Intelligent control system and method for high-purity molten steel smelting converter

Abstract

The invention relates to the technical field of steelmaking automation, in particular to an intelligent control system and method for a high-purity molten steel smelting converter, and the system comprises a data collection module which is used for collecting operation parameters related to the converter during molten steel smelting, and the operation parameters comprise molten steel temperature, gas components, oxygen amount and blowing time; the curve generation module is used for determining a change curve of the corresponding operation parameters during molten steel smelting according to the obtained operation parameters, and the change curve comprises a change curve corresponding to the molten steel temperature, the gas composition, the oxygen amount and the blowing time; the component display module is used for analyzing and processing the change curve according to the identified change curve of the operating parameters, and determining the occurrence probability of the operating parameters and a target value corresponding to the occurrence probability when the purity of the molten steel meets the requirement; and the production efficiency and the stability in the molten steel smelting process are improved.

Description

Technical Field

[0001] The invention relates to the technical field of steelmaking automation, and in particular to an intelligent control system and method for a high-purity molten steel smelting converter. Background Art

[0002] With the continuous development of steel industry technology and the increasing maturity of metallurgical process monomer technology, the production level has been further improved, the cost has been reduced, and the competitiveness has been enhanced. More stringent requirements have been put forward for the temperature control level of molten steel in the steelmaking-continuous casting process. The temperature of molten steel must be controlled within a narrow range to meet the requirements of efficient continuous casting production. Therefore, in modern continuous casting production, stable molten steel temperature conditions are the guarantee of stable continuous casting production, and appropriate molten steel superheat is an important condition for obtaining high-quality ingots. Effective control of molten steel temperature throughout the process is the key to ensuring smooth and orderly production rhythm. Therefore, improving the temperature control operation level of molten steel in the ladle throughout the process from converter tapping to refining outside the furnace and continuous casting turret, and making good preparations for molten steel temperature conditions before continuous casting are important aspects of the above technical ideas.

[0003] For example, Chinese patent publication number CN109857067A discloses a temperature coordination control system and method for multiple steelmaking processes in a big data environment, which belongs to the field of automatic control of steelmaking. The problem of reducing the temperature of the converter steel tapping while meeting the superheat condition of continuous casting is solved. The temperature coordination control method for multiple steelmaking processes proposed in the present invention uses the lowest possible converter steel tapping temperature to meet the superheat requirement of the continuous casting process. Compared with the existing method of increasing the converter steel tapping temperature to ensure the superheat of continuous casting, it can reduce the converter steel tapping temperature, reduce oxygen consumption, reduce cooling material consumption, reduce the oxygen content of molten steel, and reduce the phosphorus return of molten steel, thereby reducing the cost of the converter process.

[0004] However, the prior art does not take into account the relevant parameters of the molten steel, as well as the probability and target values ​​of the relevant parameters during operation, when controlling the molten steel, resulting in the inability to make timely adjustments based on the current situation when automatic control of steelmaking is selected. Summary of the invention

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is: an intelligent control system for a high-purity molten steel smelting converter, including: a data acquisition module for collecting operating parameters related to the converter during molten steel smelting, the operating parameters including molten steel temperature, gas composition, oxygen content and blowing time.

[0006] The curve generation module is used to determine the change curve of the corresponding operating parameters during molten steel smelting according to the acquired operating parameters, and the change curve includes the change curve corresponding to the molten steel temperature, gas composition, oxygen amount and blowing time.

[0007] The component display module is used to analyze and process the change curve of the identified operating parameters, and determine the probability of occurrence of the operating parameters and the target value corresponding to the probability of occurrence when the purity of the molten steel meets the requirements.

[0008] The data storage module is used to store the occurrence probability and target value of the operating parameters and generate a control strategy based on the stored data.

[0009] An intelligent control method for a converter for smelting high-purity molten steel comprises: S1, collecting operating parameters of the converter during smelting molten steel, including molten steel temperature, gas composition, oxygen content and blowing time.

[0010] S2, based on the collected operating parameters, draw the change curves of molten steel temperature, gas composition, oxygen content and blowing time.

