Die plate gridding-based aluminum leakage risk positioning evaluation and intelligent prevention and control method and system

Abstract

This invention provides a method and system for aluminum leakage risk location assessment and intelligent prevention based on a gridded mold. The method includes: S1, gridded liquid level monitoring; S2, gridded risk location assessment; S3, risk distribution analysis; S4, calculation of cooling deviation; and S5, emergency response for risk prevention. This invention achieves accurate location and assessment of aluminum leakage risk through gridded liquid level monitoring, and further analyzes the risk distribution trend based on the type, level, and location characteristics of the risk, providing a basis for subsequent differentiated emergency prevention strategies. This invention also solves the problem of unmeasurable cooling demand through a neural network model, thereby realizing the calculation of cooling deviation; by introducing cooling deviation into a multi-parameter sequential control strategy, an online control decision-making mechanism with cooling deviation correction as the core objective is formed, achieving proactive prevention of aluminum leakage risk.

Description

Technical Field

[0001] This invention relates to the field of industrial production safety risk prevention and control, specifically to a method and system for aluminum leakage risk location assessment and intelligent prevention and control based on a grid-based mold. Background Technology

[0002] Deep well casting is a crucial step in aluminum production. During the aluminum rod casting process, there is a risk of high-temperature molten aluminum leaking or overflowing from the mold into the casting well. When the high-temperature molten aluminum mixes with cooling water in the well and reaches a certain mass ratio or spatial constraint, a violent explosion may occur, characterized by its suddenness, destructiveness, and potential to cause mass casualties.

[0003] Analysis of the relevant accident mechanisms and on-site management processes revealed that the main causes of similar explosion accidents typically include: (1) before aluminum leakage (including aluminum liquid leakage and aluminum liquid overflow) occurs, there is a lack of effective prediction and quantitative assessment methods for potential risks, making it difficult to achieve early intervention and preventive control; (2) when aluminum leakage occurs in the casting mold, it is difficult to quickly locate the specific location of the leakage; at the same time, there is a lack of emergency response plans that match different risk conditions, leading to delayed handling and increased risk; (3) one of the key causes of aluminum leakage is the uneven cooling caused by the imbalance between the supply and demand of cold energy in the casting process. However, in actual production, the cooling required for aluminum liquid solidification cannot be directly detected online, making it difficult to correct the balance of the supply and demand of cold energy in the system in a timely manner, resulting in the accumulation of risk and triggering an accident in a short period of time. In conclusion, traditional low-intelligence, experience-based management methods are difficult to meet the needs of refined control of aluminum leakage risks in deep well casting.

[0004] In terms of aluminum leakage risk identification, existing technologies typically employ methods such as thermal imaging detection, overall liquid level fluctuation monitoring, and sound recognition for aluminum leakage identification and early warning. However, these existing technologies can only qualitatively identify whether aluminum leakage exists in the casting mold, but cannot accurately locate the leakage point or assess the leakage risk. After qualitative identification, it is still necessary to rely on staff to check for leakage points. Manual inspection is time-consuming, has significant delays, and may not be handled promptly, increasing the risk of worsening aluminum leakage and even triggering an explosion.

[0005] In terms of risk control, most existing technologies reduce leakage risk by shutting down the system or adjusting the process parameters of the cooling system. However, these control strategies heavily rely on the subjective experience of workers and lack clear control rules and objectives. The fundamental reason for this is that existing technologies cannot quantify the degree of imbalance between cooling supply and demand, making it difficult to achieve dynamic and proactive intervention in preventing aluminum leakage risks.

[0006] It is evident that existing aluminum leakage risk identification and prevention systems generally suffer from drawbacks such as difficulty in risk location and assessment, outdated risk prevention measures, and a lack of proactive prevention capabilities, and therefore require further improvement and refinement. Summary of the Invention

[0007] The purpose of this invention is to address the shortcomings of existing technologies by providing a method and system for locating, assessing, and intelligently preventing aluminum leakage risks based on a mold grid. This system enables accurate location and rapid investigation of aluminum leakage risks and constructs a preventive system that allows for location assessment, prediction, and closed-loop control of aluminum leakage risks, thereby enhancing the proactive risk control capabilities in the aluminum rod casting process.

[0008] To achieve the above objectives, the present invention adopts the following technical solution:

[0009] A method for locating, assessing, and intelligently preventing aluminum leakage risks based on a gridded mold includes:

[0010] S1. Grid-based liquid level monitoring: Based on the distribution of aluminum rod casting holes on the casting mold, the casting mold is divided into a grid matrix with several rows and columns, where each grid corresponds to an aluminum rod casting hole; during the casting process, the liquid level height of aluminum liquid in all grids is collected in real time and synchronously. When recording the liquid level height of aluminum liquid in each grid, the row and column of the grid are recorded simultaneously to mark its position characteristics on the casting mold.

[0011] S2. Grid-based risk location assessment: Risk location assessment is carried out using grids as units. Based on the aluminum liquid level height and liquid level change rate in each grid, the risk type in the grid is identified and classified into different risk levels.

[0012] S3. Risk Distribution Analysis: Based on the number, location characteristics, and adjacent location relationships of grids of different risk types and risk levels, comprehensively analyze the overall distribution of aluminum leakage risk on the casting mold to determine whether there is a diffusion or aggregation trend of aluminum leakage risk on the casting mold.

[0013] S4. Calculate the cooling deviation: During the casting process, the operating condition parameters are collected in real time to calculate the current cooling deviation ΔQ = Q1-Q2; where Q1 is the predicted cooling amount required for the solidification of aluminum liquid under the current operating condition, which is predicted based on the real-time collected operating condition parameters; Q2 is the actual cooling amount Q2 under the current operating condition, which is calculated based on the real-time collected operating condition parameters.

[0014] S5. Risk Prevention and Emergency Response: Based on whether the risk of aluminum leakage on the casting mold plate has a tendency to spread or accumulate, different emergency prevention and control strategies are triggered to eliminate the risk of aluminum leakage.

[0015] When the risk of aluminum leakage on the casting mold plate shows a tendency to spread or accumulate, an emergency shutdown strategy is triggered to eliminate the risk of aluminum leakage by stopping the casting process; when the risk of aluminum leakage on the casting mold plate does not show a tendency to spread or accumulate, a multi-parameter sequential control strategy is triggered to eliminate the risk of aluminum leakage by dynamically adjusting the system operating parameters while maintaining the casting process.

[0016] The system operating parameters include casting speed, cooling water flow rate, cooling water pressure, aluminum liquid level at the casting mold inlet, and cooling water temperature.

[0017] The multi-parameter sequential control strategy is as follows: within the adjustable threshold range of each system working parameter, the system working parameters are adjusted in multiple rounds according to the priority order of casting speed, cooling water flow rate, cooling water pressure, aluminum liquid level at the casting mold inlet, and cooling water temperature, so that the cooling deviation ΔQ approaches 0, thereby restoring the cooling supply and demand balance in the casting process.

[0018] Furthermore, in S2, when assessing the risk location of each grid, each grid is marked as a leakage risk grid or an overflow risk grid according to the height of the aluminum liquid level. In addition, based on the rate of change of liquid level, the leakage risk grid is marked as a leaking grid, a high-risk leakage grid, or a low-risk leakage grid according to the risk level, and the overflow risk grid is marked as an overflowing grid, a high-risk overflow grid, or a low-risk overflow grid according to the risk level.

