An on-line control method and system for the temperature of molten steel in a steelworks

Abstract

The application discloses an online regulation and control method and system for molten steel temperature of a steel mill, and belongs to the field of intelligent smelting processes and equipment, and comprises the following steps: obtaining relevant information of main process smelting equipment and a ladle of the steel mill, screening historical production data of the main process to obtain a data set corresponding to each process; classifying the ladle based on a ladle lining temperature to obtain temperature drop conditions of molten steel in each type of ladle; establishing a molten steel temperature prediction model by using a case-based reasoning algorithm based on the data set and the temperature drop conditions of the ladle, and establishing a molten steel temperature prediction model by using an intelligent algorithm; obtaining all subsequent process scheduled temperatures and next process predicted temperatures respectively before smelting of each furnace of molten steel, during smelting and after time information of scheduling is updated; and adjusting a production plan according to the scheduled molten steel temperature, the predicted molten steel temperature and the measured molten steel temperature. The application assists in realizing accurate control of the molten steel temperature, and is beneficial to reducing high-temperature tapping.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent smelting process and equipment, in particular to an online regulation method and system for molten steel temperature in a steelmaking plant. BACKGROUND

[0002] For the existing long process steel manufacturing process, the smelting process from converter, refining to continuous casting is a key link for obtaining qualified molten steel in the steelmaking plant. In this process, the molten steel temperature is a key parameter, and its control level affects the production rhythm, the quality of the casting blank and the energy cost and other elements. Precise control should ensure that each process, especially the terminal process-continuous casting, can obtain molten steel with temperature up to standard, while reducing the end molten steel temperature to reduce energy consumption.

[0003] To achieve precise temperature control, first of all, a reasonable temperature system should be established before smelting, that is, the molten steel temperature at the beginning and end of each process is required. The current mainstream temperature system is to evaluate the average temperature drop of the molten steel on the basis of the standard process smelting time, and to make a rough compensation for the heats with insufficient heat storage in the ladle or tundish. This system has a relatively rough consideration on the influence of the ladle heat state, and a relatively wide regulation on the reasonable temperature interval, and lacks dynamic consideration of the target temperature of each heat of molten steel at each process node under different production conditions.

[0004] In smelting, some scholars predict the molten steel temperature within each process or between processes to know the molten steel temperature in advance, so as to adjust the process control to meet the requirements of the temperature system. The prediction model includes a method based on a mechanism model, a data-driven model or a combination of multiple models. The application of the data-driven model reduces the influence of the complex molten pool reaction conditions and the non-linear relationship between numerous influencing factors on the prediction of the molten steel temperature to a certain extent, but the commonly used algorithms in the prior art are prone to local optimization, and it is difficult to realize classification optimization processing for different data conditions. In addition, the existing model rarely considers the influence of the ladle heat state on the molten steel temperature, or simply takes the relevant parameters of the ladle heat state as one of the input conditions of the algorithm.

[0005] The reasonable establishment of the temperature system and the accurate prediction of the molten steel temperature are complementary to each other, and there is currently a lack of a molten steel temperature regulation system that can be applied to online production and fully considers the temperature connection and matching between processes, so as to truly guide the on-site operators. At the same time, in online production, time is an important influencing factor of the molten steel temperature, and the regulation of the molten steel temperature and the scheduling of the production time are inseparable. The stability of the molten steel temperature depends on the dynamic scheduling means of the steelmaking plant including the plan scheduling and the crane scheduling, and is also an important target of the dynamic scheduling. In the prior art, the regulation of the molten steel temperature in the steelmaking plant is relatively isolated, lacks information association with the plan scheduling, and is also difficult to realize collaborative matching optimization of the temperature and the time. SUMMARY

[0006] In order to solve the above problems, the purpose of the present application is to provide a molten steel temperature online regulation technology for a steelmaking plant, which is based on ladle thermal state simulation analysis, considers the influencing factors of molten steel temperature in each section, establishes a data-driven molten steel temperature prediction model and a prediction model, and combines to form a molten steel temperature online regulation system, so as to provide the predetermined temperature and the predicted temperature of molten steel in each process for production operators before and during smelting, and assist them in taking control measures.

