IRON AND STEEL: PRINCIPLES, TECHNOLOGIES AND INDUSTRIAL PRACTICE

The slab surface quality is influenced by the mold level stability. Serious mold level fluctuation makes additional production cost for the derived slab machine scarfing process. Different types of mold level fluctuation in a continuous casting process can be observed online through the application of a real time fast Fourier transform method. Four types of compensators which are phase-lead, phase-lag, sliding mode and periodic mode applied to the mold level control with a control logic based on the results of FFT. Each compensator corresponds to a specific type of fluctuation. From a statistic result, the phase-lead compensator is mostly triggered at the casting of the silicon steel and the periodic mode is mostly triggered at the peritectic steel casting. A low-pass filter is also adopted to ignore the standing wave effect on the flow control actuator. The advanced compensator is put into practical use at the #5 Slab CC of China Steel Corporation in 2011 and the mold level fluctuation is reduced to about 60% of conventional PID mode. The accumulated mold level fluctuation can be down from 1000 to 80 mm per minute at the peritectic steel casting.

In the steel industry, annealing furnaces are used for reheating and heat treatment of steel strips. To realize a continuous operation of the furnace, the strips are welded together to form an infinite strip. For a high product quality, the strip has to be heated according to a predetermined heating curve while passing through the furnace. From a control point of view, this is a challenging task, particularly when a welded joint traverses the furnace. The task is further complicated by the fact, that the strip temperature usually can be monitored by radiation pyrometers only, which are located at a few discrete points. For accurate model-based control for the strip temperature, a mathematical model of the furnace is required. In the presented paper, a dynamical model of a strip annealing furnace, which is designed as a tunnel furnace, is proposed. The furnace is divided into several heating zones, where the local temperatures are used as model inputs. The presented model takes into account the most important underlying physical phenomena, i. E., the energy balance of the strip and the heat transfer by radiation. For determining the radiative heat transfer, radiation balances in the form of integral equations are used. Approximate solutions of these equations are obtained by means of the method of weighted residuals. Furthermore, the strip motion is described in a Lagrangian framework, which facilitates an accurate representation of the strip temperature, which is especially useful at welded joints. The capabilities of the model are explored in simulation studies.

An automatic control of a DRI Plant has been performed in order to provide the best possible working point, to obtain the target Level 2 Process Set Points (DRI Metallization, Carburation , Productivity) in function of Iron Ore and Reducing Gas quality. - A Process Reconstruction Model provides continuous static estimations of plant measures, efficiency analyses, virtual sensor and diagnostics- An Extended Kalman Filter provides continuous robust dynamic estimations of output process measures to feed the Regulator. - A Linear Quadratic Gaussian Regulator performs, in real time, the calculation of Optimal Controls (minimum energy, maximum gas quality Level 1 Set Points), in function of Level 2 Sets. The suite has been successfully applied to a DRI EnergIRON Plant.

The object of this study is the control of the influence of minerals, nitrogen, phosphorus and potassium during cultivation of some varieties of winter barley (hordeum vulgare L.) for beer production in the climatic conditions of Kosovo. During cultivation and production a total of six barley cultivars have been analyzed such as Bingo, Zlatko, Vannesa, Esterel, Barun while Rex was used as comparative (standard). The analysis of the breeding and production are conducted in two agro climatic regions of Kosovo (in Arbnesh Research farm of the agricultural institute of Kosovo, Peja – Dukagjin Field, and Pestova – Kosovo Fields). Experiments were carried out by the method of randomized blocks in repetitions. Area of each experimental was 10 m2. Soil analyses for minerals, nitrogen, phosphorus and potassium were conducted on two fields. In order to achieve a proper control the following factors were measured: the yield (kg/ha), the weight (1000 seeds in grams), the hectoliters weight (kg), the protein content (%), the humidity (%) and the starch. The obtained results showed that there were significant statistical differences between various levels for all investigated areas compared with the standard (Rex) and between various localities. These data served for a proper control of the quality of the beer production.

This contribution is about the active damping injection procedure for high performance converter tilting drives. Irrespective of the particular steel making process (BOF, AOD) the metallurgical reactions will cause unwanted vibrations within the mechanical equipment. The oscillations are additive to the static loads and the amplitude can affect the equipment lifetime. Therefore, it is of strong interest to suppress the process vibrations as much as possible. According to the usual fixation procedure, it is not possible to transfer dissipative energy from the equipment to the environment due to the zero velocity at the clamping. A model-based MIMO control approach exploits the opportunities given by the multiple electric drives of the converter tilting system for an active damping injection method. The presented concept shows a possible upgrade for operating drives and it is based on an extension of the current control scheme. By the way, an upgrade of the controller has a very short ROI time and it can be done during a regular maintenance break without major influence to the ongoing production process.

