Converter roughing slag detection method

Abstract

The invention relates to a converter roughing slag detection method, belongs to the field of material thermophysical property testing and data processing, and solves the problem of low judgment accuracy of a steel slag condition in a converter tapping process in the prior art. The method comprises the following steps: when a tapping signal of a converter is monitored, acquiring a plurality of wavelengths and corresponding gray values of molten steel at a tapping hole of the converter; based on each wavelength and the gray value thereof, obtaining spectral radiation brightness corresponding to each wavelength by using pre-calibrated data; obtaining the spectral emissivity of the molten steel in the current converter by using a material spectral emissivity temperature separation inversion method based on each wavelength and the spectral radiance thereof; and when the spectral emissivity of the molten steel in the converter is larger than a set threshold value, it is judged that the slag content in the molten steel exceeds a normal value, and the converter is controlled to stop tapping. The spectral emissivity and the temperature of the material can be obtained through separated calculation only depending on the heat capacity function of the material, calculation is simple, and reliability is high.

Description

Technical Field

[0001] The present invention relates to the technical fields of testing and steelmaking, and particularly relates to a method for detecting slag falling in a converter. Background Art

[0002] The converter slag generated during the converter steelmaking process has many adverse effects on the subsequent treatment of molten steel, including causing the return of silicon and phosphorus in the molten steel and the diffusion of oxygen in the top slag. Seriously, it can lead to the out-of-specification of the molten steel composition and cause waste products. Therefore, effective slag blocking operation during converter tapping to prevent the steel slag from entering the subsequent treatment process of the molten steel is an important means to ensure the quality of the molten steel. It can reduce the return of phosphorus and sulfur in the molten steel, improve the recovery rate of alloys, reduce inclusions in the steel, improve the cleanliness of the molten steel, and also provide good conditions for refining the molten steel. In addition, reducing the amount of slag falling can also reduce the consumption of related consumables such as deoxidizers and alloys, extend the service life of the ladle, and play a role in reducing production costs. In order to improve the quality of the molten steel, during the process of pouring the molten steel from the converter, the operator needs to continuously observe the changes in the molten steel. When the brightness of the poured molten steel increases, it proves that steel slag has appeared. This requires the operator to have certain work experience, so there is a great deal of subjectivity.

[0003] At the present stage, there is a technical solution for judging the timing of slag falling based on the difference in emissivity between molten steel and steel slag. However, for high-temperature spectral emissivity measurement, it is necessary to measure the temperature and emissivity of the material simultaneously, and these two parameters affect and depend on each other. Therefore, it is necessary to solve the problem of how to accurately measure these two parameters simultaneously. However, no matter whether non-contact or contact temperature measurement is used, there will always be a large systematic error in the measurement results. Therefore, there is an urgent need for a new method for detecting slag falling that can accurately separate temperature and emissivity to replace the existing technology for judging slag falling. Summary of the Invention

[0004] In view of the above analysis, the embodiments of the present invention aim to provide a method for detecting slag falling in a converter to solve the problem of low accuracy in judging the situation of steel slag during the existing converter tapping process.

[0005] The object of the present invention is mainly achieved through the following technical solutions:

[0006] The present invention provides a method for detecting slag falling in a converter, including the following steps:

[0007] When a tapping signal of the converter is detected, obtain multiple wavelengths and corresponding gray values of the molten steel at the tapping port of the converter;

[0008] Based on each of the wavelengths and its gray value, use pre-calibrated data to obtain the spectral radiance corresponding to each of the wavelengths;

[0009] Based on each of the wavelengths and their spectral radiance, the spectral emissivity of the molten steel in the current converter is obtained using the material spectral emissivity temperature separation inversion method;

[0010] When the spectral emissivity of the molten steel inside the converter is greater than the set threshold, it is determined that the slag content in the molten steel exceeds the normal value, and the converter is controlled to stop tapping.

[0011] Further, when the tapping signal appears, the wavelengths and their gray values of the molten steel in the converter are obtained, including: when the tapping signal is monitored, the rotary motion of the converter is controlled;

[0012] When the tapping inclination angle of the converter reaches the preset angle, the tapping hole is opened. When the molten steel flows out of the converter nozzle, a spectrometer is used to obtain multiple wavelengths of the flowing molten steel and their corresponding gray values.

[0013] Further, the determination that the slag content in the molten steel exceeds the normal value and the control of the converter to stop tapping include: when the spectral emissivity of the molten steel is greater than the set threshold, an alarm value is displayed on the control panel, and the alarm buzzer and alarm light are started. At the same time, the tapping hole is automatically closed according to the pre-setting or the operator manually closes the tapping hole, and the converter is rotated back to the initial position.