[0007] In order to achieve the above technical purpose, the present application provides an online regulation method for molten steel temperature in a steelmaking plant, comprising the following steps:

[0008] Obtaining the related information of the main process smelting equipment and the ladle of the steelmaking plant, and screening the data set for model training in each process according to the historical production data of the main process;

[0009] Classifying the ladles with different thermal states based on the ladle lining temperature, and obtaining the temperature drop of molten steel in each category of ladle;

[0010] Based on the data set, screening the historical heats, establishing a molten steel temperature prediction model using a case-based reasoning algorithm according to the ladle temperature drop, and establishing a molten steel temperature prediction model using K-means and BPNN algorithms;

[0011] Based on the molten steel temperature prediction and molten steel temperature prediction model, the predetermined temperature of all subsequent processes and the predicted temperature of the next process are obtained before and during smelting of each molten steel, and after updating the scheduling time information, and the production plan is adjusted according to the predetermined molten steel temperature, the predicted molten steel temperature and the measured molten steel temperature.

[0012] Preferably, in the process of obtaining the related information of the main process smelting equipment and the ladle of the steelmaking plant, the related information of the steelmaking workshop is obtained, including: overall layout, equipment parameters, production steel grade and process path, and the related information of the ladle, including: ladle lining map, brick thermal physical parameters, and ladle turnover, wherein the related information of the steelmaking workshop is used for selection of molten steel temperature influencing factors, model classification, determination of standard smelting time and standard molten steel temperature drop, and the related information of the ladle is used for numerical simulation and classification of the ladle.

[0013] Preferably, in the process of obtaining the temperature drop of molten steel in each type of ladle, according to the ladle turnover, the ladle age, the empty ladle time and the baking time are taken as the ladle thermal state, and numerical simulation is carried out for each type of ladle to obtain the temperature drop rate of each type of ladle in the static time period, and the compensation value of the molten steel temperature of each section under different ladle thermal states is obtained according to historical production data, wherein for the empty ladle time, according to the overall production rhythm and the ladle turnover procedure, the ladles with empty ladle time less than 90 min, between 90-150 min and more than 150 min are marked respectively.

[0014] Preferably, in the process of constructing the molten steel temperature prediction model, a history furnace period with better performance is screened out, all factors affecting the molten steel temperature from the converter to the continuous casting process are taken as the input, and the temperature at each process node is taken as the output to establish the data set of case-based reasoning.

[0015] The input new furnace period selects a history furnace period with the same ladle thermal state and the highest similarity of the remaining influencing factors in the data set, and the temperature of the history furnace period at each process node is taken as the target temperature of the new furnace period.

[0016] Preferably, in the process of obtaining the predicted molten steel temperature, the influencing factors of the molten steel temperature in each process section are analyzed, and based on the data set, the variables with larger Pearson correlation coefficient and meeting the mechanism analysis are selected as the dimensions for subsequent modeling except the ladle thermal state.

[0017] The K-means clustering analysis method is used to try to increase the number of clustering centers one by one by enumeration method, calculate the clustering number when the error square sum decreases obviously, and divide the data set by using the clustering number.

[0018] Based on the BPNN model, the model of the data set to which each furnace belongs is called to predict for each furnace of molten steel according to the divided data set.

[0019] For the furnace period with abnormal ladle thermal state, the corrected predicted molten steel temperature is obtained by subtracting the corresponding temperature correction value from the model predicted temperature.

[0020] Preferably, in the process of obtaining the subsequent process predetermined temperature and the next process predicted temperature, for each furnace of molten steel, the temperature prediction model is triggered before the converter smelting is completed, the liquidus temperature of the steel grade is taken as the starting point, the temperature drop of each stage is gradually back calculated according to the superheat and the production plan, the optimized temperature system of each node is formulated, the target temperature of each node is obtained and updated in time according to the ladle thermal state, and wherein the temperature prediction model is triggered in the smelting process from the converter to the continuous casting, and the molten steel temperature of the next process node is calculated based on the last process node and the current known production condition, which provides a basis for determining the temperature control parameters of each process.