[0011] S3, identify the key information in the operating parameter change curve, analyze the change in molten steel temperature, record the extreme point data, and perform corresponding processing according to the preset temperature range; determine the change trend of gas composition, calculate the change ratio of key components, and draw a curve of the change of composition coefficient over time; determine the degree of change of oxygen amount according to oxygen flow rate and contact area, and draw a curve of the change of oxygen amount.

[0012] S4, based on the change curve of the operating parameters, extract the characteristic values ​​and determine the corresponding characteristic value set when the preset molten steel purity is reached; analyze the probability distribution of the characteristic values ​​in the smelting process, identify the peak and skewed data as the target characteristic set; iteratively process the target characteristic set and output the final target value.

[0013] S5, storing the occurrence probability and target value of the operating parameters; generating a control strategy based on the stored data, and selecting the strategy with the highest similarity to the target value as the control strategy currently being processed.

[0014] The beneficial effects of the present invention are as follows: the present invention collects the operating parameters in the steelmaking process in real time, including the steelmaking temperature, gas composition, oxygen content and blowing time, and draws a change curve according to these parameters, analyzes and processes the curve, and determines the probability of occurrence of the operating parameters and the corresponding target value when the steelmaking purity reaches the requirement; the system also has a data storage function, and can generate an optimized control strategy according to the stored data; thereby, the state of the steelmaking process can be grasped in real time, providing a basis for optimized control; the purity control of the steelmaking is made more precise, and the product quality is improved; the strategy with the highest similarity to the target value is selected as the control strategy for the current processing, thereby improving the steelmaking efficiency and production benefits; and the automatic and intelligent monitoring and control of the steelmaking process is realized, which reduces manual intervention and improves production efficiency and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0016] Figure 1 The system diagram of an intelligent control system for a converter for smelting high-purity molten steel.

[0017] Figure 2 The present invention is a logic diagram of an intelligent control system for a converter for smelting high-purity molten steel.

[0018] Figure 3 The present invention is a flow chart of an intelligent control method for a converter for smelting high-purity molten steel. DETAILED DESCRIPTION

[0019] The embodiments of the present invention are described in detail below. The embodiments described below are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. If no specific techniques or conditions are specified in the embodiments, the techniques or conditions described in the literature in the art or the product specifications are used.

[0020] See also Figure 1 , Figure 2 , an intelligent control system for a high-purity molten steel smelting converter, comprising: a data acquisition module collects converter operating parameters in real time through various sensors, such as temperature sensors, component analyzers, flow meters, etc.; the sensors are connected to the data acquisition module through data cables or wireless networks to ensure the real-time and accuracy of the data; the data acquisition module transmits real-time data to a curve generation module; the curve generation module draws a curve of each operating parameter changing with time based on the received data; the curve generation module transmits the generated change curve data to a component display module; the component display module identifies and analyzes the curve data to determine the occurrence probability and target value of each operating parameter when the purity of the molten steel meets the standard; the component display module transmits the occurrence probability and target value obtained by the analysis to the data storage module; the data storage module stores the data in a database for subsequent query, analysis and use.

[0021] The data acquisition module is used to collect the operating parameters related to the converter during molten steel smelting, including molten steel temperature, gas composition, oxygen content and blowing time.

[0022] The curve generation module is used to determine the change curve of the corresponding operating parameters during molten steel smelting according to the acquired operating parameters, and the change curve includes the change curve corresponding to the molten steel temperature, gas composition, oxygen amount and blowing time.

[0023] The component display module is used to analyze and process the change curve of the identified operating parameters, and determine the probability of occurrence of the operating parameters and the target value corresponding to the probability of occurrence when the purity of the molten steel meets the requirements.

[0024] The data storage module is used to store the occurrence probability and target value of the operating parameters and generate a control strategy based on the stored data.

[0025] Use a molten steel thermometer to measure the temperature of molten steel. A molten steel thermometer is an instrument specifically used to measure the temperature of molten steel; for example, a microcomputer molten steel thermometer, which can quickly and accurately measure the temperature of molten steel. These thermometers are usually made of high-temperature resistant materials to ensure that they can still work normally under extreme temperatures.