[0014] Further, the receiving of each of the wavelengths and the corresponding spectral radiance and the obtaining of the spectral emissivity of the material in the converter using the temperature-emissivity separation inversion method include:

[0015] Based on the spectral radiance of each of the wavelengths, the blackbody temperature of the ideal blackbody at each of the wavelengths is obtained using the Planck blackbody radiation formula;

[0016] Based on the spectral radiance corresponding to each wavelength and the blackbody temperature corresponding to each wavelength, the temperature of the inverted material and the spectral emissivity corresponding to each wavelength are obtained using the EM optimization strategy.

[0017] Further, in the EM optimization strategy, in one iteration process:

[0018] The initial temperature of this iteration and the initial spectral emissivity corresponding to each wavelength are obtained;

[0019] In the maximization step of the EM optimization strategy: Based on the initial spectral emissivity of each of the wavelengths, the free energy minimization method is used as the maximum likelihood function of the EM optimization strategy to calculate the optimal temperature as the hidden parameter of the EM optimization strategy;

[0020] In the expectation step of the EM optimization strategy: Based on the optimal temperature, the spectral emissivity of each wavelength at this temperature is obtained, and the spectral emissivity fitting smoothing method is used to obtain the smoothed spectral emissivity of each wavelength as the spectral emissivity expectation value corresponding to each wavelength;

[0021] The optimal temperature is used as the initial temperature for the next iteration, and the spectral emissivity expectation value corresponding to each wavelength is used as the initial spectral emissivity for the next iteration;

[0022] When the spectral emissivity expectation value corresponding to each wavelength and the initial spectral emissivity of each wavelength satisfy the convergence condition, the obtained optimal temperature and the spectral emissivity of each wavelength are used as the inverted temperature and the spectral emissivity corresponding to each wavelength, and the iteration ends.

[0023] Furthermore, the optimal temperature is calculated by using the free energy minimization method based on the initial emissivity of each wavelength. The formula of the free energy minimization method is:

[0024]

[0025]

[0026] where T opt is the optimal temperature; ΔA is the internal energy change; C V is the heat capacity of the material; is the initial spectral emissivity of wavelength λ i ; is the spectral emissivity of the material at wavelength λ i at temperature T opt ;

[0027] Based on the optimal temperature, the spectral emissivity of each wavelength at this temperature is obtained by using the following formula

[0028]

[0029] Furthermore, the fitting and smoothing of the spectral emissivity of each wavelength by using the spectral emissivity fitting smoothing method includes:

[0030] Based on the spectral emissivity corresponding to each wavelength, the single-wavelength radiation entropy of each wavelength is obtained:

[0031] The high-order polynomial fitting method is used to obtain the smoothed wavelength radiation entropy for the single-wavelength radiation entropy of each wavelength;

[0032] Based on the smoothed wavelength radiation entropy, the fitted and smoothed spectral emissivity is obtained.

[0033] Further, the high-order polynomial fitting method includes: for each wavelength, using a kernel function as a weight to perform local polynomial kernel regression to obtain a high-order polynomial as a regression function, and using the regression function to calculate the single-wavelength radiation entropy of each of the wavelengths as the smoothed wavelength radiation entropy; wherein, the local polynomial kernel regression includes: selecting a preset number of wavelengths adjacent to this wavelength, and performing polynomial fitting based on the kernel function within this range;

[0034] Based on the smoothed wavelength radiation entropy, the following formula is used to obtain the fitted and smoothed spectral emissivity of the wavelength range to be fitted:

[0035]

[0036] Wherein, is the spectral emissivity of the smoothed wavelength λ i ; is the single-wavelength radiation entropy of the smoothed wavelength λ i .

[0037] Further, the regression function is:

[0038] H(λ i ) = a0 + a1λ i + a2λ i 2 + a3λ i 3 + … + a n λ i n

[0039] Wherein, λ i is the wavelength value of the i-th wavelength; a0, a1…a n are the coefficients of the polynomial function; n is the order of the polynomial function;

[0040] The coefficients of the polynomial function are calculated using the following formula:

[0041] θ = (X T K T KX) -1 X T K T KY

[0042] Wherein, θ is the coefficient vector of the polynomial function; X is the Vandermonde matrix with respect to the wavelength; K is the diagonalized kernel function matrix; Y is the vector of the single-wavelength radiation entropy of each of the wavelengths;

[0043] The Vandermonde matrix X with respect to the wavelength is:

[0044]

[0045] Wherein, λ is the wavelength after smoothing; i is the sequence number of the i-th wavelength; and N is the preset quantity.