[0021] Preferably, in the process of adjusting the production plan, the steelmaking production plan issued by the superior system is received, and the operation plan of the production site is generated according to the on-site production performance, logistics information and abnormal working conditions; and according to the received operation plan, the basic information of the steel grade and process path of the specific heat is obtained, and the smelting time and the transportation time between processes are planned, and when the operation plan changes, the corresponding adjustment is made.

[0022] The application discloses an online regulation system for molten steel temperature in a steelmaking plant.

[0023] The data acquisition module is used for acquiring relevant information of the main process smelting equipment and the ladle of the steelmaking plant, and screening the data set for model training of each process according to the historical production data of the main process.

[0024] The data analysis module is used for classifying the ladles with different thermal states based on the ladle lining temperature, and obtaining the temperature drop of the molten steel in each category of ladle.

[0025] The model construction module is used for screening the historical heats based on the data set, establishing the molten steel temperature prediction model by using the case-based reasoning algorithm according to the ladle temperature drop, and establishing the molten steel temperature prediction model by using the K-means and BPNN algorithms.

[0026] The online regulation module is used for obtaining the predetermined temperature of all subsequent processes and the predicted temperature of the next process before smelting of each molten steel, during smelting and after updating of the scheduling time information, and adjusting the production plan according to the predetermined molten steel temperature, the predicted molten steel temperature and the measured molten steel temperature.

[0027] The application discloses the following technical effects:

[0028] The application establishes an online regulation system for molten steel temperature in a steelmaking plant based on molten steel temperature prediction and prediction, and integrates the system with a dynamic scheduling system of the steelmaking plant, thereby providing accurate molten steel temperature recommendation and perception in the production process, assisting in realizing accurate control of the molten steel temperature, and being beneficial to reducing the phenomenon of high-temperature tapping. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings in the following description only constitute some embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of these drawings.

[0030] Figure 1 is a molten steel temperature prediction model architecture diagram according to the present application;

[0031] Figure 2 is a molten steel temperature online regulation system triggering flow chart according to the present application. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work belong to the scope of protection of the present application.

[0033] As shown in Figures 1-2 , the present application provides a molten steel temperature online regulation method in a steel plant. Based on hot state simulation analysis of a ladle, the method considers the influencing factors of molten steel temperature in each section, establishes a data-driven molten steel temperature prediction model and a prediction model, and combines to form a molten steel temperature online regulation system, to provide the molten steel predetermined temperature and the predicted temperature of each process for the production operator before and during smelting, to assist the operator to take control measures.

[0034] Among them, the predetermined temperature is the optimized temperature system and the target molten steel temperature, and the predicted temperature is the current predicted molten steel temperature at the end of each process. The operator adjusts the operation according to the predicted temperature, so that the actual molten steel temperature finally reaches the predetermined temperature target. For the molten steel temperature prediction model, a whole predetermined method based on case-based reasoning is proposed. Compared with the existing step-by-step reverse forward prediction method, the predetermined result is more in line with the actual requirements. For the molten steel temperature prediction model, a prediction method based on K-means clustering algorithm (K-means), back propagation neural network (BPNN) and ladle hot state correction is proposed. Compared with the prediction model without clustering analysis or without detailed consideration of the ladle hot state, the prediction hit rate is obviously improved, which can better guide the field production.

[0035] On this basis, the correlation between the molten steel temperature online regulation system and the dynamic scheduling system of the steel plant is established, the planning time and scheduling provided by the dynamic scheduling are received, and the regulation result of the molten steel temperature is returned, to provide decision support at the temperature level for optimizing the production process scheduling

[0036] Specifically, the application discloses a method for on-line regulation and control of molten steel temperature in a steelmaking plant, comprising the following steps:

[0037] In step S1, parameters, structures, operation conditions and other information of smelting equipment and ladles in main processes of the steelmaking plant are acquired, historical production data of the main processes are acquired, and data sets for model training of each process are obtained after screening;

[0038] In step S2, numerical simulation technology is used in combination with actually measured ladle lining temperatures to classify ladles in different thermal states, and temperature drop conditions of molten steel in ladles in various categories are obtained.

