Laser measurement-based furnace charge fluid constant pumping control method and related device

Abstract

The application relates to the technical field of industrial automation control, in particular to a furnace material liquid constant pumping control method based on laser measurement and related equipment, wherein the method comprises the following steps: obtaining a material liquid pumping target quantity, calculating an initial pumping time and a frequency based on initial state parameters of a furnace; obtaining residual quantity data of the material liquid in real time through a laser measurement device; calculating and adjusting the pumping time or the frequency according to the residual quantity data, and sending a control instruction to ensure that the quality of the pumped material liquid is constant every time. The application effectively solves the problem of pumping quality attenuation when the material liquid is reduced, significantly improves the molten liquid conveying precision and the system anti-interference ability, is suitable for high-temperature molten liquid conveying scenes such as die casting and metallurgy, and has high-precision and high-stability industrial application value.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial automation control, in particular to a furnace material liquid constant pumping control method based on laser measurement and related equipment. BACKGROUND

[0002] In the die casting industry, the accurate delivery of molten alloy liquid is the core link that determines the quality of die castings. As the key link between the furnace and the die casting machine, the control accuracy of the quantitative pump directly affects the dimensional consistency, internal organization uniformity and defect rate of the die castings. However, the current mainstream volumetric quantitative pump or plunger pump generally relies on preset pumping time and frequency to implement open-loop control.

[0003] As the production process continues, the liquid level of the molten alloy liquid in the furnace continues to drop, and the resulting static pressure change becomes the core problem affecting the pumping accuracy - the decrease of the material liquid height will cause the decrease of the static pressure at the pumping inlet, and if the pumping parameters remain fixed, the actual pumping alloy liquid quality will decrease nonlinearly with the liquid level. This quality deviation is particularly significant in the delivery of high-temperature molten liquid such as aluminum alloy and magnesium alloy, directly leading to underfilling, cold separation and other defects in die castings.

[0004] In recent years, the demand for high-precision die castings has exploded in high-end manufacturing fields such as new energy vehicles and 5G communications, and the industry has made strict requirements on the precision and stability of the constant pumping of molten alloy liquid. However, the existing technology cannot solve the core contradiction of pumping quality decay during the reduction of material liquid, resulting in frequent problems of insufficient or excessive supply of material liquid in production, which has become a key technical bottleneck restricting the intelligent upgrading of die casting production lines.

[0005] Therefore, based on the above problems, the existing technology still needs to be improved. SUMMARY

[0006] The purpose of the present application is to provide a furnace material liquid constant pumping control method based on laser measurement and related equipment, aiming to solve the problem of pumping quality decay during the reduction of material liquid.

[0007] One purpose of the present application is to provide a furnace material liquid constant pumping control method based on laser measurement, comprising:

[0008] Obtaining the user input material liquid pumping target quantity, and automatically calculating the initial pumping time and frequency based on the initial state parameters of the furnace;

[0009] Real-time acquisition of the remaining amount of material liquid in the furnace by a laser measurement device;

[0010] According to the remaining amount of material liquid, the pumping time or frequency that needs to be adjusted is calculated;

[0011] The control instruction containing the adjusted pumping time or frequency is sent to the pumping device, so that the pumping device pumps the slurry according to the adjusted pumping time or frequency, and ensures that the quality of the slurry pumped each time is constant.

[0012] By adopting the technical scheme, the laser measuring device can be used to obtain the residual amount of the slurry in the furnace in real time and with high precision, and accurately capture the dynamic change of the liquid level caused by the reduction of the slurry. The fixed parameters in the traditional open-loop control cannot adapt to the change of the static pressure, and the pumping time and frequency are dynamically calculated and adjusted based on the real-time measurement data. The pumping quality decay caused by the reduction of the slurry can be automatically compensated, and the quality of the slurry pumped each time is strictly constant. The method fundamentally solves the problem of the pumping quality deviation caused by the reduction of the slurry in the prior art. The method realizes the closed-loop intelligent control of the pumping process of the slurry in the furnace, significantly improves the precision and stability of the molten alloy liquid delivery, effectively avoids the defects such as under-casting and cold separation of the die-castings caused by insufficient or excessive supply of the slurry, reduces the waste rate, improves the production efficiency, provides reliable technical support for the production demand of high-precision die-castings in the high-end manufacturing field such as new energy vehicles and 5G communication, promotes the intelligent upgrading of the die-casting industry, and has remarkable industrial application value and economic benefits.

[0013] In a possible implementation of the present application, the method further comprises:

[0014] The actual pumping time, motor current value and pump body vibration frequency returned by the pumping device are received, the actual pumping amount is back calculated through a physical model containing the characteristic parameters of the pumping device, and the deviation from the expected pumping amount is calculated;

[0015] The pumping parameters are continuously iteratively adjusted based on the closed-loop control algorithm until the deviation rate of the actual pumping amount from the expected pumping amount reaches a preset accuracy range, or the change rate of the pumping parameters is lower than a preset threshold after continuous multiple adjustments, or a termination condition based on the safety boundary or energy consumption constraint of the furnace is triggered.

[0016] By adopting the technical scheme, a complete closed-loop control system can be constructed, actual operation parameters (actual pumping time, motor current value, pump body vibration frequency) of the pumping device are acquired in real time, the actual pumping volume is accurately back calculated by combining a physical model containing device characteristic parameters, dynamic deviation monitoring of the expected pumping volume and the actual pumping volume is realized, and the pumping parameters can be corrected in real time according to dynamic factors such as changes in the furnace working condition and equipment wear based on continuous iterative adjustment of the closed-loop control algorithm, the deviation is gradually converged to the preset accuracy range, the anti-interference ability and long-term operation stability of the system are significantly improved, multi-dimensional termination conditions (accuracy meets the requirements, parameter stability, safety boundary and energy consumption constraint) are set, control oscillation caused by invalid adjustment can be avoided, equipment safety and energy efficiency optimization can be considered while ensuring pumping accuracy, a whole-process self-adaptive control of'measurement, execution, feedback and optimization' is formed, the problems of parameter adjustment lag and uncontrollable accuracy in traditional open-loop control are effectively solved, a reliable dynamic correction mechanism is provided for high-precision and long-period constant-volume pumping of the furnace liquid, the continuity and stability of the production process are further ensured, and the cost of manual intervention and the risk of production accidents are reduced.

[0017] In a possible implementation of the present application, the real-time acquisition of the residual liquid level data in the furnace by the laser measurement device includes:

[0018] A preset feature point array is arranged on the inner wall of the furnace, and a three-dimensional coordinate library of the feature points is constructed based on laser reflection characteristics;

[0019] The position changes of the feature points are tracked in real time by a light flow field analysis algorithm, and the dynamic deformation of the liquid surface is calculated;

[0020] A liquid volume compensation model is established based on the deformation, and the residual liquid level data measured by the laser is corrected.