[0026] Gas composition is measured using a gas composition detector, which is used to detect the composition of the gas, including harmful gases and flammable gases. In the steel production process, the detection of gas composition is crucial for safety and environmental control. For example, a portable gas composition detector can easily perform real-time detection on site.

[0027] The amount of oxygen is measured using an oxygen detector, which is an instrument specifically used to measure oxygen content. In the steel smelting process, oxygen control has an important impact on reaction efficiency and product quality. The oxygen detector can monitor the oxygen concentration in real time to ensure that it is within the appropriate range.

[0028] As for the blowing time, a timer is used to calculate the corresponding time during blowing, and this time is recorded as the blowing time to determine the corresponding time point in the molten steel processing process.

[0029] The change curve of the molten steel temperature in the curve generation module is expressed as: obtaining the change of the molten steel temperature corresponding to the corresponding blowing time, and obtaining the change curve of the molten steel temperature, so as to determine the intensity of the chemical reaction in the current converter.

[0030] When paying attention to the temperature change curve of molten steel, the main focus is on the temperature change rate per unit time. The temperature change of molten steel over time is monitored by installing a temperature sensor on the furnace wall or in the molten steel. Combining the specific heat capacity and mass information, the heat absorbed by the molten steel in a certain period of time can be inferred. The heat absorbed by the molten steel at the corresponding time is obtained by calculating the product of the specific heat capacity, mass and the temperature change over the corresponding time.

[0031] The heat, specific heat capacity and mass absorbed by the molten steel are obtained, and the change in the molten steel temperature is obtained. The change in the molten steel temperature is used as the ordinate and the moment corresponding to the blowing time is used as the abscissa to obtain a change curve of the molten steel temperature.

[0032] The calculation method of the change in molten steel temperature is expressed as: Where ΔT(t) represents the change in temperature of the molten steel as a function of time t, which is the change in temperature of the molten steel from the initial time t = 0 to the current time t; Q1(t′) represents the heat absorbed by the molten steel at time t′, where t′ is the integral variable, m represents the mass of the molten steel, and c p represents the specific heat capacity of molten steel, that is, the energy required to raise the temperature of molten steel by 1°C per unit mass; dt′ represents the integral differential element, which represents a small increment of the time interval. In the process of integration, dt′ is used to accumulate the contribution of heat input at different time points to the temperature change; represents the integral symbol, which means integrating the time t′ from the initial time 0 to the current time t; then the change in the molten steel temperature calculated at this time is the change curve of the molten steel temperature.

[0033] After obtaining the variation curve of the molten steel temperature, the currently identified molten steel temperature needs to be processed, so the variation curve of the molten steel temperature also includes the following processing methods.

[0034] Record the data corresponding to each extreme point in the temperature change curve of the molten steel, and determine whether the molten steel temperature at each extreme point is within the preset temperature range. If it is within the preset temperature range, output the average temperature of the extreme point and the adjacent points; if it is greater than the upper limit of the preset temperature range, determine the slope change rate of the temperature change curve of the molten steel at the position close to the extreme point, and output the maximum value of the slope change rate; if it is less than the lower limit of the preset temperature range, obtain the median of the adjacent points of the extreme point, and output the corresponding median.

[0035] The purpose of this processing step is to conduct a detailed analysis of the temperature change curve of the molten steel in order to understand the characteristics of its temperature change and make corresponding treatments based on these characteristics. Specifically, the purpose of this step includes the following.

[0036] Determine whether the temperature of the extreme point is within the preset range: This is to determine whether the temperature of the molten steel is within a safe or ideal range. If the temperature is too high or too low, it may affect the production process and product quality.

[0037] Calculate the average temperature of the extreme point and adjacent points: When the temperature of the extreme point is within the preset range, outputting the average temperature of the extreme point and adjacent points can better reflect the temperature conditions of the area and help control the temperature more accurately.

[0038] Determine the rate of change of the slope value of the steel liquid temperature curve near the extreme point: When the temperature of the extreme point exceeds the upper limit of the preset range, analyzing the rate of change of the slope value can help understand the rate and stability of temperature change, which is very important for predicting temperature trends and taking corresponding control measures.

[0039] Get the median of the adjacent points of the extreme point: When the temperature of the extreme point is lower than the lower limit of the preset range, outputting the median of the adjacent points can provide a more robust temperature indicator and avoid deviations caused by the extreme point temperature being too low.