[0019] Furthermore, in S5, when the multi-parameter sequential control strategy is triggered, if there are leaked grids and / or high-risk leaked grids on the casting mold and the cooling deviation ΔQ > 0, then the system operating parameters are adjusted in multiple rounds according to the priority order of reducing casting speed, increasing cooling water flow rate, increasing cooling water pressure, reducing aluminum liquid level at the casting mold inlet, and reducing cooling water temperature, until the cooling deviation ΔQ is not greater than 0.

[0020] Furthermore, in S5, when the multi-parameter sequential control strategy is triggered, if there are overflowing grids and / or high-risk overflow grids on the casting mold and the cooling deviation ΔQ < 0, then the system operating parameters are adjusted in multiple rounds according to the priority order of increasing casting speed, reducing cooling water flow rate, reducing cooling water pressure, increasing the aluminum liquid level at the casting mold inlet, and increasing the cooling water temperature, until the cooling deviation ΔQ is not less than 0.

[0021] Furthermore, in S3, when the grid on the casting mold plate meets one of the following conditions, it is determined that the risk of aluminum leakage on the casting mold plate has a tendency to spread or accumulate: there are 4 or more leaked grids at the same time; or, there are 3 leaked grids at the same time and these 3 leaked grids are adjacent to each other; or, there is at least 1 overflowed grid at the edge of the casting mold plate; or, there are 3 or more overflowed grids at the non-edge of the casting mold plate at the same time.

[0022] The risk of aluminum leakage on the casting mold is considered to have no tendency to spread or accumulate when the grid on the casting mold meets one of the following conditions: there is 1 or 2 leaked grids; or, there are 3 leaked grids at the same time and these 3 leaked grids are not adjacent to each other; or, there is 1 or 2 overflowed grids at the non-edge of the casting mold; or, the number of both leaked and overflowed grids is 0 and the number of high-risk leakage grids is greater than or equal to 1; or, the number of both leaked and overflowed grids is 0 and the number of high-risk overflow grids is greater than or equal to 1.

[0023] Furthermore, S4 specifically includes the following steps:

[0024] S41. Real-time acquisition of working condition parameters during the casting process, including mold type, casting rod specifications, aluminum liquid temperature, casting speed, cooling water outlet temperature, workshop ambient temperature and humidity, cooling water inlet temperature and cooling water flow rate.

[0025] S42. Input the mold type, casting rod specifications, aluminum liquid temperature, casting speed, cooling water temperature and workshop ambient temperature and humidity into the trained neural network model to obtain the predicted cooling capacity Q1.

[0026] S43. Calculate the actual cooling capacity under the current operating conditions. Where c is the specific heat capacity of the cooling water, ρ is the density of the cooling water, V is the cooling water flow rate, T1 is the cooling water inlet temperature, and T2 is the cooling water outlet temperature.

[0027] S44. The cooling deviation ΔQ = Q1-Q2 is obtained through calculation.

[0028] Furthermore, the neural network model used to predict the cooling demand is trained using the following method:

[0029] S61. During normal and stable casting in the casting mold, collect several sets of working condition parameters.

[0030] S62. For each set of operating condition parameters, the actual cooling capacity is calculated and used as the cooling capacity required for each set of operating condition parameters; the mold type, casting rod specifications, aluminum liquid temperature, casting speed, cooling water temperature, workshop ambient temperature and humidity, and cooling capacity are packaged into a sample set according to the corresponding relationship.

[0031] S63. Using mold type, casting rod specifications, aluminum liquid temperature, casting speed, cooling water temperature and workshop ambient temperature and humidity as model inputs, and the required cooling capacity as model output, construct a neural network model.

[0032] S64. Train the neural network model built in S63 using the sample set in S62.

[0033] A leakage risk location assessment and intelligent prevention and control system based on a gridded mold, used to implement the above-mentioned methods, includes a server and a terminal controller. The server is connected to the terminal controller in sequence through a switch and a protocol gateway. The server is equipped with a leakage risk location assessment module, a risk distribution analysis module, a cooling deviation calculation module, and an emergency prevention and control strategy output module. The terminal controller is equipped with a data acquisition module, a dynamic control module, and an emergency shutdown module.

[0034] The data acquisition module is used to collect the operating status parameters and system operating parameters of the aluminum rod casting production line, and is also used to collect the aluminum liquid level height in different areas of the casting mold in a gridded manner. The operating status parameters include mold type, rod specifications, aluminum liquid temperature, casting speed, cooling water outlet temperature, workshop ambient temperature and humidity, cooling water inlet temperature, and cooling water flow rate. The system operating parameters include casting speed, cooling water flow rate, cooling water pressure, aluminum liquid level at the casting mold inlet, and cooling water inlet temperature.

[0035] The aluminum leakage risk location and assessment module is used to locate and assess the aluminum leakage risk in each grid on the casting mold in real time based on the aluminum liquid level height obtained by the data acquisition module, and record the location characteristics, risk type and risk level of each grid.

[0036] The risk distribution analysis module is used to comprehensively analyze the overall distribution of aluminum leakage risk on the casting mold based on the location characteristics, risk type and risk level of each grid recorded in the aluminum leakage risk location and assessment module, so as to determine whether there is a diffusion or accumulation trend of aluminum leakage risk on the casting mold.

[0037] The cooling deviation calculation module is used to calculate the predicted cooling capacity Q1 and the actual cooling capacity Q2 of aluminum liquid solidification under the current working condition based on the working condition parameters obtained by the data acquisition module, and finally obtain the cooling deviation ΔQ = Q1-Q2, which is provided to the dynamic control module in the terminal controller.

[0038] The emergency prevention and control strategy output module is used to trigger an emergency shutdown strategy or a multi-parameter sequential control strategy according to the analysis and judgment results of the risk distribution analysis module. When the emergency shutdown strategy is triggered, it outputs an emergency shutdown command to the emergency shutdown module in the terminal controller. When the multi-parameter sequential control strategy is triggered, it outputs a dynamic control command to the dynamic control module in the terminal controller.

[0039] The emergency stop module is used to stop the casting process of the aluminum rod casting production line when an emergency stop command is received.

[0040] The dynamic control module is used to dynamically control the system operating parameters when it receives a dynamic control command, in conjunction with the cooling deviation ΔQ provided by the cooling deviation calculation module, so that the cooling deviation ΔQ approaches 0, so as to restore the balance of cooling supply and demand in the casting process.

[0041] Furthermore, the data acquisition module is connected to the water temperature sensor, water pressure sensor, water flow meter, aluminum liquid temperature sensor, aluminum liquid level sensor, casting speed instrument, and ambient temperature and humidity sensor in the aluminum rod casting production line.

[0042] The water temperature sensor is used to collect the cooling inlet water temperature and cooling outlet water temperature of the cooling system.

[0043] The water pressure sensor is used to collect the cooling water supply pressure of the cooling system.

[0044] The water flow meter is used to collect the cooling water flow rate of the cooling system;

[0045] The aluminum liquid temperature sensor is used to collect the temperature of the aluminum liquid at the inlet of the casting mold.

[0046] The aluminum liquid level sensor is used to collect the aluminum liquid level at the inlet of the casting mold.

[0047] The casting speed instrument is used to detect the casting speed of the casting machine;

[0048] The ambient temperature and humidity sensor is used to detect the temperature and humidity of the workshop environment.

[0049] The data acquisition module is also connected to a matrix laser ranging device; the matrix laser ranging device is set above the casting mold and is used to collect the liquid level height of aluminum liquid in different areas of the casting mold in a grid-like manner.

