Dynamic regulation and control method for water cooling strength of molten iron trough based on infrared temperature field reconstruction

Abstract

The invention discloses a dynamic regulation and control method for the water cooling strength of a molten iron trough based on infrared temperature field reconstruction, and belongs to the field of metallurgical equipment cooling control. According to the method, multi-source data are collected and preprocessed through multi-dimensional measuring point layout, a global temperature field is reconstructed by adopting a radial basis function interpolation fusion algorithm, and the residual ledge thickness and the erosion rate are calculated in combination with a heat conduction law and an erosion correction model; linkage regulation and control of water cooling parameters are realized according to a multi-dimensional threshold system, and the temperature field uniformity is guaranteed through graded alarm early warning risks and a closed-loop optimization algorithm; and storing full-process data, generating a dynamic maintenance plan based on an attenuation prediction model, and performing parameter adaptive calibration to optimize the precision. A traditional high-risk detection mode is abandoned, the water-cooling regulation and control accuracy and the equipment operation stability are improved, the service life of the lining body is prolonged, the energy consumption and the maintenance cost are reduced, and safety, accuracy and intelligentization of molten iron runner operation management are achieved.

Description

Technical Field

[0001] This invention relates to the field of cooling control technology for metallurgical production equipment, and in particular to a method for dynamic regulation of the water cooling intensity of iron slurry based on infrared temperature field reconstruction. Background Technology

[0002] The molten iron trough is a core piece of equipment in metallurgical production for transporting molten iron. Its lining is made of refractory material and is subjected to the scouring, erosion, and thermal shock of high-temperature molten iron over a long period of time, making it prone to thinning due to corrosion. The remaining thickness of the lining is directly related to the safe operation of the molten iron trough. Insufficient thickness can lead to major safety accidents such as molten iron leakage. Therefore, it is necessary to monitor the condition of the lining in real time and match it with an appropriate water cooling intensity.

[0003] Current methods for controlling the cooling of molten iron troughs primarily rely on manual, periodic "trough probing" to check the lining thickness. This involves harsh working conditions, extremely high risks, and delayed results, failing to provide timely guidance for water cooling adjustments. Some methods using thermocouple temperature measurement lack consideration of the overall temperature distribution throughout the molten iron trough, leading to uneven temperature distribution. Overcooling in certain areas can cause cracking, while insufficient cooling accelerates erosion. These problems severely restrict the safe and stable operation of molten iron troughs and hinder production efficiency improvements, necessitating a precise, safe, and adaptive dynamic control method for water cooling intensity. Summary of the Invention

[0004] The present invention aims to provide a dynamic control method for the water cooling intensity of iron troughs based on infrared temperature field reconstruction, so as to solve the problems of high detection risk, low control accuracy, poor temperature field uniformity, and untimely maintenance in existing iron trough cooling control methods.

[0005] To achieve the above objectives, this application provides the following solution:

[0006] This application provides a method for dynamic control of the water cooling intensity of iron troughs based on infrared temperature field reconstruction, including the following steps:

[0007] S1. Collect multi-source data of the iron ditch through a sensor system with a multi-dimensional measurement point layout, and obtain a standardized dataset after preprocessing;

[0008] S2. Based on a standardized dataset, a temperature field reconstruction algorithm using multi-source data fusion is used to generate a global temperature field distribution. The residual trench thickness and real-time erosion rate are calculated by combining the thermal conductivity law and the erosion correction model.

[0009] S3. Based on the temperature field distribution, residual trench thickness and erosion rate, a multi-dimensional threshold system is constructed. The operating parameters of the water cooling system are dynamically adjusted through linkage control logic to achieve water cooling intensity matching.

[0010] S4. Based on the control feedback data, monitor the temperature field uniformity and parameter change trend, trigger graded alarms, and correct the control parameters through a closed-loop optimization algorithm;

[0011] S5. Store full-process data and control records, and generate dynamic maintenance plans for iron ditch based on data correlation analysis.

[0012] Optionally, the multi-dimensional measurement point layout in step S1 is specifically as follows:

[0013] A measuring point group is set up every 0.5-1m along the length of the iron trough. Each measuring point group has 2-6 measuring points distributed along the width of the lining. An infrared temperature sensor is deployed at each measuring point perpendicular to the surface of the lining.

[0014] Electromagnetic flow meters were installed on each water supply and return branch, and environmental temperature and humidity sensors were deployed around the iron ditch.