[0039] In step S3, based on the training data set, a historical heat with better performance is selected, the temperature drop condition of the ladle is combined, and a molten steel temperature prediction model is established by using a case-based reasoning algorithm.

[0040] In step S4, based on the training data set, a K-means algorithm is used for clustering, a BPNN model is established for each data set after clustering, the temperature drop condition of the ladle is combined, and a molten steel temperature prediction model is established.

[0041] In step S5, a molten steel temperature on-line regulation and control system is formed, time information provided by a dynamic scheduling system of the steelmaking plant is acquired, the molten steel temperature prediction model is called to display predicted temperatures of all subsequent processes after a molten steel smelting process of each heat, and the molten steel temperature prediction model is called to display predicted temperatures of the next process in sequence.

[0042] In step S6, the predicted molten steel temperature, the predicted molten steel temperature and the actually measured molten steel temperature are returned to the dynamic scheduling system of the steelmaking plant, the dynamic scheduling system adjusts a production plan according to a difference between the actually measured temperature and a target temperature, performs crane scheduling, and returns time information to the molten steel temperature on-line regulation and control system.

[0043] The establishment process of the molten steel temperature prediction model in step S3 comprises the following steps:

[0044] In step S31, a historical heat with better performance is selected, all factors influencing the molten steel temperature from a converter to a continuous casting process are taken as inputs, and temperatures of nodes of various processes are taken as outputs to establish a data set for case-based reasoning.

[0045] In step S32, a new heat is input, a historical heat with the same ladle thermal state category and the highest similarity of other influencing factors is selected from the data set, and temperatures of nodes of the historical heat are taken as target temperatures of the new heat.

[0046] The establishment process of the molten steel temperature prediction model in step S4 comprises the following steps:

[0047] Step S41, analyze the influencing factors of the molten steel temperature in each process section, select variables with larger Pearson correlation coefficients and conforming to mechanism analysis except the ladle thermal state as dimensions for subsequent modeling based on the data set;

[0048] Step S42, use the K-means clustering analysis method, and calculate the number of clusters when the error sum of squares decreases significantly, and divide the data set by using the number of clusters;

[0049] Step S43, for the divided data set, respectively establish a BPNN model, and call the model of the data set to which the molten steel belongs for prediction for each molten steel;

[0050] Step S44, for the ladle thermal state of the abnormal furnace, subtract the corresponding temperature correction value from the model predicted temperature to obtain the corrected predicted molten steel temperature.

[0051] The running logic of the molten steel temperature online regulation system in actual application includes:

[0052] Step S51, the system obtains the production plan provided by the dynamic scheduling system of the steelmaking plant and the ladle thermal state, and performs molten steel temperature predetermination, from the start of the converter smelting, the recommended target temperature provided by the system can be seen in advance by the scheduling personnel and the operation personnel of each process;

[0053] Step S52, if a new scheduling scheme is generated during the smelting process, the target temperature will be dynamically adjusted according to the rhythm change;

[0054] Step S53, after the specific operation of each process (such as charging, oxygen supply opening and closing, power supply opening and closing, etc.) and every fixed time, the molten steel temperature prediction model will automatically predict the temperature of the molten steel after reaching the target node;

[0055] Step S54, the field personnel determine the subsequent operation within the process and the scheduling between processes according to the deviation degree of the predicted temperature and the predetermined temperature, so that the predicted molten steel temperature of each process node is close to the predetermined molten steel temperature.

[0056] Embodiment: The embodiment of the present application provides a molten steel temperature online regulation method for a steelmaking plant, which comprises the following steps:

[0057] Step S1, obtain the parameters, structure, running conditions and the like of the main process smelting equipment and the ladle of the steelmaking plant, obtain the historical production data of the main process, and screen to obtain a data set for model training of each process.