[0040] Assume that the preset molten steel temperature range is 1500℃ to 1600℃. There is an extreme point in the recorded molten steel temperature curve, and its temperature is 1550℃, which is within the preset range; therefore, the temperature average of this extreme point and its adjacent points will be calculated, and assuming the average value is 1540℃, this value can be used to represent the temperature condition of this area.

[0041] If the temperature of another extreme point is 1650°C, which exceeds the upper limit of the preset range, in this case, the slope value change rate of the temperature change curve near this extreme point will be analyzed; assuming that the maximum value of the slope value change rate is found to be 0.5°C / s, this means that the temperature rise rate in this area is very fast, and measures may need to be taken to control the temperature rise.

[0042] If there is another extreme point with a temperature of 1450°C, which is lower than the lower limit of the preset range, in this case, the median temperature of the adjacent points of this extreme point will be obtained. Assuming the median is 1480°C, this value can be used to represent the temperature condition of this area to avoid deviations caused by extreme point temperatures being too low.

[0043] Through such processing steps, the changing characteristics of molten steel temperature can be more fully understood, and the production process and product quality can be optimized based on these characteristics.

[0044] When processing the gas components, the identified gas components may include carbon monoxide, carbon dioxide, oxygen, nitrogen, and other gases such as sulfide SO2, nitrogen oxide NOx and other gas components. At this time, under different current temperature change rates, the overall amount of gas generated will also be different, and the content of each gas in the component will also be different. Then, it is necessary to identify whether the ratio of gas components generated under the current temperature change rate is the same as the preset ratio. If different, identify the corresponding temperature change rate and related gas components to predict the impurity components that may exist after the current molten steel is smelted, and whether the purity of the molten steel can meet the preset requirements.

[0045] The gas composition change curve evaluates the efficiency and quality of the smelting process by analyzing the changes in gas elements during the smelting process. The detection content is to identify the gas elements in the molten steel, especially oxygen and nitrogen. The content of these gas elements in the molten steel has an important influence on the quality and performance of the steel. For example, too high oxygen content may increase the brittleness of the steel, while too high nitrogen content may affect the strength and toughness of the steel.

[0046] Therefore, the method for identifying the change curve of the gas composition is to determine the key components of the gas discharged during molten steel smelting, and obtain the change curve of the gas composition based on the change trend of the key components of the gas.

[0047] The key components may be the contents of oxygen, nitrogen, sulfur and carbon to determine the main components of the current exhaust gas.

[0048] The change trend of the key components is expressed as follows: for each key component existing in the gas composition, the change ratio value of the key component between different time points is calculated; the change ratio value is the ratio of the difference between the component content at the current time point and the component content at the initial time point.

[0049] Compare whether each identified key component is within the preset range, and obtain the composition coefficient of the gas component according to the change ratio value of the currently identified key component, and output the change of the composition coefficient with time point as a gas composition change curve.

[0050] The collected gas samples are sent to the laboratory for analysis; using gas analyzers, such as gas chromatographs, infrared spectrometers, etc.; the content of each key gas component in the sample is measured; the content of each key component at each sampling time point is recorded; with time as the horizontal axis and gas content as the vertical axis, a curve of the change of each key component over time is drawn; each point on the curve represents the content of the key component at a specific time point.

[0051] The setting method of the change ratio value of the key component is as follows: for each key component, the ratio of the current key component content to the preset component range is calculated; if the key component content is within the preset component range, the change ratio value is 1; if the key component content is lower than the minimum value of the preset component range, the change ratio value is the ratio of the current key component content to the minimum value of the preset component range; if the key component content is higher than the maximum value of the preset component range, the change ratio value is the ratio of the current key component content to the maximum value of the preset component range.

[0052] The weighted average of the change ratio values ​​of all key components is calculated to obtain the component coefficient. The weight of the weighted calculation is determined according to the importance of each key component and the degree of its overall impact on the molten steel.

[0053] The calculated composition coefficient is plotted as a curve along with the change in time points; the horizontal axis represents time, and the vertical axis represents the composition coefficient; each point on the curve represents the composition coefficient of the gas component at a specific time point.

