Coating information prediction method and device for silicon steel coating, equipment and medium

Abstract

The invention relates to a coating information prediction method and device for a silicon steel coating, equipment and a medium. The method comprises the steps that coating liquid basic parameters are obtained in real time; based on the coating liquid basic parameters and a preset initial quality prediction model, obtaining an initial quality prediction value of the silicon steel coating in unit area; based on the coating liquid basic parameters and a preset initial thickness prediction model, obtaining an initial thickness prediction value of the silicon steel coating in unit area; based on the initial quality predicted value and the initial thickness predicted value, a target quality predicted value and a target thickness predicted value of the coating in unit area are obtained, and the target quality predicted value and the target thickness predicted value are final credible values used for production control and quality judgment after the corresponding initial predicted values are superposed with correction values. According to the dynamic correction mechanism, systematic deviation caused by hydration reaction and magnesium carbonate generation is eliminated, online prediction and closed-loop control of coating parameters are achieved, and real-time data support is provided for technological parameter optimization.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of coating coating situation prediction, and in particular to a coating information prediction method, device and equipment for a silicon steel coating and a medium. BACKGROUND

[0002] In the production process of the silicon steel coating, the composition and state of the coating liquid change with the coating operation time, which makes it difficult to accurately control the coating weight and thickness. Specifically, 1. In the coating process of the oriented silicon steel magnesium oxide coating, the hydration degree of the magnesium oxide gradually increases with the flow of the coating liquid in the circulation system. 2. The mutual contact of the coating liquid and the air causes the continuous generation of magnesium carbonate in the recycling process of the coating liquid, which breaks the stability of the magnesium oxide coating liquid. 3. A good magnesium oxide coating liquid has specific requirements for the purity, activity, particle size, adhesion of the magnesium oxide configured in the coating liquid and the hydration degree of the coating liquid after preparation, and the change of the state of the coating liquid such as the hydration degree will significantly affect the thickness and final quality of the coating. 4. Under the condition that the process parameters are stable, the coating weight and thickness also change with the coating time due to the change of the composition and state of the coating liquid.

[0003] Therefore, a method is needed to predict the change of the coating weight and thickness to reduce the subsequent coating re-inspection workload and provide data support for the control of the coating liquid state and the coating thickness. SUMMARY

[0004] In order to predict the change of the coating weight and thickness and reduce the subsequent coating re-inspection workload, the application provides a coating information prediction method, device, equipment and medium for a silicon steel coating.

[0005] In a first aspect, the application provides a coating information prediction method for a silicon steel coating, which adopts the following technical scheme:

[0006] A coating information prediction method for a silicon steel coating comprises the following steps:

[0007] Real-time acquisition of coating liquid basic parameters, wherein the coating liquid basic parameters include the mass of magnesium oxide in the coating liquid, the mass of additives in the coating liquid, the volume of pure water required for preparing the coating liquid, the volume increase of the coating liquid, the coating operation time, the instantaneous flow of spraying and the instantaneous flow of returning;

[0008] Based on the coating liquid basic parameters and a preset initial mass prediction model, an initial mass prediction value of the silicon steel unit area coating is obtained;

[0009] Based on the coating liquid basic parameters and a preset initial thickness prediction model, an initial thickness prediction value of the silicon steel unit area coating is obtained;

[0010] Based on the initial quality prediction value and the initial thickness prediction value, the target quality prediction value and the target thickness prediction value of the coating per unit area are obtained. The target quality prediction value and the target thickness prediction value are the final reliable values ​​used for production control and quality judgment after adding corrections to the corresponding initial prediction values.

[0011] By adopting the above technical solution, the dynamic correction mechanism effectively eliminates the systematic deviations caused by hydration reaction and magnesium carbonate formation, significantly improving the agreement between the final target prediction value and the actual measurement value. It achieves online prediction and closed-loop control of coating parameters, reduces the frequency of manual re-inspection, and provides real-time data support for process parameter optimization.

[0012] Optionally, the method for constructing the initial quality prediction model includes:

[0013] Based on the coating liquid density, the coating operation time, the instantaneous spray flow rate, and the instantaneous return flow rate, the total mass of coating liquid consumed during the coating operation time is obtained;

[0014] Based on the coating quality per unit area of ​​silicon steel, the width of the silicon steel plate, the coating operation speed, and the coating operation time, the first total coating mass coated on the silicon steel plate surface within the coating operation time is obtained; the basic parameters of the coating liquid include the width of the silicon steel plate and the coating operation speed;

[0015] Based on the total mass of the coating liquid consumed and the first total coating mass, the initial mass prediction model is constructed;

[0016] The total mass of the coating solution consumed is expressed as follows:

[0017]

[0018] Where M1 is the total mass of the coating liquid consumed, ρ is the density of the coating liquid, Q1 is the instantaneous flow rate of the spray, Q2 is the instantaneous flow rate of the return, and t is the coating operation time;

[0019] The total mass of the first coating is expressed as:

[0020] M2=2pWvt

[0021] Where M2 is the total mass of the first coating, p is the coating mass per unit area of ​​silicon steel, W is the width of the silicon steel plate, v is the coating speed, and t is the coating time.

[0022] M1 is equal to M2, and the initial quality prediction model corresponding to the coating quality per unit area of ​​silicon steel is expressed as:

[0023]

[0024] By adopting the technical scheme, the dynamic running state of the coating liquid circulating system can be tracked in real time, the influence of the increase of the magnesium oxide hydration degree and the generation of magnesium carbonate on the coating liquid consumption can be accurately quantified, and the accuracy of coating quality prediction is significantly improved. The initial quality prediction model provides a quantitative basis for the state monitoring of the coating liquid circulating system, changes the coating quality control from experience judgment to accurate regulation based on dynamic data, and effectively reduces the coating quality fluctuation caused by unstable coating liquid state.

[0025] Optionally, the coating liquid density is obtained by:

[0026] The coating liquid density is obtained based on the magnesium oxide mass, the additive mass, the volume of pure water required for preparing the coating liquid, and the coating liquid volume increase.

[0027] The coating liquid density is represented as:

[0028]

[0029] wherein, ρ is the coating liquid density, M MgO is the magnesium oxide mass in the coating liquid, M1 is the additive mass, is the volume of pure water required for preparing the coating liquid, and V1 is the coating liquid volume increase.

[0030] By adopting the technical scheme, the volume change caused by the hydration reaction can be accurately captured, and the density deviation caused by only considering the magnesium oxide mass is reduced. By dynamically calculating the accurate ratio of the solid component and the liquid volume, the actual state of the coating liquid is reflected in real time, accurate physical property parameter basis is provided for coating quality prediction, and the reliability of the coating weight prediction model is significantly improved.

[0031] Optionally, the construction method of the initial thickness prediction model comprises:

[0032] Based on the coating thickness per unit area of the silicon steel plate, the density of the magnesium oxide coating, the width of the silicon steel plate, the coating running speed, and the coating running time, a second total coating mass coated on the surface of the silicon steel plate within the coating running time is obtained.