[0050] Furthermore, the dynamic control module is connected to the casting machine controller, cooling pump controller, electric valve controller, smelting furnace controller, and cooling tower controller, respectively;

[0051] The casting machine controller is used to adjust the casting speed of the casting machine;

[0052] The cooling pump controller is used to control the frequency of the cooling pump in the cooling system in order to adjust the cooling water flow rate;

[0053] The electric valve controller is used to control the opening degree of the electric valve in the cooling system to regulate the cooling water pressure.

[0054] The furnace controller is used to control the flow rate of molten aluminum at the furnace outlet in order to adjust the molten aluminum level at the casting mold inlet.

[0055] The cooling tower controller is used to control the frequency and number of cooling towers in the cooling system in order to regulate the cooling water inlet temperature.

[0056] This invention provides a method and system for locating, assessing, and intelligently controlling aluminum leakage risks based on a gridded mold. Through gridded liquid level monitoring, it achieves precise location and assessment of aluminum leakage risks. Furthermore, it analyzes the distribution trend of risks based on their type, level, and location characteristics, providing a basis for selecting differentiated emergency control strategies. Further, this invention solves the problem of unmeasurable cooling demand through a neural network model, thereby enabling the calculation of cooling supply deviation. By introducing cooling supply deviation into a multi-parameter sequential control strategy, it forms an online control decision-making mechanism with cooling supply deviation correction as its core objective, achieving proactive prevention of aluminum leakage risks. Based on the above technical solutions, this invention constructs a preventive system for aluminum leakage risks that is locatable, predictable, and controllable in a closed loop, improving the proactive risk control capabilities in the aluminum rod casting process.

[0057] Compared with existing technologies, the aluminum leakage risk location assessment and intelligent prevention and control method and system based on mold gridding provided by this invention has the following advantages:

[0058] 1. This invention achieves precise location of aluminum leakage risks, compensating for the shortcomings of qualitative identification of aluminum leakage risks and eliminating the lag in manual inspection. By dividing the casting mold into grids and performing real-time liquid level monitoring and risk assessment for each grid, this invention significantly shortens the manual inspection time for aluminum leakage risk location, improves the accuracy and timeliness of emergency response, and thus reduces the risk of explosion.

[0059] 2. This invention enables pre-emptive graded early warning and trend analysis, compensating for the shortcomings of post-event handling of aluminum leakage risks. The invention marks all grids on the casting mold according to different risk types and levels, facilitating graded early warning and trend analysis of risks. This transforms risk management from passive alarm after aluminum leakage to proactive early warning and prevention control before leakage, helping to improve stability and robustness under complex working conditions and minimizing the probability of aluminum leakage.

[0060] 3. This invention solves the problem of unquantifiable cooling supply deviation, providing a clear target for subsequent risk control and adjustment strategies. Addressing the difficulty in directly detecting the cooling requirement for molten aluminum solidification online, this invention uses a cooling supply-demand balance relationship—where the actual cooling supply of the cooling system equals the cooling requirement for molten aluminum solidification—to back-calculate training data. Combined with a neural network model, this achieves reliable prediction of the cooling requirement, thus solving the problem of quantifiable cooling supply deviation. It transforms the unmeasurable cooling requirement into a quantifiable decision-making basis, providing a clear direction and target for preventing aluminum leakage risks.

[0061] 4. This invention provides differentiated emergency control strategies to improve the continuity and efficiency of the casting process. After analyzing the distribution trend of aluminum leakage risk, it outputs matching disposal strategies according to the different levels of urgency. When the risk is still controllable, a multi-parameter sequential control strategy is triggered. While maintaining continuous casting production, the system's operating parameters are dynamically controlled sequentially. This avoids introducing new process risks through experience-based manual adjustments while ensuring process safety boundaries, quickly restoring the balance between cold supply and demand, achieving online dynamic reduction of aluminum leakage risk, reducing the defects and economic losses caused by downtime, and ensuring the smoothness and efficiency of the casting process. Attached Figure Description

[0062] Figure 1 This is a schematic diagram of the grid division of the casting mold in a method for locating, assessing, and intelligently preventing aluminum leakage risks based on mold gridding provided in Embodiment 1 of the present invention.

[0063] Figure 2 This is a schematic diagram of the structure of an aluminum rod casting production line in the existing technology.

[0064] Figure 3 This is a flowchart illustrating the multi-parameter sequential control strategy for leakage risk in Embodiment 1 of the present invention.

[0065] Figure 4 This is a flowchart illustrating the multi-parameter sequential control strategy for overflow-type risks in Embodiment 1 of the present invention.

[0066] Figure 5 This is a schematic diagram of a leakage risk location assessment and intelligent prevention and control system based on a gridded mold provided in Embodiment 2 of the present invention. Detailed Implementation

[0067] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0068] Example 1

[0069] This invention provides a method for aluminum leakage risk location assessment and intelligent prevention based on a gridded mold, including:

[0070] S1. Grid-based liquid level monitoring: Based on the distribution of aluminum rod casting holes on the casting mold, the casting mold is divided into a grid matrix with several rows and columns, where each grid corresponds to an aluminum rod casting hole; during the casting process, the liquid level height of aluminum liquid in all grids is collected in real time and synchronously. When recording the liquid level height of aluminum liquid in each grid, the row and column of the grid are recorded simultaneously to mark its position characteristics on the casting mold.

[0071] S2. Grid-based risk location assessment: Risk location assessment is carried out using grids as units. Based on the aluminum liquid level height and liquid level change rate in each grid, the risk type in the grid is identified and classified into different risk levels.

[0072] S3. Risk Distribution Analysis: Based on the number, location characteristics, and adjacent location relationships of grids of different risk types and risk levels, comprehensively analyze the overall distribution of aluminum leakage risk on the casting mold to determine whether there is a diffusion or aggregation trend of aluminum leakage risk on the casting mold.

[0073] S4. Calculate the cooling deviation: During the casting process, the operating condition parameters are collected in real time to calculate the current cooling deviation ΔQ = Q1-Q2; where Q1 is the predicted cooling amount required for the solidification of aluminum liquid under the current operating condition, which is predicted based on the real-time collected operating condition parameters; Q2 is the actual cooling amount Q2 under the current operating condition, which is calculated based on the real-time collected operating condition parameters.

[0074] S5. Risk Prevention and Emergency Response: Based on whether the risk of aluminum leakage on the casting mold plate has a tendency to spread or accumulate, different emergency prevention and control strategies are triggered to eliminate the risk of aluminum leakage.

[0075] When the risk of aluminum leakage on the casting mold plate shows a tendency to spread or accumulate, an emergency shutdown strategy is triggered to eliminate the risk of aluminum leakage by stopping the casting process; when the risk of aluminum leakage on the casting mold plate does not show a tendency to spread or accumulate, a multi-parameter sequential control strategy is triggered to eliminate the risk of aluminum leakage by dynamically adjusting the system operating parameters while maintaining the casting process.

[0076] The system operating parameters include casting speed, cooling water flow rate, cooling water pressure, aluminum liquid level at the casting mold inlet, and cooling water temperature.

[0077] The multi-parameter sequential control strategy is as follows: within the adjustable threshold range of each system working parameter, the system working parameters are adjusted in multiple rounds according to the priority order of casting speed, cooling water flow rate, cooling water pressure, aluminum liquid level at the casting mold inlet, and cooling water temperature, so that the cooling deviation ΔQ approaches 0, thereby restoring the cooling supply and demand balance in the casting process.