[0054] The purpose of the current step is to monitor and evaluate the changes in key gas components: by calculating the change ratio values ​​of key components at different time points, the changing trend of gas components during the production process can be understood; this is crucial to controlling product quality and production efficiency.

[0055] Determine whether the gas composition is within the preset range: By comparing whether each key component identified is within the preset range, it can be determined whether the production process is in a normal state; if the gas composition is outside the preset range, measures may need to be taken to make adjustments.

[0056] Evaluate the overall impact of gas composition on molten steel: By calculating the weighted average of the change ratio values ​​of all key components and obtaining the composition coefficient, the overall impact of gas composition on molten steel can be comprehensively evaluated; this is of great significance for optimizing the production process and product quality.

[0057] Provide real-time data support and visual display: By plotting the changes in composition coefficients over time, real-time data support and visual display can be provided to production personnel, helping them to better understand the changes in gas composition during the production process and make more informed decisions.

[0058] For example, suppose that in the steel production process, it is necessary to monitor the changes in the three key gas components of oxygen, carbon dioxide and carbon monoxide; by collecting gas samples and analyzing them, the content of these three components at different time points can be obtained; then, based on the preset component range, the change ratio of each component can be calculated; finally, the composition coefficient is obtained through weighted calculation, and its changes over time are plotted into a curve; if the curve shows that the composition coefficient suddenly increases at a certain point in time, this may mean that the gas composition has undergone abnormal changes, and production personnel can take corresponding measures to make adjustments accordingly.

[0059] The oxygen amount change curve is determined by obtaining the current input oxygen flow rate and the contact area between oxygen and molten steel to determine the degree of change of the current oxygen amount. The degree of change of the oxygen amount is used as the output oxygen amount change curve to determine whether the current input oxygen amount can make the chemical reaction proceed normally.

[0060] The degree of change of the oxygen amount is expressed as follows: the oxygen flux is calculated based on the oxygen flow rate and the contact area between oxygen and molten steel; the reaction rate is obtained based on the oxygen flux, reaction rate constant and carbon concentration in the molten steel; the oxygen flux and reaction rate are used as the degree of change of the output oxygen amount.

[0061] Oxygen flow FO2: Use a flow meter to monitor the flow of oxygen in real time.

[0062] The contact area A between oxygen and molten steel: Estimate the actual contact area between oxygen and molten steel based on the design parameters of the spray gun and the position of the spray gun in the molten steel.

[0063] To calculate the oxygen flux, the amount of oxygen passing through a unit area per unit time, the following formula can be used: JO2 = FO2 / A; where JO2 represents the oxygen flux, FO2 is the oxygen flow rate, and A is the contact area between oxygen and molten steel.

[0064] To construct an oxygen quantity variation curve, the oxygen flux variation curve over time can be drawn by recording the data of oxygen flow rate and contact area at different time points; this curve can help analyze the variation trend of oxygen quantity and adjust the oxygen flow rate accordingly.

[0065] The chemical reaction rate can be estimated based on the oxygen flux and the carbon content in the molten steel. In general, the reaction rate can be expressed as: r = k × JO2 × C; where r is the reaction rate, k is the reaction rate constant, and C is the carbon concentration in the molten steel.

[0066] By comparing the actual reaction rate with the expected reaction rate, if it is found that the reaction rate is lower than expected, it is necessary to increase the oxygen flow rate or adjust the spray gun position to increase the contact area; conversely, if the reaction rate is too high, the oxygen flow rate should be appropriately reduced.

[0067] Use a PID controller or other type of controller to automatically adjust the oxygen flow to maintain a steady reaction rate.

[0068] The data from the temperature sensor, oxygen flow meter and pressure sensor are integrated into the control system to form a closed-loop control system.

[0069] By using temperature changes as feedback signals, the oxygen flow adjustment strategy is further optimized to ensure the normal progress of the chemical reaction.

[0070] If r<target r, increase FO2 or adjust the gun position to increase A; if r>target r, reduce FO2 or adjust the gun position to reduce A.

[0071] Continuously adjust until the optimal reaction rate is achieved and maintained within the target range, and use historical data to continuously optimize control parameters and improve control accuracy.