[0033] Based on the total coating liquid consumption mass and the second total coating mass, the initial thickness prediction model is constructed.

[0034] The second total coating mass is represented as:

[0035]

[0036] wherein, M3 is the second total coating mass, h is the coating thickness per unit area of the silicon steel plate, ρ Coatingfor the density of the magnesium oxide coating, W is the width of the silicon steel sheet, v is the coating run speed, and t is the coating run time;

[0037] The M1 and the M3 are equal, and the initial thickness prediction model corresponding to the coating thickness per unit area of the silicon steel is expressed as:

[0038]

[0039] By adopting the above technical solution, the prediction error caused by the change of the coating liquid state can be dynamically corrected, the coating thickness prediction deviation problem caused by the increase of the hydration degree and the generation of magnesium carbonate in the coating liquid circulation process is effectively solved, and the accuracy of the coating thickness prediction is improved.

[0040] Optionally, the target mass prediction value and the target thickness prediction value of the unit area coating are obtained based on the initial mass prediction value and the initial thickness prediction value, including:

[0041] The actual mass of the unit area coating is obtained by sampling at fixed time intervals within the coating run time, and the mass deviation of the actual mass from the initial mass prediction value is calculated;

[0042] The mass deviation and the initial mass prediction value are polynomial fitted to obtain a mass fitting polynomial;

[0043] The mass fitting polynomial is superimposed on the initial mass prediction value and the initial thickness prediction value to obtain the target mass prediction value and the target thickness prediction value of the unit area coating, respectively.

[0044] By adopting the above technical solution, the dynamic calibration of the coating weight and thickness prediction value is realized, the adaptability of the prediction model to the change of the coating liquid state is improved, the coating re-inspection frequency is reduced, and reliable data support is provided for real-time adjustment of the coating process parameters.

[0045] Optionally, the target mass prediction value and the target thickness prediction value of the unit area coating are obtained by superimposing the mass fitting polynomial on the initial mass prediction value and the initial thickness prediction value, including:

[0046] The mass fitting polynomial and the initial mass prediction value are taken as the input of a preset target weight prediction model to obtain the target mass prediction value;

[0047] The mass fitting polynomial and the initial thickness prediction value are taken as the input of a preset target thickness prediction model to obtain the target thickness prediction value;

[0048] The target weight prediction model is expressed as:

[0049]

[0050] wherein p Correction is the target mass prediction value, and Δ is the mass fitting polynomial;

[0051] The target thickness prediction model is represented as:

[0052]

[0053] wherein h Correction is the target thickness prediction value.

[0054] By adopting the above technical solution, the prediction deviation caused by magnesium oxide hydration and magnesium carbonate generation in the coating liquid circulation process is effectively eliminated, and the stability of coating weight and thickness prediction is improved. The dynamic correction mechanism updates the prediction model parameters in real time through real-time mass data, so that the prediction value always tracks the actual state change of the coating liquid, reduces the systematic error caused by the change of the coating liquid composition, and provides more accurate data support for coating process control.

[0055] Optionally, after the real-time acquisition of the coating liquid basic parameters, the method further comprises:

[0056] real-time acquisition of a first coating liquid temperature corresponding to the injection port and a second coating liquid temperature corresponding to the return port;

[0057] based on a preset temperature-viscosity relationship model, the first coating liquid temperature and the second coating liquid temperature, respectively correcting the injection instantaneous flow and the return instantaneous flow, and based on the corrected injection instantaneous flow and return instantaneous flow, calculating the total mass of the coating liquid consumed.

[0058] By adopting the above technical solution, the temperature drift error of the flow sensor is corrected, the viscosity distribution difference caused by the temperature gradient in the coating liquid flow process is compensated, and the accurate calculation of the coating liquid consumption mass is realized. The calculation error of the coating liquid consumption caused by viscosity fluctuation is reduced.

[0059] In a second aspect, the application provides a coating information prediction device for a silicon steel coating, which adopts the following technical solution:

[0060] A coating information prediction device for a silicon steel coating, comprising:

[0061] A first acquisition module for real-time acquisition of coating liquid basic parameters, the coating liquid basic parameters including magnesium oxide mass in the coating liquid, additive mass in the coating liquid, volume of pure water required for preparing the coating liquid, coating liquid volume increase, coating running time, injection instantaneous flow and return instantaneous flow;

[0062] The first prediction module is configured to obtain an initial mass prediction value of the coating per unit area of the silicon steel based on the coating liquid basic parameters and a preset initial mass prediction model.

[0063] The second prediction module is configured to obtain an initial thickness prediction value of the coating per unit area of the silicon steel based on the coating liquid basic parameters and a preset initial thickness prediction model.

[0064] The fitting correction module is configured to obtain a target mass prediction value and a target thickness prediction value of the coating per unit area based on the initial mass prediction value and the initial thickness prediction value, wherein the target mass prediction value and the target thickness prediction value are final reliable values for production control and quality determination after a correction value is added to the corresponding initial prediction value.

[0065] In a third aspect, the present application provides an electronic device, which adopts the technical scheme as follows:

[0066] An electronic device includes a processor and a memory, wherein the processor is coupled to the memory.

[0067] The processor is configured to execute a computer program stored in the memory, so that the electronic device executes the method according to any one of the first aspect.

[0068] In a fourth aspect, the present application provides a computer readable storage medium, which adopts the technical scheme as follows:

[0069] A computer readable storage medium includes a computer program or instructions, when the computer program or instructions are run on a computer, so that the computer executes the method according to any one of the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0070] Figure 1 FIG. 1 is a flowchart of a coating information prediction method of a silicon steel coating according to an embodiment of the present application.

[0071] Figure 2 FIG. 2 is a schematic diagram of fitting of a weight difference per unit area curve according to an embodiment of the present application.

[0072] Figure 3 FIG. 3 is a structural block diagram of a coating information prediction device of a silicon steel coating according to an embodiment of the present application.

[0073] Figure 4 FIG. 4 is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0074] The principles and features of the present application are described below, and the examples are only used to explain the present application, and are not used to limit the scope of the present application.

[0075] The application will be further described in detail below with reference to the accompanying drawings.

[0076] The applicant finds that the dynamic change of the coating liquid composition is a core factor affecting the prediction accuracy, especially the change of the coating liquid density and fluidity caused by the hydration reaction and the generation of carbonates. By analyzing the operation data of the coating system, it is observed that the difference between the spraying flow and the return flow directly affects the consumption of the coating liquid, and the coating quality is directly related to the consumption of the coating liquid. Based on this, a dynamic prediction model is proposed, which acquires the basic parameters and operation parameters of the coating liquid in real time, constructs an initial prediction model, and superimposes a bias compensation mechanism to form a closed-loop control system.

[0077] Therefore, the embodiment of the application provides a coating information prediction method for a silicon steel coating, which can be executed by a device, which can be a server or a terminal device. The server can be a physical server, a server cluster composed of multiple physical servers, or a distributed system, and can also be a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a desktop computer, etc., but is not limited thereto.