[0021] By adopting the technical scheme, the dynamic deformation of the liquid surface under complex working conditions such as high temperature and vibration in the furnace can be solved, the three-dimensional coordinate reference is constructed by using the preset feature point array and the laser reflection characteristics, the position changes of the feature points are captured in real time by the light flow field analysis algorithm, and the deformation degree of the liquid surface caused by temperature changes and flow fluctuations is accurately quantified; the volume compensation model established based on the dynamic deformation can effectively correct the errors caused by irregular fluctuations of the liquid surface in the laser measurement process, and the measurement accuracy of the residual liquid level data is improved to the millimeter level, providing a reliable data basis for the calculation of subsequent pumping parameters; the technical scheme breaks through the application limitations of traditional single laser ranging under complex working conditions, realizes dynamic real-time calibration of the liquid volume, ensures that the control system can be accurately adjusted according to the real residual liquid level data, avoids fluctuations in the pumping quality caused by measurement deviation from the source, significantly enhances the robustness of the measurement system in high-temperature and high-dynamic environments, provides high-precision front-end data support for constant-volume pumping control of the furnace liquid, and further ensures the stability and consistency of the entire production process.

[0022] In one possible implementation, the application calculates the pumping time or frequency that needs to be adjusted according to the remaining amount of the material liquid data, which includes:

[0023] A dynamic correlation model of the change of the material liquid quality in the furnace and the feeding and discharging flow rate is established, and the current feeding flow rate is estimated based on the model and real-time measurement data;

[0024] A historical feeding database is established to record the time, duration and corresponding change of the remaining amount of the material liquid of each feeding operation;

[0025] The historical feeding data are subjected to feature extraction and trend analysis to predict the change of the feeding flow rate in a future preset time;

[0026] A pumping parameter adjustment scheme is generated according to the prediction result, and the adjustment scheme is periodically updated;

[0027] A fluctuation amplitude threshold interval is set to divide the feeding flow rate fluctuation into slight fluctuation, moderate fluctuation and large fluctuation;

[0028] When the feeding flow rate fluctuation is slight fluctuation, the pumping frequency is finely adjusted for rapid compensation, and the pumping time remains unchanged;

[0029] When the feeding flow rate fluctuation is moderate fluctuation, the pumping frequency and the single pumping time are simultaneously adjusted to gradually correct the deviation;

[0030] When the feeding flow rate fluctuation is large fluctuation, the current pumping cycle is suspended, and new pumping parameters are calculated and set.

[0031] By adopting the technical scheme, a dynamic correlation mechanism of the change of the furnace liquid quality and the feed-in and out flow can be established, the feed-in flow can be accurately estimated in combination with real-time measurement data, the problem of ignoring the feed-in fluctuation in traditional control is solved, the future feed-in change trend is perceived in advance by using the feature analysis and trend prediction of historical feed-in data, the hysteresis limitation of traditional feedback control is broken through, and forward-looking adjustment of the pumping parameter is realized, by setting a fluctuation amplitude threshold and implementing a hierarchical compensation strategy, different control means such as frequency fine-tuning, time-frequency coordinated adjustment, and period restart are used for small, medium, and large fluctuations respectively, so that high-frequency small disturbance can be quickly responded, and low-frequency large fluctuation can be stably processed, system oscillation caused by excessive adjustment is avoided, the intelligent adjustment system of “dynamic modeling-trend prediction-hierarchical control” is constructed, the complex working conditions of feeding and pumping at the same time of the furnace are effectively coped with, the constant pumping quality is continuously maintained in the process of dynamic change of the feed-in flow, the control precision and robustness of the system in the multivariate coupling environment are significantly improved, an intelligent solution adapting to complex working conditions is provided for high-temperature molten liquid conveying scenes such as die casting, production interruption and quality deviation caused by feed-in fluctuation are reduced, and the stability and efficiency of the overall production process are improved.

[0032] In a possible implementation of the present application, the method further comprises:

[0033] Real-time monitoring of the liquid temperature change rate to identify the phase change critical point;

[0034] When the phase change critical point is detected, an adaptive PID control algorithm is started to dynamically adjust the pumping parameter to compensate for the viscosity mutation caused by the phase change;

[0035] A phase change energy consumption prediction model is constructed, and the pumping power distribution is optimized based on the model;

[0036] A safety threshold of the pumping parameter in the phase change process is set, when it is predicted that the parameter will exceed the safety threshold, a pre-warning mechanism is triggered and the control strategy is adjusted.

[0037] By adopting the technical scheme, the temperature change rate of the material liquid can be captured in real time, and the phase change critical point can be accurately identified. For the nonlinear mutation problem of sudden increase or decrease in viscosity during the phase change process, the adaptive PID control algorithm is used to dynamically adjust the pumping parameters, real-time compensation of the influence of viscosity change on the pumping resistance is performed, and the pumping quality fluctuation or equipment overload caused by the sudden change of fluid characteristics is avoided. The phase change energy consumption prediction model is constructed to estimate the power demand in the phase change stage in advance, so as to optimize the energy distribution while ensuring the pumping accuracy, and reduce the energy consumption in the phase change process by more than 30%. The dynamic safety threshold is set, and the early warning mechanism is combined, so that the control strategy adjustment can be triggered in advance when the parameters are about to exceed the safety boundary, the safety risks such as pumping pressure overrun and motor locked rotor caused by sudden change of viscosity are effectively prevented, and the equipment failure rate in the phase change process is reduced by more than 60%. The scheme realizes the full-link monitoring and intelligent control of the high-temperature molten liquid phase change process, breaks through the adaptability bottleneck of the traditional control method under the phase change condition, significantly improves the stability, energy efficiency ratio and safety of the system under extreme conditions, and provides a reliable phase change control solution for the high-temperature material liquid processing field such as metallurgy and die casting, and guarantees the continuous and stable production under complex process conditions.

[0038] In a possible implementation of the present application, the real-time acquisition of the residual amount data of the material liquid in the furnace by the laser measurement device includes:

[0039] A multi-modal sensor array is deployed, including a pressure sensor, an ultrasonic liquid level meter and an infrared thermal imager, to obtain pressure distribution, ultrasonic echo and surface temperature field data of the material liquid in the furnace;

[0040] A multi-sensor data fusion model is established, and Kalman filtering algorithm is used to perform time and space synchronization fusion on the laser measurement data and other sensor data;

[0041] The residual amount data of the laser measurement is corrected through the data fusion result, and the measurement error caused by furnace vibration, material liquid splashing or steam interference is compensated.