[0078] Combination Figure 1 As shown, S1 specifically includes:

[0079] Using the aluminum rod casting holes on the casting mold as the dividing unit, the space on the casting mold is divided into a grid matrix of n rows and m columns according to the row and column distribution of the aluminum rod casting holes, forming a total of n×m grids, with each grid corresponding to one aluminum rod casting hole.

[0080] Assign a unique code (i, j) to each grid cell, where i is the row number of the grid cell and j is the column number of the grid cell, i = 1, 2, …, n, j = 1, 2, …, m;

[0081] A matrix-type laser rangefinder is installed above the casting mold to synchronously monitor the molten aluminum level in n×m grids in real time. Specifically, at second t, the molten aluminum level in the grid at row i and column j is recorded as follows: .

[0082] Furthermore, in S2, when assessing the risk location of each grid, each grid is marked as a leakage risk grid or an overflow risk grid according to the height of the aluminum liquid level. In addition, based on the rate of change of liquid level, the leakage risk grid is marked as a leaking grid, a high-risk leakage grid, or a low-risk leakage grid according to the risk level, and the overflow risk grid is marked as an overflowing grid, a high-risk overflow grid, or a low-risk overflow grid according to the risk level.

[0083] By performing gridded liquid level monitoring in S1, this invention can promptly detect abnormal aluminum liquid levels at different locations on the casting mold, so as to accurately locate the risk of aluminum leakage by the location of the abnormal aluminum liquid level, and classify and grade the risk of aluminum leakage by the degree of the abnormal aluminum liquid level; providing a data basis for gridded risk location assessment of aluminum leakage risk in S2.

[0084] Specifically, in S2 of this embodiment, the following process is used to perform risk location assessment on the grid on the casting mold:

[0085] S21. For each grid, record the height of the molten aluminum level within the grid. Relative to liquid level height reference value In comparison; if Then mark the grid as a leakage risk grid; if If the value is not specified, the grid is marked as an overflow-type risk grid; wherein, the liquid level height reference value The average height of the molten aluminum level is obtained statistically during normal and stable casting.

[0086] S22. For each grid cell, calculate its current liquid level deviation amplitude. and the rate of change of liquid level within Δt seconds ;

[0087] S23. For leakage-type risk grids, risk levels are classified according to the following criteria:

[0088] If the liquid level deviation of a single grid If so, mark it as a leaked mesh;

[0089] If the liquid level deviation of a single grid And the duration window ΔT seconds, or the rate of change of liquid level in a single grid. If so, mark it as a high-risk grid for leakage;

[0090] If the liquid level deviation between two or more adjacent grids is If the duration window is ΔT seconds, then all these grids are marked as high-risk leakage grids; if the liquid level change rate of two or more adjacent grids is... These grids are then marked as high-risk grids for leakage;

[0091] All other leakage-risk grids are marked as low-risk leakage grids;

[0092] in, and All of these are set thresholds for assessing leakage risk levels;

[0093] S24. For overflow-type risk grids, risk levels are classified according to the following criteria:

[0094] If the liquid level deviation of a single grid If so, mark it as an overflowing mesh;

[0095] If the liquid level deviation of a single grid And the duration window ΔT seconds, or the rate of change of liquid level in a single grid. If so, mark it as a high-risk overflow grid;

[0096] If the liquid level deviation between two or more adjacent grids is If the duration window is ΔT seconds, then all these grids are marked as high-risk overflow grids; if the liquid level change rate of two or more adjacent grids is... These grids will then be marked as high-risk overflow grids;

[0097] All other overflow-type risk grids are marked as low-risk overflow grids;

[0098] in, and These are all set thresholds for assessing leakage risk levels.

[0099] As an improvement, an n×m aluminum leakage risk location matrix can be constructed in real time based on the above risk location assessment results, so as to more intuitively understand the location distribution and changing trend of grids of different risk types and risk levels; in addition, different audible and visual alarms or early warning signals can be issued for different types and risk levels of aluminum leakage risks.

[0100] By conducting a gridded risk location assessment in S2, this invention distinguishes and marks all grids on the casting mold according to different risk types and risk levels. On the one hand, this facilitates the rapid and accurate location of risk points using the positional characteristics of the grids. On the other hand, it also facilitates a higher-dimensional understanding of the location distribution and changing trends of aluminum leakage risk on the casting mold, so as to provide a basis for analysis and decision-making in subsequent risk prevention and control stages.

[0101] Furthermore, in S3, when the grid on the casting mold plate meets one of the following conditions, it is determined that the risk of aluminum leakage on the casting mold plate has a tendency to spread or accumulate: there are 4 or more leaked grids at the same time; or, there are 3 leaked grids at the same time and these 3 leaked grids are adjacent to each other; or, there is at least 1 overflowed grid at the edge of the casting mold plate; or, there are 3 or more overflowed grids at the non-edge of the casting mold plate at the same time.

[0102] The risk of aluminum leakage on the casting mold is considered to have no tendency to spread or accumulate when the grid on the casting mold meets one of the following conditions: there is 1 or 2 leaked grids; or, there are 3 leaked grids at the same time and these 3 leaked grids are not adjacent to each other; or, there is 1 or 2 overflowed grids at the non-edge of the casting mold; or, the number of both leaked and overflowed grids is 0 and the number of high-risk leakage grids is greater than or equal to 1; or, the number of both leaked and overflowed grids is 0 and the number of high-risk overflow grids is greater than or equal to 1.

[0103] Based on the risk distribution analysis in S3, this invention can determine the severity of aluminum leakage risk on the casting mold based on whether there is a diffusion or aggregation trend, providing a clear and reliable decision-making basis for subsequent differentiated emergency control strategies. Specifically, when the aluminum leakage risk on the casting mold shows a diffusion or aggregation trend, it often indicates that the aluminum leakage risk has a tendency to deteriorate rapidly, with more serious consequences and a higher overall risk level, making it difficult to improve or eliminate the risk through simple adjustments. Conversely, when there is aluminum leakage risk on the casting mold but it has not yet formed a diffusion or aggregation trend, the aluminum leakage risk on the casting mold is in a generally controllable state, with potentially less severe consequences and a lower overall risk level, making it more feasible to improve or eliminate the risk by adjusting system operating parameters. Therefore, for these two different risk situations, this invention adopts two different emergency control strategies in S5. It can ensure production safety through emergency shutdown strategies when the risk is out of control, and can eliminate the risk by restoring the cold supply and demand balance of the casting process through multi-parameter sequential control strategies when the risk is controllable or has not yet occurred.

[0104] To better explain the cooling deviation calculation process in S4 and the multi-parameter sequential control strategy in S5, a detailed explanation will be provided below, combining the structure and working principle of the aluminum rod casting production line. Please refer to... Figure 2 This is a schematic diagram of the structure of an existing aluminum rod casting production line. Existing aluminum rod casting production lines generally include an aluminum rod casting system and a cooling system. The aluminum rod casting system includes a melting furnace, casting mold, casting machine, etc., with the casting mold positioned above a casting well. The cooling system includes a low-level water tank, a low-level water tank pump, a cooling tower, a high-level water tank, a cooling pump, electric valves, etc., connected in sequence, used to supply cooling to the casting mold to counteract the heat released during the solidification of the high-temperature molten aluminum during the casting process.