[0078] The present scheme is mainly used in the production process of the silicon steel coating, and the corresponding relationship between the change of the coating liquid flow and the change of the coating mass per unit area is used to observe the liquid consumption per unit time. The relationship between the coating thickness and the coating consumption is determined by establishing a relationship model.

[0079] In the production process of the silicon steel coating, when the magnesium oxide coating liquid and the coating amount are in a stable state, the coating liquid in the liquid collecting tank no longer increases or decreases. The magnesium oxide coating liquid is mainly composed of pure water, magnesium oxide, additives, etc. In order to improve the adhesion of the magnesium oxide coating after coating and promote the formation of the magnesium silicate bottom layer in the subsequent high-temperature annealing stage, appropriate TiO2, boride and other additives can also be added.

[0080] As shown in Figure 1 A coating information prediction method for a silicon steel coating is provided, and an electronic device is used as the execution subject. The main process of the method is described as follows (steps S101-S104):

[0081] Step S101: Real-time acquisition of coating liquid basic parameters, including the mass of magnesium oxide in the coating liquid, the mass of additives in the coating liquid, the volume of pure water required for preparing the coating liquid, the volume increase of the coating liquid, the coating operation time, the instantaneous flow of spraying, and the instantaneous flow of returning;

[0082] Step S102: Based on the coating liquid basic parameters and the preset initial mass prediction model, an initial mass prediction value of the silicon steel coating per unit area is acquired;

[0083] Step S103: obtaining an initial thickness prediction value of the coating per unit area of the silicon steel based on the coating liquid basic parameters and a preset initial thickness prediction model;

[0084] Step S104: obtaining a target mass prediction value and a target thickness prediction value of the coating per unit area based on the initial mass prediction value and the initial thickness prediction value, the target mass prediction value and the target thickness prediction value being final reliable values for production control and quality determination after a correction amount is superimposed on the corresponding initial prediction values.

[0085] The coating liquid basic parameters refer to core variables reflecting the composition and flow state of the coating liquid, and can be collected in real time by a mass sensor, a flow meter and a timing device for dynamic calculation of the coating liquid density and consumption. The coating liquid basic parameters can be collected in real time by an online monitoring system, in which the magnesium oxide mass and the additive mass are used to calculate the coating liquid density, the pure water volume and the volume increase amount reflect the preparation state of the coating liquid, and the volume increase amount refers to the volume increase amount of the coating liquid caused by the addition of magnesium oxide and additives during the preparation of the coating liquid, which can be directly read out by a measuring container. The additives can be titanium dioxide, borides and the like.

[0086] The coating running time refers to the running time of the coating process, and refers to the cumulative length of time from the start of coating to the current time, which can be measured by a high-precision timer. The spraying instantaneous flow rate refers to the instantaneous flow rate value measured by the spraying flow meter of the coating liquid circulating supply system, i.e. the volume of coating liquid sprayed to the surface of the silicon steel per unit time, which can be monitored in real time by an electromagnetic flow meter for dynamic calculation of the consumption change of the coating liquid.

[0087] The return instantaneous flow rate refers to the instantaneous flow rate value measured by the return flow meter of the coating liquid circulating supply system, i.e. the volume of coating liquid returned to the circulating system per unit time without being attached to the surface of the silicon steel, which can be monitored synchronously by an ultrasonic flow meter to form a dynamic difference with the spraying flow rate.

[0088] The initial mass prediction model refers to a mathematical model established based on the balance relationship between the coating liquid consumption and the coating quality, which is used to convert the flow parameters into the coating mass prediction value. The initial thickness prediction model refers to a calculation model for converting the mass prediction result into physical thickness, which is used to realize the mapping of the mass parameters to the geometric parameters. The target mass prediction value refers to the final prediction result after deviation correction, which is used to eliminate the model deviation caused by the change of the coating liquid state.

[0089] The coating running time is combined with the flow parameter to calculate the coating liquid consumption, and an initial mass prediction model is established by balancing the total mass of the coating liquid consumption and the total mass of the coating to establish a dynamic equation. An initial thickness prediction model divides the mass prediction value by the density of the magnesium oxide coating to convert it into a thickness parameter. The initial mass prediction value and the initial thickness prediction value are further self-corrected to obtain the final coating information, i.e., a target mass prediction value and a target thickness prediction value.

[0090] In this embodiment, the key parameters such as the magnesium oxide mass and the additive ratio are collected in real time to dynamically update the coating liquid density calculation value. The real-time coating liquid consumption is calculated based on the difference between the injection flow and the return flow. A physical correlation model of mass and thickness is established to improve the parameter conversion accuracy.

[0091] The influence of the coating liquid state change on the prediction model can be compensated in real time, so that the prediction error is reduced to within the process control allowable range. The dynamic correction mechanism effectively eliminates the systematic deviation caused by the hydration reaction and the generation of magnesium carbonate, so that the degree of agreement between the final target prediction value and the actual measured value is significantly improved. Online prediction and closed-loop control of the coating parameters are realized, the manual recheck frequency is reduced, and real-time data support is provided for process parameter optimization.

[0092] In this embodiment, the construction method of the initial mass prediction model comprises:

[0093] Based on the coating liquid density, the coating running time, the injection instantaneous flow and the return instantaneous flow, the total mass of the coating liquid consumption in the coating running time is obtained;

[0094] Based on the coating mass per unit area of the silicon steel, the silicon steel plate width, the coating running speed and the coating running time, a first total coating mass on the surface of the silicon steel plate in the coating running time is obtained; the coating liquid basic parameters include the silicon steel plate width and the coating running speed;

[0095] Based on the total mass of the coating liquid consumption and the first total coating mass, the initial mass prediction model is constructed;

[0096] The total mass of the coating liquid consumption is represented as:

[0097]

[0098] Wherein, M1 is the total mass of the coating liquid consumption, p is the coating liquid density, Q1 is the injection instantaneous flow, Q2 is the return instantaneous flow, and t is the coating running time;

[0099] The first total coating mass is represented as:

[0100] M2=2pWvt

[0101] Wherein, M2 is the first coating total mass, p is the coating mass per unit area of silicon steel, W is the width of the silicon steel sheet, v is the coating running speed, and t is the coating running time;

[0102] The M1 is equal to the M2, and the initial mass prediction model corresponding to the coating mass per unit area of silicon steel is expressed as:

[0103]

[0104] The coating liquid density can be calculated by the magnesium oxide mass, the additive mass, the pure water volume, and the volume increase, and is used to accurately characterize the actual physical properties of the coating liquid in the circulating system. By collecting the dynamic difference between the spraying flow and the return flow in real time, and combining the coating liquid density parameter for operation, the actual total mass of the coating liquid consumed in the coating process can be accurately calculated. Meanwhile, the total coating area is obtained based on the product of the width of the silicon steel sheet, the running speed, and the running time, and the product of the coating mass per unit area and the total area is taken as the theoretical total coating mass. By establishing the conservation relationship between the total mass of the coating liquid consumed and the theoretical total coating mass, the prediction formula of the coating mass per unit area, i.e., the initial mass prediction model, is derived.