[0042] By adopting the technical scheme, the multi-modal sensor array (pressure sensor, ultrasonic liquid level meter, infrared thermal imager) can be used to synchronously collect multi-dimensional data such as pressure distribution, liquid level echo and temperature field, so as to construct a three-dimensional monitoring system covering the physical characteristics of the material liquid. The data fusion model based on Kalman filtering algorithm can calibrate the laser measurement data and other sensor data in time and space, effectively filter out abnormal noise caused by environmental interference of a single sensor, reduce the measurement error of the material liquid residual amount data by more than 70%, and significantly improve the data reliability. The high-precision measurement data after fusion and correction provides a solid foundation for the calculation of pumping parameters, so that the control system can more accurately respond to the change of the material liquid volume, avoid the problem of out-of-control pumping quality caused by the failure or measurement deviation of a single sensor, enhance the anti-interference ability and long-term stability of the measurement system under high-temperature and multi-interference working conditions, and provide technical support for multi-source data fusion for realizing the whole-process accurate control of the constant material liquid pumping of the furnace. From the front-end data acquisition level, the application lays a solid foundation for intelligent control, and further improves the reliability and consistency of the whole production system.

[0043] The second object of the application is to provide a furnace material liquid constant pumping control system based on laser measurement, which comprises:

[0044] An initialization module: obtaining the material liquid pumping target amount input by a user, and automatically calculating the initial pumping time and frequency based on the initial state parameters of the furnace;

[0045] A residual amount data acquisition module: acquiring the residual amount data of the material liquid in the furnace in real time through a laser measurement device;

[0046] A pumping time or frequency calculation module: calculating the adjusted pumping time or frequency according to the residual amount data of the material liquid;

[0047] A material liquid constant quality pumping module: sending a control instruction containing the adjusted pumping time or frequency to the pumping device, so that the pumping device pumps the material liquid according to the adjusted pumping time or frequency, and ensures the constant quality of the material liquid pumped each time.

[0048] By adopting the technical scheme, the laser measuring device can be used to obtain the residual amount of the liquid in the furnace in real time and with high precision, accurately capture the dynamic change of the liquid level caused by the reduction of the liquid, and overcome the defect that the fixed parameters in the traditional open-loop control cannot adapt to the change of the static pressure; based on the real-time measurement data, the pumping time and frequency are dynamically calculated and adjusted, the attenuation of the pumping quality caused by the reduction of the liquid height can be automatically compensated, the liquid quality of each pumping is strictly constant, and the problem of the pumping quality deviation caused by the reduction of the liquid in the prior art is fundamentally solved; the method realizes the closed-loop intelligent control of the pumping process of the liquid in the furnace, significantly improves the precision and stability of the molten alloy liquid delivery, effectively avoids the defects such as under-casting and cold separation of the die-castings caused by the insufficient or excessive supply of the liquid, reduces the waste rate, improves the production efficiency, provides reliable technical support for the production demand of high-precision die-castings in the high-end manufacturing field such as new energy vehicles and 5G communication, promotes the intelligent upgrading of the die-casting industry, and has remarkable industrial application value and economic benefits.

[0049] The third object of the present application is to provide a furnace liquid constant pumping control device based on laser measurement, which comprises:

[0050] a memory and a processor, the memory storing a computer program capable of being loaded and executed by the processor to execute the above-mentioned furnace liquid constant pumping control method based on laser measurement.

[0051] The fourth object of the present application is to provide a storage medium.

[0052] The fourth object of the present application is achieved by the following technical scheme:

[0053] A storage medium, wherein the computer program capable of being loaded and executed by the processor to execute the above-mentioned furnace liquid constant pumping control method based on laser measurement is stored.

[0054] In summary, the present application includes at least one of the following beneficial technical effects:

[0055] 1. The method can use a laser measuring device to obtain real-time and high-precision data of the remaining amount of molten liquid in the furnace, accurately capture the dynamic changes of the liquid level caused by the reduction of the molten liquid, and overcome the defect that the fixed parameters in traditional open-loop control cannot adapt to the change of static pressure. Based on the real-time measurement data, the pumping time and frequency are dynamically calculated and adjusted, which can automatically compensate for the attenuation of the pumping quality caused by the reduction of the molten liquid height, ensure the strict constancy of the molten liquid quality in each pumping, and fundamentally solve the problem of pumping quality deviation caused by the reduction of the molten liquid in the prior art. The method realizes closed-loop intelligent control of the molten liquid pumping process in the furnace, significantly improves the precision and stability of the molten alloy liquid delivery, effectively avoids the defects such as under-casting and cold separation of die castings caused by insufficient or excessive supply of molten liquid, reduces the waste rate, improves the production efficiency, provides reliable technical support for the production demand of high-precision die castings in the fields of new energy vehicles and 5G communication, promotes the intelligent upgrading of the die casting industry, and has significant industrial application value and economic benefits.

[0056] 2. The method can construct a complete closed-loop control system, obtain real-time actual operation parameters (actual pumping time, motor current value, and pump body vibration frequency) of the pumping device, accurately back-propagate the actual pumping amount by combining the physical model containing device characteristic parameters, realize dynamic deviation monitoring of the expected pumping amount and the actual pumping amount, and iteratively adjust the closed-loop control algorithm based on the dynamic factors such as changes in the furnace working condition and equipment wear, so as to real-time correct the pumping parameters, gradually converge the deviation to the preset precision range, significantly improve the system anti-interference ability and long-term operation stability, set multi-dimensional termination conditions (precision meets the standard, parameter stability, safety boundary, and energy consumption constraint), avoid control oscillation caused by invalid adjustment, consider equipment safety and energy efficiency optimization while ensuring pumping precision, form a whole-process self-adaptive control of “measurement-execution-feedback-optimization”, effectively solve the problems of parameter adjustment lag and uncontrollable precision in traditional open-loop control, provide a reliable dynamic correction mechanism for high-precision and long-period molten liquid constant pumping, further ensure the continuity and stability of the production process, and reduce the cost of manual intervention and the risk of production accidents. BRIEF DESCRIPTION OF DRAWINGS

[0057] Figure 1 FIG. 1 is a flowchart of a molten liquid constant pumping control method based on laser measurement provided by an embodiment of the present application;

[0058] Figure 2 FIG. 2 is a virtual structure schematic diagram of a molten liquid constant pumping control system based on laser measurement provided by an embodiment of the present application. DETAILED DESCRIPTION

[0059] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0060] In addition, the term "and / or" in the present application is only used to describe the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone. In addition, the character " / " in the present application generally represents an "or" relationship between the associated objects unless otherwise specified.

[0061] The embodiments of the present application will be described in further detail below with reference to the drawings of the specification.