[0046] Further, in the first iteration process, the initial temperature is the maximum temperature T among the blackbody temperatures of each wavelength. max ;

[0047] In the first iteration process, the initial spectral emissivity of each wavelength is obtained using the Planck blackbody radiation formula based on the maximum temperature.

[0048] Compared with the prior art, the present invention can at least achieve one of the following beneficial effects:

[0049] 1. The present invention uses the change in material emissivity to determine the time when steel slag appears, which is not affected by environmental and subjective human factors, and has high judgment accuracy, making the ladle slagging accuracy of the converter higher and it is easier to obtain better steel.

[0050] 2. The emissivity-temperature separation inversion algorithm used in the present invention can separately calculate the spectral emissivity and temperature of the material only by relying on the heat capacity function of the material, and this function is known for most materials, with simple calculation and high reliability, solving the problem of inaccurate measurement of the spectral emissivity of materials in high-temperature environments.

[0051] 3. The present invention uses a hyperspectral emissivity fitting and smoothing method and a physical modeling method for the spectral emissivity characteristics of materials based on probability, clarifying the relationship between entropy and the spectral emissivity of materials, making the smoothed emissivity have a very clear physical meaning and satisfying various physical constraint conditions of emissivity, and using a high-order polynomial to fit the spectral radiation entropy, meeting the requirements of smoothing and continuity, and having a fast operation speed.

[0052] In the present invention, the above technical solutions can also be combined with each other to achieve more preferred combination schemes. Other features and advantages of the present invention will be described in the subsequent specification, and some advantages can be made obvious from the specification or understood by implementing the present invention. The objectives and other advantages of the present invention can be realized and obtained from the content specifically pointed out in the specification and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The drawings are only used for the purpose of showing specific embodiments and are not considered as a limitation to the present invention. Throughout the drawings, the same reference numerals represent the same components.

[0054] Figure 1 It is a schematic flow chart of a converter ladle slag detection method in an embodiment of the present invention.

[0055] Figure 2Schematic diagram of the spectral emissivity of molten steel and slag at 1600°C in the embodiments of the present invention;

[0056] Figure 3 Schematic diagram of the EM optimization strategy process in the embodiments of the present invention;

[0057] Figure 4 Schematic diagram of the process of the spectral emissivity fitting smoothing method in the embodiments of the present invention. Detailed implementation manners

[0058] The preferred embodiments of the present invention will be specifically described below with reference to the accompanying drawings. The accompanying drawings form a part of this application and are used together with the embodiments of the present invention to explain the principle of the present invention, rather than to limit the scope of the present invention.

[0059] A specific embodiment of the present invention discloses a method for detecting slag entrainment in a converter, as Figure 1 shown, including the following steps:

[0060] Step S1: When a tapping signal of the converter is detected, obtain multiple wavelengths and corresponding gray values of the molten steel at the tapping port of the converter;

[0061] Specifically, after the converter steelmaking completes the steelmaking operation, the system issues a signal indicating that steelmaking is completed and tapping is ready; when the tapping signal is detected, control the converter to rotate to a predetermined angle, open the tapping port of the converter, and at this time, the molten steel pours out along the furnace mouth; use a spectrometer to obtain multiple wavelengths of the flowing molten steel and their corresponding gray values.

[0062] In this embodiment, as Figure 2 shown, for the change of the emissivity of molten steel and slag at a high temperature of 1600 degrees, the wavelength range set by the spectrometer for obtaining the wavelength value of molten steel is selected as 7 - 14 μm.

[0063] Furthermore, for the spectrometer used to obtain data, the signal-to-noise ratio at both ends of the set wavelength region for obtaining data is relatively low, and the gray value will be negative after subtracting the background, so it needs to be removed.

[0064] Step S2: Based on each of the wavelengths and their gray values, use the pre-calibrated data to obtain the spectral radiance corresponding to each of the wavelengths;

[0065] Specifically, the pre-calibrated data is the corresponding data of gray value - spectral radiance value.

[0066] Step S3: Based on each of the wavelengths and their spectral radiance, use the material spectral emissivity temperature separation inversion method to obtain the spectral emissivity of the molten steel in the current converter;

[0067] Furthermore, according to Planck's quantum hypothesis, the following formula is used to obtain the blackbody temperature of each of the wavelengths when the material is an ideal blackbody:

[0068]

[0069] wherein, is the spectral radiance of a perfect black body at wavelength λ i at temperature ; C1 is the first radiation constant with a value of 3.74×10 -16 (W×m 2 ); C2 is the second radiation constant with a value of 1.4398×10 -2 (m×K).

