- creasing. Therefore, the effective use of iron resources orig- inating from iron ore for the dilution of scrap has recently become a principal subject in the EF steel industry, together with the improvement of scrap recovery technology. Recent electric furnaces are operated with a high mixing rate of pig iron, from 20 to 50%, as iron feed, and other furnaces have been combined with a converter to enable the hybrid steel- making process. 2) Since it is very important for the steel in- dustry in Japan to utilize the iron stock effectively as valu- able resource, the optimum ratio of EF product to BF product and that of scrap utilized in the EF process to that utilized in the converter were investigated in this study in terms of the efficiency of the use of iron resources originating from iron ore for the dilution. A model of the iron and steel recycling system including material flow was constructed, and the ther- modynamic property of exergy was applied to estimate the measure of optimization for the steel production system. 2. The Concept of Exergy

The electric energy consumption prediction (EECP) is an essential and complex task in intelligent power management system. EECP plays a significant role in drawing up a national energy development policy. Therefore, this study proposes an Electric Energy Consumption Prediction model utilizing the combination of Convolutional Neural Network (CNN) and Bi-directional Long Short-Term Memory (Bi-LSTM) that is named EECP-CBL model to predict electric energy consumption. In this framework, two CNNs in the first module extract the important information from several variables in the individual household electric power consumption (IHEPC) dataset. Then, Bi-LSTM module with two Bi-LSTM layers uses the above information as well as the trends of time series in two directions including the forward and backward states to make predictions. The obtained values in the Bi-LSTM module will be passed to the last module that consists of two fully connected layers for finally predicting the electric energy consumption in the future. The experiments were conducted to compare the prediction performances of the proposed model and the state-of-the-art models for the IHEPC dataset with several variants. The experimental results indicate that EECP-CBL framework outperforms the state-of-the-art approaches in terms of several performance metrics for electric energy consumption prediction on several variations of IHEPC dataset in real-time, short-term, medium-term and long-term timespans.

ABSTRACTŠtore Steel Ltd. makes more than 1400 steel grades. The highest costs in steel production from scrap stems from the electric arc furnace electric energy consumption. Electrical energy is used to produce heat energy generated by the burning arc between the graphite electrodes and steel scrap. In general, the balanced heat input of all electrodes is essential. Based on the input of thermal energy from all electrodes, also the possibility of occurrence of hot and cold spots in the electric arc furnace can be determined. Perception of and the elimination of the unequal heat load of electrodes have a major impact on reducing operating costs and increasing the efficiency of the electric arc furnace production. Most authors have modeled the arc furnace as an electrical equivalent circuit, where the electric arc is modeled using the macroscopic approach. In this paper, the microscopic approach to the electric arc model is described, where a set of equations (electrical neutrality, Dalton law, Saha–Eggert) was solved using differential evolution algorithm. The results of modeling were practically confirmed by measuring electric parameters (voltage, current, active power) during the electric arc furnace operation. In November 2016, the investment in a new electrode controller using implemented logic will be carried out.

Purpose of reseach. Development of a forecast model of energy consumption and assessment of factors influencing its consumption. The obtained forecast estimates of energy consumption will improve the quality and efficiency of management decisions at all levels of administrative management.Methods. The article presents an analytical review of the existing methods of cognitive modelling and forecasting of electric power consumption, the description of the software implementation of the information-computing system that allows to make a forecast of electric power consumption by the population of the administrative-territorial formation. The approach to the description of factors of electric power consumption by both population and various branches of national economy, as well as organisations engaged in rendering various services has been proposed. Special software has been developed, which allows to obtain model results of electric power consumption in an automated mode, to carry out factor analysis of power consumption. The experimental verification of the work of the programme of cognitive modelling and forecasting of electric power consumption by the population of Lgovsky district of Kursk region is given. The developed software also makes it possible to evaluate the adequacy of the obtained results and promptly adjust the model parameters.Results. As a result of the research a fuzzy cognitive map of energy consumption for a municipal entity was developed. The concepts of the subject area describing the influence of various groups of factors on the level of electric energy consumption were identified. Forecast estimates of electricity consumption were obtained, which were based on the data for the retrospective period. Adequacy indicators based on the calculation of statistical criteria are determined for the obtained estimates.Conclusion. The results of the study have shown that the combination of cognitive and statistical methods allows to achieve an adequate solution when solving the problem of energy consumption forecasting.

The prediction of household or building electricity load energy consumption is a problem that is being tackled and many solutions such as statistical, machine learning and physical modes have been developed to better improve prediction accuracy. Electricity consumption predictions play an important role for both consumers (end users) and producers of electricity. The Electric utility industry uses electric load predictions to assist in electric power supply and load balancing, hence ensuring that the electricity provided reaches its customers, while meeting the standards of the quality set. For consumers, understanding and prediction of their electricity consumption offer consumers the capability to plan and manage their electricity expenses, even more so for off-grid systems where power cuts due to in availability of or insufficient energy supply lead to disruptions of planned activities and supplement with other energy sources is not possible or costly to do so. Using the dataset of an individual household electricity consumption, this article analyses the dataset and explores the application of statistical and machine learning methods to develop regression models for electric load energy consumption prediction. The study further compares the accuracy of the different regression model designs. The developed load prediction models are designed to perform electric load consumption prediction on 24h ahead. Results of the study show low prediction accuracy in all of the models. This is evident in the high MAPE scores close to 100%, which indicate that the prediction error of the model is large, furthermore, low <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathrm{R}^{\wedge}2$</tex> scores close to 0 or below, of which indicate that the prediction models provide little explanation of the variation in the target variable, hence resulting in low accuracy.