[0072] The method for realizing the occurrence probability of the operating parameters and the target value corresponding to the occurrence probability includes: extracting the characteristic values ​​of the molten steel temperature, gas composition and oxygen content based on the change curve of the operating parameters, determining the corresponding multiple groups of characteristic values ​​of the molten steel temperature, gas composition and oxygen content when the molten steel reaches the preset molten steel purity, recording them as the initial characteristic set, and determining the probability distribution of the initial characteristic set in the smelting process.

[0073] At this time, the way to mark the initial feature set is to convert the molten steel temperature, gas composition and oxygen content into dimensionless feature values ​​and then mark them separately to determine the probability distribution of the corresponding initial feature set under the corresponding molten steel purity.

[0074] The probability distribution calculation method is to calculate the current data by using the kernel density estimation on the initial feature set currently obtained, such as Among them, f h (x) represents the kernel density estimation function, x represents the element in the initial feature set to be estimated, and x i represents the element of the i-th initial feature set, n represents the number of elements in the initial feature set, K() represents the kernel function, which is usually a symmetric function used to weight the contribution of the elements near each initial feature set, h represents the bandwidth function, which controls the width of the kernel function and affects the smoothness of the estimate, i = 1, 2,,, n; for the bandwidth function, h = 1.07×σ x ×n -1 / 5 ; Among them, σ x represents the standard deviation of the initial feature set.

[0075] According to the calculated value of the kernel density estimation function, we can know the probability distribution of molten steel temperature, gas composition and oxygen content under different molten steel purity, which can serve as a marker for the relevant status of the current molten steel smelting. At the same time, we extract the obvious peaks and skewed data in the data output by the kernel density estimation function, and identify these data to determine whether these peaked and skewed data can meet the current expectations of molten steel smelting, and whether there will be abnormalities and other situations in the current operating parameters when these values ​​appear, so as to find the target values ​​that need to be monitored and identified.

[0076] According to the probability distribution of the initial feature set, the data with kurtosis and skewness in the data after the initial feature set is calculated is used as the target feature set, the target feature set is iterated, and the data after the iteration is output as the target value.

[0077] The calculation method for iterating the target feature set is to first initialize the predicted values ​​corresponding to all elements in the target feature set. The value of the initialized predicted value is the mean of the elements in the target feature set. The residuals between all elements in the target feature set and the corresponding predicted values ​​are calculated, and the decision tree corresponding to the residuals is trained. The result output by the decision tree is the estimated value related to the residuals, so that the decision tree fits the residuals as much as possible. When the residuals are minimized, the predicted values ​​can accurately match the values ​​of the elements in the target feature set. The decision tree and the predicted values ​​are continuously iterated, and the data after the iteration is output as the target value. The number of iterations is m, m=1, 2,,,M.

[0078] The residual between all elements in the target feature set and the corresponding predicted values ​​is expressed as follows.

[0079] Among them, r j represents the residual of the jth element in the target feature set, yj represents the value of the jth element in the target feature set, represents the predicted value of the jth element in the target feature set after the mth iteration, It represents the predicted value of the j-th element in the target feature set after the m-1th iteration, where j represents the number of elements in the target feature set, j=1,2,,,J. At this time, in order to prevent the denominator from being zero, the set predicted values ​​are not 1.

[0080] When the residual corresponding to the target feature set is less than the preset residual, the predicted value is updated. At this time, the updated predicted value is the output target value.

[0081] Among them, λ represents the learning rate, h j,m Represents the estimated value of the decision tree corresponding to the jth element in the target feature set after the mth iteration.

[0082] By continuously training new decision trees to fit the residuals and gradually updating the predicted values, the estimated value of the decision tree is an estimate of the difference between the current predicted value and the true value. By updating the predicted values ​​in a weighted manner, the model can better fit the training data after each round of iteration; the learning rate controls the update amplitude in each round of iteration, which helps to avoid overfitting; the target value obtained in this way can reflect the trend and related conditions of the current operating parameters, which is convenient for judging whether the current molten steel smelting process is normal and whether relevant processing is required.