Ironmaking blast furnace uses fossil fuel and hence produces top gas as a by-product of ironmaking which is usually composed of 22-26% CO, 16-19% CO2, 1-4% H2 and 58-60% of N2 by volume. Blast furnace gas (BFG) is commonly used for heating, which resulted not only in very low energy efficiency, but also in the increasing CO2 emission. Meanwhile, CO and H2 are important reducing agents for iron ore reduction in BFs. Therefore, it would be a most promising technology to maximize the reuse of CO and H2 in BFG by the top gas recycling (TGR) to reduce coke consumption and production costs, thus significantly reduces CO2 emissions. TGR relies on the removal of the CO2 contained in BFG so that the useful components – CO+H2 can be recycled back into the furnace and reused as reducing agents. In addition, injecting oxygen (O2) into the furnace instead of preheated air, removes unwanted nitrogen (N2) from the gas, facilitating CO2 Capture and Storage (CCS). However, there are still two issues, i. E., supply of oxygen and storage of CO2, which are difficult to overcome due to economic and technical barriers. In addition, process for CO2 capture is also needed to be further optimized. To address the issues, this study aims to develop an alternative TGR technology in order to optimize the CO2 separation from BFG and maximize utilizing carbon and hydrogen in BFs. The study will demonstrate a process to capture CO2 from BFG with calcium looping packed bed carbonators, convert it to CO and H2 using non-thermal plasma with steam addition, and then directly re-inject the production gas into blast furnace. This research involves numerical and experimental simulation that will generate important information on technical and economic feasibility applying TGR technology in ironmaking blast furnace. It deals with several novel technologies, such as calcium looping packed bed carbonators for CO2 capture and non-thermal plasma reactor for CO2 dissociation; And investigation on the effects of new TGR process on blast furnace operation. Those technologies will significantly improve carbon utilization and energy efficiency in ironmaking blast furnaces and remarkably reduce CO2 emission and ironmaking cost.

South Africa has about 80% of the world chrome ore reserves, located mainly in the Bushveld Igneous Complex. Abundant reserves of coal were also found in the vicinity of the Bushveld Complex. Consequently various ferrochrome producers have their production plants set along the Complex, making use of the advantage of short transport of both chrome ore and reductants. From 1970’s, majority of ferrochrome alloys has been produced in South Africa after successful marketing of charge chrome. Since 2008, due to the crisis of electricity shortage, ferrochrome producers in South Africa have been forced to reduce their electricity usage. With a limited supply of electricity, and sharp increase of electricity tariffs, South Africa’s share of world ferrochrome production has declined to 38% in 2011. The situation is set to continue and could go from bad to worse in the near future. Consequently, all ferrochrome producers in South Africa are looking for any alternatives that can use less electricity. An investigation of the effect of various reductants on the electric energy consumption is under consideration. A group of reductants, consisting of 6 types of coke and 2 types of anthracite, is selected to investigate the effect on energy consumption of ferrochrome production in submerged arc furnace. An excel-based simulation is used to calculate all charges and energy usage with similar production conditions used by most ferrochrome producers in South Africa. There is no clear indication to show a close relationship between the fixed carbon content and energy consumption among the selected reductants. A similar result is obtained for the contents of ash, volatile and water in the reductants. The combination of ash and added flux quartzite depicts a very close relationship with the energy consumption. Considering the fact that almost all contents of ash go to slag, the higher amount of ash and flux quartzite added in the furnace, the more electric energy is required and more slag will be produced.

The contribution presents the results of a collaborative R &D effort of two private companies and two national research institutions, joined at the European level. It was aimed to develop an enhanced on-line predictor of the strip thickness in the rolling gap. The issue dealt with the absence of a reliable delay-free measurement of the outgoing strip thickness or the gap size, making the thickness control a challenging task. Although several satisfactory solutions are used for decades and modern control theory is exploited as well, the pervasive competition in the field of metal strip processing emphasizes the need of a novel, more precious measuring method. The solution developed within the completed project is based on a parallel run of several adaptive Bayesian predictors whose outputs are continuously mixed to provide the best available rolling gap size prediction. The system was already tested in open loop in a real industrial environment for two reversing cold rolling mills, processing steel and copper alloys strips, respectively.