[0105] The initial mass prediction model effectively captures the state change of the coating liquid in the circulating system caused by the hydration reaction and the generation of magnesium carbonate by dynamically monitoring the flow, converts the time-varying flow parameter into the cumulative mass parameter, and overcomes the defect that the traditional static model cannot adapt to the dynamic change of the coating liquid composition.

[0106] Therefore, the application can track the dynamic running state of the coating liquid circulating system in real time, accurately quantify the influence of the increase of the hydration degree of magnesium oxide and the generation of magnesium carbonate on the consumption of the coating liquid, and significantly improve the accuracy of the coating mass prediction. The initial mass prediction model provides a quantitative basis for the state monitoring of the coating liquid circulating system, changes the coating mass control from experience judgment to accurate regulation based on dynamic data, and effectively reduces the coating mass fluctuation caused by the unstable state of the coating liquid.

[0107] In the embodiment, the coating liquid density is obtained by:

[0108] The coating liquid density is obtained based on the magnesium oxide mass, the additive mass, the volume of the pure water required for preparing the coating liquid, and the volume increase of the coating liquid.

[0109] The coating liquid density is expressed as:

[0110]

[0111] Wherein, p is the coating liquid density, M MgO is the magnesium oxide mass in the coating liquid, M1 is the additive mass, V1 is the volume increase of the coating liquid.

[0112] It is easy to understand that the magnesium oxide mass refers to the mass of magnesium oxide solid put in when the coating liquid is prepared, which can be weighed by an electronic balance or the like, and is used to characterize the basic contribution of solid components to the density of the coating liquid. The additive mass refers to the total mass of auxiliary components added in addition to magnesium oxide, which can be accurately controlled by an automatic metering device, and is used to reflect the adjusting effect of additives on the physical properties of the coating liquid. The volume of pure water required for preparing the coating liquid refers to the volume of solvent added in the initial preparation stage, which can be monitored in real time by a flowmeter, and is used to determine the basic solvent volume of the coating liquid. The volume increase of the coating liquid refers to the volume expansion caused by the hydration reaction of magnesium oxide, which is used to quantify the actual influence of the hydration reaction on the volume of the coating liquid.

[0113] By combining the total mass of solid components with the change in liquid volume, the density of the coating liquid is calculated, and a mass-volume ratio model is constructed. The sum of the mass of magnesium oxide and the mass of additives is used as the numerator, which directly reflects the contribution of solid components to the density; the sum of the volume of pure water and the volume increase is used as the denominator, which accurately characterizes the actual liquid volume after the hydration reaction. The calculation model of the density of the coating liquid particularly considers the volume expansion effect caused by the hydration reaction of magnesium oxide, and by introducing the volume increase parameter, the hydration degree is quantified as a calculable volume increase. In the process of recycling the coating liquid, as the hydration degree deepens, the volume increase parameter can be dynamically updated, so that the density calculation can reflect the change of the coating liquid in real time.

[0114] By establishing a dynamic volume calculation model containing the volume increase, the volume change caused by the hydration reaction can be accurately captured. At the same time, by including the mass of additives in the calculation of solid components, the density deviation caused by only considering the mass of magnesium oxide is reduced. The present application effectively solves the prediction deviation problem of the density of the coating liquid caused by the change of the hydration reaction and the addition ratio of additives. By dynamically calculating the accurate ratio of solid components and liquid volume, the actual state of the coating liquid is reflected in real time, which provides an accurate physical property parameter basis for coating weight prediction, and significantly improves the reliability of the coating weight prediction model.

[0115] In the embodiment, the construction method of the initial thickness prediction model comprises:

[0116] Based on the coating thickness per unit area of the silicon steel sheet, the density of the magnesium oxide coating, the width of the silicon steel sheet, the coating running speed, and the coating running time, a second total coating mass coated on the surface of the silicon steel sheet within the coating running time is obtained;

[0117] Based on the total consumption mass of the coating liquid and the second total coating mass, the initial thickness prediction model is constructed;

[0118] The second total coating mass is expressed as:

[0119]

[0120] wherein M3 is the second total coating mass, h is the coating thickness per unit area of the silicon steel sheet, p is the density of the magnesium oxide coating, W is the width of the silicon steel sheet, v is the coating running speed, and t is the coating running time; Coating wherein M3 is the second total coating mass, h is the coating thickness per unit area of the silicon steel sheet, p is the density of the magnesium oxide coating, W is the width of the silicon steel sheet, v is the coating running speed, and t is the coating running time;

[0121] The M1 and the M3 are equal, and the initial thickness prediction model corresponding to the coating thickness per unit area of the silicon steel sheet is represented as:

[0122]

[0123] In the embodiment, the coating thickness per unit area of the silicon steel sheet refers to the vertical height of the magnesium oxide coating per unit area, which is used to characterize the geometric features of the coating film. The density of the magnesium oxide coating refers to the mass-volume ratio of the solid magnesium oxide film layer, which can be measured by a laboratory density tester and is used to establish the conversion relationship between the coating liquid consumption and the solid coating mass. The width of the silicon steel sheet refers to the lateral dimension of the coated steel strip, which can be obtained in real time by an online measurement system of the production line and is used to calculate the coating area per unit time. The coating running speed refers to the moving rate of the steel strip in the coating section, which can be obtained by converting the rotational speed of the encoder measurement transmission roller and is used to determine the time-area parameters of the coating process.

[0124] By multiplying the coating thickness per unit area of the silicon steel sheet and the density of the magnesium oxide as a mass conversion coefficient, combining the product of the width of the silicon steel sheet and the coating running speed as an area change rate, and multiplying the coating running time, the second total coating mass is obtained. The total mass and the total mass of the coating liquid consumption establish a dynamic balance relationship to form the initial thickness prediction model. In the running process, the total mass of the coating liquid consumption is obtained by integrating the difference between the injection flow and the return flow, and the second total coating mass is derived by the coating thickness, the material density and the process parameters. When the two are equal, the predicted value of the coating thickness is obtained by solving the model equation. By combining the dynamic consumption data of the coating liquid circulation system with the physical properties of the solid coating, the real-time prediction of the coating thickness is realized.

[0125] In this embodiment, the measured density parameter of the magnesium oxide coating is introduced, and the mass conservation equation of coating liquid consumption and coating film formation is established, which can dynamically correct the prediction error caused by the change of coating liquid state. The problem of coating thickness prediction deviation caused by the increase of hydration degree and the generation of magnesium carbonate in the coating liquid circulation process is effectively solved. By establishing a dynamic balance model between coating liquid consumption and coating film mass, the influence of coating liquid state change on density parameter is compensated in real time, and the accuracy of coating thickness prediction is improved. The density of magnesium oxide coating is used as the key conversion parameter to accurately distinguish the physical property difference between liquid coating liquid and solid coating, so that the prediction model can adapt to the dynamic change of coating liquid composition. By integrating process operation parameters and material physical property parameters, online tracking prediction of coating thickness is realized, which provides real-time data support for coating liquid state control.