[0105] During operation, the high-temperature molten aluminum from the smelting furnace flows into the casting mold and is then distributed to the casting holes of various aluminum rods. The high-temperature molten aluminum solidifies into cast rods after being cooled by cooling water supplied by the cooling system. Under normal and stable casting conditions, the actual cooling capacity Q2 of the cooling system is basically equal to the cooling capacity Q3 required for the molten aluminum to solidify into cast rods, and the casting mold is in a state of cooling supply and demand balance. When the two are not equal, a cooling deviation ΔQ = Q3-Q2 will occur. If the cooling deviation exists for a long time, it will lead to molten aluminum leakage (ΔQ>0) or molten aluminum overflow (ΔQ<0), thereby increasing the possibility of an explosion accident caused by the contact of high-temperature molten aluminum with water.

[0106] Therefore, the fundamental cause of aluminum leakage risk is uneven cooling due to the imbalance between cold supply and demand during the casting process. Correspondingly, the key objective of eliminating aluminum leakage risk is to restore the system's cold supply and demand balance. Thus, in the multi-parameter sequential control strategy of S5, this invention uses the cooling deviation ΔQ approaching 0 as the control target, transforming the unmeasurable cold demand into a quantifiable decision-making basis, providing a clear direction and objective for the prevention and control of aluminum leakage risk.

[0107] In the process of calculating the cooling supply deviation ΔQ, although the actual cooling capacity Q2 of the cooling system can be obtained through detection and calculation, the required cooling capacity Q3 cannot be directly detected. To solve the problem of difficulty in obtaining the required cooling capacity Q3, this invention uses a neural network model in S4 based on real-time operating condition parameters to predict the required cooling capacity Q3, obtaining the predicted value of the required cooling capacity Q1, and further obtaining the cooling supply deviation ΔQ = Q1-Q2.

[0108] Specifically, S4 includes the following steps:

[0109] S41. Real-time acquisition of working condition parameters during the casting process, including mold type, casting rod specifications, aluminum liquid temperature, casting speed, cooling water outlet temperature, workshop ambient temperature and humidity, cooling water inlet temperature and cooling water flow rate.

[0110] S42. Input the mold type, casting rod specifications, aluminum liquid temperature, casting speed, cooling water temperature and workshop ambient temperature and humidity into the trained neural network model to obtain the predicted cooling capacity Q1.

[0111] S43. Calculate the actual cooling capacity under the current operating conditions. Where c is the specific heat capacity of the cooling water, ρ is the density of the cooling water, V is the cooling water flow rate, T1 is the cooling water inlet temperature, and T2 is the cooling water outlet temperature.

[0112] S44. The cooling deviation ΔQ = Q1-Q2 is obtained through calculation.

[0113] In S4, the neural network model used to predict the required cooling capacity is trained using the following method:

[0114] S61. During normal and stable casting in the casting mold, collect several sets of working condition parameters.

[0115] The criteria for judging the normal and stable casting state of the casting mold are: within a continuous time window of W seconds, no leaked mesh, overflowed mesh, high-risk leaked mesh or high-risk overflowed mesh appears.

[0116] S62. For each set of operating condition parameters, the actual cooling capacity is calculated and used as the cooling capacity required for each set of public condition parameters; the mold type, casting rod specifications, aluminum liquid temperature, casting speed, cooling water temperature, workshop ambient temperature and humidity, and cooling capacity are packaged into a sample set according to the corresponding relationship.

[0117] S63. Using mold type, casting rod specifications, aluminum liquid temperature, casting speed, cooling water temperature and workshop ambient temperature and humidity as model inputs, and the required cooling capacity as model output, construct a neural network model.

[0118] Specifically, the neural network model includes a class embedding layer, a temporal coding layer, a feature cross layer, and a residual fully connected regression head; the construction process of the neural network model is as follows:

[0119] S631, based on mold type Casting rod specifications As inputting categories, static category vectors are obtained through a category embedding layer. ;

[0120] S632, based on the temperature of the molten aluminum Casting speed Cooling water outlet temperature Workshop ambient temperature and workshop environmental humidity Construct a continuous input sequence for a timing window of length L:

[0121] ;

[0122] Construct a missing mask sequence while constructing a continuous input sequence:

[0123]

[0124] in, 1 indicates that the variable is valid at that moment, and 0 indicates that the variable is missing. When any variable is missing, the missing value is imputed by a combination of "forward filling of the most recent valid value and setting the normalized value to zero" or an equivalent imputation method.

[0125] S633. The missing mask sequence is concatenated with the continuous input sequence as the input to the temporal coding layer, and the continuous input sequence is processed by the temporal coding layer (TCN). and missing mask sequence Extracting dynamic features:

[0126] ;

[0127] S634. After concatenating the static category vector and dynamic features, input them into the feature cross layer and the residual fully connected regression head to output the predicted cooling requirement for aluminum liquid solidification.

[0128] S64. Train the neural network model built in S63 using the sample set in S62.

[0129] During normal, stable casting, the casting mold is essentially in a state of heat supply and demand balance, meaning the actual heat supply is roughly equivalent to the heat demand. This invention uses the heat supply under normal, stable casting conditions as an approximate label for the heat demand in the neural network model, enabling the trained model to predict the heat demand based on operating condition parameters.

[0130] It should be noted that the aluminum rod casting process involves numerous parameters. This invention selects mold type, rod specifications, molten aluminum temperature, casting speed, cooling water temperature, and workshop ambient temperature and humidity as the comprehensive basis for predicting cooling demand, which is an optimized selection based on the principle of physical heat transfer. Specifically, mold type directly affects the heat transfer area and path; rod specifications determine the solidification volume and latent heat; molten aluminum temperature determines the sensible heat; casting speed determines the amount of solidification per unit time; cooling water temperature indirectly reflects the actual heat transfer intensity; and workshop ambient temperature and humidity affect the heat transfer between the molten aluminum on the mold surface and the surrounding air. Therefore, the above data combination is highly correlated with cooling demand. Using this data combination as the comprehensive basis for cooling demand prediction improves prediction accuracy and model interpretability by selecting highly correlated parameters as input to the prediction model. Furthermore, avoiding parameters with weak or no correlation helps address the problems of large training load, high computational cost, and long computation time in neural network models, and is more conducive to the timely requirements of risk prevention.

[0131] like Figure 3 As shown, in S5, when the multi-parameter sequential control strategy is triggered, if there are leaked grids and / or high-risk leaked grids on the casting mold and the cooling deviation ΔQ > 0, then the system operating parameters are adjusted in multiple rounds according to the priority order of reducing casting speed, increasing cooling water flow rate, increasing cooling water pressure, reducing aluminum liquid level at the casting mold inlet, and reducing cooling water temperature, until the cooling deviation ΔQ is not greater than 0.

[0132] In other words, for leakage-type risks, the multi-parameter sequential control strategy specifically includes the following steps:

[0133] S5101. Determine whether there is a leaked mesh on the casting mold. If yes, proceed to S5102; otherwise, proceed to S5103.

[0134] S5102: Determine whether the emergency shutdown policy has been triggered. If yes, end; otherwise, proceed to S5104.

[0135] S5103: Determine if there is a high-risk leakage mesh on the casting mold. If yes, proceed to S5104; otherwise, end.

[0136] S5104. Update the current cooling deviation ΔQ. Determine if the cooling deviation ΔQ is greater than 0. If ΔQ > 0, proceed to S5105; otherwise, end.