[0015] Optionally, the temperature field reconstruction algorithm for multi-source data fusion in step S2 is specifically a radial basis function interpolation fusion algorithm, with the following formula:

[0016]

[0017] Where (x,y,z) are arbitrary spatial coordinates in the global domain, (x j ,y j ,z j Let ) represent the coordinates of the j-th measuring point, and m represent the total number of measuring points;

[0018] For the fusion weights, σ s Standard deviations for various data types;

[0019] It is a Gaussian radial basis function, with ε ranging from 0.7 to 1.3;

[0020] a k The coefficients are polynomials, which are solved using the least squares method.

[0021] Optionally, the real-time erosion rate in step S2 is calculated based on the first derivative of the residual trench thickness with respect to time, as shown in the formula:

[0022]

[0023] Where k is the erosion coefficient, with a value of 0.0008-0.006 mm / (h·℃); T m For real-time molten iron temperature, T c This refers to the cooling water temperature.

[0024] Optionally, the multi-dimensional threshold system in step S3 includes:

[0025] Temperature thresholds include low threshold, medium threshold, and high threshold, and satisfy the condition that low threshold < medium threshold < high threshold;

[0026] Residual thickness safety threshold, erosion rate warning threshold, and temperature field uniformity threshold;

[0027] The specific logic of the coordinated control is as follows:

[0028] When the temperature is greater than or equal to the high threshold, or the erosion rate is greater than or equal to the warning threshold, the backup water cooling branch is turned on and switched to counterflow mode, and the solenoid valve opening is adjusted to 80%-100%.

[0029] When the temperature is between the medium and high thresholds, the erosion rate is less than the warning threshold and the residual thickness is greater than the safety threshold, maintain the main branch in the downstream mode and adjust the opening to 40%-80%.

[0030] When the temperature is ≤ the low threshold and the residual thickness is ≥ 1.5 times the safety threshold, close some branches and adjust the opening to 20%-40%;

[0031] When the residual thickness is less than or equal to the safety threshold, immediately shut down the water-cooled branch in the corresponding area.

[0032] Optionally, the operating parameters of the water cooling system in step S3 include the solenoid valve switch status, solenoid valve opening degree, and water flow direction;

[0033] The opening degree is adjusted using PWM pulse width modulation, and the water flow direction is switched via a bidirectional solenoid valve.

[0034] Optionally, the temperature field uniformity in step S4 is quantified by the ratio of the difference between the highest temperature, lowest temperature and average temperature across the entire region, with the qualified threshold set to 0.8.

[0035] The closed-loop optimization algorithm adopts the gradient descent method, which minimizes the temperature field non-uniformity by iteratively correcting the control parameters. The iteration step size coefficient is 0.03-0.12, and the control parameters include the solenoid valve opening degree and the water flow direction coefficient.

[0036] Optionally, the graded alarm mechanism described in step S4 includes three levels:

[0037] Level 1 warning is triggered when the temperature is between the medium and high thresholds, or when the residual thickness is between the safety threshold and 1.2 times the safety threshold. The alarm method is a workstation audible and visual alarm and a remote text notification.

[0038] Level 2 emergency, triggered when temperature ≥ high threshold or erosion rate ≥ early warning threshold, alarm methods include audible and visual alarm, voice notification and cloud platform push, and control measures are recorded simultaneously.

[0039] Level 3 critical condition, triggered when residual thickness is ≤ safety threshold, alarm method is all-channel alarm and equipment shutdown reminder, and maintenance process is forcibly locked.

[0040] Optionally, the stored data in step S5 includes standardized datasets, temperature field distribution data, residual trench thickness and erosion rate data, control parameter records, and alarm logs;

[0041] The dynamic maintenance plan is generated based on a residual thickness attenuation prediction model, and the model formula is as follows:

[0042]

[0043] Where Δt is the prediction time step; β is the temperature influence coefficient, ranging from 0.2 to 0.6; T avg The average temperature over the entire temperature field;

[0044] When the predicted residual thickness approaches the safety threshold, a maintenance reminder is generated, which includes the maintenance area and a suggested time window.

[0045] Optionally, it also includes a parameter adaptive calibration step:

[0046] The erosion coefficient and the three-dimensional thermal conductivity of the refractory material are calibrated based on the ratio of the actual residual trench thickness detected periodically to the model calculation value.

[0047] By calibrating and updating parameter values, we ensure that the algorithm's calculation accuracy continues to optimize over time.

[0048] Through the above technical solutions, the beneficial effects of the present invention are as follows: The present invention completely avoids the safety risks of traditional manual inspection through non-contact multi-dimensional monitoring, and achieves precise adaptation of water cooling intensity by relying on the reconstruction of the whole-domain temperature field and multi-dimensional linkage control, effectively avoiding the problem of excessive cooling and cracking of the lining or accelerated erosion. Combined with dynamic maintenance plan and parameter adaptive calibration, it not only ensures the long-term stable and continuous operation of the iron trough, but also extends the service life of the lining and related equipment, while reducing production energy consumption and maintenance costs, and realizing the safe, precise and intelligent upgrade of iron trough operation and management. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention, and the embodiments in the accompanying drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1A flowchart illustrating a method for dynamic control of water cooling intensity in a molten iron trough based on infrared temperature field reconstruction, provided in an embodiment of this application.