[0058] In this step, the information required to be obtained includes the relevant information of the steelmaking plant, such as the overall layout, equipment parameters, production steel grade, process path, etc., and the ladle-related ladle lining map, brick thermal property parameters, ladle turnover, etc. Among them, the relevant information of the steelmaking plant is used for the selection of molten steel temperature influencing factors, the classification of the model, the determination of the standard smelting time and the standard molten steel temperature drop, and the ladle-related information is mainly used for the numerical simulation and classification of the ladle. The screening of historical production data mainly includes the elimination of blank values and abnormal values, data normalization, etc.

[0059] In step S2, numerical simulation technology is used to classify ladles of different thermal states in combination with the measured ladle lining temperature on site, and the temperature drop of molten steel in each category of ladles is obtained.

[0060] In this step, since the ladle is a container for holding, transporting and refining molten steel outside the furnace, its thermal state has a significant impact on the temperature of the molten steel throughout the entire process. Therefore, it is necessary to classify the ladles according to their thermal states for subsequent analysis and modeling. According to the ladle turnover, the thermal state of the ladle is mainly composed of ladle age, empty ladle time and baking time. For example, in a certain steel plant, the ladle needs to be repaired every 20 times of turnover, which is divided into minor repair, medium repair and major repair, and different maintenance operations are performed until the lining bricks are completely removed and new ones are laid. Considering that the ladle needs to be baked before it is put into operation again after repair, and that the ladle has a large heat storage space for the first 4 heats before it is put into operation, the ladle is divided into the first 4 heats before the new ladle is put into operation, the 5th heat to the repair after the new ladle is put into operation, the first 4 heats after the minor repair, the 5th heat to the repair after the minor repair, etc. For the empty ladle time, according to the overall production rhythm and the ladle turnover procedure, the ladles with empty ladle time less than 90 min, between 90-150 min and more than 150 min are marked as A, B and C respectively. Then, numerical simulation is performed on each category of ladles to obtain the temperature drop rate of each category of ladles during the static period, and the historical production data is corrected to obtain the compensation value of the molten steel temperature of each section of the ladle in different thermal states.

[0061] In step S3, based on the training data set, the historical heats with better performance are selected, the ladle temperature drop is combined, and the case-based reasoning algorithm is used to establish the molten steel temperature prediction model.

[0062] In this step, based on the filtered historical production data set obtained in step S1 and the ladle classification result obtained in step S2, a case-based reasoning method is used to establish a molten steel temperature prediction model. The specific establishment process includes the following steps:

[0063] In step S31, the historical heats with better performance are selected, all factors affecting the temperature of the molten steel from the converter to the continuous casting process are taken as inputs, and the temperature at each process node is taken as output to establish a case-based reasoning data set.

[0064] The screening criteria are: the continuous casting superheat meets the standard, the temperature of each process node meets the temperature system, the smelting time or the holding time of each section is shorter, and in these heats, the LF power supply of the heats with the LF refining-continuous casting path under the middle 50% (25%-75%) is reserved, and the heats without RH oxygen blowing operation under the converter-RH refining-continuous casting path. The factors that can be known before the smelting process of the molten steel temperature mainly include: ladle thermal state, pouring sequence, planned processing time in each process, planned holding time between processes, refining heating power, converter tapping port times, alloy addition amount, etc. The process nodes mainly include converter endpoint, refining start, refining end, continuous casting start, etc.

[0065] In step S32, a new heat is input, a certain historical heat with the same ladle thermal state category and the highest similarity of the remaining influencing factors in the data set is selected, and the temperature of the historical heat at each process node is used as the target temperature of the new heat. The similarity can be calculated by using the Euclidean distance similarity, and the Euclidean distance calculation formula is shown as formula (1), and the similarity calculation formula is shown as formula (2).

[0066]

[0067] In the formula, d(X, Y) represents the Euclidean distance between the new heat and the heat in the case base, S sim (X, Y) is the Euclidean distance similarity, m is the number of influencing factors in a heat, xj represents the jth influencing factor of the new heat, y j represents the jth influencing factor of the heat in the case base, w j represents the weight of the jth influencing factor.

