626results about "Steel manufacturing process aspects" patented technology

The invention relates to metallurgical steelmaking technologies, and aims to provide a converter intelligent tapping device and a control method. The device comprises a converter steelmaking secondarycontrol system, a converter deslag detection system, a taphole slag blocking system, a converter opening monitoring module, a tapping monitoring module, a communication control module and a deep learning host, wherein the converter opening monitoring module is provided with a converter opening monitoring probe and a converter opening monitoring and processing unit; the tapping monitoring module is composed of a tapping monitoring probe and a tapping monitoring and processing unit; and the communication control module is connected with the deep learning host, the converter opening monitoring module, the tapping monitoring module, the converter steelmaking secondary control system, the converter deslag detection system and the taphole slag blocking system, so that two-way intercommunicationof data information and control signals is achieved. Intelligent prediction and control of converter tapping can be achieved, the tapping time is shortened, and the tapping efficiency is improved; the tapping process is reproducible, traceable and high in safety; and the molten steel quality is improved while the yield of molten steel is increased.

The invention relates to a robust random-weight neural network-based molten-iron quality multi-dimensional soft measurement method which belongs to the blast-furnace smelting automatic control field, in particular to a Cauchy distribution weighted M-estimation random-weight neural network (M-RVFLNs) based method for multi-dimensional parameter-dynamic soft measurement of the molten-iron quality in the blast-furnace smelting process. According to the method of the invention, the principal component analysis (PCA) method is adopted to chose main parameters which affect the blast-furnace molten iron quality as model input variables, a molten-iron quality multi-dimensional dynamic prediction model which has an output self-feedback structure and takes into account input-output data at different moments is constructed, and it is possible to carry out multi-dimensional dynamic soft measurement of the main parameters Si content, P content, S content and molten iron temperature which represent the blast-furnace molten iron quality. The method of the invention comprises the following steps of (1) choosing auxiliary variables and determining model input variables and (2) training and using the M-RVFLNs soft measurement model.

The invention discloses a low-carbon ultra-low sulfur steel smelting method. The method comprises the following technological processes: pouring molten iron into a tank, pretreatment of molten iron, oxygen remaining operation of a converter, RH furnace vacuum decarburization, alloying and vacuum degassing in an RH furnace, deoxidization of aluminium yarn/aluminium wire in an LF refining furnace, slagging by using lime and CCM. Through combination of molten iron desulphurization and drossing, optimization of a converter steel tapping and oxygen remaining system and a slagging system, the vacuum decarburization, alloying and degassing control in the RH furnace, and the aluminium wire diffusing deoxidation slagging and aluminium wire sediment deoxidation in the LF furnace, deep decarburization and desulphurization are carried out. With an emphasis on combination of convertor steel tapping and oxygen remaining operations, RH deep decarburization and LF furnace deep desulphurization, technical optimization is carried out; due to reasonable ladle argon bottom blowing control in the smelting process, carbon content control and metallurgical thermodynamics and dynamics conditions of desulphurization are given a full play; the RH furnace decarburization is stable; LF furnace slagging and desulphurization and recarburization control effects are obvious; and low-carbon ultra-low sulfur steel smelting is achieved.

A method for controlling the foaming of slag in an electric arc furnace is disclosed. The furnace comprises at least one electrode column. Current is applied to the electrode column, causing an arc to form between the tip of the electrode column and the scrap, melting the scrap. Impurities in the molten scrap metal rise to the surface forming slag. A meter determines the total harmonic distortion associated with the system. If the total harmonic distortion is greater than a predetermined set point, and the scrap metal is sufficiently molten, then a foaming agent is added thereto.

A blast furnace control system may include a hardware processor that generates a deep learning based predictive model for forecasting hot metal temperature, where the actual measured HMT data is only available sparsely, and for example, measured at irregular interval of time. HMT data points may be imputed by interpolating the HMT measurement data. HMT gradients are computed and a model is generated to learn a relationship between state variables and the HTM gradients. HMT may be forecasted for a time point, in which no measured HMT data is available. The forecasted HMT may be transmitted to a controller coupled to a blast furnace, to trigger a control action to control a manufacturing process occurring in the blast furnace.