[0051] Figure 2 A schematic diagram of a sensing system structure with a multi-dimensional measurement point layout provided in an embodiment of this application;

[0052] Figure 3 This is a schematic diagram of the structure of a computer device provided in one embodiment of this application;

[0053] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0054] It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of the application. Rather, these embodiments are provided to make the disclosure more thorough and complete, and to fully convey the scope of the disclosure to those skilled in the art.

[0055] The foregoing and other technical contents, features, and effects of the present invention will be clearly presented in the following detailed description of the embodiments with reference to the accompanying drawings 0.5-1. All structural contents mentioned in the following embodiments are with reference to the accompanying drawings.

[0056] Exemplary embodiments of the present invention will now be described with reference to the accompanying drawings.

[0057] In one exemplary embodiment, such as Figure 1 As shown, a method for dynamic control of water cooling intensity in iron troughs based on infrared temperature field reconstruction is provided. This method, through steps S1 to S5, achieves precise adaptive adjustment of water cooling intensity through multi-source data fusion, global temperature field reconstruction, coordinated control, and closed-loop optimization. The specific steps are detailed below:

[0058] Step S1: Collect multi-source data of the iron ditch through a sensor system with a multi-dimensional measurement point layout, and obtain a standardized dataset after preprocessing.

[0059] In one specific embodiment, the first step is to establish a multi-dimensional measurement point layout and a sensing system, such as... Figure 2As shown, measuring point groups are set at fixed intervals along the length of the iron trough. For example, when the length of the iron trough is 15m, a measuring point group can be set every 0.5m, forming 30 evenly distributed measuring point groups. Each measuring point group has multiple measuring points evenly distributed along the width of the lining. If the lining width is 1000mm, 4 measuring points can be distributed to cover the entire width of the lining. Each measuring point is equipped with an infrared temperature sensor perpendicular to the lining surface to ensure that the sensor detection direction is perpendicular to the lining surface and accurately captures the lining temperature signal. Electromagnetic flow meters are synchronously configured in each water supply and return branch. For example, two DN35 branch water pipes branching from the DN65 main water pipe are each equipped with an electromagnetic flow meter to capture the water flow data of the branch in real time. Environmental temperature and humidity sensors are evenly deployed around the iron trough, for example, one at each end and one in the middle of the iron trough, to comprehensively collect the temperature and humidity information of the surrounding environment.

[0060] After the sensing system is set up, the data acquisition process is initiated. The acquired multi-source data covers three core categories: first, real-time lining temperature data obtained from the infrared temperature sensor; second, real-time branch flow data collected from the electromagnetic flowmeter; and third, ambient temperature and humidity data recorded by the ambient temperature and humidity sensor. Simultaneously, preset basic parameters are imported. These parameters include the initial refractory lining thickness, the three-dimensional thermal conductivity of the refractory material, the design temperature of the molten iron, the pipe diameter, and the pipe friction coefficient. For example, the initial refractory lining thickness can be set to 1000mm, the molten iron design temperature to 1450℃, and the pipe friction coefficient to 0.02. After acquisition, a sliding window weighted average noise reduction algorithm is used to preprocess the original multi-source data. The mathematical expression of the algorithm is:

[0061]

[0062] in, x represents the i-th standardized data point after preprocessing. i+k This represents the (i+k)th data point in the original data sequence; The weighting coefficients are based on an exponential decay law, where n represents the half-width of the sliding window, ranging from 2 to 6. For example, when n=4, the window covers 9 consecutive data points, and the weights decrease exponentially from the center of the window to both sides. This algorithm performs weighted smoothing on the original data, effectively filtering out electromagnetic interference and random noise present in the industrial environment, ultimately outputting a high-quality standardized dataset, providing a reliable data foundation for subsequent temperature field reconstruction.

[0063] Step S2: Based on the standardized dataset, a temperature field reconstruction algorithm using multi-source data fusion is used to generate the global temperature field distribution. The residual trench thickness and real-time erosion rate are calculated by combining the thermal conductivity law and the erosion correction model.