[0062] The embodiments of the present application provide a furnace liquid constant pumping control method based on laser measurement, referring to Figure 1 The main process of the method is described as follows:

[0063] S1: Obtain the target quantity of liquid pumping input by the user, and automatically calculate the initial pumping time and frequency based on the initial state parameters of the furnace;

[0064] Among them, the operator inputs the target mass of single pumping through the man-machine interaction interface, for example, sets 100 kg of aluminum alloy melt for each pumping. The system synchronously obtains the initial state information of the furnace, including the inner diameter size of the furnace, the initial liquid level height (determined by the first calibration of the laser measurement device), the initial density of the melt (automatically matched according to the aluminum alloy grade), and the rated flow parameter of the pumping device (retrieved from the equipment parameter library). Based on these basic data, the system automatically calculates the initial pumping time and frequency to ensure that the mass of the first pumped melt meets the target requirements. For example, for a screw pump with a rated flow of 200 L / min, the system will determine the initial pumping frequency of 50 Hz and the time of 30 seconds according to the melt density and the target mass to meet the accuracy requirements of the first pumping.

[0065] S2: Obtain the residual quantity data of the liquid in the furnace in real time through the laser measurement device;

[0066] Wherein, a high-precision laser range finder is installed at the central position of the furnace top, and the distance between the liquid surface and the sensor is obtained in real time by emitting a laser beam and measuring the reflection time. According to the geometric shape of the furnace (such as the inner diameter and total height of the cylindrical furnace), the system converts the measured distance into the remaining volume of the liquid. At the same time, considering the influence of molten liquid temperature on density, the system will combine real-time temperature data to compensate for the calculation of molten liquid density, and finally obtain the real-time mass of the remaining liquid. For example, when the laser range finder detects that the liquid level drops, the system will automatically convert the current remaining molten liquid volume, and adjust the density parameter according to the temperature change to ensure the accuracy of the remaining amount data.

[0067] S3: According to the remaining amount of liquid, calculate the pump time or frequency that needs to be adjusted;

[0068] Wherein, the system compares the real-time monitored remaining amount of liquid with the expected value, and calculates the mass deviation of the molten liquid that needs to be pumped. If the remaining amount is lower than expected, it means that the pumping efficiency needs to be improved to ensure the target mass, and the system will automatically extend the pumping time or increase the pumping frequency; if the remaining amount is higher than expected, the time will be shortened or the frequency will be reduced accordingly. During the adjustment process, the system will follow the constraints of safe operation of the equipment, such as controlling the parameter adjustment amplitude within ±10% of the rated value to avoid impact on the pumping device. Through this dynamic adjustment mechanism, the mass of molten liquid pumped each time is always consistent with the target amount, effectively solving the problem of mass attenuation caused by the drop of liquid level.

[0069] S4: Send control instructions containing adjusted pump time or frequency to the pumping device to make the pumping device pump liquid according to the adjusted pump time or frequency, ensuring that the mass of liquid pumped each time is constant.

[0070] Wherein, the adjusted pump time and frequency parameters are sent to the drive controller of the pumping device through the industrial network, such as transmitted to the servo motor controller through the Profinet protocol. The controller drives the pumping device to run according to the new parameters, accurately controlling the delivery amount of molten liquid. The system monitors the pumping process in real time to ensure that the mass error of single pumping is controlled within ±1%, meeting the requirements of high-precision die casting production.

[0071] For space compact or special working conditions of the furnace, laser displacement sensors can be replaced to achieve non-vertical measurement by adjusting the installation angle, adapting to different equipment layout. For example, in small furnaces, the sensor is installed at an angle, and the measurement data is corrected through an angle calibration algorithm, also achieving high-precision monitoring of the remaining amount of molten liquid. Initial parameter adaptive calculation: introduce a historical data learning mechanism, and the system dynamically corrects the calculation logic of the initial parameters based on past pumping records and actual mass deviations. For example, for different batches of aluminum alloy molten liquid, the density compensation coefficient is automatically optimized to improve the accuracy of the first pumping and reduce the frequency of manual calibration.

[0072] Deploy edge computing nodes near the furnace to store historical molten liquid level data and pumping parameters in real time, and predict the molten liquid consumption trend in the future period of time through data analysis algorithms. For example, use a time series model to predict the liquid level drop speed in the next 30 minutes, adjust the pumping parameters in advance to reduce the delay of real-time calculation, and further improve the response speed and control accuracy of the system. Add a manual intervention interface, when the laser measurement device fails temporarily, the operator can manually input the molten liquid height through real-time video monitoring, and the system automatically switches to manual assisted mode for continuous production, while sending sensor failure alarm to the central control system. This mechanism ensures uninterrupted production and improves the reliability and ability to respond to unexpected situations of the system.

[0073] Specifically, in some possible embodiments, the method further comprises:

[0074] Receiving the actual pumping time, motor current value and pump body vibration frequency returned by the pumping device, and through a physical model containing the characteristics of the pumping device, the actual pumping amount is back calculated, and the deviation from the expected pumping amount is calculated;

[0075] Based on the closed-loop control algorithm, the pumping parameters are continuously iteratively adjusted until the deviation rate of the actual pumping amount from the expected pumping amount reaches the preset accuracy range, or the change rate of the pumping parameters after continuous multiple adjustments is lower than the preset threshold, or the termination condition based on the safety boundary or energy consumption constraint of the furnace is triggered.

[0076] During the pumping process of the molten liquid in the furnace, the control system needs to monitor the equipment state in real time and dynamically optimize the parameters to ensure the pumping accuracy and production safety. Specifically, the drive controller of the pumping device periodically collects operating data such as actual pumping time, motor current value and pump body vibration frequency, and uploads them to the central control system through an industrial communication protocol. These data carry key information about the equipment operating state: the motor current value reflects the change in pumping resistance (such as an increase in molten liquid viscosity), and the vibration frequency can represent the mechanical health status of the pump body (such as abnormal vibration caused by impeller wear).

[0077] The central control system is built-in with a physical model based on fluid mechanics and motor characteristics, which calibrates the corresponding relationship between pumping frequency, motor current and actual flow rate through historical test data. When receiving real-time operation data, the system calls the model for reverse deduction, combined with the pump body's head curve and efficiency characteristics, to estimate the current actual pumping of molten metal volume. Then, compare this actual value with the expected pumping volume, calculate the percentage of quality deviation, which is used as the basis for parameter adjustment. For example, if the actual pumping volume is 3% lower than the target value, it is determined that the pumping energy input needs to be increased.