The invention discloses a blast furnace burden distribution intelligent monitoring system and an adjusting method. The blast furnace burden distribution intelligent monitoring system is characterizedin that a communication server is used for transmitting setting parameters and real time data of a Programmable Logic Controller (PLC) of a system under a trough of a blast furnace, a PLC of a furnacetop burden distribution system and a PLC of a blast furnace body to a data acquisition server; and a function server is used for processing and modeling the data of the data acquisition server, establishing a burden distribution process monitoring module and a burden distribution result monitoring module and passing back results to each client through a browser/server framework. An operator optimizes the time sequence of burden distribution and adjusts the parameters of burden distribution through the burden distribution process monitoring module and the burden distribution result monitoringmodule on each client, so that the accurate control over the burden distribution is realized; and the design and optimization of burden distribution parameters are carried out by using the off-line function of the burden distribution result monitoring module, so that the airflow stability of the burden distribution process is ensured, the reasonable distribution of airflow is realized, and the optimized parameters are used for the operation control of the practical blast furnace. The system realizes the whole internet availability after being installed at one place, and is convenient to use and simple to maintain.

The invention provides a converter steelmaking process cost optimal control method and system based on the BP neural network. The method comprises the steps that control parameters affecting the cost are selected according to a converter steelmaking process; a modeling sample set is established; a uniformization sample set is obtained; three layers of BP neural network algorithms are established; modeling is conducted on data obtained through converter steelmaking simulation experiments by adoption of the BP neural network algorithms, and neural network parameters are obtained; by use of a genetic algorithm, models established through the BP neural network algorithms are optimized, extremums of the established models are obtained, and optimal control parameters are determined according to the extremums of the established models; and the minimum cost value of the converter steelmaking process is determined according to the comparison results of obtained optimal control parameter cost values and the minimum cost values in the modeling sample set. The problem of the high cost of the converter steelmaking can be solved by use of the converter steelmaking process cost control method and system based on the BP neural network.

A blast furnace control system may include a hardware processor that generates a deep learning based predictive model for forecasting hot metal temperature, where the actual measured HMT data is only available sparsely, and for example, measured at irregular interval of time. HMT data points may be imputed by interpolating the HMT measurement data. HMT gradients are computed and a model is generated to learn a relationship between state variables and the HTM gradients. HMT may be forecasted for a time point, in which no measured HMT data is available. The forecasted HMT may be transmitted to a controller coupled to a blast furnace, to trigger a control action to control a manufacturing process occurring in the blast furnace.

The invention discloses a method for treating converter mouth slag through oxygen blowing by an oxygen lance. The method includes the steps of arranging the position of the oxygen lance and setting the oxygen pressure. Specifically, the method includes the steps of building a mathematic model between the position h of the oxygen lance and an impact converter mouth diameter Dimpact, the number M of spraying holes of the oxygen lance, the included angle alpha between the spraying holes and the center of the oxygen lance as well as the expansion angle beta of spraying holes of a Laval nozzle and determining the position h of the oxygen lance through the mathematic model, wherein h is equal to Dimpact/[(M+1)*tan(alpha+beta)] according to the mathematic model; and setting the oxygen pressure of the oxygen lance to be 60-65% of the normal work pressure of the oxygen lance, and carrying out slag removal operation in the sample waiting process after converting, lance lifting, temperature measuring and sampling. According to the method, by reasonably arranging the position of the oxygen lance and setting the work pressure and the oxygen blowing time, the converter mouth slag is treated through oxygen blowing by the oxygen lance, and damage to converter equipment and converter linings is avoided. Compared with methods of treating the slag through a converter disassembly machine, the method has the advantages that 25 minutes is saved each time, the production and operation efficiency is greatly improved, the operation is safe and fast, and the slag clearing effect is good.

The invention provides a blast furnace hearth sidewall monitoring method. According to the monitoring method, the angular heat flow densities of hearth sidewall monitoring positions are obtained, thetemperature is detected by a thermocouple arranged in carbon bricks of the hearth, based on the angular heat flow densities, the temperature, and the angular heat flow density of the hearth lining, the heat resistant values of different positions can be calculated, and the damages of the hearth sidewall can be diagnosed. According to the method, the change rules of onsite detected variables such as temperature are used to judge the situation of a blast furnace hearth; through the thermal resistant values of different positions and change rates in a certain period, the hearth can be diagnosed,the wrong judgment caused by onsite interference, fluctuation of the furnace situations, and change of cooling conditions can be eliminated, and multiple types of damages can be effectively diagnosedat the same time.