[0126] In this embodiment, the target mass prediction value and the target thickness prediction value of the unit area coating are obtained based on the initial mass prediction value and the initial thickness prediction value, which includes:

[0127] Sampling at fixed time intervals within the coating running time to obtain the actual mass of the unit area coating, and calculating the mass deviation of the actual mass from the initial mass prediction value;

[0128] Polynomial fitting is performed on the mass deviation and the initial mass prediction value to obtain a mass fitting polynomial;

[0129] The mass fitting polynomial is superimposed on the initial mass prediction value and the initial thickness prediction value to obtain the target mass prediction value and the target thickness prediction value of the unit area coating, respectively.

[0130] It is easy to understand that fixed time interval sampling refers to periodically collecting coating samples to obtain actual mass data. The mass deviation refers to the difference between the actual measurement value and the model prediction value, which is used to quantify the dynamic influence of coating liquid state change on the prediction model. Polynomial fitting refers to a method of converting discrete deviation data into a continuous function relationship. Specifically, a least squares method can be used to perform quadratic or cubic polynomial regression analysis on the deviation sequence, which can represent the nonlinear law of deviation change with coating time. The superposition operation refers to the mathematical combination of the fitting function and the initial prediction value, which realizes the fusion of the initial model prediction and the dynamic deviation correction.

[0131] Specifically, during the operation of the coating device, the sampling device is triggered periodically to collect coating samples, for example, the edge area of the silicon steel strip is intercepted every ten minutes for weighing measurement. After the actual mass data is measured, the theoretical value output by the initial mass prediction model is compared point by point, and a deviation data set containing a time stamp is generated. The deviation data set is input into a polynomial fitting algorithm, and a function expression of the deviation with respect to time is obtained by curve fitting. The fitting function is superimposed into the calculation formula of the initial mass prediction model to form a composite prediction model containing a dynamic correction term. For the thickness prediction value, since there is a physical correlation between mass and thickness, the fitting function of the mass deviation simultaneously acts on the initial thickness prediction model through the density conversion relationship. In this way, the initial prediction model retains the influence of the basic parameters of the coating liquid and superimposes a correction term reflecting the dynamic changes of the coating liquid state.

[0132] For example, according to the coating operation time of the magnesium oxide coating liquid, the actual coating weight on the upper and lower surfaces of the magnesium oxide coating can be sampled and weighed every 15 minutes within 6 hours, and the average value of each time point is obtained to obtain the unit area coating weight difference of each coating time, as shown in Table 1, which is used to represent the difference between the unit area coating weight and the target within the coating time. The fitting result is shown in Table 2. Figure 2

[0133] Table 1

[0134]

[0135]

[0136] By periodically sampling to establish a deviation data set and combining a polynomial fitting method to construct a dynamic correction function, the cumulative influence of the coating liquid state change on the coating can be more accurately described. The application effectively solves the prediction model deviation problem caused by the dynamic change of the coating liquid composition, and realizes the dynamic calibration of the coating weight and thickness prediction values. By real-time sampling to obtain actual coating data and establishing a deviation correction function to continuously optimize the initial model, the adaptability of the prediction model to the change of the coating liquid state is improved. Thus, the prediction error caused by the change of the hydration degree of the coating liquid is reduced, the coating reinspection frequency is reduced, and reliable data support is provided for real-time adjustment of the coating process parameters.

[0137] In this embodiment, the mass fitting polynomial is superimposed into the initial mass prediction value and the initial thickness prediction value to obtain a target mass prediction value and a target thickness prediction value of the unit area coating, respectively, which includes:

[0138] The mass fitting polynomial and the initial mass prediction value are input into a preset target weight prediction model to obtain the target mass prediction value;

[0139] ​The mass fitting polynomial and the initial thickness prediction value are taken as inputs of a preset target thickness prediction model to obtain a target thickness prediction value;

[0140] The target weight prediction model is represented as:

[0141]

[0142] wherein p Correction is the target mass prediction value, and Δ is the mass fitting polynomial;

[0143] The target thickness prediction model is represented as:

[0144]

[0145] wherein h Correction is the target thickness prediction value.

[0146] In the embodiment, the mass fitting polynomial can be represented as:

[0147] Δ = A0 + A1t + A2t 2 + A3t 3 + A4t 4 + A5t 5

[0148] wherein Δ is a fitting polynomial corresponding to a weight deviation value, and the unit can be grams per square meter; A0, A1, A2, A3, A4 and A5 are polynomial coefficients; and t is a coating running time, and the unit can be hours.

[0149] The mass fitting polynomial captures the systematic deviation rule caused by the coating liquid state change through a mathematical method, and the polynomial is simultaneously input to the weight and thickness prediction models. In the weight prediction model, the mass fitting polynomial is directly superimposed on the initial mass prediction value to form a composite prediction model with real-time correction function; in the thickness prediction model, the mass fitting polynomial is associated with the thickness prediction value through the coating density parameter to realize the physical quantity conversion of the mass deviation to the thickness prediction value. The two models respectively use different function structures to process the influence of the mass deviation on the coating parameters, the weight prediction uses a direct superposition method for rapid correction, and the thickness prediction maintains the dimensional consistency of physical quantities through parameter transmission.

[0150] By establishing a dynamic quality deviation model and integrating it into the prediction system, real-time response to the changes in coating liquid state is achieved, solving the problem of prediction accuracy decay in long-term coating process for static models. The prediction deviation caused by magnesium hydroxide hydration and magnesium carbonate generation during coating liquid circulation is effectively eliminated, improving the stability of coating weight and thickness prediction. The dynamic correction mechanism updates the prediction model parameters in real time through real-time mass data, so that the prediction value always tracks the actual state changes of the coating liquid, reducing systematic errors caused by changes in coating liquid composition and providing more accurate data support for coating process control.

[0151] In this embodiment, after the real-time acquisition of the coating liquid basic parameters, the following steps are further included:

[0152] Real-time acquisition of the first coating liquid temperature corresponding to the injection port and the second coating liquid temperature corresponding to the return port;

[0153] Based on the preset temperature-viscosity relationship model, the first coating liquid temperature and the second coating liquid temperature, the injection instantaneous flow rate and the return instantaneous flow rate are corrected respectively, and the total mass of the coating liquid consumed is calculated based on the corrected injection instantaneous flow rate and return instantaneous flow rate.

[0154] The first coating liquid temperature refers to the real-time temperature data of the coating liquid flowing through the injection port, which can be collected by a thermocouple or an infrared thermometer, and is used to represent the viscosity change of the coating liquid in the injection stage. The second coating liquid temperature refers to the real-time temperature data of the coating liquid flowing through the return port, which can be continuously monitored by an embedded temperature sensor, and is used to reflect the temperature distribution difference of the coating liquid in the circulation loop. The temperature-viscosity relationship model refers to the mathematical relationship describing the change of coating liquid viscosity with temperature, which can be established by using Arrhenius equation or polynomial fitting model, and is used to convert temperature data into viscosity compensation coefficient, so as to eliminate the interference of temperature on the measurement value of flow sensor.

