Neste trabalho, desenvolveu-se um novo método para a detecção e classificação dos distúrbios que afetam a qualidade de energia elétrica em sistemas elétricos industriais na presença de fornos elétricos a arco. Durante o processo de fusão dos fornos elétricos a arco, ocorrem diversos eventos que afetam o sistema elétrico ao qual estão inseridos, tendo como características: forma de onda do sinal de corrente altamente desequilibradas e com grande distorção devido aos harmônicos, efeitos de cintilação; bem como afundamento e elevação nos sinais de tensão. O método ora proposto foi aplicado a sinais reais, permitindo a detecção e classificação dos distúrbios múltiplos na forma de onda do sinal de tensão, proveniente da operação dos fornos elétricos a arco. Para tal, foi usada como base do algoritmo, uma técnica baseada na Transformada Wavelet, aplicada aos sinais não-estacionários de uma instalação industrial com três fornos elétricos a arco. / A new method for the detection and classification of the disturbances that affect the electric power quality in industrial electric systems with electric arc furnaces was developed in this work. During the fusion process of the electric arc furnaces, may occur several events that affect the electric system to which it is inserted may occur, having as characteristic: waveform of the signal of current highly unbalanced and with great distortion due to the harmonic, scintillation effects; as well as sag and swell in the voltage signals.The method proposed was applied to real signals, allowing the detection and classification of the multiple disturbances in the waveform of the voltage signal originating from the operation of the electric arc furnace. For this purpose, a technique based on Wavelet Transform will be used and applied to the not-stationary signals of an industrial installation with three electric arc furnaces.

O presente trabalho tem como objetivo a caracterização do carvão da mina de Leão IIcamada inferior, com a finalidade de gerar carvão pulverizado para a injeção nas ventaneiras do alto-forno na indústria siderúrgica. Esta prática visa diminuir o custo do coque, através do aumento da taxa de substituição do coque por carvões brasileiros de baixo rank. Para o estudo foram realizados ensaios de beneficiamento, análise imediata e elementar, poder calorífico, análise petrográfica e índice de inchamento livre. Na 1ª etapa de ensaios de beneficiamento, usaram-se densidades entre: 1,3-2,4g/cm3, com intervalo de diferença de 0,1 e 0,2 para as densidades de 1,3-1,9 e 2,0-2,4 g/cm3, respectivamente.Verificou-se que a fração de densidade 1,3 g/cm3 apresentou uma menor recuperação (2,0%) e um teor de cinzas de 4,3%. As maiores massas foram obtidas nas densidades de 1,4; 1,5 e 2,4 g/cm3, com recuperação de 11,1%;14,3% e 22,2% e teores de cinzas de 14,9%; 24,3% e 83,4%, respectivamente. Foram escolhidas as frações de densidades 1,4 e 1,5 g/cm3 com maior recuperação na 1ª etapa e com teor de cinzas em torno de 12 a 18%, conforme as curvas lavabilidade. Na 2ª etapa foi flutuado o carvão nas densidades de interesse escolhidas para o PCI e obtiveram-se recuperações mássicas de 13,11% e 27,00%, respectivamente. Assim as recuperações das médias ponderadas corresponderam a 22,46% para os carvões de interesse com alto teor de matéria carbonosa enquanto para as frações indesejadas obteve-se 80,45% em massa. A partir da análise imediata foram obtidos os teores das frações densidades flutuadas de 1,4 e 1,5 g/cm3, obteve-se os seguintes teores: U%- 14,45%; Cz(%bs)-15,22%;MV(bs)-36,04%; MV(bsic)- 42,51%; CF(bs) -50,54% e CF(bsic)- 59,6%. As médias da análise elementar obtida das duas frações foram: enxofre total-0,54%, hidrogênio total-5,44% e carbono total-56.74%, respectivamente. O poder refletor da vitrinita médio foi de 0,46 a 0,48% Ro, sendo classificado como sub-betuminosoB de acordo com a ASTM. O valor do poder calorífico médio foi de 5462 kcal/kg. A fração FL 1,4 é levemente aglutinante com 0,5 de perfil de inchamento, pois concentrou mais vitrinita e perfil zero (0) para fração de FL 1,5g/cm3 e NB, não sendo aglutinante. A partir dos resultados o carvão de leão II apresenta altos voláteis, bom poder calorífico superior; baixo S<1% e teores de cinzas e umidade razoáveis após beneficiamento em densidade de 1,4 a 1,5g/cm3. Assim pode ser usado para PCI, mas há necessidade de se fazer a devida mistura com carvão de médio e/ou alto rank para ajustar algumas propriedades de interesse de acordo com as exigências de cada alto-forno. / The present study concerns itself with the characterization of coal from the mine of Leão II (lower layer), for use in the steel industry with the purpose of generating pulverized coal for injection in the blast furnace. This practice is intended to reduce the cost of coke, by increasing the rate of replacement of Coke by Brazilian low-rank coals. For the study, were carried processing trials, proximate(%) and ultimate(%) analysis, calorific value, petrographic analysis and free swelling index. In the first step of processing tests, it was used density in between densities: 1.3-2, 4 g/cm3, with range of difference of 0.1 and 0.2 to 1.3 densities-1.9 and -2.0 2.4 g/cm3, respectively. It was found that the fraction of 1.3 g/cm3 density showed a minor recovery (2.0%) and an ash content of 4.3%. The biggest masses were obtained in 1.4; 1.5 and 2.4 g/cm3 densities, with 11.1%; 14.3% and 22.2% recovery and 14.9%; 24.3% and 83.4% ash content, respectively. There were chosen the fractional density 1.4 and 1.5 g/cm3 with greater recovery on the first step and with ash content around 12 to 18%, as the washability curves. The second stage was floated coal in densities of interest chosen for PCI and led to mass recalls of 13.11% and 27.00%, respectively. So recoveries of weighted averages corresponded to 22.46% to the coals with high carbon content while the unwanted fraction was 80.45% by mass. From the immediate analysis were obtained the contents of the fractions of floated densities of 1.4 and 1.5 g/cm3, the following levels: U%- 14.45%; as (% bd)- 15.22%; VM (bd)- 36.04%; VM (daf)- 42.51%; FC(bd)- 50.54% and FC (daf)- 59.6%. The averages of the elemental analysis obtained from two fractions were: total sulphur- 0.54%, total hydrogen- 5.44% and total carbon- 56.74%, respectively. The vitrinite reflector medium power was of 0.46 to 0.48% Ro, being classified as Sub-bituminous B in accordance with the ASTM standard. The value of the calorific average was of 5462 kcal/kg. The fraction FL 1.4 is slightly binder with 0.5 swelling profile because it obtained more vitrinite and focused profile of zero (0) for FL1.5 g/cm3 fraction and NB not agglutinative. From the results the Leão II coal has it was obvious presenting high volatile, good upper calorific value; low levels of S<1% and ashes and humidity reasonable after processing in density of 1.4 to 1.5 g/cm3. So it can be used for PCI, but there is need to do proper mixture with medium carbon and/or high rank to set some properties of interest according to the requirements of each blast furnace.