[0070] Furthermore, based on the spectral radiance corresponding to each wavelength and the black body temperature corresponding to each wavelength, the EM optimization strategy is used to obtain the temperature of the material after inversion and the spectral emissivity corresponding to each wavelength.

[0071] Specifically, the EM optimization strategy is an iterative optimization strategy. Each iteration in its calculation method is divided into two steps. One is the expectation step, i.e., the E step, and the other is the maximization step, i.e., the M step, which is to solve the parameter estimation problem in the case of missing data (including latent variables).

[0072] Furthermore, the basic idea of the EM optimization strategy is as follows: First, based on the given observed data, estimate the values of the model parameters (initialization); then, based on the parameter values estimated in the previous step, estimate the values of the missing data, and then re-estimate the parameter values according to the estimated missing data plus the previously observed data, and then iterate repeatedly until convergence, at which point the iteration ends.

[0073] In this embodiment, as Figure 3 shown, the EM optimization strategy includes the following steps in one iteration process:

[0074] In one iteration process:

[0075] Step S301: Obtain the initial temperature of this iteration and the initial spectral emissivity corresponding to each wavelength;

[0076] Specifically, in the first iteration process, the initial temperature is the maximum temperature T among the black body temperatures of each wavelength max .

[0077] In the first iteration process, the initial spectral emissivity of each wavelength is obtained based on the maximum temperature using the following formula:

[0078]

[0079]

[0080] wherein, is the initial spectral emissivity at wavelength λ i ; T max is the maximum value of the temperature; I obs (λ i ) is the spectral radiance corresponding to the wavelength λ to be separated i ; I B (λ i , T max ) is the spectral radiance of a perfect blackbody at wavelength λ i at temperature T max ; e is the natural constant

[0081] Step S302, in the maximization step of the EM optimization strategy: using the free energy minimization method as the maximum likelihood function of the EM optimization strategy based on the initial spectral emissivity of each wavelength to calculate the optimal temperature as the hidden parameter of the EM optimization strategy

[0082] Specifically, when calculating the emissivity at a certain wavelength point based on measured data, it often implies that the data is under a certain temperature condition, and it also implies that the emissivities of other wavelength points can be calculated based on this temperature, that is, the obtained emissivity information is related to the temperature

[0083] In this embodiment, the emissivity-temperature separation algorithm based on free energy minimization means calculating the entropy using the spectral emissivity and calculating the internal energy using the material heat capacity, and obtaining the optimal temperature through free energy minimization

[0084] In statistics, the KL divergence is generally used to measure the "distance" between two probability distribution functions, and can describe the relative distance between two distributions P i and Q i to a certain extent

[0085]

[0086] Similarly, in thermodynamics, the KL divergence corresponds to the sum of the entropy changes of the system and the environment in these two different states

[0087]

[0088] where ΔS is the change in entropy; ΔU env is the change in the internal energy of the environment; ΔS sys is the entropy of the system; ΔS env is the entropy of the environment; T0 is the temperature

[0089] Specifically, according to the definition of Helmholtz free energy

[0090] A = U - TS

[0091] Assume that the volume and energy of the system can be neglected relative to the environment, and the temperature change of the environment before and after the energy exchange with the system can be neglected. Then, the total change in free energy of the system and the environment before and after the state change can be defined as:

[0092] ΔA = ΔU - TΔS = ΔU - TD KL (ε2||ε1)

[0093] Where, ΔU is the change in internal energy of the system and can be calculated through the heat capacity. The formula is:

[0094] ΔU = C v (T - T0)

[0095] Where, C v is the heat capacity of the material, which is generally a function of temperature: C v (T) = f(T); Many scientists have accurately measured the heat capacity values of various substances at various temperatures by experimental methods and obtained the empirical expressions representing the relationship between heat capacity and temperature.

[0096] Furthermore, the optimal temperature T opt ;

[0097] Specifically, the formula of the free energy minimization method is:

[0098]

[0099]

[0100]

[0101] Where, T opt is the optimal temperature and is the hidden parameter of the EM optimization strategy; ΔA is the change in internal energy; C V is the heat capacity of the material; is the initial spectral emissivity at wavelength λ i ; I obs (λ i ) is the spectral radiance corresponding to the wavelength λ i to be separated; I B (λ i , T opt ) is the spectral radiance of the ideal blackbody at wavelength λ i at temperature T opt .