[0155] Specifically, during the coating liquid circulation process, the temperature difference of the coating liquid between the injection port and the return port will cause flow measurement error. By collecting the temperature data of the two places in real time, combined with the temperature-viscosity relationship model, the actual viscosity value of the coating liquid can be dynamically calculated. For example, when the temperature rises and the viscosity decreases, the volume flow rate measured by the flow sensor needs to be compensated according to the viscosity change and converted into equivalent mass flow rate. The corrected injection instantaneous flow rate and return instantaneous flow rate are substituted into the total mass of coating liquid consumed calculation formula, which can eliminate the coating liquid mass calculation deviation caused by temperature fluctuation, so as to ensure the accuracy of the input parameters of the initial mass prediction model.

[0156] By establishing a temperature-viscosity dynamic compensation mechanism, not only the temperature drift error of the flow sensor is corrected, but also the viscosity distribution difference caused by temperature gradient in the coating liquid flow process is compensated, realizing the accurate calculation of the coating liquid consumption quality. The flow measurement distortion problem caused by the change of coating liquid temperature is effectively solved, and the coating liquid consumption calculation error caused by viscosity fluctuation is reduced. Through real-time temperature monitoring and dynamic flow correction, the input data reliability of the coating weight and thickness prediction model is significantly improved, providing a foundation guarantee for accurate control of coating quality.

[0157] After obtaining the basic parameters of the coating liquid in real time, the following steps are included: establishing a relationship model between temperature and magnesium oxide hydration degree, obtaining the correction coefficient of temperature influence on hydration degree through experimental data; studying the influence of temperature on the generation rate of magnesium carbonate, and establishing a corresponding correction model; integrating the temperature compensation factor into the original prediction model to obtain a new coating weight and thickness prediction formula: p_new = p_original * f(T) * g(H) * h(C); continuously collecting temperature, flow and actual coating data during production, and using machine learning algorithm to continuously optimize the parameters of the prediction model.

[0158] Among them, the temperature correction function f(T) refers to the function of dynamically adjusting the coating weight and thickness prediction value according to the change of coating liquid temperature, which can be realized by fitting a polynomial function through coating experiment data under different temperature gradients, and is used to compensate the influence of temperature on the flowability of coating liquid and the measurement accuracy of flow. The hydration degree correction function g(H) refers to the function of quantifying the influence of temperature on the hydration reaction process of magnesium oxide, which is realized by using an exponential function of the relationship between hydration reaction rate and temperature measured under constant temperature water bath conditions in the laboratory, and is used to correct the change of coating liquid viscosity and adhesion performance caused by the change of hydration degree. The magnesium carbonate influence correction function h(C) refers to the function of characterizing the influence of temperature on the generation rate of magnesium carbonate, which is realized by establishing a linear regression model of temperature and magnesium carbonate generation amount through high temperature accelerated test, and is used to offset the cumulative effect of coating liquid component degradation on coating quality. The machine learning algorithm refers to the training method for dynamically updating model parameters, which uses gradient boosting tree algorithm to incrementally learn the real-time collected process data, and automatically adjusts the weight coefficient of temperature compensation function through feature importance analysis.

[0159] Specifically, a temperature sensor is arranged in the coating production line to monitor the coating liquid temperature in real time, the temperature data is input into a pre-constructed magnesium oxide hydration degree model, and a correction coefficient of the hydration reaction rate at the current temperature is calculated. Simultaneously, coating liquid sampling analysis is performed, a database of the correspondence between the temperature and the magnesium carbonate precipitation amount is established, and a compensation curve of the magnesium carbonate generation rate is generated. The temperature correction function, the hydration correction function and the magnesium carbonate correction function are multiplied and combined with the original prediction model output value to form a composite compensation mechanism. In the continuous production process, the coating thickness detector feeds back the actual coating data in real time, the control system compares these data with the predicted values to generate a training sample set, and the parameter combination of the three correction functions is iteratively updated through the gradient boosting tree algorithm, so that the prediction model gradually adapts to the dynamic changes of the coating liquid state.

[0160] The application effectively reduces the coating weight prediction deviation caused by the temperature fluctuation of the coating liquid, and improves the response sensitivity of the thickness prediction model to the changes of the coating liquid composition. Through the composite temperature compensation mechanism, the coating weight prediction error is controlled within the process allowable range, and the coating re-inspection frequency caused by temperature changes is reduced. The dynamic learning mechanism ensures the stability of the prediction model in the continuous production process, and provides real-time data support for the state monitoring of the magnesium oxide coating liquid.

[0161] Based on the same technical concept, the application also provides a coating information prediction device for a silicon steel coating, as shown in Figure 3 The coating information prediction device 200 for the silicon steel coating mainly includes:

[0162] The first acquisition module 201 is configured to acquire coating liquid basic parameters in real time, and the coating liquid basic parameters include the mass of magnesium oxide in the coating liquid, the mass of additives in the coating liquid, the volume of pure water required for preparing the coating liquid, the volume increase of the coating liquid, the coating running time, the instantaneous flow of spraying, and the instantaneous flow of returning.

[0163] The first prediction module 202 is configured to acquire an initial mass prediction value of the coating on the unit area of the silicon steel based on the coating liquid basic parameters and a preset initial mass prediction model.

[0164] The second prediction module 203 is configured to acquire an initial thickness prediction value of the coating on the unit area of the silicon steel based on the coating liquid basic parameters and a preset initial thickness prediction model.

[0165] The fitting correction module 204 is configured to acquire a target mass prediction value and a target thickness prediction value of the coating on the unit area based on the initial mass prediction value and the initial thickness prediction value, and the target mass prediction value and the target thickness prediction value are final reliable values for production control and quality determination after adding correction amounts to the corresponding initial prediction values.

[0166] Optionally, the construction device of the initial mass prediction model comprises:

[0167] a first obtaining sub-module, configured to obtain total coating liquid consumption mass in the coating operation time based on coating liquid density, the coating operation time, spraying instantaneous flow and return instantaneous flow;

[0168] a second obtaining sub-module, configured to obtain first total coating mass on the surface of the silicon steel sheet in the coating operation time based on silicon steel sheet width, coating operation speed and the coating operation time according to silicon steel sheet unit area coating mass;

[0169] a first constructing sub-module, configured to construct the initial mass prediction model based on the total coating liquid consumption mass and the first total coating mass;

[0170] the total coating liquid consumption mass is expressed as:

[0171]

[0172] wherein, M1 is the total coating liquid consumption mass, p is the coating liquid density, Q1 is the spraying instantaneous flow, Q2 is the return instantaneous flow, and t is the coating operation time;

[0173] the first total coating mass is expressed as:

[0174] M2=2pWvt

[0175] wherein, M2 is the first total coating mass, p is the silicon steel sheet unit area coating mass, W is the silicon steel sheet width, v is the coating operation speed, and t is the coating operation time;

[0176] the M1 is equal to the M2, and the initial mass prediction model corresponding to the silicon steel sheet unit area coating mass is expressed as:

[0177]

[0178] Optionally, the coating liquid density is obtained by:

[0179] a third obtaining sub-module, configured to obtain the coating liquid density based on magnesium oxide mass, additive mass, volume of pure water required for preparing the coating liquid, and coating liquid volume increase;

[0180] the coating liquid density is expressed as:

[0181]

[0182] wherein, p is the coating liquid density, M MgO is the magnesium oxide mass in the coating liquid, M1 is the additive mass, V1 is the volume of pure water required for the prepared coating liquid.