The reduction behaviour of the olivine iron ore pellet MPBOwas studied in laboratory scale at KTH as well as in the LKABexperimental blast furnace. Initially, a newreduction-under-load, or so-called reduction/softening/melting,test equipment was developed. Experiments using differentreducing conditions, corresponding to different radialpositions of the blast furnace, were conducted. The experimentsincluded different temperature profiles, reducing atmospheresand mechanical loads applied on the sample bed to simulate thevarying conditions in the blast furnace process. The progressof reduction was investigated, as well as the processes ofsintering and contraction during reduction. A model of thecarburisation (pick-up of carbon by the reduced iron) andmelt-down process during rapid contraction was presented. Laboratory testing of MPBO pellets was compared with resultsfrom the LKAB experimental blast furnace. The reduction of ironore pellets in the experimental blast furnace was surveyed by adissection of the furnace after quenching. The high temperaturephenomena occurring when reducing the MPBO pellet, with limitedsoftening and a short temperature range of the melting process,resulting in a thin cohesive zone, were found to be the same inlaboratorytests and in the experimental blast furnace. Thereduction down through the burden of the experimental blastfurnace was similar, but not identical to the results of theRUL experiments. The differences were found to be due todifferent reducing conditions. Therefore, it was concluded thata simulation of the reduction occurring in the blast furnacecan be performed in laboratory scale, provided the experimentalconditions are correctly chosen. Finally, a modification to further improve the properties ofthe MPBO pellets was examined. With the aim to improve theblast furnace process, coating of blast furnace pellets wasinvestigated in laboratory scale, as well as in the LKABexperimental blast furnace. Olivine, dolomite and quartzitewere used as coating agents. In laboratory scale the stickingprevention action of the different coating materials wasverified, in established test methods as well as in new testmethods, modified for blast furnace conditions. Testing of thecoated pellets in the experimental blast furnace revealedseveral advantages; significantly reduced blast furnace fluedust generation, improved gas utilisation and a smoother blastfurnace operation with a potential for a lowered fuel rate. <b>Keywords:</b>Olivine, pellets, pellet testing,reduction/softening/melting, MPBO, blast furnace, reduction,quenching, dissection, coating, sticking, coated pellets.