[0068] This embodiment uses 3667 data of a certain steel plant for illustration. Through step S31, 1622 heats are screened, and 1000 of them are randomly selected as the data set of case reasoning, and the remaining 622 heats are used as the test set for verification. The model prediction results and the actual process node molten steel temperature of the test set are compared, and the hit rate within the error allowable range is shown in Table 1. In order to ensure the actual application effect, the error range of the molten steel temperature at the refining end node is selected as [-5℃, 5℃], and the error range of the molten steel temperature at the remaining pre-process nodes is selected as [-7℃, 7℃].

[0069] Table 1

[0070] Converter end point Tapping argon station Refining start Refining end Full node hit 91.48％ 92.12％ 93.57％ 94.86％ 90.03％

[0071] In step S4, based on the training data set, the K-means algorithm is used for clustering, and a BPNN model is established for each class of data set after clustering. Combined with the ladle temperature drop, a molten steel temperature prediction model is established.

[0072] In this step, based on the screened historical production data set obtained in step S1 and the ladle classification result obtained in step S2, a molten steel temperature prediction model is established, and the model structure is as shown in Figure 1 The specific establishment process includes the following steps:

[0073] Step S41, analyze the influencing factors of molten steel temperature in each process section, and based on the data set, select variables with larger Pearson correlation coefficients and conforming to mechanism analysis except for the ladle thermal state as the dimensions for subsequent modeling.

[0074] Step S42, using K-means clustering analysis method, the number of clustering centers is increased by enumeration method, and when the error square sum decreases obviously, the number of clusters is calculated, and the data set is divided by using the number of clusters;

[0075] Step S43, for the divided data set, respectively establish BPNN model, and for each molten steel, call the model of its own data set for prediction;

[0076] Step S44, for the ladle heat state of the abnormal furnace, subtract the corresponding temperature correction value from the model predicted temperature to obtain the corrected predicted molten steel temperature.

[0077] This embodiment uses 3667 data of a certain steel plant for illustration. Taking the process from the start of LF refining to the end of LF refining as an example, the molten steel temperature influencing factors with larger Pearson correlation coefficients are selected as shown in Table 2.

[0078] Table 2

[0079]

[0080]

[0081] The factors in Table 2 are used as the dimensions of Kmeans clustering analysis, the number of clustering centers is increased by enumeration method, the error square sum is calculated, the number of clusters is selected as 4, and the data set is divided into 4 categories. For each data set, use other influencing factors in Table 2 except for the ladle thermal state as the input of BP neural network model. The BP neural network model is established. For the furnace with ladle emptying time longer than A or less than 5 times, subtract the corresponding temperature correction value from the BP model predicted temperature to obtain the corrected predicted molten steel temperature. 2500 furnaces are used as training data, and the remaining 1167 furnaces are used as test data for verification, and the molten steel temperature prediction hit rate of the method in different error ranges is shown in Table 3.

[0082] Table 3

[0083] [-5℃,5℃] [-7℃,7℃] [-10℃,10℃] 87.23％ 93.14％ 97.86％

[0084] Step S5: Establish an online steel temperature control system, obtain time information provided by the steel plant's dynamic scheduling system, and before, during, and after each heat of steel is smelted, call the steel temperature pre-set model to display the pre-set temperature of all subsequent processes, and sequentially call the steel temperature prediction model to display the predicted temperature of the next process.

[0085] In this step, the triggering procedure of the online steel temperature control system for each heat of molten steel is as follows: Figure 2 As shown, the temperature pre-determining model is triggered before the end of converter smelting. Starting from the liquidus temperature of the steel grade, it progressively calculates the temperature drop at each stage based on superheat, production plan, etc., and formulates optimized temperature regimes for each node. Compensation is then made based on the thermal state of the ladle to obtain the target temperature for each node and update it in a timely manner. The temperature prediction model is triggered during the smelting process from converter to continuous casting. Based on the previous process node and the currently known production conditions, it calculates the molten steel temperature for the next process node, providing a basis for determining temperature control parameters for each process and providing guidance for operators in temperature adjustment operations. Specifically, it includes the following steps:

[0086] Step S51: The system obtains the production plan and ladle thermal status provided by the steel plant's dynamic scheduling system, and makes a steel temperature reservation. Starting from the converter smelting, the scheduling personnel and operators of each process can see the recommended target temperature provided by the system in advance.