Disclosed is a slag foaming killing material which comprises: a mixture comprising 20 to 40 mass% (inclusive) of a carbon powder having a grain size of 0.2 to 2 mm (inclusive) and 30 to 60 mass% (inclusive) of water; and a container comprising a moisture-impermeable combustible substance and having the mixture contained therein.

A soft measurement system and method for the quality index of blast furnace multi-element molten iron, the system includes: a data acquisition unit, a data preprocessing unit, and a soft measurement unit; Filter, remove noise and normalize the parameters required by the dynamic soft measurement of the quality index; use the dynamic soft measurement model of the quality index of the blast furnace multicomponent molten iron to perform dynamic soft measurement of the quality index of the blast furnace multicomponent The parameters required for the dynamic soft measurement of the molten iron quality index are used as input, and the multivariate molten iron quality index of the blast furnace is output as the output, and the output self-feedback is adopted to dynamically and online recursively predict the multivariate molten iron quality index of the blast furnace. The invention considers the hysteresis characteristics of the blast furnace smelting process and the time series relationship between input and output variables, and utilizes the recursive subspace intelligent modeling technology to realize the dynamic online soft measurement of the multivariate molten iron quality indicators in the blast furnace smelting process.

The invention relates to an automatic steelmaking method suitable for low-slag smelting, and belongs to the technical field of metallurgy steelmaking. The technical scheme is as follow: slag remaining amount and furnace slag information of a converter are collected, and an early-stage deslagging target of converting heat is determined; a heat converting control mode is determined according to feeding raw material information, the slag remaining amount and the tapping target requirements of the converter; a first-stage system of the converter is controlled to carry out automatic converting, and a one-key type steelmaking of the low-slag smelting process is realized; and online adjustment is carried out on the model calculation module by utilizing a self-learning function of a model, so that the hit rate of a converting endpoint is improved, production cost is reduced, and labor productivity is improved. According to the automatic steelmaking method, the online application can be realized only by relying on an existing sublance system and a computer system, cost is low, maintenance is convenient, and the stability is good; and when the blowing end point C error is within the range of +/- 0.02% and the temperature error is within the range of +/-10 DEG C, the hit rate can reach more than 92%.

Systems and methods are provided for capturing CO₂ from a combustion source using molten carbonate fuel cells (MCFCs). The fuel cells are operated to have a reduced anode fuel utilization. Optionally, at least a portion of the anode exhaust is recycled for use as a fuel for the combustion source. Optionally, a second portion of the anode exhaust is recycled for use as part of an anode input stream. This can allow for a reduction in the amount of fuel cell area required for separating CO₂ from the combustion source exhaust and/or modifications in how the fuel cells are operated.

The invention provides a blast furnace distribution process distribution matrix optimization method and a blast furnace distribution process distribution matrix optimization system. The method comprises the following steps: an ideal burden surface profile is set; by taking minimizing the deviation between a burden surface profile of a furnace burden in the next batch and the ideal burden surface profile as an objective, a distribution matrix is optimized; and the optimal distribution matrix obtained through optimization acts as the distribution matrix of the furnace burden in the next batch. The system comprises a setting unit, an optimization unit and an output unit. A basis is provided for the adjustment of the distribution matrix of a blast furnace, and operating personnel can adjust the distribution matrix in distribution operation of the blast furnace according to the distribution matrix obtained through the optimization by the method and the system. The burden surface profile under the distribution matrix optimized by the method and the system is close to the ideal burden surface profile. A blast furnace gas flow is distributed reasonably, a utilization ratio of a raw material is raised, and the effects of high yield and high efficiency of the blast furnace are achieved.

The invention relates to the technical field of blast furnace smelting and particularly relates to a method for establishing an operating model of a blast furnace. The method comprises the steps of acquiring original data in the production process of the blast furnace; acquiring a process parameter of the blast furnace on the basis of the original data; acquiring state mode recognition results of the blast furnace in various control links in a cluster analysis way on the basis of the original data and the process parameter; acquiring at least one furnace condition on the basis of the mode recognition results; determining a regulating scheme corresponding to each of the furnace conditions; and establishing the operating model of the blast furnace on the basis of the furnace conditions and the regulating schemes. Due to the establishment of the operating model of the blast furnace, different furnace conditions can be accurately and rapidly handled according to the operating model, it is not needed to judge the furnace conditions by experienced operating personnel on site, and furnace condition abnormality caused by temporal judgment faults of the operating personnel and blast furnace damage caused when effective handling measures are not taken in time are also avoided.