[0102] Step S303, in the expected step of the EM optimization strategy: Based on the optimal temperature, obtain the spectral emissivity of each wavelength at this temperature, and use the spectral emissivity fitting smoothing method to obtain the smoothed spectral emissivity of each wavelength as the spectral emissivity expectation value corresponding to each wavelength;

[0103] Specifically, use the following formula to calculate the spectral emissivity of each wavelength at the optimal temperature

[0104]

[0105] Furthermore, use the spectral emissivity fitting smoothing method to perform fitting smoothing on the spectral emissivity of each wavelength;

[0106] Specifically, during the measurement process, due to reasons such as detector noise, wavelength calibration, and stray radiation interference, the measured data is always accompanied by a large amount of fluctuations. Therefore, the calculated material spectral emissivity is also accompanied by a large amount of fluctuations. It is also very crucial to effectively fit and smooth the spectral emissivity of the material in this embodiment.

[0107] Furthermore, as Figure 4 shown, the spectral emissivity fitting smoothing method includes:

[0108] Step S3031, obtain the single-wavelength radiation entropy of each wavelength based on the spectral emissivity corresponding to each wavelength;

[0109] Specifically, according to the probability characteristics of the emissivity and the Lebesgue measure, the entropy of the system at this wavelength can be measured by the following formula:

[0110]

[0111] where g(m) is the state degeneracy when exactly m particles radiate and de-excite, and its numerical value is indicating that there are g(m) microscopic states with the same energy.

[0112] Furthermore, from the above formula, it can be deduced that:

[0113] H λ =-N[ε λ lnε λ +(1 - ε λ )ln(1 - ε λ )]

[0114] Therefore, the single-wavelength radiation entropy of a single particle at each wavelength is:

[0115]

[0116] where, is the single - wavelength radiation entropy at wavelength λ i ; is the spectral emissivity corresponding to wavelength λ; To avoid the situation where the spectral emissivity is greater than or equal to 1 or less than or equal to 0 (in these cases i the calculation result is a complex number), take the modulus of the calculation result to obtain

[0117] It can be seen from this that the entropy of the system is equal to the entropy of a single particle multiplied by the total number of particles N, and its magnitude is related to the emissivity of the material in different wavelength bands.

[0118] Step S3032: Use the high - order polynomial fitting method for the single - wavelength radiation entropy of each wavelength to obtain the smoothed wavelength radiation entropy;

[0119] Specifically, the high - order polynomial fitting method includes: for each wavelength, use the kernel function as the weight to perform local polynomial kernel regression to obtain a high - order polynomial as the regression function, and use the regression function to calculate the single - wavelength radiation entropy of each wavelength as the smoothed wavelength radiation entropy; where, the local polynomial kernel regression includes: select a preset number of wavelengths adjacent to this wavelength, and perform polynomial fitting based on the kernel function within this range.

[0120] It should be noted that the preset number is the number of data points participating in the calculation on both sides of the smoothing point, which controls how many adjacent data points are used in the calculation; the larger the preset number range, the smoother the result. Preferably, the preset number N is selected to be 10 - 50.

[0121] Furthermore, according to solid - state physics theory, the spectral emissivity reflects the micro - scale photo - acoustic coupling characteristics of the material, and it should exhibit certain short - range continuity, smoothness, and certain long - range correlation in hyperspectral data. The smoothness of the material spectral emissivity means that its derivative is continuous when the wavelength changes. The short - range correlation and long - range correlation are the external manifestations of the material energy - band structure. When the material is a mixture, the long - range correlation and short - range correlation weaken, but still maintain relatively high spectral continuity and smoothness.

[0122] Furthermore, the continuity and smoothness of the material spectral emissivity are due to the fact that the radiation entropy H λ at different wavelengths has smooth and continuous characteristics at the micro - scale.

[0123] Specifically, high - order polynomial fitting is a method commonly used in data analysis and machine learning. It can fit a set of data into a high - order polynomial model. This method can improve the fitting accuracy of the data to a certain extent.

[0124] ​The basic idea of high-order polynomial fitting is to find an optimal polynomial function to fit a given dataset, thereby minimizing the error between the fitting function and the original dataset. To achieve this goal, we need to select a suitable polynomial function and solve the coefficients of the polynomial function using the least squares method.