[0183] Optionally, the construction device of the initial thickness prediction model comprises:

[0184] The fourth acquisition sub-module is configured to acquire a second total coating mass coated on the surface of the silicon steel sheet within the coating operation time based on the coating thickness per unit area of the silicon steel sheet, the density of the magnesium oxide coating, the width of the silicon steel sheet, the coating operation speed, and the coating operation time.

[0185] The second construction sub-module is configured to construct the initial thickness prediction model based on the total consumption mass of the coating liquid and the second total coating mass.

[0186] The second total coating mass is expressed as:

[0187]

[0188] wherein M3 is the second total coating mass, h is the coating thickness per unit area of the silicon steel sheet, ρ Coating is the density of the magnesium oxide coating, W is the width of the silicon steel sheet, v is the coating operation speed, and t is the coating operation time.

[0189] The M1 is equal to the M3, and the initial thickness prediction model corresponding to the coating thickness per unit area of the silicon steel sheet is expressed as:

[0190]

[0191] Optionally, the fitting correction module 204 comprises:

[0192] The acquisition actual mass sub-module is configured to sample at fixed time intervals within the coating operation time to obtain an actual mass of the coating per unit area and calculate a mass deviation of the actual mass from the initial mass prediction value.

[0193] The mass fitting sub-module is configured to perform polynomial fitting on the mass deviation and the initial mass prediction value to obtain a mass fitting polynomial.

[0194] The correction sub-module is configured to superimpose the mass fitting polynomial on the initial mass prediction value and the initial thickness prediction value to obtain a target mass prediction value and a target thickness prediction value of the coating per unit area, respectively.

[0195] Optionally, the correction sub-module comprises:

[0196] The first as sub-module is configured to take the mass fitting polynomial and the initial mass prediction value as inputs of a preset target weight prediction model to obtain the target mass prediction value.

[0197] The second sub-module is configured to input the mass fitting polynomial and the initial thickness prediction value into a preset target thickness prediction model to obtain a target thickness prediction value.

[0198] The target weight prediction model is represented as:

[0199]

[0200] wherein p Correction is the target mass prediction value, and Δ is the mass fitting polynomial.

[0201] The target thickness prediction model is represented as:

[0202]

[0203] wherein h Correction is the target thickness prediction value.

[0204] Optionally, the first obtaining module is followed by:

[0205] The second obtaining module is configured to obtain a first coating liquid temperature corresponding to the injection port and a second coating liquid temperature corresponding to the return port in real time.

[0206] The correction module is configured to correct the injection instantaneous flow rate and the return instantaneous flow rate based on a preset temperature-viscosity relationship model, the first coating liquid temperature, and the second coating liquid temperature, and to calculate the total mass of the coating liquid consumed based on the corrected injection instantaneous flow rate and the return instantaneous flow rate.

[0207] In one example, the modules in any of the above apparatuses can be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASICs), or one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0208] For another example, when the modules in the apparatuses can be implemented in the form of a processing element scheduler, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor that can invoke programs. For another example, the modules can be integrated together to be implemented in the form of a system-on-a-chip (SOC).

[0209] In this application, various objects such as messages / information / devices / network elements / systems / apparatus / actions / operations / processes / concepts may be named. It is understood that these specific names do not constitute a limitation on the relevant objects. The names may be changed depending on the scenario, context, or usage habits. The understanding of the technical meaning of the technical terms in this application should be mainly determined from their functions and technical effects embodied / performed in the technical solution.

[0210] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0211] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0212] Based on the same technical concept, this application also provides an electronic device, such as... Figure 4 As shown, the electronic device 300 includes a processor 301 and a memory 302, and may further include one or more of an information input / output (I / O) interface 303, a communication component 304, and a communication bus 305.

[0213] The processor 301 is configured to control overall operations of the electronic device 300 to complete all or part of the steps of the coating information prediction method of the silicon steel coating described above. The memory 302 is configured to store various types of data to support the operations of the electronic device 300. The data can include, for example, instructions for operating any application or method on the electronic device 300, and application-related data. The memory 302 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as one or more of a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk, or an optical disk.

[0214] The I / O interface 303 provides an interface between the processor 301 and other interface modules, which can be a keyboard, a mouse, a button, and the like. The buttons can be virtual buttons or physical buttons. The communication component 304 is configured to test wired or wireless communication between the electronic device 300 and other devices. The wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G, or 4G, or a combination of one or more of them, so the corresponding communication component 304 can include a Wi-Fi component, a Bluetooth component, and an NFC component.

[0215] The communication bus 305 can include a path for transmitting information between the above-mentioned components. The communication bus 305 can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 305 can be divided into an address bus, a data bus, a control bus, etc.

[0216] The electronic device 300 can be implemented with one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic elements for performing the coating information prediction method of the silicon steel coating given in the above-described embodiments.

[0217] The electronic device 300 can include, but not limited to, a mobile terminal of a digital broadcasting receiver, a PDA (Personal Digital Assistant), a PMP (Portable Multimedia Player), and the like, and a stationary terminal such as a digital TV, a desktop computer, and the like, and can also be a server or the like.

[0218] Based on the same technical concept, the present application also provides a computer readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps of the coating information prediction method of the silicon steel coating described above.

[0219] The computer readable storage medium can include a U disk, a mobile hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and the like, various media capable of storing program codes.

[0220] The term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such a process, method, article or device.

[0221] In addition, the terms "first", "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include at least one of the features. In the description of the present application, the meaning of "a plurality of" is at least two, for example, two, three, etc., unless otherwise explicitly specified.

[0222] In the description of the specification, the description using the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" etc. means that the particular feature, structure, material or characteristic being described is included in at least one embodiment or example of the present application. The illustrative appearances of the above-mentioned terms in various places in the specification are not necessarily referred to the same embodiment or example. Moreover, the particular features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples. Furthermore, the terms "comprise", "comprising", "include", "including", "contain", "containing" or variations thereof are used inclusively and do not exclude the additional inclusion of unrecited features, structures, materials, or characteristics.

[0223] Although the embodiments of the present application have been shown and described above, it is understood that the above-described embodiments are exemplary and are not to be interpreted as limiting the present application, and ordinary skilled people in the art can make changes, modifications, replacements and variations to the above-described embodiments within the scope of the present application.