[0064] In one specific embodiment, a global temperature field reconstruction is first performed by fusing multi-source data. A radial basis function (RBF) interpolation fusion algorithm is used to transform discrete measurement point data into a continuous temperature field. The radial basis function, in the temperature field reconstruction algorithm of multi-source data fusion, uses the Euclidean distance from any point in the entire iron ditch to each measurement point as the independent variable. It quantifies the contribution of different measurement points to the temperature at any point in the entire region through Gaussian and other functional forms, and then combines fusion weights and polynomial terms to realize the transformation from discrete measurement point data to a continuous temperature field. The mathematical expression of this algorithm is:

[0065]

[0066] Where (x,y,z) are arbitrary spatial coordinates across the entire region, and T(x,y,z) represents the temperature value at any spatial location across the entire iron ditch region, (x j ,y j ,z j Let ) represent the coordinates of the j-th measuring point, and m represent the total number of measuring points;

[0067] For the fusion weights, σ s These represent the standard deviations of temperature data, flow rate data, and ambient temperature and humidity data, respectively.

[0068] It is a Gaussian radial basis function, where ε is a shape parameter with a value of 0.7-1.3;

[0069] a k The coefficients are polynomials, solved using the least squares method. The least squares method determines the coefficients by minimizing the sum of squared errors between the reconstructed temperature and the measured temperature. For example, the error between the "measured temperature value" and the "preliminary calculated temperature value" for all measuring points is calculated, and a0 to a3 is adjusted to minimize the sum of squared errors, ensuring a high degree of fit between the reconstructed temperature field and the actual operating conditions. This algorithm can transform discrete measuring point temperature data into a continuous temperature distribution covering the entire iron trough area, clearly presenting the temperature differences between different regions.

[0070] After reconstructing the global temperature field, the residual trench thickness was calculated using the thermal conductivity law and the erosion correction model. The thermal conductivity law employs a three-dimensional unsteady-state thermal conductivity principle, deriving the internal heat flow state of the lining through the temperature field distribution. The erosion correction model introduces an erosion coefficient to correlate heat flow intensity with lining thickness attenuation. Specifically, the residual trench thickness at different spatial locations and operating times was obtained through integral calculations. The formula is:

[0071]

[0072] in, The temperature gradient represents the three-dimensional temperature gradient, reflecting the degree of heat flow concentration at that location on the lining; the stronger the heat flow, the greater the erosion rate. k is the erosion coefficient, ranging from 0.0008 to 0.006 mm / (h·℃). T m For real-time molten iron temperature, T c This refers to the cooling water temperature.

[0073] Finally, the real-time erosion rate is calculated based on the first derivative of the residual trench thickness with respect to time, and the mathematical expression is as follows:

[0074]

[0075] Step S3: Based on the temperature field distribution, residual trench thickness and erosion rate, a multi-dimensional threshold system is constructed, and the operating parameters of the water cooling system are dynamically adjusted through linkage control logic to achieve water cooling intensity matching.

[0076] In one specific embodiment, a multi-dimensional threshold system is first constructed, encompassing four key threshold categories to comprehensively cover the requirements for determining the operational status of the molten iron trough. The first category is the temperature threshold, which is divided into low, medium, and high thresholds based on the heat resistance characteristics and cooling requirements of different areas of the molten iron trough, satisfying the condition that low threshold < medium threshold < high threshold. For example, the low threshold is 300℃ (initial operation of the trough), the medium threshold is 600℃ (mid-operation), and the high threshold is 900℃ (late-stage operation). The second category is the residual thickness safety threshold, determined based on the minimum safe bearing thickness of the refractory material, ensuring that protective measures are triggered promptly when the lining thickness falls below this value. The third category is the erosion rate warning threshold, derived by combining the lining's designed service life and the average daily erosion rate. The fourth category is the temperature field uniformity threshold, which determines the qualification standard by quantifying the differences in temperature distribution across the entire area; its calculation expression is as follows:

[0077]

[0078] Where η is the temperature field uniformity coefficient, T max The highest temperature in the entire region, T min The lowest temperature in the entire region, T avg The average temperature across the entire region.