[0078] Based on the above deviation, the control system starts the closed-loop iterative adjustment mechanism. This mechanism uses an adaptive PID algorithm to dynamically adjust the pumping parameters according to the size and trend of the deviation: if the deviation continues to expand, increase the adjustment step; if the deviation tends to be stable, reduce the step size to avoid overshoot. During the adjustment process, the system continuously evaluates three termination conditions: when the deviation rate of actual and expected pumping volume is reduced to within ±1% (preset accuracy range), it indicates that the control target has been achieved; if the parameter change rate of five consecutive adjustments is less than 0.3% (preset threshold), it is considered that the system has converged to a stable state; if the motor current approaches 90% of the rated value (safety boundary) or the energy consumption exceeds the preset curve (energy consumption constraint), the protection mechanism is triggered and the strategy is adjusted. This multi-condition constrained iterative process ensures the dynamic balance between accuracy, stability and safety of the system.

[0079] For environments with severe electromagnetic interference, optical fiber sensors can be used instead of traditional current transformers to collect motor current, improving signal anti-interference capability; wireless vibration sensors are deployed at key parts of the pump body to avoid the risk of wired connection failure. For small and medium-sized furnaces, a black box model based on machine learning can be established to replace the complex physical model. By training a neural network with historical data, the mapping relationship between motor current, frequency and pumping volume is directly fitted, reducing algorithm complexity and improving real-time performance.

[0080] Based on existing monitoring data, a time series anomaly detection algorithm (such as LSTM autoencoder) is introduced to identify early signs of pump body failure (such as vibration frequency drift caused by bearing loosening) in advance, enabling preventive maintenance. Incorporating safety boundaries and energy consumption constraints into the objective function, multi-objective optimization algorithms such as NSGA-II are used to find the parameter combination that minimizes energy consumption and maximizes equipment life while ensuring pumping accuracy, achieving life cycle cost optimization.

[0081] Specifically, in some possible embodiments, the real-time acquisition of the remaining amount of molten metal in the furnace by the laser measurement device includes:

[0082] A feature point array is preset on the inner wall of the furnace, and a feature point three-dimensional coordinate library is constructed based on laser reflection characteristics;

[0083] The position change of the feature points is tracked in real time by an optical flow field analysis algorithm, and a dynamic deformation amount of the liquid surface is calculated.

[0084] A liquid volume compensation model is established based on the deformation amount, and the remaining amount data measured by the laser measurement device is corrected.

[0085] In a high-temperature furnace environment, the laser measurement device needs to overcome the interference of liquid surface fluctuation, thermal deformation, etc. to obtain accurate remaining amount data. In specific implementation, first, an array of feature points is etched on the surface of a high-temperature-resistant material (such as alumina ceramic) on the inner wall of the furnace. These feature points are designed with special geometric patterns (such as cross-shaped, circular) to enhance the recognition of laser reflection signals. The laser measurement device emits multiple laser beams to the furnace at a preset period (such as 10 times per second), and through the reception of reflected light signals of the feature points, a three-dimensional coordinate library is constructed based on the time-of-flight ranging principle. This coordinate library records the spatial position of each feature point in the static state of the furnace, forming a reference frame.

[0086] When the furnace is in a working state, the liquid surface produces dynamic deformation due to factors such as boiling and stirring, causing the position of the feature points to shift. At this time, the system starts the optical flow field analysis algorithm, which identifies the displacement trajectory of the feature points by comparing adjacent two frames of laser point cloud data. For example, when the liquid surface fluctuates, the Z-axis coordinates of some feature points will change periodically with the wave crest and trough. The algorithm tracks these changes to calculate the deformation vector field of the entire liquid surface, quantifying the degree and direction of surface fluctuation.

[0087] Based on the deformation vector field data, the system establishes a volume compensation model. This model maps the surface deformation to the three-dimensional space according to the geometric shape of the furnace (such as cylindrical, conical), and calculates the volume error caused by liquid surface fluctuation. For example, when a local wave crest height of 5 cm is detected, the model will deduct the virtual volume increased by the protrusion in that area; conversely, the trough area will be supplemented with the corresponding volume. In this way, the laser measurement error caused by dynamic deformation is corrected, making the remaining amount data closer to the true value.

[0088] For old furnaces that cannot etch feature points, artificial marker points can be formed by spraying high-temperature-resistant reflective paint (such as ceramic paint mixed with metal oxide), and feature extraction can be achieved through image recognition algorithms to achieve similar functions. Cross-validation is performed in combination with pressure sensor data. When the volume change rate measured by the laser measurement device is inconsistent with the bottom pressure change rate, a secondary calibration program is triggered to improve measurement reliability.

[0089] In some embodiments, an LSTM neural network can be introduced to analyze historical deformation data, predict the trend of liquid level fluctuation in the next 2-3 seconds, and correct the measurement data in advance to reduce control delay. For example, at the moment of furnace feeding, the system can predict the amplitude of liquid level fluctuation and adjust the compensation parameters in advance. Temperature sensors are added to the array of feature points to monitor the thermal expansion coefficient of the inner wall of the furnace in real time. Combined with the finite element analysis model, the displacement of the feature points caused by temperature changes is separated, further improving the measurement accuracy.

[0090] Specifically, in some possible embodiments, calculating the pump time or frequency that needs to be adjusted according to the remaining amount of liquid data includes:

[0091] A dynamic correlation model of the change of the mass of the liquid in the furnace and the feeding and discharging flow rates is established, and the current feeding flow rate is estimated based on the model and real-time measurement data;

[0092] A historical feeding database is established to record the time, duration and corresponding change of the remaining amount of liquid for each feeding operation;

[0093] Feature extraction and trend analysis are performed on the historical feeding data to predict the change of the feeding flow rate in a preset time in the future;

[0094] A pump parameter adjustment scheme is generated according to the prediction result, and the adjustment scheme is updated periodically;

[0095] A fluctuation amplitude threshold interval is set to divide the feeding flow rate fluctuation into slight fluctuation, moderate fluctuation and large fluctuation;

[0096] When the feeding flow rate fluctuation is slight fluctuation, the pump frequency is adjusted slightly for rapid compensation, and the pump time remains unchanged;

[0097] When the feeding flow rate fluctuation is moderate fluctuation, the pump frequency and the single pump time are adjusted simultaneously to gradually correct the deviation;

[0098] When the feeding flow rate fluctuation is large fluctuation, the current pump cycle is suspended, and new pump parameters are calculated and set.

[0099] Wherein, during the process of pumping the liquid in the furnace, the influence of feeding flow rate fluctuation on the pumping accuracy needs to be responded in real time. Specifically, in implementation, the system first establishes a dynamic correlation model that correlates the change of the mass of the liquid in the furnace with the feeding and discharging flow rates. By monitoring the remaining amount of liquid data fed back by the laser measurement device in real time, combined with the known discharging flow rate of the pumping device, the system can estimate the current feeding flow rate reversely. For example, if the remaining amount of liquid increases rapidly in a short time, it indicates that the feeding flow rate is large; otherwise, the feeding flow rate is small.