In the steel industry, it is important to know the current temperature of the product to monitor and control the production process. The temperature evolution is often decisive for the production process, because many mechanical properties of the product are temperature-dependent, e. G., the yield stress. Furthermore, the temperature evolution also influences the characteristics of the final product. However, thermal models, which are used for scheduling the production process, may not always capture the real production conditions with sufficient accuracy. Therefore, observers have to be designed which corrects the model predictions, e. G., with the help of available pyrometer measurements. Thus, the production schedule can be adapted based on a more accurate temperature estimation and a better quality of the product can be achieved. In this paper, an integrated thermal model of the hot rolling of steel plates is presented. This dynamical model takes into account the effect of all significant production steps on the temperature evolution of the plate, i. E., roll passes, descaling passes, and air cooling periods. Consequently, the model structure is switching. Based on this model, an extended Kalman filter (EKF) is developed. The EKF uses surface temperature measurements from pyrometers installed along the production line. However, these pyrometers measure only at discrete points in time, when the plate passes the pyrometer. Thus, the EKF has not only to cope with the switching model structure but also with scarce measurements. To test the performance of the developed EKF, experiments with supplementary pyrometer measurements were carried out so that the observed temperature can be compared to measurements which were not used as inputs of the filter.

Phase diagrams with allotropy of iron and titanium as a rule have the complicated geometrical structure. A use of computer models of phase diagrams permits to overcome the difficulties of its geometrical understanding. The goal of work is to demonstrate the opportunities of computer models of phase diagrams at the study of real ternary systems. As an example we consider T-x-y diagram including binary system A-B with continuous range of solid solutions, binary systems В-С and А-С of peritectic type and two modifications of component B. The high-temperature modification of component B forms the continuous range of solid solutions with component A, but the low-temperature modification forms the continuous range of solid solutions with component C. On the base of obtained computer model the analysis of phase regions structure and the decoding of horizontal and vertical sections were made. The substitution of peritectic interaction to eutectic one in the binary system А-С gives the change of three-phase reaction. Vertical diagrams of material balances (for given mass center) and horizontal ones (for given temperature on isopleth) give the possibility to determine the effect of sign change of mass increment and to study the crystallization stages.

In this paper, highly accurate laser welding system for very thin steel strips of 0.1 ~ 0.4 mm thickness is presented. For continuous processing lines in cold rolling mills and galvanizing lines, laser line welding is becoming a popular method, since a demand for ultra-high-strength steel, such as TRIP, CP, and DP steels is increasing and the thickness of steel strips is getting thinner in automotive industries. Actually, in laser line welding system, cutting edge quality before laser welding is very important [1, 2].. POSCO has developed a new type of thin steel strip welding system that uses the welding laser for laser cutting squared edges. The developed laser welding system is currently used in Pohang steel works. As shown in Figure 1 and Figure 2, the welding seam smoothness tells the developed system is excellently suited for application in real continuous production line. [1] U. Reisgen, M. Schleser, O. Mokrov, and E. Ahmed, Optimization of Laser Welding of DP/TRIP Steel Sheets Using Statistical Approach, , Optics & Laser Technology, 44, 255-262, (2012) [2] W. I. Cho, S. J, Na, Claus Thomy, and Frank Vollertsen, Numerical Simulation of Molten Pool Dynamics in High Power Disk Laser Welding, J. Of Materials Processing Technology, 212, 262-275, (2012)

In this study, 4140 steel powders produced by atomization were sintered by vacuum hot pressing technique with different process parameters. The effect of process parameters on the densification were determined in detail by using microstructural and mechanical techniques namely hardness and wear tests.

The basic physicochemical properties of the iron ore materials define the high reducibility, high temperature of the softening beginning and the small temperature interval of the softening and melting. One of the most important metallurgical characteristics of the iron ore materials is the softening interval. In other words, the temperatures of the softening and melting beginning. The softening interval determines the location and form of the cohesive zone which, in turn, substantially defines the flow dynamics conditions of the blast furnace process. Influence of the softening and melting interval of iron ore materials on technical and economical parameters of a number of blast furnaces of the Russian plants is calculated using the two-dimensional mathematical model of the blast furnace process. The analysis of influence of the softening and melting temperatures on the location and form of the cohesive zone is made using the complex of the two-dimensional mathematical models of the blast furnace process. In addition, the degree of iron reduction in the bulk of the furnace is also considered. The possibility of taking into account the quality influences the softening properties on the location and form of cohesive zone in the conditions of the radial non uniform taking into account the blast furnace profile is established.