Claims

1. A method for predicting coating information of silicon steel coatings, characterized in that, include: Real-time acquisition of basic parameters of the coating solution, including the mass of magnesium oxide in the coating solution, the mass of additives in the coating solution, the volume of pure water required to prepare the coating solution, the increase in the volume of the coating solution, the coating operation time, the instantaneous spray flow rate, and the instantaneous return flow rate; Based on the basic parameters of the coating liquid and the preset initial quality prediction model, the initial quality prediction value of the coating per unit area of ​​silicon steel is obtained. Based on the basic parameters of the coating liquid and the preset initial thickness prediction model, the initial thickness prediction value of the coating per unit area of ​​silicon steel is obtained. Based on the initial quality prediction value and the initial thickness prediction value, the target quality prediction value and the target thickness prediction value of the coating per unit area are obtained. The target quality prediction value and the target thickness prediction value are the final reliable values ​​used for production control and quality judgment after adding corrections to the corresponding initial prediction values.

2. The method according to claim 1, characterized in that, The method for constructing the initial quality prediction model includes: Based on the coating liquid density, the coating operation time, the instantaneous spray flow rate, and the instantaneous return flow rate, the total mass of coating liquid consumed during the coating operation time is obtained; Based on the coating quality per unit area of ​​silicon steel, the width of the silicon steel plate, the coating operation speed, and the coating operation time, the first total coating mass coated on the silicon steel plate surface within the coating operation time is obtained; the basic parameters of the coating liquid include the width of the silicon steel plate and the coating operation speed; Based on the total mass of the coating liquid consumed and the first total coating mass, the initial mass prediction model is constructed; The total mass of the coating solution consumed is expressed as follows: Where M1 is the total mass of the coating liquid consumed, ρ is the density of the coating liquid, Q1 is the instantaneous flow rate of the spray, Q2 is the instantaneous flow rate of the return, and t is the coating operation time; The total mass of the first coating is expressed as: M2=2pWvt Where M2 is the total mass of the first coating, p is the coating mass per unit area of ​​silicon steel, W is the width of the silicon steel plate, v is the coating speed, and t is the coating time. M1 is equal to M2, and the initial quality prediction model corresponding to the coating quality per unit area of ​​silicon steel is expressed as:

3. The method according to claim 2, characterized in that, Obtaining the density of the coating liquid includes: The density of the coating solution is obtained based on the mass of magnesium oxide, the mass of additives, the volume of pure water required to prepare the coating solution, and the increase in the volume of the coating solution. The density of the coating solution is expressed as: Where ρ is the density of the coating liquid, M MgO M1 represents the mass of magnesium oxide in the coating solution, and M2 represents the mass of the additive. V1 represents the volume of pure water required to prepare the coating solution, and V1 represents the increase in the volume of the coating solution.

4. The method according to claim 2 or 3, characterized in that, The method for constructing the initial thickness prediction model includes: Based on the coating thickness per unit area of ​​the silicon steel plate, the density of the magnesium oxide coating, the width of the silicon steel plate, the coating operation speed, and the coating operation time, the total second coating mass coated on the silicon steel plate surface during the coating operation time is obtained; Based on the total mass of the coating liquid consumed and the total mass of the second coating, the initial thickness prediction model is constructed; The total mass of the second coating is expressed as: Where M3 is the total mass of the second coating, h is the coating thickness per unit area of ​​the silicon steel plate, and ρ Coating Where is the density of the magnesium oxide coating, W is the width of the silicon steel plate, v is the coating speed, and t is the coating time; M1 is equal to M3, and the initial thickness prediction model corresponding to the coating thickness per unit area of ​​silicon steel is expressed as:

5. The method according to claim 4, characterized in that, The step of obtaining the target quality prediction value and target thickness prediction value of the coating per unit area based on the initial quality prediction value and the initial thickness prediction value includes: During the coating operation time, samples are taken at fixed time intervals to obtain the actual mass of the coating per unit area, and the mass deviation between the actual mass and the initial mass prediction value is calculated. The quality deviation is fitted with the initial quality prediction value using a polynomial to obtain the quality fitting polynomial. The quality fitting polynomial is superimposed on the initial quality prediction value and the initial thickness prediction value to obtain the target quality prediction value and target thickness prediction value of the coating per unit area, respectively.

6. The method according to claim 5, characterized in that, The step of superimposing the quality fitting polynomial onto the initial quality prediction value and the initial thickness prediction value to obtain the target quality prediction value and target thickness prediction value per unit area of ​​coating, respectively, includes: The mass fitting polynomial and the initial mass prediction value are used as inputs to a preset target weight prediction model to obtain the target mass prediction value. The mass fitting polynomial and the initial thickness prediction value are used as inputs to a preset target thickness prediction model to obtain the target thickness prediction value. The target weight prediction model is expressed as follows: Where, p Correction Let Δ be the predicted value of the target quality, and let Δ be the quality fitting polynomial. The target thickness prediction model is expressed as follows: Among them, h Correction This is the predicted value for the target thickness.

7. The method according to claim 1, characterized in that, After acquiring the basic parameters of the coating solution in real time, the following is also included: Real-time acquisition of the temperature of the first coating liquid corresponding to the injection port and the temperature of the second coating liquid corresponding to the return port; Based on a preset temperature-viscosity relationship model, the temperature of the first coating liquid, and the temperature of the second coating liquid, the instantaneous flow rate of the injection and the instantaneous flow rate of the return are corrected respectively, so as to calculate the total mass of coating liquid consumed based on the corrected instantaneous flow rate of the injection and the instantaneous flow rate of the return.

8. A coating information prediction device for silicon steel coating, characterized in that, include, The first acquisition module is used to acquire the basic parameters of the coating solution in real time. The basic parameters of the coating solution include the mass of magnesium oxide in the coating solution, the mass of additives in the coating solution, the volume of pure water required to prepare the coating solution, the increase in the volume of the coating solution, the coating operation time, the instantaneous flow rate of the spray and the instantaneous flow rate of the return. The first prediction module is used to obtain the initial quality prediction value of the silicon steel coating per unit area based on the coating liquid basic parameters and the preset initial quality prediction model. The second prediction module is used to obtain the initial thickness prediction value of the silicon steel coating per unit area based on the coating liquid basic parameters and the preset initial thickness prediction model. The fitting correction module is used to obtain the target quality prediction value and the target thickness prediction value of the coating per unit area based on the initial quality prediction value and the initial thickness prediction value. The target quality prediction value and the target thickness prediction value are the final reliable values ​​used for production control and quality judgment after adding correction amounts to the corresponding initial prediction values.

9. An electronic device, characterized in that, It includes a processor and a memory, wherein the processor is coupled to the memory; The processor is configured to execute a computer program stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It includes a computer program or instructions that, when run on a computer, cause the computer to perform the method as described in any one of claims 1 to 7.

Images