[0079] Next, the linkage control logic is executed. Based on the comparison results of the multi-dimensional threshold system and real-time monitoring parameters, the adjustment direction and specific values ​​of the water cooling system operating parameters are determined. The water cooling system operating parameters include the solenoid valve on / off status, solenoid valve opening degree, and water flow direction. Different control strategies correspond to different operating conditions: When the real-time temperature of a certain area is greater than or equal to the high threshold, or the real-time erosion rate is greater than or equal to the warning threshold, it is determined that there is an overheating risk or an accelerated erosion trend in that area, and the enhanced cooling mode needs to be activated. At this time, the backup water cooling branch is opened and the water flow direction is switched to counter-flow mode. Counter-flow mode can increase the heat transfer temperature difference between the cooling water and the lining, improve cooling efficiency, and at the same time, the solenoid valve opening degree is adjusted to 80%-100%, for example, to 90% to maximize the cooling flow rate; When the real-time temperature of the area is between the medium and high thresholds, the erosion rate is less than the warning threshold, and the residual thickness is greater than the safety threshold, it is determined that the operating condition of that area is stable, and the normal cooling mode needs to be maintained. At this time, the main branch is kept in co-flow mode. Co-flow mode is suitable for continuous heat dissipation when the operating condition is stable, and the solenoid valve opening degree is adjusted to 40%-80%, for example, to 60% to balance the cooling. However, efficiency and energy consumption are considered. When the real-time temperature of the area is ≤ the low threshold and the residual thickness is ≥ 1.5 times the safety threshold, it is determined that the area is overcooled and the cooling intensity needs to be reduced to avoid the lining cracking due to excessive temperature difference. At this time, some non-core water cooling branches are shut down and the opening of the solenoid valves of the remaining branches is adjusted to 20%-40%, for example, to 30%, to reduce the cooling flow. When the residual thickness of the area is ≤ the safety threshold, it is determined that the lining of the area has reached the safety limit and the cooling operation needs to be terminated immediately to prevent the risk caused by lining damage. At this time, all water cooling branches in the corresponding area are shut down.

[0080] The solenoid valve opening adjustment adopts PWM pulse width modulation, which controls the valve opening time by adjusting the duty cycle of the pulse signal, thereby achieving precise opening adjustment. For example, when the opening needs to be adjusted to 60%, the PWM signal period is set to 10ms, with a high level lasting for 6ms and a low level lasting for 4ms, so that the valve is in the open state for 60% of the time in one cycle, and the adjustment accuracy can reach ±3%. The water flow direction switching is achieved through a bidirectional solenoid valve. This valve has a built-in bidirectional valve core, and the flow direction of cooling water can be switched by changing the conduction direction of the valve core, without the need for additional pipeline modification.

[0081] Finally, the PLC control system transmits and executes control commands. Communication between the PLC control system and various sensors and actuators is achieved through an 8-port RS485 serial server using the Modbus-RTU communication protocol. This protocol is a commonly used serial communication protocol in the industrial field, characterized by stable data transmission and strong anti-interference capabilities. For example, temperature data collected by infrared temperature sensors and flow data collected by electromagnetic flowmeters are transmitted to the RS485 serial server via shielded signal cables, and then aggregated to the PLC. The shielded signal cables effectively block electromagnetic interference in the industrial environment, reducing data transmission errors. Long-distance data and command transmission is achieved through a wired transmission network constructed using a single-mode dual-core photoelectric converter, a 4-core single-mode armored optical cable, and an industrial switch. The single-mode dual-core photoelectric converter converts electrical signals into optical signals, which are then transmitted through the 4-core single-mode armored optical cable. Optical cables offer advantages such as long transmission distance and strong anti-interference capabilities. For example, when the distance between the iron ditch and the control center is 500m, this wired network can achieve attenuation-free data transmission. Simultaneously, a 5G redundant network is used, automatically switching to the 5G network to transmit data and commands when the wired network is interrupted due to a fault, ensuring uninterrupted data transmission. After receiving parameter commands from the linkage control logic output, the PLC control system drives the solenoid valve to perform switching, opening adjustment, and water flow direction switching operations, ultimately achieving dynamic matching of water cooling intensity.

[0082] Step S4: Based on the control feedback data, monitor the temperature field uniformity and parameter change trends, trigger a graded alarm, and correct the control parameters through a closed-loop optimization algorithm.

[0083] In one specific embodiment, the temperature distribution difference is first quantified based on the temperature field uniformity coefficient η. This coefficient ranges from 0 to 1, with a value closer to 1 indicating a more uniform temperature field distribution. The acceptable threshold is set at 0.8, meaning that when η ≥ 0.8, the temperature field uniformity is considered to meet the requirements. Simultaneously, the changing trends of key parameters such as residual trench thickness and real-time erosion rate are continuously tracked. For example, the residual thickness and erosion rate of each area are recorded every 5 minutes, and the trend curves are used to determine whether there are any abnormal situations of continuous increase or decrease in parameters, providing a trend basis for alarm triggering and parameter optimization. Next, a graded alarm mechanism is executed. The real-time temperature, residual trench thickness, and erosion rate output in step S2 are compared with the multi-dimensional threshold system set in step S3, triggering the corresponding level of alarm operation.