[0087] Step S52: If a new scheduling scheme is generated during the smelting process, the target temperature will be dynamically adjusted according to the change in rhythm.

[0088] Step S53: After each specific operation (such as feeding, oxygen supply on / off, power supply on / off, etc.) and at fixed intervals, the molten steel temperature prediction model will automatically predict the temperature of the molten steel after it reaches the target node.

[0089] In step S54, on-site personnel determine subsequent intra-process operations and inter-process scheduling based on the degree of deviation between the predicted temperature and the predetermined temperature, ultimately ensuring that the predicted molten steel temperature at each process node is close to the predetermined molten steel temperature.

[0090] Step S6: The predetermined, predicted, and actual molten steel temperatures are returned to the steel plant's dynamic scheduling system. Based on the difference between the actual and target temperatures, the system adjusts the production plan, schedules overhead cranes, and returns the time information to the online molten steel temperature control system.

[0091] In this step, the steelmaking plant dynamic scheduling system receives the steelmaking production plan issued by the superior system, and generates the production site operation plan according to the field production performance, logistics information and abnormal working conditions. The molten steel temperature regulation system will receive the operation plan, obtain the basic information such as steel grade, process path and the time information such as smelting time in each process and transportation time between processes of the specific heat, and re-calculate when the operation plan changes. After the calculation is completed, the molten steel temperature regulation system will transmit the results and the actual measured values of the temperature of each heat tracked to the dynamic scheduling system for analysis and adjustment. The temperature prediction and prediction results can be displayed separately, or can be integrated into the Gantt chart interface of the dynamic scheduling system.

[0092] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other processing devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other processing devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions specified in the flowchart block or blocks. Figure One one or more flow or blocks Figure One means for functionally implementing the one or more flows or blocks

[0093] In the description of the present application, it is to be understood that the terms "first", "second", "third" and the like, merely identify features being described but do not imply or imply relative importance or a number of the indicated technical features. Therefore, a feature defined as "first", "second", "third" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly specified.

[0094] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A method for on-line regulation of the temperature of the steel melt in a steelworks, characterized in that, The method comprises the following steps: obtaining relevant information of main process smelting equipment and ladles in a steelmaking plant, screening data sets for model training of each process according to historical production data of the main process; classifying ladles in different thermal states based on the temperature of the ladle lining, and obtaining the temperature drop of molten steel in each category of ladle; based on the data set, screening historical heats, establishing a molten steel temperature prediction model using a case-based reasoning algorithm according to the temperature drop of the ladle, and establishing a molten steel temperature prediction model using K-means and BPNN algorithms; based on the molten steel temperature prediction model and the molten steel temperature prediction model, obtaining the scheduled temperature of all subsequent processes and the predicted temperature of the next process before and during the smelting of each heat of molten steel, and updating the scheduling time information, and adjusting the production plan according to the scheduled molten steel temperature, the predicted molten steel temperature and the measured molten steel temperature.

2. The method for online control of molten steel temperature in a steelmaking plant according to claim 1, wherein: in the process of obtaining relevant information of main process smelting equipment and ladles in a steelmaking plant, obtaining relevant information of the steelmaking plant, including: overall layout, equipment parameters, production steel grade and process path, and relevant information of the ladle, including: ladle lining map, brick thermal property parameters, and ladle turnover, wherein the relevant information of the steelmaking plant is used for selection of molten steel temperature influencing factors, classification of models, determination of standard smelting time and standard molten steel temperature drop, and the relevant information of the ladle is used for numerical simulation and classification of the ladle.