During the last decade, power quality has been recognised asa global problem. Among different types of power qualityproblems, voltage sags have been identified to be one of themost severe problems for different process industries. The mostcommon reason to voltage sags is lightning strikes in powerlines. Protection equipment, usually located at switchyards,disconnect faulted power lines as soon as possible, which isapproximately 100 ms. Thus, the duration of voltage sags areapproximately 100 ms. The sensitivity to voltage sags ofelectrical equipment in process industries can be observed asfor instance malfunction, automatic turnoff or damages. This thesis gives an overview of three metals processes withfocus on the sensitivity to voltage sags and interruptions. Theinherent energy in the process is used to find the sensitivity.This energy may also be used to obtain "ride-through" for theprocesses. The three metals processes are a blast furnaceprocess, a hot rolling mill process and a cold rolling millprocess. The main attention in this thesis is paid to the blastfurnace process, which is powered by a line-operatedsynchronous machine. The thesis shows that the protection equipment forelectrical machines can be adjusted to avoid unnecessaryshutdowns. It is also explained why there are high torque andcurrents during voltage sags as well as after voltage sags. Itis shown that the first peak torque and current during thevoltage sags is almost proportional to the voltage change, thatis the voltage magnitude before the voltage sag minus thevoltage magnitude during the voltage sag. The first peak torqueand current after the voltage sag depends sinusoidal-like onthe duration of the voltage sag and almost proportional to thevoltage change during the voltage sag. There is no fluxsaturation during voltage sags, however after voltage sagssaturation is very likely to occur. The thesis explains why andalso how the flux is changed during and after voltage sags. The duration of voltage sags is in many cases set by theprotection equipment located in switchyards. It is shown thatthe durations of voltage sags can be changed to durations thatwill cause less peak torque and current after voltage sags forline-operated AC-machines. It is also shown how this istheoretically achieved. <b>Keywords:</b>Rolling mill, Blast furnace, Power Quality,Synchronous machine, Asynchronous machine, Voltage sag, Voltageinterruption, Ride-through, Process disturbances, Simulation,Modelling

The thesis utilizes the analogy in dynamic equations between a mechanical and an electrical system to convert the steam-turbine, micro-turbine, wind-turbine and hydro-turbine generator mechanical model to equivalent electrical circuit models respectively. And based on the round rotor type and permanent magnetic rotor type synchronous generators¡¦ dynamic equations, as well as their electromagnetic torque equations, the equivalent electrical interface circuits were derived respectively. By using the interface circuit, the circuit model of synchronous generator and the equivalent electrical circuit model of turbine-generator mechanism can thus be combined into the electromechanical integrated circuit model (Thevenin¡¦s analogy circuit model and Norton's analogy circuit model). The electromechanical integrated circuit model is helpful for analyzing the energy conversion, power transmission and interactions between the mechanical and electrical systems for a turbine generator unit. In order to learn about these electromechanical interactions by using the proposed electromechanical integrated circuit model, the thesis has made a study on the torsional vibrations for a small gas turbine generator unit and for a large steam turbine generator unit respectively. By way of the frequency scanning and eigenvalue calculation, it is found that the torsional mode frequencies can be changed due to the electromechanical integration. Moreover, the small unit was more affected by the electromechanical integration than the large unit. Finally, we studied the effect of operations of an Electric Arc Furnaces (EAF) on torsional vibrations of a low capacity turbine generator. The electric system studied belongs to a practical steel plant in an industrial park. Based on the electromechanical integrated equivalent circuit model, a flywheel coupling shaft was designed. It is found by simulations that the coupling shaft can be quite effective in alleviating vibrations caused by the system unbalance arising from the EAF operations.