[0084] Specifically, the first level is a Level 1 warning, triggered when the real-time temperature of a certain area is between the medium and high thresholds, or the residual trough thickness is between the safety threshold and 1.2 times the safety threshold. When a Level 1 warning is triggered, the workstation immediately activates an audible and visual alarm, with a flashing red alarm light accompanied by a buzzer, and simultaneously pushes a text notification to the staff's remote terminal, informing them of the specific warning area and abnormal parameter values. The second level is a Level 2 emergency, triggered when the real-time temperature of a certain area is ≥ the high threshold, or the real-time erosion rate is ≥ the warning threshold. When a Level 2 emergency alarm is triggered, in addition to the audible and visual alarm of the Level 1 warning, a voice notification function is added. The workstation announces via voice, "Temperature in area XX exceeds limit, enhanced cooling has been activated," and simultaneously uploads the alarm information and current control measures to the cloud platform to ensure that remote management personnel have real-time access to the situation. The third level is a Level 3 critical, triggered when the residual trough thickness of a certain area is ≤ the safety threshold. When a Level 3 critical alarm is triggered, in addition to activating all-channel alarms, a shutdown reminder signal is sent to the production equipment associated with the molten iron trough, forcibly locking the maintenance process in that area and prohibiting the continued transport of molten iron until maintenance personnel confirm that the hazard has been eliminated.

[0085] Finally, a closed-loop optimization algorithm is implemented, using gradient descent to iteratively correct the control parameters to minimize temperature field non-uniformity and bring the uniformity coefficient to a acceptable threshold. The core of gradient descent is to calculate the gradient of temperature field non-uniformity with respect to the control parameters and update the parameters along the negative gradient direction. The iterative formula is:

[0086]

[0087] Among them, K n+1 K is the control parameter after the (n+1)th iteration. n The control parameters for the nth iteration include the solenoid valve opening coefficient and the water flow direction coefficient (a water flow direction coefficient of 1 represents downstream flow and -1 represents upstream flow); α is the iteration step size coefficient, ranging from 0.03 to 0.12; D is the temperature field non-uniformity, D = 1 - η. This is the partial derivative of the temperature field non-uniformity with respect to the nth control parameter, reflecting the degree of influence of parameter changes on the non-uniformity. For example, in the initial control parameters, the solenoid valve opening coefficient K0 = 0.6 in a certain region (corresponding to an opening of 60%) is calculated as follows: The parameter K1 after the first iteration is 0.6 - 0.08 × 0.3 = 0.576, corresponding to a solenoid valve opening of 57.6%. The adjusted parameter is sent to the PLC control system to drive the solenoid valve to adjust its opening. New temperature data is then collected to calculate the uniformity coefficient. If the target is still not met, the iteration continues until the uniformity coefficient η ≥ 0.8. During the iteration process, the impact of changes in the control parameters on the residual thickness and erosion rate is monitored in real time. For example, if the erosion rate in a certain area increases abnormally after the opening adjustment, the step size coefficient is appropriately reduced to avoid new risks arising from parameter correction and to ensure the optimization process is safe and controllable.

[0088] Step S5: Store the entire process data and control records, and generate a dynamic maintenance plan for the iron ditch based on data correlation analysis.

[0089] In one specific embodiment, the data stored in step S5 includes a standardized dataset, temperature field distribution data, residual trench thickness and erosion rate data, control parameter records, and alarm logs. Data storage utilizes an industrial-grade database (such as MySQL or SQL Server), automatically writing data at fixed intervals via a real-time communication interface between the workstation computer and the PLC control system. Simultaneously, data compression algorithms are employed to reduce storage footprint; for example, temperature field distribution data is averaged and compressed using a spatial grid, retaining high-precision data for critical areas while appropriately reducing sampling density in non-critical areas. The storage system supports data backtracking and querying; for instance, the temperature field distribution at a specific moment or the historical erosion rate curve of a certain area can be retrieved via a timestamp, providing comprehensive data support for subsequent data analysis and model optimization.

[0090] In one specific embodiment, a dynamic maintenance plan is generated based on a residual thickness attenuation prediction model. This model, adjusted for current residual thickness, real-time erosion rate, and temperature effects, predicts changes in residual thickness over a future period. The formula is as follows:

[0091]

[0092] Where Δt is the prediction time step; β is the temperature influence coefficient, with a value range of 0.2-0.6.

[0093] When the predicted residual thickness approaches the safety threshold, a maintenance reminder is generated, which includes the maintenance area and a suggested time window.

[0094] Furthermore, in order to regularly compare actual detection data with model calculation results, a parameter adaptive calibration step is also included. This step dynamically corrects the core parameters to maintain the long-term accuracy of the algorithm.

[0095] In one specific embodiment, the calibration cycle and data acquisition method are first determined. Considering the operational intensity and maintenance convenience of the trough, a regular inspection cycle is set, for example, the actual residual thickness of the lining is measured every 15 days. During the inspection, a non-contact laser thickness gauge or borehole sampling method is used to collect the actual residual trough wall thickness in typical areas (such as high-temperature erosion zones, edge stability zones, etc.) selected throughout the trough, denoted as δ. real (x,y), and simultaneously extract the model from the same time and region to calculate the residual thickness δ. calc (x,y) form pairs of calibration samples.