3. The method for online control of molten steel temperature in a steelmaking plant according to claim 2, wherein: in the process of obtaining the temperature drop of molten steel in each category of ladle, according to the ladle turnover, the ladle age, the empty ladle time and the baking time are taken as the thermal state of the ladle, and numerical simulation is performed on each category of ladle to obtain the temperature drop rate of each category of ladle in the static time period, and the historical production data is corrected to obtain the compensation value of the molten steel temperature of each section of different ladle thermal state, wherein for the empty ladle time, according to the overall production rhythm and the ladle turnover procedure, the ladles with empty ladle time less than 90 min, between 90-150 min and more than 150 min are marked respectively.

4. The method for online control of molten steel temperature in a steelmaking plant according to claim 3, wherein: in the process of constructing the molten steel temperature prediction model, the historical heats with better performance are screened out, all factors affecting the molten steel temperature from the converter to the continuous casting process are taken as the input, and the temperature at each process node is taken as the output to establish a data set for case-based reasoning; input a new heat, select a historical heat with the same ladle thermal state category and the highest similarity of the remaining influencing factors in the data set, and use the temperature at each process node of the historical heat as the target temperature of the new heat.

5. The method for online control of molten steel temperature in a steelmaking plant according to claim 4, wherein: In the process of obtaining the predicted molten steel temperature, the influencing factors of the molten steel temperature in each process section are analyzed, based on the data set, the variables other than the ladle thermal state with larger Pearson correlation coefficient and conforming to the mechanism analysis are selected as the dimensions for subsequent modeling; The K-means clustering analysis method is used, the number of clustering centers is increased by enumeration method, the clustering number when the error square sum decreases obviously is calculated, and the data set is divided by using the clustering number; Based on the BPNN model, the model of the data set to which the molten steel belongs is called for each molten steel to perform prediction according to the divided data set; For the ladle heat state of the abnormal furnace, the corresponding temperature correction value is subtracted from the model predicted temperature to obtain the corrected predicted molten steel temperature.

6. The online regulation method for molten steel temperature of a steelmaking plant according to claim 5, characterized in that: In the process of obtaining the predetermined temperature of all subsequent processes and the predicted temperature of the next process, for each molten steel, the liquidus temperature of the steel grade is taken as the starting point, the temperature drop of each stage is inversely calculated according to the superheat and the production plan before the end of the converter smelting, and the optimized temperature system of each node is formulated, and the target temperature of each node is obtained and updated in time according to the ladle thermal state, wherein the temperature prediction model is triggered during the smelting process from the converter to the continuous casting, and the molten steel temperature of the next process node is calculated according to the current known production conditions based on the last process node, to provide a basis for determining the temperature regulation parameters of each process.

7. The online regulation method for molten steel temperature of a steelmaking plant according to claim 6, characterized in that: In the process of adjusting the production plan, the steelmaking production plan issued by the superior system is received, and the job plan of the production site is generated according to the field production performance, logistics information and abnormal working conditions; And according to the received job plan, the steel grade and process path basic information of the specific furnace, and the smelting time and inter-process transportation time information of each process are obtained, and when the job plan changes, the corresponding adjustment is made.

8. An online temperature control system for molten steel in a steel plant, characterized in that, The system is used to realize the online regulation method for molten steel temperature of a steelmaking plant as claimed in any one of claims 1-7, and the system comprises: A data acquisition module is used to obtain the related information of the main process smelting equipment and the ladle of the steelmaking plant, and to obtain the data set for model training of each process by screening based on the historical production data of the main process; A data analysis module is used to classify the ladles of different thermal states based on the ladle lining temperature, and to obtain the temperature drop of the molten steel in each category of ladle; A model construction module is used to select the historical furnace based on the data set, to establish the molten steel temperature prediction model by using the case-based reasoning algorithm according to the ladle temperature drop, and to establish the molten steel temperature prediction model by using the K-means and BPNN algorithms; An online regulation module is configured to, based on the molten steel temperature prediction model and the molten steel temperature prediction model, obtain the subsequent all-process scheduled temperature and the next-process predicted temperature respectively before each molten steel smelting, during the smelting and after the scheduling time information is updated, and adjust the production plan according to the scheduled molten steel temperature, the predicted molten steel temperature and the measured molten steel temperature.

Images