[0125] Specifically, we can select an nth-degree polynomial function with respect to the independent variable λ i to fit the data. In this embodiment, this polynomial function as the regression function can be expressed as:

[0126] H(λ i ) = a0 + a1λ i + a2λ i 2 + a3λ i 3 + … + a n λ i n

[0127] where λ i is the wavelength value of the i-th wavelength; a0, a1 … a n are the coefficients of the polynomial function, and n is the order of the polynomial function. By using the least squares method, we can solve the optimal values of these coefficients, thereby obtaining an optimal polynomial function to fit the data.

[0128] It should be noted that high-order polynomial fitting may cause overfitting problems in some cases. Overfitting means that when fitting the data, in order to achieve a better fitting effect, the original dataset is overfitted, resulting in a poor prediction effect for new data. Therefore, when performing high-order polynomial fitting, it is necessary to select the order of the polynomial according to the specific situation and make appropriate adjustments and optimizations.

[0129] Specifically, the order of the polynomial is from 2 to 11, preferably an odd order.

[0130] In this embodiment, the order is preferably 5.

[0131] As can be seen from the above, as long as the polynomial coefficients of the radiation entropy at different wavelengths are obtained, the material emissivity can be fitted and smoothed. However, if a fixed polynomial is used, this fitting method has strong long-range correlation, which does not conform to the actual material characteristics.

[0132] Therefore, in this embodiment, the kernel regression method is used to fit the entropy, so that while meeting the requirements of smoothness and continuity, the hard constraint of its long-range correlation is reduced.

[0133] Specifically, traditional linear regression can only fit a straight line. Kernel regression is a regression method based on non-linear mapping, which is a method that only uses multiple data points near the data point for regression. Its essence is to use the kernel function as a weight function to establish a non-linear regression model.

[0134] Further, according to the least squares method, local polynomial kernel regression is to solve the coefficients θ of the polynomial function to minimize the following objective function J(θ):

[0135] J(θ) = (Xθ - Y) T K(Xθ - Y)

[0136] where θ is the coefficient vector of the polynomial function; X is the Vandermonde matrix with respect to the wavelength; K is the diagonalized kernel function matrix; and Y is the single-wavelength radiation entropy vector of each of the wavelengths.

[0137] Therefore, according to the following coefficient calculation formula of the polynomial function can be obtained:

[0138]

[0139] Further, the Vandermonde matrix X with respect to the wavelength is:

[0140]

[0141] where λ is the smoothed wavelength; i is the serial number of the i-th wavelength currently selected; and N is the preset quantity.

[0142] It should be noted that in the matrix, if i - N is less than 1, it is calculated starting from 1; if i + N is greater than the total number of wavelengths within the wavelength range to be fitted, it is only calculated up to the last wavelength.

[0143] Further, the diagonalized kernel function matrix K is:

[0144] K = diag(k(λ,λ i-N ),…k(λ,λ i ),…k(λ,λ i+N ))

[0145] where k(λ,λ i ) is the kernel function; λ is the smoothed wavelength; i is the serial number of the i-th wavelength currently selected; and N is the preset quantity.

[0146] Specifically, the kernel function defines the similarity measurement method of the input data in the feature space. Commonly used kernel functions include Gaussian kernel function, polynomial kernel function, sigmoid kernel function, etc.

[0147] Preferably, the Gaussian kernel function regression model is selected in the present invention:

[0148]

[0149] Among them, σ is the standard deviation.

[0150] Specifically, the Gaussian kernel function can be regarded as a weight negatively correlated with the distance from the center; when smoothing, adjusting the standard deviation is to adjust the influence degree of the surrounding wavelengths on the current wavelength. Increasing σ increases the influence degree of the distant wavelengths on the central wavelength, and the filtering result becomes smoother.

[0151] Furthermore, the single-wavelength radiation entropy vector of the wavelength is:

[0152]

[0153] Among them, is the single-wavelength radiation entropy of wavelength λ i ; i is the serial number of the i-th wavelength currently selected; N is the preset quantity.

[0154] So far, after substituting the wavelength λ i into the regression function H(λ), the calculated H(λ i ) is the entropy of the wavelength after smoothing

[0155] Step S3033: Obtain the fitted and smoothed spectral emissivity based on the smoothed wavelength radiation entropy.

[0156] Specifically, based on the smoothed wavelength radiation entropy, the following formula is used to obtain the fitted and smoothed spectral emissivity of the wavelength range to be fitted:

[0157]

[0158] Among them, is the spectral emissivity of the smoothed wavelength λ i , which is the expected value of the spectral emissivity of the wavelength λ i in this iteration; is the single-wavelength radiation entropy of the smoothed wavelength λ i .