[0137] S5105. Determine whether the current casting speed is greater than the set lower limit of the speed threshold. If so, reduce the casting machine frequency by 0.1Hz to reduce the casting speed, and return to S5104 after adjustment to enter the next round of adjustment cycle; otherwise, proceed to S5106.

[0138] S5106. Determine whether the current cooling water flow rate is less than the set upper limit of the flow rate threshold. If so, increase the cooling pump frequency by 0.5Hz to increase the cooling water flow rate, and return to S5104 after adjustment to enter the next round of control cycle; otherwise, proceed to S5107.

[0139] S5107. Determine whether the current cooling water pressure is less than the set pressure threshold limit. If so, increase the opening of the electric valve by 1% to increase the cooling water pressure, and return to S5104 after adjustment to enter the next control cycle; otherwise, proceed to S5108.

[0140] S5108. Determine whether the current aluminum liquid level at the inlet of the casting mold is greater than the set lower limit of the liquid level threshold. If so, reduce the aluminum liquid flow rate at the outlet of the melting furnace to lower the aluminum liquid level at the inlet of the casting mold, and return to S5104 after adjustment to enter the next round of control cycle; otherwise, proceed to S5109.

[0141] S5109. Determine whether the current cooling water inlet temperature is greater than the set lower limit of the temperature threshold. If so, increase the cooling tower frequency by 1Hz. When the cooling tower frequency reaches the set upper limit of the frequency and the number of cooling towers in operation is less than the maximum number of units, add one cooling tower to operate to reduce the cooling water inlet temperature. After adjustment, return to S5104 to enter the next round of control cycle.

[0142] like Figure 4 As shown, in S5, when the multi-parameter sequential control strategy is triggered, if there are overflowing grids and / or high-risk overflow grids on the casting mold and the cooling deviation ΔQ < 0, then the system operating parameters are adjusted in multiple rounds according to the priority order of increasing casting speed, reducing cooling water flow rate, reducing cooling water pressure, increasing the aluminum liquid level at the casting mold inlet, and increasing the cooling water temperature, until the cooling deviation ΔQ is not less than 0.

[0143] In other words, for overflow-type risks, the multi-parameter sequential control strategy specifically includes the following steps:

[0144] S5201. Determine whether there is an overflowing mesh on the casting mold. If yes, proceed to S5202; otherwise, proceed to S5203.

[0145] S5202: Determine whether the emergency shutdown policy has been triggered. If yes, end; otherwise, proceed to S5204.

[0146] S5203: Determine if there is a high-risk overflow grid on the casting mold. If yes, proceed to S5204; otherwise, end.

[0147] S5204. Update the current cooling deviation ΔQ, and determine whether the cooling deviation ΔQ is less than 0. If ΔQ < 0, proceed to S5205; otherwise, end.

[0148] S5205. Determine whether the current casting speed is less than the set upper limit of the speed threshold. If so, increase the casting machine frequency by 0.1Hz to increase the casting speed, and return to S5204 after adjustment to enter the next round of adjustment cycle; otherwise, proceed to S5206.

[0149] S5206. Determine whether the current cooling water flow rate is greater than the set lower limit of the flow rate threshold. If so, reduce the cooling pump frequency by 0.5Hz to reduce the cooling water flow rate, and return to S5204 after adjustment to enter the next control cycle; otherwise, proceed to S5207.

[0150] S5207. Determine whether the current cooling water pressure is greater than the set lower limit of the pressure threshold. If so, reduce the opening of the electric valve by 1% to reduce the cooling water pressure, and return to S5204 after adjustment to enter the next control cycle; otherwise, proceed to S5208.

[0151] S5208. Determine whether the current aluminum liquid level at the inlet of the casting mold is lower than the set upper limit of the liquid level threshold. If so, increase the aluminum liquid flow rate at the outlet of the melting furnace to increase the aluminum liquid level at the inlet of the casting mold, and return to S5204 after adjustment to enter the next round of control cycle; otherwise, proceed to S5209.

[0152] S5209. Determine whether the current cooling water inlet temperature is lower than the set upper limit of the temperature threshold. If so, reduce the cooling tower frequency by 1Hz. When the cooling tower frequency reaches the set lower limit of the frequency and the number of cooling towers in operation exceeds 1, reduce the operation of 1 cooling tower to increase the cooling water inlet temperature. After adjustment, return to S5204 to enter the next round of control cycle.

[0153] It should be noted that the priority order of adjusting various system operating parameters in the multi-parameter sequential control strategy of this invention is an optimal solution obtained after scientific research, analysis, demonstration, and experimentation, based on the formation mechanism and physical constraints of aluminum leakage risk. This is one of the key improvements of this invention. Specifically, during the casting process, the casting speed has the greatest impact on the cooling capacity and the fastest adjustment response, so it is given the highest priority. The cooling water flow rate is the most direct factor affecting the cooling capacity, and its adjustment response speed is second only to the casting speed, so it is given the second highest priority. Although the cooling water pressure cannot directly change the cooling capacity, it can affect the stability of the cooling water flow rate and local heat exchange, avoiding cavitation or fluctuations in the cooling water, so it is given the third highest priority. The aluminum liquid level at the casting mold inlet indirectly affects the geometric boundary conditions of aluminum leakage risk, so it is given the fourth highest priority. The cooling water inlet temperature has a slow adjustment response, and the energy consumption and system inertia generated during its adjustment process are also relatively large, so it is given the lowest priority.

[0154] Furthermore, in S5, before implementing the emergency shutdown strategy or the multi-parameter sequential control strategy, an emergency measure is taken to seal the aluminum rod casting holes in the leaked and overflowed meshes. This operation can prevent the risk from worsening immediately, and then the emergency shutdown strategy or the multi-parameter sequential control strategy can be implemented as appropriate to further eliminate the risk.

[0155] Example 2

[0156] like Figure 5 As shown, this embodiment of the invention provides a leakage aluminum risk location assessment and intelligent prevention and control system based on a gridded mold, which is used to implement the method described in Embodiment 1.

[0157] Specifically, the aluminum leakage risk location assessment and intelligent prevention and control system based on the grid-based mold includes a server and a terminal controller. The server is connected to the terminal controller in sequence through a switch and a protocol gateway. The server is equipped with an aluminum leakage risk location assessment module, a risk distribution analysis module, a cooling deviation calculation module, and an emergency prevention and control strategy output module. The terminal controller is equipped with a data acquisition module, a dynamic control module, and an emergency shutdown module.

[0158] The data acquisition module is used to collect the operating status parameters and system operating parameters of the aluminum rod casting production line, and is also used to collect the aluminum liquid level height in different areas of the casting mold in a gridded manner. The operating status parameters include mold type, rod specifications, aluminum liquid temperature, casting speed, cooling water outlet temperature, workshop ambient temperature and humidity, cooling water inlet temperature, and cooling water flow rate. The system operating parameters include casting speed, cooling water flow rate, cooling water pressure, aluminum liquid level at the casting mold inlet, and cooling water inlet temperature.

[0159] The aluminum leakage risk location and assessment module is used to locate and assess the aluminum leakage risk in each grid on the casting mold in real time based on the aluminum liquid level height obtained by the data acquisition module, and record the location characteristics, risk type and risk level of each grid.

[0160] The risk distribution analysis module is used to comprehensively analyze the overall distribution of aluminum leakage risk on the casting mold based on the location characteristics, risk type and risk level of each grid recorded in the aluminum leakage risk location and assessment module, so as to determine whether there is a diffusion or accumulation trend of aluminum leakage risk on the casting mold.