[0083] The generation method of the control strategy includes: according to the obtained probability of occurrence of the operating parameters and the target value corresponding to the probability of occurrence, a strategy set whose similarity with the target value is greater than a preset similarity threshold is selected from the preset strategies, and the strategy with the highest similarity between the strategy set and the operating parameters is used as the control strategy for the current processing to judge the current processing process.

[0084] The method for calculating the similarity at this time is to calculate the target value and the corresponding value in the preset strategy to obtain the Pearson correlation coefficient, and the preset similarity threshold set at this time is 0.7. All strategies greater than the threshold after calculation are combined into a strategy set, and the probability of occurrence of the relevant operating parameters in the strategy set is compared with the probability of occurrence of the current operating parameters. The part with the maximum value of the calculated Pearson coefficient is output to determine the current control strategy.

[0085] like Figure 3 As shown, the present invention also provides an intelligent control method for a high-purity molten steel smelting converter, comprising: S1, collecting operating parameters of the converter during molten steel smelting, including molten steel temperature, gas composition, oxygen content and blowing time.

[0086] S2, based on the collected operating parameters, draw the change curves of molten steel temperature, gas composition, oxygen content and blowing time.

[0087] S3, identify the key information in the operating parameter change curve, analyze the change in molten steel temperature, record the extreme point data, and perform corresponding processing according to the preset temperature range; determine the change trend of gas composition, calculate the change ratio of key components, and draw a curve of the change of composition coefficient over time; determine the degree of change of oxygen amount according to oxygen flow rate and contact area, and draw a curve of the change of oxygen amount.

[0088] S4, based on the change curve of the operating parameters, extract the characteristic values ​​and determine the corresponding characteristic value set when the preset molten steel purity is reached; analyze the probability distribution of the characteristic values ​​in the smelting process, identify the peak and skewed data as the target characteristic set; iteratively process the target characteristic set and output the final target value.

[0089] S5, storing the occurrence probability and target value of the operating parameters; generating a control strategy based on the stored data, and selecting the strategy with the highest similarity to the target value as the control strategy currently being processed.

[0090] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations on the present invention. A person skilled in the art may make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention and they are still covered by the protection scope of the present invention.

Claims

1. An intelligent control system for a high-purity molten steel smelting converter, characterized in that: include: The data acquisition module is used to collect the operating parameters related to the converter during molten steel smelting, including molten steel temperature, gas composition, oxygen content and blowing time; A curve generation module is used to determine the change curve of the corresponding operating parameters during molten steel smelting according to the acquired operating parameters, and the change curve includes the change curve corresponding to the molten steel temperature, gas composition, oxygen amount and blowing time; The component display module is used to analyze and process the change curve of the identified operating parameters, and determine the occurrence probability of the operating parameters and the target value corresponding to the occurrence probability when the purity of the molten steel meets the requirements; The data storage module is used to store the occurrence probability and target value of the operating parameters and generate a control strategy based on the stored data.

2. The intelligent control system for a high-purity molten steel smelting converter according to claim 1 is characterized in that: The temperature variation curve of the molten steel is expressed as follows: the heat absorbed by the molten steel, the specific heat capacity, and the mass are obtained to obtain the temperature variation of the molten steel. The temperature variation of the molten steel is taken as the ordinate, and the moment corresponding to the blowing time is taken as the abscissa to obtain the temperature variation curve of the molten steel.

3. The intelligent control system for a high-purity molten steel smelting converter according to claim 2 is characterized in that: The calculation method of the change in molten steel temperature is expressed as: Where ΔT(t) represents the change in temperature of the molten steel as a function of time t; Q1(t′) represents the heat absorbed by the molten steel at time t′, t′ is the integral variable, m represents the mass of the molten steel, c p represents the specific heat capacity of molten steel; dt′ represents the integral differential element, which represents the small increment of the time interval. Represents the integral symbol.

4. The intelligent control system for a high-purity molten steel smelting converter according to claim 2 is characterized in that: The change curve of molten steel temperature also includes the following processing methods: Record the data corresponding to each extreme point in the temperature change curve of the molten steel, and determine whether the molten steel temperature at each extreme point is within the preset temperature range. If it is within the preset temperature range, output the average temperature of the extreme point and the adjacent points; if it is greater than the upper limit of the preset temperature range, determine the slope change rate of the temperature change curve of the molten steel at the position close to the extreme point, and output the maximum value of the slope change rate; if it is less than the lower limit of the preset temperature range, obtain the median of the adjacent points of the extreme point, and output the corresponding median.