In recent years, there has been a growing trend for the use of industrial by-products in the production of blended cements because of economical, environmental, ecological and diversified product quality reasons. Granulated blast furnace slag, a by-product of the transformation of iron ore into pig-iron in a blast furnace, is one of these materials which is used as a cementitious ingredient. The aim of this study is to investigate the properties of Portland slag cement (CEMII/B-S) by using separate grinding and intergrinding of granulated blast furnace slag and Portland cement clinker. For this purpose, granulated blast furnace slag was used as mineral admixture replacing 30% of the clinker. Clinker and granulated blast furnace slag were ground to four different Blaine fineness values of 3000 cm2/g, 3500 cm2/g, 4000 cm2/g and 4500 cm2/g by intergrinding and separate grinding in a laboratory ball mill. Then, eight Portland slag cement mixes and four Portland cement control mixes were prepared, in order to determine and compare 2-, 7-, 28-, and 90-day compressive and flexural strengths, normal consistencies and setting times. It was found that for the Blaine fineness values of 3000 cm2/g, 3500 cm2/g and 4000cm2/g, the 2-, 7-, 28-, and 90-day compressive strength of the interground Portland slag cements had higher values than the separately ground Portland slag cements. However, for the Blaine fineness values of 4500 cm2/g, separately ground Portland slag cement specimens had slightly higher 2-, 7-, 28-, and 90-day compressive strength values than the interground ones.

Electric Arc Furnace (EAF) modeling has been a common research area to date. This thesis work proposes a new field-data based EAF-specific model. The data used in developing the proposed model is obtained from field measurements of instantaneous voltages and currents of EAF the plants. This model presents the dynamic behavior of the EAF system including all its parts, which are the EAF transformer, the secondary circuit, the electrodes moving and the arc itself. It consists of a cascade connected variable-resistance and &ndash / inductance combination to represent the variation in time of the fundamental frequency, and a current source in parallel with it to inject the harmonics and interharmonics content of the EAF current. The proposed model is capable of representing both AC and DC EAFs, whose controllers&rsquo / set points are the impedance values seen from the low voltage (LV) side of the EAF transformer. The validity of the proposed model has been verified by comparing EMTDC/PSCAD simulations of the model with the field measurements. The results obtained have shown quite satisfactory correlation between the behavior of the proposed model and v the actual EAF operation. To show the advantages of the model while developing FACTS solutions for power quality (PQ) problem mitigation of a given busbar supplying single- or multi-EAF installations, various applications are presented.

In this work, feasibility of utilizing municipal incinerator fly ash (MIFA) of different sources as a substitute of raw materials for steel-making in a mini-mill was studied. Also studied included the efficacy of this mode of melting treatment and recycling. Under the condition of adding 1 wt% MIFA to scrap iron/scrap steel for partial replacement of lime, the L9 orthogonal arrays of Taguchi methods were utilized to investigate the effects of MIFA melting treatment of different sources by an electric arc furnace (EAF). Four experimental factors (i.e., scrap iron mass, lime mass, coke mass, and MIFA mass) were selected to study their effects on Pb leaching of EAF dust and slag. Test results for MIFA obtained from Plants K1, K2, and K3 show that EAF dust remains hazardous as it is originally a listed waste. On the other hand, slag so generated remains nonhazardous based on the TCLP results. The experimental results of EAF dust and slag were further subjected to the analysis of variance (ANOVA) and regular analysis. Using this process, the optimal operating conditions with respect to the leached Pb concentration would be as follows: (1) Plant K1(injection mode operation)--87 tons of scrap iron, 1.4 tons of lime, 0.4 ton of coke, and 0.9 ton of MIFA; (2) Plant K2(injection mode operation)--90 tons of scrap iron, 1.4 tons of lime, 0.7 ton of coke, and 1.1 tons of MIFA; (3) Plant K2(one-time-charge mode operation)--90 tons of scrap iron, 1.4 tons of lime, 0.9 ton of coke, and 0.7 ton of MIFA; (4) Plant K3(injection mode operation)--85 tons of scrap iron, 1.4 tons of lime, 0.7 ton of coke, and 0.9 ton of MIFA; and (5) Plant K3(one-time-charge mode operation)--87 tons of scrap iron, 1.5 tons of lime, 0.4 ton of coke, and 0.7 ton of MIFA. In this study, using MIFA from Plant K2 as an example, it was found that it required 29-35 kg of lime per ton of steel billets produced when MIFA was added. Under a normal operation of EAF steel-making, however, it required 35-45 kg of lime per ton of steel billets produced based on the past experience. In average, when MIFA is added, it needs only 32 kg of lime per ton of steel billets produced as compared to 40 kg of lime for regular steel-making. In other words, it would result in a reduction of 8 kg of lime per ton of steel billets produced by using this novel process. Based on an average monthly production of 20,973 tons of steel billets and a unit cost of 2,200 NT$ per ton of lime, a monthly saving of lime cost would be 369,125 NT$. Namely, about 4.4 million NT$ per year. In addition, it was also found that using this novel process to melt MIFA would not deteriorate the quality of steel billets and bars produced.