[0096] Next, the calibration factor is calculated, using the ratio of the actual thickness to the model-calculated thickness as the basis for parameter correction. The formula for calculating the calibration factor γ is:

[0097]

[0098] γ reflects the degree of deviation between the model's calculated value and the actual value: when γ > 1, it indicates that the residual thickness calculated by the model is too small, meaning the degree of erosion is overestimated; when γ < 1, it indicates that the model's calculated value is too large, meaning the degree of erosion is underestimated. For example, the actual measured thickness δ in a high-temperature erosion zone... real =46mm, thickness δ calculated by the same model calc =45mm, then the calibration factor γ = 1.022 for this area, indicating that the model's estimate of the erosion degree in this area is slightly higher than the actual situation.

[0099] Subsequently, the core parameters were updated based on the calibration factor. The parameters that needed to be corrected included the erosion coefficient k and the three-dimensional thermal conductivity λ of the refractory material. x ,λ y ,λ z The parameter update formula is:

[0100] k new =k old ·γ

[0101] λ x,new =λ x,old ·γ

[0102] λ y,new =λ y,pld ·γ

[0103] λ z,new =λ z,old ·γ

[0104] Where, k old , λ x,ild , λ y,ild and λ z,old k is the parameter value before the update. new , λ x,new , λ y,newand λ z,new This is the updated value.

[0105] Finally, accuracy verification is performed by substituting the updated parameters into the model from step S2 to recalculate the residual thickness and comparing the deviation rate with the actual detected value. If the deviation rate drops to within 5%, the calibration is considered effective; if it still exceeds the tolerance, the testing cycle is shortened and the number of testing points is increased until the accuracy meets the standard. Through continuous adaptive parameter calibration, the algorithm's calculation accuracy can be gradually optimized over time, providing a reliable parameter basis for the generation of subsequent water cooling control and maintenance plans.

[0106] In summary, this invention achieves continuous display of the temperature distribution across the entire iron trough through multi-dimensional measurement point layout and precise acquisition of multi-source data, combined with a temperature field reconstruction algorithm based on multi-source data fusion. It accurately captures changes in residual trough thickness and real-time erosion rate using the thermal conductivity law and erosion correction model. Furthermore, it dynamically matches water cooling intensity through a multi-dimensional threshold system and linked control logic, coupled with a graded alarm mechanism to promptly warn of various operational risks. A closed-loop optimization algorithm continuously corrects control parameters to ensure temperature field uniformity. The system systematically stores full-process data and generates dynamic maintenance plans based on a residual thickness attenuation prediction model, supplemented by parameter adaptive calibration to ensure long-term algorithm accuracy. Overall, this invention achieves non-contact safety monitoring of the iron trough's operating status, completely abandoning traditional high-risk detection methods, significantly improving the accuracy and adaptability of water cooling control, effectively avoiding problems such as excessive cooling and cracking of the lining or insufficient cooling accelerating erosion, enhancing equipment operational stability and continuity. Simultaneously, by scientifically predicting maintenance needs and optimizing maintenance strategies, it extends the service life of the lining and related equipment, reduces production energy consumption and maintenance costs, and promotes the upgrading of iron trough operation management towards precision, intelligence, and safety.

[0107] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 3As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database is used for dynamic control data of iron trough water cooling intensity based on infrared temperature field reconstruction. The I / O interfaces are used for information exchange between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for dynamic control of iron trough water cooling intensity based on infrared temperature field reconstruction.

[0108] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0109] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0110] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0111] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0112] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0113] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0114] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0115] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0116] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A dynamic regulation method for water cooling intensity of a trough based on infrared temperature field reconstruction, characterized in that, The method comprises the following steps: S1, collecting the multi-source data of the runner through a multi-dimensional measuring point layout sensing system, and obtaining a standardized data set after preprocessing; S2, generating a global temperature field distribution based on the standardized data set using a multi-source data fusion temperature field reconstruction algorithm, and calculating the residual bank thickness and real-time erosion rate by combining the heat conduction law and the erosion correction model; S3, constructing a multi-dimensional threshold system according to the temperature field distribution, the residual bank thickness and the erosion rate, and dynamically adjusting the operating parameters of the water cooling system through a linkage control logic to realize water cooling intensity matching; S4, monitoring the temperature field uniformity and parameter change trend based on the control feedback data, triggering a hierarchical alarm and correcting the control parameters through a closed-loop optimization algorithm; S5, storing the full-process data and control records, and generating a dynamic maintenance plan for the runner based on data correlation analysis.