One of the most effective ways of the technical progress in the blast furnace manufacture is the improvement of physical and chemical properties of the iron ore materials and coke that allows to increase the smelting of pig-iron and to reduce the coke specific consumption without building of new blast furnaces and concentrating plants. The coke consumption in blast furnace smelting makes about 40 % of costs for pig-iron production. Therefore, essential reduction of the coke consumption at the minimization of the general power costs should be considered as the important problem at the enhancement the blast furnace smelting technology. In the paper following approach to the decision of practical problems of the blast furnace is considered: The laboratory researches at the experimental installations and definition the qualitative characteristics the iron ore raw materials and coke; Analytical research by means of the mathematical models; Pilot and industrial tests.

Iron has probably been produced in many places around the world many thousand years before Christ including in India. It has been assumed that the technological development has been a continuous chain from primitive to successively more sophisticated iron making process. Production of iron is traditionally divided into the direct method and indirect method. In the direct method iron ore is transformed in one step and mostly also in the same furnace into iron. In the primitive iron making the spongy end product was forged out to a small bar. In the modern iron making the end product is either pure sponge iron or hot metal. In the indirect method, iron ore is first reduced into a liquid hot metal which after solidification is called cast iron or pig iron. In early iron making the removal of carbon was made in separate furnace in so called finery. In modern iron making the refining of the liquid hot metal is made with oxygen in separate vessel, called converter. The paper presents early iron making furnaces, theories about the Introduction of the primitive blast furnaces and their subsequent developments leading to the recent developments of the blast furnaces. The paper also discusses alternative direct reduction smelting reduction based iron making processes vis-a vis the basic knowledge and novel research that can help achieve better and new processes in iron making.Corresponding author : Tel. 00 91 657 2345244, 2345242, telefax 00 91 657 2345245, email: sprakash@nmlindia.org sprksh@yahoo.com

In this paper, new control algorithm reducing camber during roughing mill process is proposed. Using previous studies on mathematical model of camber in roughing mill process, mechanisms of camber generation is presented as linear state space form. Assuming that information of strip curvature is measured behind rolling position, control algorithm based on optimal control theory is designed considering measurement time delay. FEM (Finite Element Method) simulator composed of commercial FEM software DEFORM 3D combined with MATLAB is used to evaluate effectiveness of proposed control algorithm. Simulation results where reasonable parameters are applied show that proposed control algorithm improves camber generation.

The thermal processes occurring in the continuous casting machine, significantly affect the quality of the continuous ingot. The automatic control system of the continuous casting should help to more accurately maintain the prescribed temperature regime and thus produce a higher-quality of continuous-cast semi-finished product. The paper is devoted to problems of development of the mathematical model for model predictive control (MPC) algorithms of the secondary cooling modes. We consider the continuous wide slab and the water-cooling walls of the mold, in their longitudinal cross-section, which is parallel to the narrow faces and passes through the middle of the wide faces. Since the thermal physical parameters depend on the temperature of the cast metal and the copper material of the mold walls, the description of processes of the heat transfer is based on the nonlinear partial differential equations. The position of the boundary between liquid and solid phase is given by the conditions of temperatures equality and the Stefan conditions for the two-dimensional case. The boundary conditions for the wide faces of the ingot in the secondary cooling zone take into account the convective and radiative heat exchange components, as well as the ambient temperature dependence and the heat-transfer coefficients on the locations of the nozzles and water discharge in them. Mathematical modeling can answer the question of when and how needed to change the cooling water discharge. The MPC principle is the following. At each step of the receipt of new measurements, the model calculates the value of control inputs (water rate), which are needed to obtain required temperature in the future. This requires some time to carry out calculations on the model and the execution of new values of cooling water discharge. Model takes into account data coming from the measuring devices and uses the information about the surface temperature of the ingot at the outlet of the secondary cooling zone to adapt to the realities of the process. The calculated values of water discharge enter to cooling water flow regulators. Additionally, the model predicts the dynamics of the temperature field, temperature gradients and form of the liquid pool. It can be used for determining of shell thickness and control of the soft reduction mechanism position.