<p>The reduction behaviour of the olivine iron ore pellet MPBOwas studied in laboratory scale at KTH as well as in the LKABexperimental blast furnace. Initially, a newreduction-under-load, or so-called reduction/softening/melting,test equipment was developed. Experiments using differentreducing conditions, corresponding to different radialpositions of the blast furnace, were conducted. The experimentsincluded different temperature profiles, reducing atmospheresand mechanical loads applied on the sample bed to simulate thevarying conditions in the blast furnace process. The progressof reduction was investigated, as well as the processes ofsintering and contraction during reduction. A model of thecarburisation (pick-up of carbon by the reduced iron) andmelt-down process during rapid contraction was presented.</p><p>Laboratory testing of MPBO pellets was compared with resultsfrom the LKAB experimental blast furnace. The reduction of ironore pellets in the experimental blast furnace was surveyed by adissection of the furnace after quenching. The high temperaturephenomena occurring when reducing the MPBO pellet, with limitedsoftening and a short temperature range of the melting process,resulting in a thin cohesive zone, were found to be the same inlaboratorytests and in the experimental blast furnace. Thereduction down through the burden of the experimental blastfurnace was similar, but not identical to the results of theRUL experiments. The differences were found to be due todifferent reducing conditions. Therefore, it was concluded thata simulation of the reduction occurring in the blast furnacecan be performed in laboratory scale, provided the experimentalconditions are correctly chosen.</p><p>Finally, a modification to further improve the properties ofthe MPBO pellets was examined. With the aim to improve theblast furnace process, coating of blast furnace pellets wasinvestigated in laboratory scale, as well as in the LKABexperimental blast furnace. Olivine, dolomite and quartzitewere used as coating agents. In laboratory scale the stickingprevention action of the different coating materials wasverified, in established test methods as well as in new testmethods, modified for blast furnace conditions. Testing of thecoated pellets in the experimental blast furnace revealedseveral advantages; significantly reduced blast furnace fluedust generation, improved gas utilisation and a smoother blastfurnace operation with a potential for a lowered fuel rate.</p><p><b>Keywords:</b>Olivine, pellets, pellet testing,reduction/softening/melting, MPBO, blast furnace, reduction,quenching, dissection, coating, sticking, coated pellets.</p>

<p>During the last decade, power quality has been recognised asa global problem. Among different types of power qualityproblems, voltage sags have been identified to be one of themost severe problems for different process industries. The mostcommon reason to voltage sags is lightning strikes in powerlines. Protection equipment, usually located at switchyards,disconnect faulted power lines as soon as possible, which isapproximately 100 ms. Thus, the duration of voltage sags areapproximately 100 ms. The sensitivity to voltage sags ofelectrical equipment in process industries can be observed asfor instance malfunction, automatic turnoff or damages.</p><p>This thesis gives an overview of three metals processes withfocus on the sensitivity to voltage sags and interruptions. Theinherent energy in the process is used to find the sensitivity.This energy may also be used to obtain "ride-through" for theprocesses. The three metals processes are a blast furnaceprocess, a hot rolling mill process and a cold rolling millprocess. The main attention in this thesis is paid to the blastfurnace process, which is powered by a line-operatedsynchronous machine.</p><p>The thesis shows that the protection equipment forelectrical machines can be adjusted to avoid unnecessaryshutdowns. It is also explained why there are high torque andcurrents during voltage sags as well as after voltage sags. Itis shown that the first peak torque and current during thevoltage sags is almost proportional to the voltage change, thatis the voltage magnitude before the voltage sag minus thevoltage magnitude during the voltage sag. The first peak torqueand current after the voltage sag depends sinusoidal-like onthe duration of the voltage sag and almost proportional to thevoltage change during the voltage sag. There is no fluxsaturation during voltage sags, however after voltage sagssaturation is very likely to occur. The thesis explains why andalso how the flux is changed during and after voltage sags.</p><p>The duration of voltage sags is in many cases set by theprotection equipment located in switchyards. It is shown thatthe durations of voltage sags can be changed to durations thatwill cause less peak torque and current after voltage sags forline-operated AC-machines. It is also shown how this istheoretically achieved.</p><p><b>Keywords:</b>Rolling mill, Blast furnace, Power Quality,Synchronous machine, Asynchronous machine, Voltage sag, Voltageinterruption, Ride-through, Process disturbances, Simulation,Modelling</p>