[0161] The cooling deviation calculation module is used to calculate the predicted cooling capacity Q1 and the actual cooling capacity Q2 of aluminum liquid solidification under the current working condition based on the working condition parameters obtained by the data acquisition module, and finally obtain the cooling deviation ΔQ = Q1-Q2, which is provided to the dynamic control module in the terminal controller.

[0162] The emergency prevention and control strategy output module is used to trigger an emergency shutdown strategy or a multi-parameter sequential control strategy according to the analysis and judgment results of the risk distribution analysis module. When the emergency shutdown strategy is triggered, it outputs an emergency shutdown command to the emergency shutdown module in the terminal controller. When the multi-parameter sequential control strategy is triggered, it outputs a dynamic control command to the dynamic control module in the terminal controller.

[0163] The emergency stop module is used to stop the casting process of the aluminum rod casting production line when an emergency stop command is received.

[0164] The dynamic control module is used to dynamically control the system operating parameters when it receives a dynamic control command, in conjunction with the cooling deviation ΔQ provided by the cooling deviation calculation module, so that the cooling deviation ΔQ approaches 0, so as to restore the balance of cooling supply and demand in the casting process.

[0165] Furthermore, the data acquisition module is connected to the water temperature sensor, water pressure sensor, water flow meter, aluminum liquid temperature sensor, aluminum liquid level sensor, casting speed instrument, and ambient temperature and humidity sensor in the aluminum rod casting production line.

[0166] The water temperature sensor is used to collect the cooling inlet water temperature and cooling outlet water temperature of the cooling system.

[0167] The water pressure sensor is used to collect the cooling water supply pressure of the cooling system.

[0168] The water flow meter is used to collect the cooling water flow rate of the cooling system;

[0169] The aluminum liquid temperature sensor is used to collect the temperature of the aluminum liquid at the inlet of the casting mold.

[0170] The aluminum liquid level sensor is used to collect the aluminum liquid level at the inlet of the casting mold.

[0171] The casting speed instrument is used to detect the casting speed of the casting machine;

[0172] The ambient temperature and humidity sensor is used to detect the temperature and humidity of the workshop environment.

[0173] The data acquisition module is also connected to a matrix laser ranging device; the matrix laser ranging device is set above the casting mold and is used to collect the liquid level height of aluminum liquid in different areas of the casting mold in a grid-like manner.

[0174] Furthermore, the dynamic control module is connected to the casting machine controller, cooling pump controller, electric valve controller, smelting furnace controller, and cooling tower controller, respectively;

[0175] The casting machine controller is used to adjust the casting speed of the casting machine;

[0176] The cooling pump controller is used to control the frequency of the cooling pump in the cooling system in order to adjust the cooling water flow rate;

[0177] The electric valve controller is used to control the opening degree of the electric valve in the cooling system to regulate the cooling water pressure.

[0178] The furnace controller is used to control the flow rate of molten aluminum at the furnace outlet in order to adjust the molten aluminum level at the casting mold inlet.

[0179] The cooling tower controller is used to control the frequency and number of cooling towers in the cooling system in order to regulate the cooling water inlet temperature.

[0180] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for locating, assessing, and intelligently preventing aluminum leakage risks based on a gridded mold, characterized in that, include: S1. Grid-based liquid level monitoring: Based on the distribution of aluminum rod casting holes on the casting mold, the casting mold is divided into a grid matrix with several rows and columns, where each grid corresponds to an aluminum rod casting hole; during the casting process, the liquid level height of aluminum liquid in all grids is collected in real time and synchronously. When recording the liquid level height of aluminum liquid in each grid, the row and column of the grid are recorded simultaneously to mark its position characteristics on the casting mold. S2. Grid-based risk location assessment: Risk location assessment is carried out using grids as units. Based on the aluminum liquid level height and liquid level change rate in each grid, the risk type in the grid is identified and classified into different risk levels. S3. Risk Distribution Analysis: Based on the number, location characteristics, and adjacent location relationships of grids of different risk types and risk levels, comprehensively analyze the overall distribution of aluminum leakage risk on the casting mold to determine whether there is a diffusion or aggregation trend of aluminum leakage risk on the casting mold. S4. Calculate the cooling deviation: During the casting process, the operating condition parameters are collected in real time to calculate the current cooling deviation ΔQ = Q1 - Q2; where Q1 is the predicted cooling amount required for the solidification of aluminum liquid under the current operating condition, which is predicted based on the real-time collected operating condition parameters; Q2 is the actual cooling amount Q2 under the current operating condition, which is calculated based on the real-time collected operating condition parameters. S5. Risk Prevention and Emergency Response: Based on whether the risk of aluminum leakage on the casting mold plate has a tendency to spread or accumulate, different emergency prevention and control strategies are triggered to eliminate the risk of aluminum leakage. When the risk of aluminum leakage on the casting mold plate shows a tendency to spread or accumulate, an emergency shutdown strategy is triggered to eliminate the risk of aluminum leakage by stopping the casting process; when the risk of aluminum leakage on the casting mold plate does not show a tendency to spread or accumulate, a multi-parameter sequential control strategy is triggered to eliminate the risk of aluminum leakage by dynamically adjusting the system operating parameters while maintaining the casting process. The system operating parameters include casting speed, cooling water flow rate, cooling water pressure, aluminum liquid level at the casting mold inlet, and cooling water temperature. The multi-parameter sequential control strategy is as follows: within the adjustable threshold range of each system working parameter, the system working parameters are adjusted in multiple rounds according to the priority order of casting speed, cooling water flow rate, cooling water pressure, aluminum liquid level at the casting mold inlet, and cooling water temperature, so that the cooling deviation ΔQ approaches 0, so as to restore the cooling supply and demand balance in the casting process. S4 specifically includes the following steps: S41. Real-time acquisition of working condition parameters during the casting process, including mold type, casting rod specifications, aluminum liquid temperature, casting speed, cooling water outlet temperature, workshop ambient temperature and humidity, cooling water inlet temperature and cooling water flow rate. S42. Input the mold type, casting rod specifications, aluminum liquid temperature, casting speed, cooling water temperature and workshop ambient temperature and humidity into the trained neural network model to obtain the predicted cooling capacity Q1. S43. Calculate the actual cooling capacity under the current operating conditions. Where c is the specific heat capacity of the cooling water, ρ is the density of the cooling water, V is the cooling water flow rate, T1 is the cooling water inlet temperature, and T2 is the cooling water outlet temperature. S44. The cooling deviation ΔQ = Q1-Q2 is obtained through calculation.

2. The method for locating, assessing, and intelligently controlling aluminum leakage risks based on a gridded mold as described in claim 1, characterized in that, In S2, when assessing the risk location of each grid, each grid is marked as a leakage risk grid or an overflow risk grid according to the height of the aluminum liquid level. Furthermore, based on the rate of change of the liquid level, leakage risk grids are marked as leaked grids, high-risk leakage grids, or low-risk leakage grids according to their risk level, and overflow risk grids are marked as overflowed grids, high-risk overflow grids, or low-risk overflow grids according to their risk level.