[0159] Step S304: Use the optimal temperature as the initial temperature for the next iteration, and the expected values of the spectral emissivities corresponding to each wavelength as the initial spectral emissivities for the next iteration, and perform the next iteration;

[0160] Step S305: When the expected values of the spectral emissivities corresponding to each wavelength and the initial spectral emissivities of each wavelength satisfy the convergence condition, the obtained optimal temperature and the spectral emissivities of each wavelength are used as the temperature after inversion and the spectral emissivities corresponding to each wavelength, and the iteration ends.

[0161] Specifically, the method for judging the convergence condition is to calculate the value of, and when it is less than a specific value, it is judged to meet the convergence condition.

[0162] It should be noted that the specific value can be adjusted according to the actual situation and accuracy requirements; in this embodiment, the specific value is taken as 1E-8.

[0163] Furthermore, in this embodiment, during the iterative process of the EM optimization strategy, the Anderson acceleration algorithm is preferably used for acceleration, which can greatly improve the algorithm convergence speed.

[0164] Step S4: When the spectral emissivity of the molten steel inside the converter is greater than the set threshold, it is judged that the slag content in the molten steel exceeds the normal value, and the converter is controlled to stop tapping.

[0165] Specifically, when the spectral emissivity of the molten steel is greater than the set threshold, it means that the slag content exceeds the standard; the spectral emissivity alarm value is displayed on the control screen and shown in red font; the alarm buzzer and alarm light are started to remind the operator to throw a slag blocking ball into the converter, and at the same time, the tapping hole is automatically closed or manually closed by the operator according to the pre-setting, and the converter is rotated back to the initial position.

[0166] In summary, a converter slag detection method according to an embodiment of the present invention has the following beneficial effects:

[0167] 1. The present invention uses the change of material emissivity to judge the time when slag appears in molten steel, which is not affected by environmental and subjective human factors, has high judgment accuracy, makes the converter slag tapping accuracy higher, and is easy to obtain better steel.

[0168] 2. The emissivity-temperature separation inversion algorithm used in the present invention can separately calculate and obtain the spectral emissivity and temperature of the material only relying on the heat capacity function of the material, and this function is known for most materials, with simple calculation and high reliability, solving the problem of inaccurate measurement of the spectral emissivity of materials in high-temperature environments.

[0169] 3. The present invention uses a hyperspectral emissivity fitting and smoothing method and a physical modeling method for the spectral emissivity characteristics of materials based on probability, clarifying the relationship between entropy and the spectral emissivity of materials, making the smoothed emissivity have a very clear physical meaning, meeting various physical constraint conditions of emissivity, using a high-order polynomial to fit the spectral radiation entropy, meeting the requirements of smoothing and continuity, and having a fast operation speed.