2. The method of claim 1, wherein, The multi-dimensional measuring point layout in step S1 is specifically as follows: A measuring point group is arranged every 0.5-1m along the length direction of the runner, 2-6 measuring points are arranged along the width direction of the lining in each measuring point group, and an infrared temperature sensor is arranged vertically to the surface of the lining at each measuring point; An electromagnetic flowmeter is arranged in each water supply and return branch, and an environmental temperature and humidity sensor is arranged around the runner.

3. The method of claim 2, wherein, The multi-source data fusion temperature field reconstruction algorithm in step S2 is specifically a radial basis function interpolation fusion algorithm, and the formula is: Wherein, (x, y, z) is the global arbitrary space coordinates, (x j ,y j ,z j ) is the jth measuring point coordinates, and m is the total measuring point number. For fusion weight, σ s is the standard deviation of each type of data; Gaussian radial basis function, ε takes the value of 0.7-1.3; a k are polynomial coefficients, solved by least squares.

4. The method of claim 3, wherein, The real-time erosion rate in step S2 is calculated based on the first derivative of the residual bank thickness with respect to time, and the formula is: Wherein, k is the erosion coefficient, the value is 0.0008-0.006 mm / (h·℃); T m T is the real-time hot metal temperature, c T is the cooling water temperature.

5. The method of claim 4, wherein, The multi-dimensional threshold system in step S3 includes: Temperature thresholds, including a low threshold, a medium threshold and a high threshold, and satisfying low threshold < medium threshold < high threshold; A residual thickness safety threshold, an erosion rate warning threshold and a temperature field uniformity threshold; The linkage control logic is specifically as follows: When the temperature is greater than or equal to the high threshold, or the erosion rate is greater than or equal to the warning threshold, the standby water cooling branch is started and switched to the counter-flow mode, and the electromagnetic valve opening degree is adjusted to 80%-100%; When the temperature is between the medium threshold and the high threshold, the erosion rate is less than the warning threshold, and the residual thickness is greater than the safety threshold, the main branch is kept in the forward-flow mode, and the opening degree is adjusted to 40%-80%; When the temperature is less than or equal to the low threshold, and the residual thickness is greater than 1.5 times the safety threshold, some branches are closed, and the opening degree is adjusted to 20%-40%; When the residual thickness is less than or equal to the safety threshold, the water cooling branch in the corresponding area is immediately closed.

6. The method of claim 5, wherein, The water cooling system operating parameters in step S3 include the electromagnetic valve opening and closing state, the electromagnetic valve opening degree and the water flow direction; The opening degree adjustment adopts a PWM pulse width modulation mode, and the water flow direction switching is realized through a bidirectional electromagnetic valve.

7. The method of claim 6, wherein, The temperature field uniformity in step S4 is quantified by the difference ratio of the maximum temperature, the minimum temperature and the average temperature in the whole domain, and the qualified threshold is 0.8; The closed-loop optimization algorithm adopts a gradient descent method, and the control parameters are iteratively corrected to minimize the temperature field uniformity, and the iteration step length coefficient is 0.03-0.12, and the control parameters include the electromagnetic valve opening degree and the water flow direction coefficient.

8. The method of claim 7, wherein, The hierarchical alarm mechanism in step S4 includes three levels: Primary warning, trigger condition is temperature between medium threshold and high threshold, or residual thickness between safety threshold and 1.2 times safety threshold, alarm mode is workstation sound and light alarm and remote text notification; Secondary emergency, trigger condition is temperature ≥ high threshold or erosion rate ≥ warning threshold, alarm mode is sound and light alarm, voice notification and cloud platform push, synchronous record control measures; Critical level three, trigger condition is residual thickness ≤ safety threshold, alarm mode is full channel alarm and equipment shutdown reminder, forced locking maintenance process.

9. The method of claim 8, wherein, The stored data in step S5 includes standardized data set, temperature field distribution data, residual ridge thickness and erosion rate data, control parameter record, alarm log; The dynamic maintenance plan is generated based on the residual thickness attenuation prediction model, and the model formula is: wherein, Δt is the prediction time step; β is the temperature influence coefficient, the value range is 0.2-0.6; T avg is the global average temperature of the temperature field; When the predicted residual thickness approaches the safety threshold, a maintenance reminder including the maintenance area and the recommended time window is generated.

10. The method according to claim 9, wherein, It also includes a parameter self-adaptive calibration step: Based on the ratio of the actual residual ridge thickness detected periodically to the calculated value of the model, the erosion coefficient and the three-dimensional thermal conductivity of the refractory are calibrated; By updating the parameter value through calibration, the calculation accuracy of the algorithm is continuously optimized with the running time.

Images