3. The method for locating, assessing, and intelligently controlling aluminum leakage risks based on a gridded mold as described in claim 2, characterized in that, In S5, when the multi-parameter sequential control strategy is triggered, if there are leaked grids and / or high-risk leaked grids on the casting mold and the cooling deviation ΔQ > 0, the system operating parameters are adjusted in multiple rounds according to the priority order of reducing casting speed, increasing cooling water flow rate, increasing cooling water pressure, reducing aluminum liquid level at the casting mold inlet, and reducing cooling water temperature, until the cooling deviation ΔQ is not greater than 0.

4. The method for locating, assessing, and intelligently controlling aluminum leakage risks based on a gridded mold as described in claim 3, characterized in that, In S5, when the multi-parameter sequential control strategy is triggered, if there are overflowing grids and / or high-risk overflow grids on the casting mold and the cooling deviation ΔQ < 0, then the system operating parameters are adjusted in multiple rounds according to the priority order of increasing casting speed, reducing cooling water flow rate, reducing cooling water pressure, increasing the aluminum liquid level at the casting mold inlet, and increasing the cooling water temperature, until the cooling deviation ΔQ is not less than 0.

5. The method for locating, assessing, and intelligently controlling aluminum leakage risks based on a gridded mold as described in claim 2, characterized in that, In S3, when the grid on the casting mold plate meets one of the following conditions, it is determined that the risk of aluminum leakage on the casting mold plate has a tendency to spread or accumulate: there are 4 or more leaked grids at the same time; or, there are 3 leaked grids at the same time and these 3 leaked grids are adjacent to each other; or, there is at least 1 overflowed grid at the edge of the casting mold plate; or, there are 3 or more overflowed grids at the non-edge of the casting mold plate at the same time. The risk of aluminum leakage on the casting mold is considered to have no tendency to spread or accumulate when the grid on the casting mold meets one of the following conditions: there is one or two leaked grids; or, there are three leaked grids at the same time and these three leaked grids are not adjacent to each other; or, there is one or two overflowed grids at the non-edge of the casting mold; or, the number of both leaked and overflowed grids is 0 and the number of high-risk leakage grids is greater than or equal to 1. Alternatively, the number of leaked and overflowed grids is 0, and the number of high-risk overflow grids is greater than or equal to 1.

6. The method for locating, assessing, and intelligently controlling aluminum leakage risks based on a gridded mold as described in claim 5, characterized in that, The neural network model used to predict cooling demand was trained using the following method: S61. During normal and stable casting in the casting mold, collect several sets of working condition parameters. S62. For each set of operating condition parameters, the actual cooling capacity is calculated and used as the cooling capacity required for each set of operating condition parameters; the mold type, casting rod specifications, aluminum liquid temperature, casting speed, cooling water temperature, workshop ambient temperature and humidity, and cooling capacity are packaged into a sample set according to the corresponding relationship. S63. Using mold type, casting rod specifications, aluminum liquid temperature, casting speed, cooling water temperature and workshop ambient temperature and humidity as model inputs, and the required cooling capacity as model output, construct a neural network model. S64. Train the neural network model built in S63 using the sample set in S62.

7. A system for locating, assessing, and intelligently controlling aluminum leakage risks based on a grid-based mold, used to implement the method described in any one of claims 1 to 6, characterized in that, The system includes a server and a terminal controller. The server is connected to the terminal controller via a switch and a protocol gateway. The server is equipped with a leakage risk location and assessment module, a risk distribution analysis module, a cooling deviation calculation module, and an emergency prevention and control strategy output module. The terminal controller is equipped with a data acquisition module, a dynamic control module, and an emergency shutdown module. The data acquisition module is used to collect the operating status parameters and system operating parameters of the aluminum rod casting production line, and is also used to collect the aluminum liquid level height in different areas of the casting mold in a gridded manner. The operating status parameters include mold type, rod specifications, aluminum liquid temperature, casting speed, cooling water outlet temperature, workshop ambient temperature and humidity, cooling water inlet temperature, and cooling water flow rate. The system operating parameters include casting speed, cooling water flow rate, cooling water pressure, aluminum liquid level at the casting mold inlet, and cooling water inlet temperature. The aluminum leakage risk location and assessment module is used to locate and assess the aluminum leakage risk in each grid on the casting mold in real time based on the aluminum liquid level height obtained by the data acquisition module, and record the location characteristics, risk type and risk level of each grid. The risk distribution analysis module is used to comprehensively analyze the overall distribution of aluminum leakage risk on the casting mold based on the location characteristics, risk type and risk level of each grid recorded in the aluminum leakage risk location and assessment module, so as to determine whether there is a diffusion or accumulation trend of aluminum leakage risk on the casting mold. The cooling deviation calculation module is used to calculate the predicted cooling capacity Q1 and the actual cooling capacity Q2 of aluminum liquid solidification under the current working condition based on the working condition parameters obtained by the data acquisition module, and finally obtain the cooling deviation ΔQ = Q1-Q2, which is provided to the dynamic control module in the terminal controller. The emergency prevention and control strategy output module is used to trigger an emergency shutdown strategy or a multi-parameter sequential control strategy according to the analysis and judgment results of the risk distribution analysis module. When the emergency shutdown strategy is triggered, it outputs an emergency shutdown command to the emergency shutdown module in the terminal controller. When the multi-parameter sequential control strategy is triggered, it outputs a dynamic control command to the dynamic control module in the terminal controller. The emergency stop module is used to stop the casting process of the aluminum rod casting production line when an emergency stop command is received. The dynamic control module is used to dynamically control the system operating parameters when it receives a dynamic control command, in conjunction with the cooling deviation ΔQ provided by the cooling deviation calculation module, so that the cooling deviation ΔQ approaches 0, so as to restore the balance of cooling supply and demand in the casting process.

8. The aluminum leakage risk location assessment and intelligent prevention and control system based on mold gridding according to claim 7, characterized in that, The data acquisition module is connected to the water temperature sensor, water pressure sensor, water flow meter, aluminum liquid temperature sensor, aluminum liquid level sensor, casting speed instrument, and ambient temperature and humidity sensor in the aluminum rod casting production line. The water temperature sensor is used to collect the cooling inlet water temperature and cooling outlet water temperature of the cooling system. The water pressure sensor is used to collect the cooling water supply pressure of the cooling system. The water flow meter is used to collect the cooling water flow rate of the cooling system; The aluminum liquid temperature sensor is used to collect the temperature of the aluminum liquid at the inlet of the casting mold. The aluminum liquid level sensor is used to collect the aluminum liquid level at the inlet of the casting mold. The casting speed instrument is used to detect the casting speed of the casting machine; The ambient temperature and humidity sensor is used to detect the temperature and humidity of the workshop environment. The data acquisition module is also connected to a matrix laser ranging device; the matrix laser ranging device is set above the casting mold and is used to collect the liquid level height of aluminum liquid in different areas of the casting mold in a grid-like manner.

9. The aluminum leakage risk location assessment and intelligent prevention and control system based on mold gridding according to claim 8, characterized in that, The dynamic control module is connected to the casting machine controller, cooling pump controller, electric valve controller, smelting furnace controller, and cooling tower controller, respectively. The casting machine controller is used to adjust the casting speed of the casting machine; The cooling pump controller is used to control the frequency of the cooling pump in the cooling system in order to adjust the cooling water flow rate; The electric valve controller is used to control the opening degree of the electric valve in the cooling system to regulate the cooling water pressure. The furnace controller is used to control the flow rate of molten aluminum at the furnace outlet in order to adjust the molten aluminum level at the casting mold inlet. The cooling tower controller is used to control the frequency and number of cooling towers in the cooling system in order to regulate the cooling water inlet temperature.

Images