The invention discloses a method for predicting the steel-making temperature of a converter and a server. The method comprises the following steps: acquiring steel-making information of a current converter, wherein the steel-making information includes filling data, batch charging data, steel grade information data and furnace gas data of the current heat; calculating to obtain a first molten steel temperature according to a theoretical model which is established according to the filling data, batch charging data, steel grade information data and furnace gas data, as well as the material balance and thermal balance principle; acquiring information of current heat according to the steel-making information of the current converter, and acquiring first preset number of historical heat information; calculating a second molten steel temperature according to the current heat information and the first preset number of historical heat information; and acquiring a molten steel predicted temperature of the current heat according to the first molten steel temperature and the second molten steel temperature. The server is used for executing the method. According to the method for predicting the steel-making temperature of a converter and the server, the cost for controlling the final temperature of steel-making of a converter can be reduced.

The invention discloses a multi-information fusion modeling method for the shapes of burden surfaces in the burden distribution process of a blast furnace. The content of the multi-information fusion modeling method comprises: establishing a furnace burden stacking model according to hydromechanics and statistics rules, adopting a mechanism method to deduce a model function for burden surfaces in the burden distribution process of the blast furnace, establishing a parameterized prediction model, solving original burden surface shape parameters, and determining a furnace burden stacking equation; based on the fitting of radar scattered data regressed by a support vector machine, obtaining an original burden surface shape; according to data measured by a radar, solving the original burden surface shape parameters of the prediction model; using the radar to measure the height information of different radii of burden surfaces in multiple point positions of the blast furnace, and adopting the support vector machine which is applied to a regression problem to fit the height scattered points of the burden surfaces so as to obtain a burden surface function curve; determining related parameters according to a burden distribution mechanism relationship and the radar data, and correcting the prediction model according to the parameters; based on simulation results of a discrete element process, correcting the burden surface shape parameters; and obtaining a new burden surface shape function, using the burden surface shape function as an output result, and feeding the output result back as the next original burden surface.

The invention discloses a production process for refining large blast furnace cokes by using high-ratio low-heat-state coals. The production process comprises the following operation steps: (a) the oxygen enrichment rate of a blast furnace is 5%; and the gas quantity of a bosh is adjusted; (b) the area of a wind port is reduced; the wind speed is guaranteed about 265 m/s; and the air blowing kinetic energy is about 11000 kg.m/s; (c) the proper theoretic combustion temperature of the wind port is 2180-2230 DEG C; the wind temperature is kept in 1130-1170 DEG C; and the moisture is increased to 20-30 g/t; (d) the coal ratio is increased to 150 kg/t; the coke ratio is kept invariable; the fuel ratio is increased to 565 kg/t; the proper gas quantity is kept; and the furnace top temperature is not lower than 120 DEG C; and (e) the distribution system and the ore batch weight are adjusted; the distribution mode is PWC12345 43321 PWO12345 33321; the ore batch is 62 t-64 t; and according to the top temperature, the ore coke batch weight is adjusted according to the same ratio to guarantee two coal gas flows of the center and the edge of the blast furnace. Meanwhile, a corresponding smelting operation system is formulated according to the method.

The invention provides a method and system for estimating charge surface shapes in the blast furnace charge distribution process. The method includes the steps that the blast furnace charge distribution process parameters, the furnace charge parameters and the charge distribution matrix are obtained; the abscissa of the pile cuspidal point of a charge pile formed by the furnace charge at each chute dip angle is computed; the charge surface shape of the furnace charge at the first-chute dip angle is computed; and according to the volume constraint that the furnace charge at each chute dip angle meets, the charge surface shapes of the furnace charge at the previous chute dip angles and the charge surface shape of the basic charge surface, the charge surface shape at the second-chute dip angle is obtained, charge surface shape computing is conducted in a successive iteration manner till the nth charge surface shape of all the charge surface shapes at the chute dip angles is distributed and obtained, and the estimation result of the charge surface shapes in the blast furnace charge distribution process is obtained. The system comprises a collection module, an abscissas computing module, a first charge surface shape computing module and a charge surface estimation module. According to the method and system for estimating the charge surface shapes in the blast furnace charge distribution process, a computing basis is provided for the charge surface shapes under different charge distribution systems, precise estimation of the charge surface shapes in a blast furnace is achieved, the charge distribution system is adjusted in real time according to the furnace conditions, furnace charge distribution is changed, the gas utilization rate is increased, and therefore the blast furnace can operate stably and smoothly.