This paper aims to develop a hybrid model for forecasting electrical energy consumption of households based on a Particle Swarm Optimization (PSO) algorithm associated with the Grey and Adaptive Neuro-Fuzzy Inference System (ANFIS). This paper proposes a new Grey-ANFIS-PSO model that is based on historical data from smart meters in order to estimate and improve the accuracy of forecasting electrical energy consumption. This accuracy will be characterized by coefficients such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE). The PSO will allow to optimally design the Neuro-fuzzy forecasting. This method is implemented on Cameroon consumption data over the 24-years period in order to forecast energy consumption for the next years. Using this model, we were able to estimate that electricity consumption will be 1867 GWH in 2028 with 0.20158 RMSE and 0.62917% MAPE. The simulation results obtained show that implementation of this new optimized Neuro-fuzzy model on consumption data for a long period presents better results on prediction of electrical energy consumption compared to single artificial intelligence models of literature such as Support Vector Machine (SVM) and Artificial Neural Network (ANN).Keywords: Forecast model, PSO, ANFIS model, Grey model, electricity consumptionJEL Classifications: C22, C25, C32, C41, C45DOI: https://doi.org/10.32479/ijeep.11761

The article presents neuro-fuzzy models of the ANFIS (Adaptive Neuro-Fuzzy Inference System) type for operational forecasting of electric energy consumption of the urban system. ANFIS allows adjusting adaptively to the features of time series (TS), automatically take into account dynamically changing trend and seasonal components of TS, as well as solving the problem of their non-stationarity. The article deals with the construction and use of neuro-fuzzy models such as ANFIS (Adaptive Neuro-Fuzzy Inference System) for the operational forecasting of electricity consumption. The proposed ANFIS-models allow adaptive adjustment to the peculiarities of specific TS in the process of its forecasting, automatically take into account the dynamically changing trend and the seasonal component of TS within the “sliding window”, allowing, under conditions of uncertainty and incompleteness of TS, to solve the problem of its non-stationarity for operational forecasting electrical energy. A method for training ANFIS for operational forecasting of electric energy consumption is described. Experimental studies of the use of ANFIS for predicting electric energy consumption in one of the Russian regions were conducted, according to which it was possible to increase the accuracy of the forecast in comparison with neural network models.

Energy saving in steel production is of great significance for its large economical scale of 1500 mil t/year and high-energy consumption. Steady trend of last years is an increase of steel production in electric arc furnaces (EAF) with a very high consumption of electricity up to 750 kWh/ton. The intention to reduce so much energy consumption they can reach by many ways. One of such way is a transforming heat energy of liquid steel to electricity and destine it to steel electric arc process. Under certain conditions, it may lead to “zero” consumption of electric power in the process. The development of these conditions leads to the formation of energy-efficient heat schemes, with a minimum electricity consumption from the external network.

Developing a viable data-driven policy for the management of electrical-energy consumption in campus residences is contingent on the proper knowledge of the electricity usage pattern and its predictability. In this study, an adaptive neuro-fuzzy inference systems (ANFIS) was developed to model the electrical energy consumption of students’ residence using the University of Johannesburg, South Africa as a case study. The model was developed based on the environmental conditions vis-à-vis meteorological parameters namely temperature, wind speed, and humidity of the respective days as the input variables while electricity consumption (kWh) was used as the output variable. The fuzzy c-means (FCM) is a type of clustering technique that is preferred owing to its speed boost capacity. The best FCM-clustered ANFIS-model based on a range of 2–10 clusters was selected after evaluating their performance using relevant statistical metrics namely; mean absolute percentage error (MAPE), root mean square error (RMSE), and mean absolute deviation (MAD). FCM-ANFIS with 7 clusters outperformed all other models with the least error and highest accuracy. The RMSE, MAPE, MAD, and R2-values of the best models are 0.043, 0.65, 1.051, and 0.9890 respectively. The developed model will assist in optimizing energy consumption and assist in designing and sizing alternative energy systems for campus residences.

In the steel industry, the optimization of production processes has become increasingly important in recent years. Large amounts of historical data and various machine learning methods can be used to reduce energy consumption and increase overall time efficiency. Using data from more than two thousand electric arc furnace (EAF) batches produced in SIJ Acroni steelworks, the consumption of electrical energy during melting was analysed. Information on the consumed energy in each step of the electric arc process is essential to increase the efficiency of the EAF. In the paper, four different modelling approaches for predicting electrical energy consumption during EAF operation are presented: linear regression, k-NN modelling, evolving and conventional fuzzy modelling. In the learning phase, from a set of more than ten regressors, only those that have the greatest impact on energy consumption were selected. The obtained models that can accurately predict the energy consumption are used to determine the optimal duration of the transformer profile during melting. The models can predict the optimal energy consumption by selecting pre-processed training data, where the main steps are to find and remove outlier batches with the highest energy consumption and identify the influencing variables that contribute most to the increased energy consumption. It should be emphasised that the electrical energy consumption was too high in most batches only because the melting time was unnecessarily prolonged. Using the proposed models, EAF operators can obtain information on the estimated energy consumption before batch processing depending on the scrap weight in each basket and the added additives, as well as information on the optimal melting time for a given EAF batch. All models were validated and compared using 30% of all data, with the fuzzy model in particular providing accurate prediction results. It is expected that the use of the developed models will lead to a reduction in energy consumption as well as an increase in EAF efficiency.