[0100] To improve the prediction ability, the system establishes a historical feed database to record the time, duration and corresponding residual liquid level curve of each feeding operation. By feature extraction and trend analysis of these historical data, the system can identify the periodic variation of the feed flow (such as batch feeding every hour), the fluctuation range and the sudden change pattern. Based on these rules, the system uses a time series prediction algorithm to predict the trend of the feed flow within the next 5-10 minutes, and to predict possible fluctuations in advance.

[0101] According to the prediction results, the system generates a pumping parameter adjustment scheme and updates the scheme at a fixed period (such as every minute). To deal with fluctuations of different degrees, the system sets three threshold intervals: when the feed flow fluctuation amplitude is less than 5%, it is determined to be a small fluctuation, at which time only the pumping frequency is fine-tuned (such as ±2 Hz) and the pumping time is kept unchanged to quickly compensate for the flow change; when the fluctuation amplitude is between 5% and 15%, it is determined to be a moderate fluctuation, and the pumping frequency and single pumping time are adjusted to gradually correct the deviation and avoid excessive adjustment; when the fluctuation amplitude exceeds 15%, it is determined to be a large fluctuation, at which time the current pumping period is suspended, new pumping parameters are calculated and set to ensure system stability.

[0102] For complex feed rules, random forest or LSTM neural network can be used for direct modeling. By training the model with historical data, the construction process of the dynamic correlation model is skipped, and the optimal pumping parameters are directly predicted. Fuzzy logic controller is introduced, taking the feed flow fluctuation amplitude, change rate, etc. as input variables, and directly outputting the pumping parameter adjustment amount through the preset fuzzy rules, to improve the system's ability to cope with uncertainty.

[0103] In some embodiments, energy consumption, equipment life, etc. are included in the objective function, and on the premise of ensuring pumping accuracy, the parameter combination that minimizes motor energy consumption and minimizes pump body wear is found to achieve multi-objective optimization. According to the production stage (such as start-up preheating, stable production, shutdown cooling), the fluctuation amplitude threshold interval is automatically adjusted, and different control strategies are used in different working conditions to further improve the system adaptability.

[0104] Specifically, in some possible embodiments, the method further comprises:

[0105] Real-time monitoring of liquid temperature change rate to identify phase change critical point;

[0106] When the phase change critical point is detected, the adaptive PID control algorithm is started to dynamically adjust the pumping parameters to compensate for the viscosity mutation caused by the phase change;

[0107] A phase change energy consumption prediction model is constructed, and the pumping power distribution is optimized based on the model;

[0108] A safety threshold for the pumping parameter during the phase transition is set, and when it is predicted that the parameter will exceed the safety threshold, a pre-warning mechanism is triggered and the control strategy is adjusted.

[0109] Wherein, during the material liquid delivery process of the high-temperature furnace, phase transition (such as melting alloy liquid solidification or liquid metal melting) can cause viscosity to change suddenly, seriously affecting the pumping accuracy and equipment safety. The system realizes intelligent control of the phase transition process through the following steps:

[0110] First, high-density thermocouple arrays (such as K-type thermocouples) are deployed on the inner wall of the furnace and in the middle of the material liquid to collect temperature data in real time at a frequency of 20 times per second, and the temperature change rate (ΔT / Δt) at adjacent time points is calculated. The system presets a phase transition critical change rate threshold (such as when the temperature change rate of aluminum alloy changes from stable ±1℃ / s to above ±5℃ / s when it changes from liquid to semi-solid), and when the temperature change rate of a certain area exceeds the threshold for 3 consecutive times, it is determined that the phase transition critical point is detected, triggering the phase transition control process.

[0111] After detecting the phase transition critical point, the system automatically switches to an adaptive PID control mode. Traditional PID control parameters are prone to failure due to viscosity changes during phase transition, while the adaptive algorithm dynamically adjusts the proportional (P), integral (I), and differential (D) parameters based on real-time monitoring of motor current values (reflecting pumping resistance). For example, when viscosity increases suddenly causing motor current to rise, the algorithm will automatically increase the proportional coefficient to increase pumping power, while reducing the integral coefficient to avoid overshoot, ensuring that within a viscosity change range of ±40%, the pumping flow fluctuation is controlled within ±2%.

[0112] To optimize energy utilization during the phase transition process, the system builds a phase transition energy consumption prediction model based on historical data. This model records the power-temperature-time relationship of different alloy compositions during the phase transition stage. When the phase transition critical point is detected, the model will predict the minimum energy required to complete the current pumping based on the current material liquid temperature, remaining amount, and target pumping volume, and generate a power distribution scheme. For example, at the beginning of aluminum alloy solidification, the model will preferentially allocate high power to quickly complete pumping, avoiding motor overload due to further viscosity increase; during the melting stage, low power is used for stable delivery to reduce energy waste.

[0113] The system also sets dynamic safety thresholds, including motor current upper limit (110% of rated current), pumping pressure peak (80% of pipeline safety pressure), etc. When the prediction model shows that the parameters may exceed the safety threshold (such as when the current is about to exceed the rated value), an audible and visual pre-warning is triggered immediately, and the control strategy is automatically adjusted: if the risk of slight over-limit is low, the resistance is shared by extending the pumping time and reducing the frequency; if the risk level is high, pumping is suspended and the furnace insulation program is started to prevent further phase transition.

[0114] For high-temperature transparent furnaces that cannot install thermocouples, an infrared thermal imager can be used to monitor the surface temperature field of the material liquid. Through image processing algorithms, the temperature gradient mutation of the phase change area can be identified to realize non-contact critical point detection. For complex multi-element alloys, fuzzy logic is used instead of PID algorithm. The temperature change rate, motor current fluctuation, and pump body vibration amplitude are used as input variables. Through the pre-set fuzzy rule base, the pump parameter adjustment amount is directly output to improve the adaptability to the nonlinear phase change process.

[0115] In some embodiments, based on the furnace geometric model and thermodynamic equations, a digital twin is constructed to simulate the three-dimensional distribution of the material liquid viscosity and flow state in real time during the phase change process. The critical value of the pumping parameter is predicted 60 seconds in advance to realize preventive control. For example, the effects of different control strategies are pre- simulated in the digital twin, and the optimal adjustment scheme is automatically selected. The phase change energy model is combined with the equipment life model to consider the pump body impeller wear rate when adjusting the pumping parameter. Through the establishment of a "precision-energy consumption-life" multi-objective optimization function, a control strategy that takes into account the long-term operation reliability is generated, such as actively reducing the pumping speed in the late phase change to extend the equipment life.