5. The intelligent control system for a high-purity molten steel smelting converter according to claim 1 is characterized in that: The method of identifying the change curve of the gas composition is to determine the key components of the gas discharged during molten steel smelting, and obtain the change curve of the gas composition according to the change trend of the key components of the gas; The change trend of the key components is expressed as follows: for each key component present in the gas composition, the change ratio value of the key component between different time points is calculated; Compare whether each identified key component is within the preset range, and obtain the composition coefficient of the gas component according to the change ratio of the currently identified key component, and plot the change of the calculated composition coefficient with time points into a curve; the horizontal axis represents time, and the vertical axis represents the composition coefficient; each point on the curve represents the composition coefficient of the gas component at a specific time point.

6. The intelligent control system for a high-purity molten steel smelting converter according to claim 1 is characterized in that: The change curve of the oxygen amount determines the change degree of the current oxygen amount by obtaining the current input oxygen flow rate and the actual contact area between oxygen and molten steel, and uses the change degree of the oxygen amount as the output oxygen amount change curve.

7. The intelligent control system for a high-purity molten steel smelting converter according to claim 6 is characterized in that: The degree of change of the oxygen amount is expressed as follows: the oxygen flux is calculated based on the oxygen flow rate and the contact area between oxygen and molten steel; the reaction rate is obtained based on the oxygen flux, reaction rate constant and carbon concentration in the molten steel; the oxygen flux and reaction rate are used as the degree of change of the output oxygen amount.

8. The intelligent control system for a high-purity molten steel smelting converter according to claim 1 is characterized in that: The method for realizing the occurrence probability of the operating parameter and the target value corresponding to the occurrence probability includes: extracting characteristic values ​​of molten steel temperature, gas composition and oxygen content based on the change curve of the operating parameter, determining multiple sets of corresponding characteristic values ​​of molten steel temperature, gas composition and oxygen content when the molten steel reaches a preset molten steel purity, recording them as initial characteristic sets, and determining the probability distribution of the initial characteristic sets in the smelting process; According to the probability distribution of the initial feature set, the data with kurtosis and skewness in the data after the initial feature set is calculated is used as the target feature set, the target feature set is iterated, and the data after the iteration is output as the target value.

9. The intelligent control system for a high-purity molten steel smelting converter according to claim 1 is characterized in that: The method for generating the control strategy includes: selecting a strategy set from the preset strategies whose similarity with the target value is greater than a preset similarity threshold according to the obtained occurrence probability of the operating parameter and the target value corresponding to the occurrence probability, and using the strategy with the highest similarity between the strategy set and the operating parameter as the control strategy currently being processed.

10. An intelligent control method for a high-purity molten steel smelting converter, using the intelligent control system for a high-purity molten steel smelting converter as claimed in claim 1, characterized in that: include: S1, collecting the operating parameters of the converter during molten steel smelting, including molten steel temperature, gas composition, oxygen content and blowing time; S2, drawing the variation curves of molten steel temperature, gas composition, oxygen content and blowing time according to the collected operating parameters; S3, identify key information in the operating parameter change curve, analyze the change in molten steel temperature, record extreme point data, and perform corresponding processing according to the preset temperature range; determine the change trend of gas composition, calculate the change ratio value of key components, and draw a curve of composition coefficient change over time; determine the degree of change of oxygen amount according to oxygen flow rate and contact area, and draw a curve of oxygen amount change; S4, based on the change curve of the operating parameters, extract the characteristic values ​​and determine the corresponding characteristic value set when the preset molten steel purity is reached; analyze the probability distribution of the characteristic values ​​in the smelting process and identify the peak state and skewed data as the target characteristic set; Iterate the target feature set and output the final target value; S5, storing the occurrence probability and target value of the operating parameters; generating a control strategy based on the stored data, and selecting the strategy with the highest similarity to the target value as the control strategy currently being processed.

Images