Strip tracking problems at the hot strip mill sometimes occur at the finishing mill during tail-out. These errors can cause pinches. To reduce the severity and number of pinches, a tilt controller has been designed to minimize DS-OS roll force differences, using a Model Predictive Controller (MPC), during tail-out. This paper describes its design and implementation. The controller comprises a strip tracking model that includes, asymmetric work-roll flattening, strip lateral motion and strip camber. Using this model, the MPC computes real-time the right tilt corrections to minimize the roll force difference during tail-out, while respecting the constraints on the maximal tilt control actions. Also constraints on the rate of variation of the tilt corrections are taken into account. The controller has been implemented in the Level 1 software of the hot strip mill of Tata Steel in IJmuiden, the Netherlands. Results of the first hot commissioning trial are presented.

In continuous steel casting plants, it is very important to stabilize the molten steel level in the mold for quality of the final products and productivity. Standing waves in the mold cause periodic molten steel level fluctuations, which can deteriorate the product quality and productivity. Nevertheless, there has not been any effective countermeasure for them, because they are different from other disturbances in the sense that they are not changes in the flow rate of the molten steel and hence should not be controlled by manipulating the mass flow. This paper proposes a standing wave model and a mold stabilization technique that can prevent harmful influence of standing waves. This technique can prevent interactions between the mold level control and the standing waves in the mold by removing standing wave components from the original mold level signal. On-line experimental results have shown good performance and this technique has been in practical use in a continuous steel casting process in JFE.

This paper presents a technique for controlling the pressure of molten metal when using a new type of the iron production method called greensand mold press casting to realize high productivity and obtain high-quality castings. In this method, molten metal is poured into a lower mold first, and the after upper mold is moved down toward the lower mold, then both molds are matched. Casting defects such as metal penetration are often caused by the high pressure in the high-velocity pressing part. Therefore, we proposed a pressure control method with a mathematical model of molten metal pressure, and with it we achieved confirmation simulation of the successful pressing production at the different pressing condition such as mold shape. Results show that the proposed pressing production considering the model predictive control: MPC can realize sound, penetration-free casting production under the pressure constraint for defect prevention. Here the computational fluid dynamical analysis, simple simulation of model predictive pressing control and water pressing experiment using feed-forward motion input calculated by off-line MPC are presented in this paper. At the final part, discussions of the theoretical control design for the process as well as its practical confirmation simulation result of molten metal pressing on the fluid analysis are concluded.

It is possible to increase the efficiency of the computer-based system of diagnostics of the elements state of the firebrick lining by means of use in their structure of the module, in real time revealing the structural changes of non-stationary controlled signals. These changes as a rule testify the occurred deviations in the technical condition of the controlled object. The received information about the revealed structural features in analyzed signals it is expedient to use for forecasting of further dynamics of controlled parameters. The mathematical description and the computer program “Hearth Erosion” of calculation the two-dimensional temperature fields in any vertical and horizontal section of the blast furnace hearth are developed. Calculation is carry out the decision of the equations of heat conductivity with use of indications of the big number of sensing transducers of temperature (to 1000), built in the furnaces firebrick lining between the firebrick blocks. The continuous control of the temperature change in each point allows to define the remained thickness the firebrick lining and to warn, in case of need, the furnace personnel about the beginning of the firebrick lining erosion. The continuous control of change of temperature in the firebrick lining is made on the basis of mathematical model. The system of collecting, processing and transfer information from sensing transducers of temperature or thermal streams in a program database “Hearth Erosion” is used. Programs are installed on blast furnaces of Chinese National Republic: Jinan, Jiyuan and Liuzhou Iron and Steel Works together with Russian-Chinese company SKKONT.

Molten slags comprising Al2O3, SiO2, CaO, FenO, MgO, MnO etc. Produced in BF processes generally exhibit a network structure. The network structure of the oxides of various elements influences the rheological characteristics of slag. In order to understand the effect of alumina on the blast furnace operation, the flow characteristics of the slag in high temperature zone is presented in the paper. The paper presents a rheological model of blast furnace slag system containing high alumina and also describes some results of a model study of the momentum transfer to the blast furnace slag systems possessing network structure.