[0170] The above are only the preferred specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A method for detecting slag carry-over in a converter, characterized in that, It includes the following steps: When a tapping signal of the converter is detected, obtain multiple wavelengths and corresponding gray values of the molten steel at the tapping opening of the converter; Based on each of the wavelengths and its gray value, use pre-calibrated data to obtain the spectral radiance corresponding to each of the wavelengths; Based on each of the wavelengths and its spectral radiance, use the material spectral emissivity temperature separation inversion method to obtain the spectral emissivity of the molten steel in the current converter; When the spectral emissivity of the molten steel inside the converter is greater than a set threshold, it is determined that the slag content in the molten steel exceeds the normal value, and the converter is controlled to stop tapping; 2. The method according to claim 1, wherein When the tapping signal appears, obtaining the wavelength and its gray value of the molten steel in the converter includes: when the tapping signal is detected, controlling the converter to rotate; When the tapping inclination angle of the converter reaches a preset angle, open the tapping opening. When the molten steel flows out of the converter opening, use a spectrometer to obtain multiple wavelengths of the flowing molten steel and their corresponding gray values; 3. The method according to claim 2, wherein The determination that the slag content in the molten steel exceeds the normal value and controlling the converter to stop tapping includes: when the spectral emissivity of the molten steel is greater than the set threshold, display an alarm value on the control screen and start the alarm buzzer and alarm lamp. At the same time, perform an automatic tapping opening closing operation or an operator manual tapping opening closing operation according to the pre-setting, and return the converter to the initial position; 4. The method according to claim 1, wherein Receiving each of the wavelengths and the corresponding spectral radiance and using the temperature-emissivity separation inversion method to obtain the spectral emissivity of the material in the converter includes: Based on the spectral radiance of each of the wavelengths, use the Planck blackbody radiation formula to obtain the blackbody temperature of the ideal blackbody for each of the wavelengths; Based on the spectral radiance corresponding to each wavelength and the blackbody temperature corresponding to each wavelength, use the EM optimization strategy to obtain the temperature of the inverted material and the spectral emissivity corresponding to each wavelength; 5. The method according to claim 4, characterized in that, For the EM optimization strategy, in one iteration process: Obtain the initial temperature of this iteration and the initial spectral emissivity corresponding to each wavelength; In the maximization step of the EM optimization strategy: based on the initial spectral emissivity of each of the wavelengths, use the free energy minimization method as the maximum likelihood function of the EM optimization strategy to calculate the optimal temperature as the hidden parameter of the EM optimization strategy; In the expectation step of the EM optimization strategy: based on the optimal temperature, obtain the spectral emissivity of each of the wavelengths at this temperature and use the spectral emissivity fitting smoothing method to obtain the smoothed spectral emissivity of each of the wavelengths as the spectral emissivity expectation value corresponding to each wavelength; The optimal temperature is used as the initial temperature for the next iteration, and the spectral emissivity expectation values corresponding to each wavelength are used as the initial spectral emissivities for the next iteration; When the spectral emissivity expectation values corresponding to each of the wavelengths and the initial spectral emissivities of each of the wavelengths satisfy the convergence condition, the obtained optimal temperature and the spectral emissivities corresponding to each of the wavelengths are used as the inverted temperature and the spectral emissivities corresponding to each of the wavelengths, and the iteration ends; 6. The method according to claim 5, characterized in that, Calculating the optimal temperature based on the initial emissivity of each of the wavelengths using the free energy minimization method, where the formula of the free energy minimization method is: Among them, T opt is the optimal temperature; ΔA is the change in internal energy; C V is the heat capacity of the material; is the initial spectral emissivity at wavelength λ i ; is the spectral emissivity of the material at wavelength λ i at temperature T opt ; Based on the optimal temperature, the spectral emissivity of each of the wavelengths at this temperature is obtained using the following formula 7. The method according to claim 5 or 6, characterized in that Using the spectral emissivity fitting smoothing method to perform fitting smoothing on the spectral emissivities of each of the wavelengths includes: Obtain the single-wavelength radiation entropy for each of the wavelengths based on the spectral emissivity corresponding to each of the wavelengths: Use the high-order polynomial fitting method for the single-wavelength radiation entropy of each of the wavelengths to obtain the smoothed wavelength radiation entropy; Obtain the fitted and smoothed spectral emissivity based on the smoothed wavelength radiation entropy.

8. The method according to claim 7, wherein The high-order polynomial fitting method includes: for each wavelength, use a kernel function as a weight to perform local polynomial kernel regression to obtain a high-order polynomial as a regression function, and use the regression function to calculate the single-wavelength radiation entropy for each of the wavelengths as the smoothed wavelength radiation entropy; wherein, the local polynomial kernel regression includes: select a preset number of wavelengths adjacent to this wavelength, and perform polynomial fitting based on the kernel function within this range; Use the following formula based on the smoothed wavelength radiation entropy to obtain the fitted and smoothed spectral emissivity for the wavelength range to be fitted: Among them, is the spectral emissivity at the smoothed wavelength λ i ; is the single-wavelength radiation entropy at the smoothed wavelength λ i .

9. The method according to claim 8, wherein The regression function is: H(λ i ) = a0 + a1λ i + a2λ i 2 + a3λ i 3 +…+ a n λ i n where λ i is the wavelength value of the i-th wavelength; a0, a1…a n are the coefficients of the polynomial function; n is the order of the polynomial function; The coefficients of the polynomial function are calculated using the following formula: θ = (X T K T KX) -1 X T K T KY wherein, θ is the coefficient vector of the polynomial function; X is the Vandermonde matrix with respect to the wavelength; K is the diagonalized kernel function matrix; Y is the vector of the single-wavelength radiation entropy for each of the wavelengths; The Vandermonde matrix X with respect to the wavelength is: wherein, λ is the smoothed wavelength; i is the i-th wavelength serial number; N is the preset number.

10. The method according to claim 5, wherein In the first iteration process, the initial temperature is the maximum temperature T among the blackbody temperatures of the respective wavelengths max ; In the first iteration process, the initial spectral emissivity for each wavelength is obtained using the Planck blackbody radiation formula based on the maximum temperature.

Images