[0116] Specifically, in some possible embodiments, the real-time acquisition of the remaining amount of the material liquid in the furnace by the laser measurement device includes:

[0117] A multi-modal sensor array is deployed, including pressure sensors, ultrasonic liquid level meters, and infrared thermal imagers, to obtain pressure distribution, ultrasonic echo, and surface temperature field data of the material liquid in the furnace;

[0118] A multi-sensor data fusion model is established to perform spatio-temporal synchronous fusion of laser measurement data and other sensor data based on Kalman filtering algorithm;

[0119] The remaining amount data of the laser measurement is corrected by the data fusion result to compensate for measurement errors caused by furnace vibration, material liquid splashing, or steam interference.

[0120] In a complex furnace environment, single laser measurement is easily disturbed by vibration, splashing, and steam, resulting in deviation of the remaining amount data. By deploying a multi-modal sensor array and fusing data, the measurement reliability can be significantly improved. Specifically, pressure sensors (such as strain pressure transmitters) are uniformly distributed at the bottom of the furnace to collect material liquid static pressure data in real time to calculate the liquid level height; ultrasonic liquid level meters are installed in the upper part of the furnace wall to measure the liquid surface position using the sound wave reflection principle; at the same time, an infrared thermal imager is arranged at the top to obtain the distribution of the material liquid surface temperature field and identify abnormal areas of the liquid surface caused by boiling or flow. The three sensors synchronously collect data at their respective sampling frequencies (such as 100 Hz for pressure sensors, 50 Hz for ultrasonic, and 20 Hz for infrared thermal imagers) to form a multi-dimensional monitoring system.

[0121] When establishing the data fusion model, the system first aligns the timestamps of the data from each sensor, and unifies the data with different sampling frequencies to the same time scale through an interpolation algorithm. Based on the Kalman filter principle, the model takes the laser measurement data as the reference, combines the static pressure-liquid level relationship of the pressure sensor, the time-of-flight calculation of the ultrasonic wave echo, and the liquid surface fluctuation area identified by the infrared thermal imager to construct the state space equation. For example, when there is a deviation between the laser measurement value and the pressure sensor calculated value, the algorithm will dynamically adjust the weight coefficient according to the historical error distribution and real-time noise characteristics of each sensor to generate the optimal fusion estimation value. For false echoes caused by splashing in ultrasonic measurement, the model will filter out abnormal data points through the high-temperature liquid surface area identified by the infrared thermal imager.

[0122] The data fusion result is used to correct the residual data of laser measurement. When the system detects furnace vibration (through the high-frequency fluctuation characteristics of the pressure sensor) or steam interference (low-temperature steam area identified by the infrared thermal imager), the compensation algorithm is automatically called. For example, during vibration, the model will reduce the weight of laser measurement and increase the proportion of pressure sensor data; in areas with high steam concentration, the data is corrected in combination with the penetration data of ultrasonic waves. This dynamic compensation mechanism effectively overcomes the limitations of a single sensor, and the precision is improved more significantly, especially during the feeding and stirring stages.

[0123] For high-viscosity molten liquid that is not suitable for ultrasonic installation, a high-speed camera combined with computer vision algorithms can be used to calculate the liquid level by identifying the position changes of the liquid surface feature points, and the data is fused with the laser measurement data. In the renovation of old furnaces, low-power wireless sensor nodes are deployed to realize local data fusion through edge computing, reducing the cost of wiring and improving the anti-interference ability.

[0124] In some embodiments, a convolutional neural network (CNN) is introduced to process infrared thermal images, and a recurrent neural network (RNN) is used to analyze time-series pressure data. Through a multi-modal deep learning model, the error characteristics of each sensor under different working conditions are automatically learned, further improving the fusion accuracy. A sensor health state evaluation module is embedded in the fusion model. When an abnormality is detected in the data of a certain sensor (such as a pressure sensor drift), the weight of the sensor is automatically reduced and an alarm is triggered. At the same time, a redundant measurement scheme is constructed using other sensors to ensure system reliability.

[0125] Another embodiment of the present application provides a laser measurement-based furnace liquid constant pumping control system, wherein, referring to Figure 2 The laser measurement-based furnace liquid constant pumping control system comprises:

[0126] The initialization module 100 acquires the liquid pumping target amount input by the user, and automatically calculates the initial pumping time and frequency based on the initial state parameters of the furnace;

[0127] Residual quantity data acquisition module 200: real-time acquisition of residual quantity data of the liquid in the furnace by the laser measuring device;

[0128] Pumping time or frequency calculation module 300: according to the residual quantity data of the liquid, the pumping time or frequency that needs to be adjusted is calculated;

[0129] Liquid quality constant pumping module 400: send the control instruction containing the adjusted pumping time or frequency to the pumping device, so that the pumping device pumps the liquid according to the adjusted pumping time or frequency, and ensures the constant quality of the liquid pumped each time.

[0130] The laser measurement-based furnace liquid constant pumping control system provided by the embodiment can realize the steps of the foregoing embodiment and achieve the same technical effects as the foregoing embodiment, and the principle analysis can be referred to the related description of the steps of the laser measurement-based furnace liquid constant pumping control method.

[0131] The embodiment of the application also provides a laser measurement-based furnace liquid constant pumping control device, which comprises a memory and a processor, and the memory stores a computer program capable of being loaded and executed by the processor and performing the laser measurement-based furnace liquid constant pumping control method.

[0132] The embodiment of the application also provides a storage medium, which stores a computer program capable of being loaded and executed by the processor and performing the laser measurement-based furnace liquid constant pumping control method.

[0133] The storage medium provided by the embodiment can realize the steps of the foregoing embodiment and achieve the same technical effects as the foregoing embodiment, and the principle analysis can be referred to the related description of the method steps.

[0134] The storage medium may, for example, 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 various storage medium capable of storing program codes.

[0135] The steps of the method or algorithm described in combination with the embodiments disclosed herein can be directly implemented by hardware, a software module executed by a processor, or a combination of the two. The software module can be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0136] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" and the like means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Also, the specific features, structures, materials or characteristics described can be combined in any appropriate manner in any one or more embodiments or examples. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples, without contradiction.

[0137] In addition, the term defining the features of "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 specifically limited, only for the purpose of description, and cannot be understood as indicating or implying the relative importance or implicitly indicating the number of the indicated technical features.

[0138] Therefore, any process or method descriptions in the flowchart or otherwise described herein can be understood as representing code modules, segments, or portions of code that include one or more executable instructions for implementing the specified logical functions or processes, and the scope of the preferred embodiments of the present application includes additional implementation in which the functions are performed in a different order, including substantially simultaneously, or in reverse order, as will be understood by those skilled in the art of the embodiments to which the present application pertains.