In the last more than four thousand years, the iron making technology has come a long way. The transition from bloom in bloomeries to molten iron produced in blast furnaces was the first major step in advancing iron making technology. An average modern blast furnace has an inner volume of about 3500 m3 and produces about 8000 t of hot metal per day, thus consuming about 12000 t of iron bearing material and 4000 t of coke per day. With, the advent of electronically controlled automation and control systems, blast furnaces were equipped with so called basic automation systems to automatically control the furnace operation with logical sequences. However, these systems are beset with process complexities in finding the proper algorithms for the problems at hand. So the experience of the operators and the engineers still had a major impact on the performance of a blast furnace. It entails rule based expert systems that are closer to how people think. These may include systems based on fuzzy logic as also the vague, heuristic knowledge coupled with phenomenological models. It is expedient to take away the “magic” of determining set-points and replace it with science and to experiment with these technologies to investigate whether these models can also be used directly by an expert system. The paper seeks to examine various real time expert process control systems that can effectively interpret patterns in data and then use the information for the further reasoning process. It also discusses various aspects of turning a rule based system into a model based system vis-a vis the complexities of multi-input and multi-output, multi-phase, multi-component, counter-current yet an energy efficient and predominant production unit for liquid hot metal blast furnace operations.

Freed from the strict sinusoidal motion of classic mechanical presses, the servo press allows new tool designs. This free motion capability of the servo press can be used to integrate follow up processes into the sheet metal forming tool. With this approach, specialized equipment like dedicated welding stations become unnecessary. Moreover, by introducing transformability to the design, this integration can be made reversible. Depending on the production situation, it is possible to run the process integrated or externally. This tool design allows the manufacturer to adapt to demand more easily, thus increasing its competitiveness. However the complexity of such sheet metal forming tools is significantly increased. Thus tool manufacturers have yet to embrace the new capabilities of the servo press. In this paper we will show the design for a transformable sheet metal forming tool. The tool will produce a clamp, consisting of a ring and a holder. Ring and holder are formed, combined and welded within the tool. We will discuss the issues concerning mechanical and control systems design and how we have solved them. We will put a special emphasis on the challenges we faced with handling of forming parts, synchronization, placement and power supply of the integrated welding station.

A mathematical model for the simulation of ENERGIRON direct reduction furnaces has been developed. To simulate the reduction steps an un-reacted shrinking core model has been used. A 1D version of the model has been implemented in a software currently working on an industrial plant that solves in real time a system of partial differential equations. The outputs of the software are used to foresee the product features at the outlet of the reactor and to take any action needed to improve them. The software also has the capability to adapt itself to the different iron ore mixtures fed to the reactor.

Steel industries are important for the development of countries from social, technical aspects, where this industries may make &create many problems in area of environments, more than 370 million tons/year of some steel solid wastes accumulated across the world. This problem were devoted an attention lately by steel producers& environmental agents from economical, steel prices& environmental aspects. Libyan iron& steel company (LISCO) is one of companies that suffering from accumulation of millions of tones/year of some solid wastes. The integrated plant produce 70,000 tons /year of slag,40,000tones/year of sludge,16,000tones of dust,11,000 tones/year of scales,8500 tones/year of lime&5000 tones/year of refractory's. All of these quantities are accumulated for last 22 years of operations, not used for any others purposes& application, these millions of tones are dumped randomly at waste site &some of them are mixed with each other, there; Fore it is not possible to be sorted out after two decades of operations. Such huge amount are creating sever problems& causing environmental hazards for time being. There was a successful attempt to convert all of these solid wastes into useful products in field of building materials, slag was used as hollow building blocks , glaze for ceramics, concrete, road pavements. Sludge, Dust, under size of prefluxed pellets, scales& oxides were used for recharge to DRI, cement industries, glaze ceramics, roof tiles& other building materials. Lime were used wall paint, floor tiles. All of such products were tested in order to meet the domestic stander, economical study were performed for all products &it was visible as well.

In this study, the use of advanced statistical methods to improve the product quality in a hot sheet rolling plant is considered. Machine learning models were developed to predict and explain the relationships between the variation in process variables and the type and occurrence of observed product defects in historic process data. The analysis included the use of variable importance methods used in conjunction with multilayer perceptron models, as well as those in tree-based models, such as random forests. It is shown that these models could subsequently be used for online analysis of process operations on the industrial plant and various schemes are considered to this effect.

In this paper, we analyzed the performance of a three-dimensional imaginary robot that can come into use in workshops of different factories. The problem basically is based on folding-setting-sharing under certain details colors which are in a one plane but in random coordinates within the workshop. In this paper we took three kinds of details in red, yellow and black which have the same dimensions but different heights according to color. Performance of this scheme have been oriented towards an optimal solution that of possible shortest way of its work.

With over 20 years of providing historians for many and varied applications, Vista Control Systems has learned much about the uses of and, at times, extremely valuable, data. From leading to a deeper understanding of the process to finding the cause of very expensive plant upsets, good and complete historical data can be priceless. This paper will describe the tools we supply and outline some customer experiences to support these statements.

« Back To Technical Program