[0139] The embodiments of the present specific embodiment are the preferred embodiments of the present application, and are not limited to the protection scope of the present application, so that: any equivalent changes made according to the structure, shape, principle of the present application should be covered within the protection scope of the present application.

Claims

1. A method for constant mass pumping control of a smelting vessel's liquid based on laser measurements, characterized by, The method comprises: acquiring the target amount of the material liquid pumped by the user input, and automatically calculating the initial pumping time and frequency based on the initial state parameters of the furnace; real-time acquisition of the residual amount of the material liquid in the furnace by a laser measuring device; According to the residual amount of the material liquid, the pumping time or frequency that needs to be adjusted is calculated, and the specific steps include: establishing a dynamic correlation model of the change of the mass of the material liquid in the furnace with the feeding and discharging flow, estimating the current feeding flow based on the model and the real-time measurement data; a historical feeding database is established to record the time, duration and corresponding residual amount of the material liquid of each feeding operation; feature extraction and trend analysis are performed on the historical feeding data to predict the feeding flow change in the future within a preset time; a pumping parameter adjustment scheme is generated according to the prediction result, and the adjustment scheme is updated periodically; set the fluctuation amplitude threshold interval, divide the feeding flow fluctuation into slight fluctuation, medium fluctuation and large fluctuation; when the feeding flow fluctuation is slight, the pumping frequency is adjusted slightly for rapid compensation, and the pumping time remains unchanged; when the feeding flow fluctuation is medium, the pumping frequency and the single pumping time are adjusted at the same time, and the deviation is corrected gradually; when the feeding flow fluctuation is large, the current pumping period is suspended, and the new pumping parameters are recalculated and set; sending a control instruction containing the adjusted pumping time or frequency to the pumping device to make the pumping device pump the material liquid according to the adjusted pumping time or frequency, so as to ensure the constant quality of the material liquid pumped each time; The method further comprises: Real-time monitoring of the temperature change rate of the material liquid to identify the phase change critical point; When the phase change critical point is detected, an adaptive PID control algorithm is started to dynamically adjust the pumping parameters to compensate for the viscosity mutation caused by the phase change; Constructing a phase change energy consumption prediction model, and optimizing the pumping power distribution based on the model; Set the safety threshold of the pumping parameters in the phase change process, when it is predicted that the parameters will exceed the safety threshold, trigger the early warning mechanism and adjust the control strategy.

2. The laser measurement based molten bath level constant pumping control method of claim 1, wherein, The method further comprises: Receiving the actual pumping time, motor current value and pump body vibration frequency returned by the pumping device, inversely calculating the actual pumping amount through a physical model containing the characteristics of the pumping device, and calculating the deviation from the expected pumping amount; Based on the closed-loop control algorithm, continuously iteratively adjust the pumping parameters until the deviation rate of the actual pumping amount from the expected pumping amount reaches the preset accuracy range, or the change rate of the pumping parameters is lower than the preset threshold after continuous adjustment for multiple times, or the termination condition based on the safety boundary or energy consumption constraint of the furnace is triggered.

3. The laser measurement based molten bath level constant pumping control method of claim 1, wherein, Real-time acquisition of the residual amount of the material liquid in the furnace by a laser measuring device comprises: Pre-set feature point array on the inner wall of the furnace, and build a three-dimensional coordinate library of the feature points based on the laser reflection characteristics; Real-time tracking of the position change of the feature points through the optical flow field analysis algorithm, and calculating the dynamic deformation of the material liquid surface; Based on the deformation, a material liquid volume compensation model is established to correct the residual amount data measured by the laser.

4. The laser measurement based molten bath level constant pumping control method of claim 1, wherein, Real-time acquisition of the residual amount of the material liquid in the furnace by a laser measuring device comprises: Deploy a multi-modal sensor array including pressure sensors, ultrasonic level meters and infrared thermographs to acquire the pressure distribution, ultrasonic echo and surface temperature field data of the material liquid in the furnace; A multi-sensor data fusion model is established to synchronize and fuse laser measurement data and other sensor data in time and space based on Kalman filtering algorithm; The residual data of laser measurement is corrected by the data fusion result to compensate for the measurement error caused by furnace vibration, liquid splashing or steam interference.

5. A laser measurement based constant level pumping control system for a smelting vessel, characterised in that, It comprises: An initialization module that obtains the target amount of liquid pumped by the user and automatically calculates the initial pumping time and frequency based on the initial state parameters of the furnace; A residual data acquisition module that acquires real-time residual data of the liquid in the furnace through a laser measurement device; A pumping time or frequency calculation module that calculates the adjusted pumping time or frequency based on the residual data of the liquid, including: establishing a dynamic correlation model of the change of the liquid mass in the furnace and the feed and discharge flow, estimating the current feed flow based on the model and real-time measurement data; establishing a historical feed database to record the time, duration and corresponding change of the residual liquid of each feed operation; performing feature extraction and trend analysis on the historical feed data to predict the feed flow change in the future within a preset time; generating a pumping parameter adjustment scheme based on the prediction result and updating the adjustment scheme periodically; setting a fluctuation amplitude threshold interval to divide the feed flow fluctuation into slight fluctuation, moderate fluctuation and large fluctuation; when the feed flow fluctuation is slight, fine-tune the pumping frequency for quick compensation and keep the pumping time unchanged; when the feed flow fluctuation is moderate, adjust the pumping frequency and single pumping time simultaneously to gradually correct the deviation; when the feed flow fluctuation is large, pause the current pumping period, recalculate and set new pumping parameters; A constant liquid mass pumping module that sends control instructions containing adjusted pumping time or frequency to the pumping device to make the pumping device pump liquid according to the adjusted pumping time or frequency, ensuring constant liquid mass for each pumping; further comprising: real-time monitoring of liquid temperature change rate to identify the phase change critical point; when the phase change critical point is detected, an adaptive PID control algorithm is started to dynamically adjust the pumping parameters to compensate for the viscosity mutation caused by phase change; a phase change energy consumption prediction model is constructed to optimize the pumping power distribution based on the model; set the safety threshold of the pumping parameters during phase change, when the parameters are predicted to exceed the safety threshold, trigger the early warning mechanism and adjust the control strategy.

6. A laser measurement-based molten bath level constant pumping control apparatus, characterized by, It comprises: A memory and a processor, the memory stores a computer program that can be loaded and executed by the processor to implement the furnace liquid constant pumping control method based on laser measurement of any one of claims 1-4.

7. A storage medium, characterized by A memory and a processor, the memory stores a computer program that can be loaded and executed by the processor to implement the furnace liquid constant pumping control method based on laser measurement of any one of claims 1-4.

Images