Intelligent cooling system for rapid cooling of castings

Abstract

The application discloses an intelligent cooling system for rapidly cooling castings and relates to the technical field of casting cooling. The application discloses an intelligent cooling system for rapidly cooling castings and relates to the technical field of casting cooling. The application discloses an intelligent cooling system for rapidly cooling castings and relates to the technical field of casting cooling. The application discloses an intelligent cooling system for rapidly cooling castings and relates to the technical field of casting cooling. The application discloses an intelligent cooling system for rapidly cooling castings and relates to the technical field of casting cooling. The application discloses an intelligent cooling system for rapidly cooling castings and relates to the technical field of casting cooling. The application discloses an intelligent cooling system for rapidly cooling castings and relates to the technical field of casting cooling.

Description

Technical Field

[0001] This invention relates to the field of casting cooling technology, and in particular to an intelligent cooling system for rapidly cooling castings. Background Technology

[0002] In the casting process, the cooling stage of the casting has a critical impact on the structural integrity, dimensional accuracy, and internal defect control of the final product. Traditional casting cooling methods mainly rely on fixed air cooling, natural convection, or simple spraying, lacking customized control for different casting structures and thermal field characteristics. This can easily lead to problems such as localized overcooling, thermal stress concentration, warping deformation, and even cracks, resulting in increased casting scrap rates and reduced production efficiency.

[0003] Currently, automated cooling systems suffer from the following problems: limited structural recognition capabilities, inability to establish heat conduction models based on casting geometry and material properties, and a lack of targeted cooling solutions; cooling control is usually based only on overall temperature thresholds or time control, failing to achieve differentiated adjustment for different areas; waste occurs during the use of cooling media (such as water or air), and there is a lack of recycling and energy efficiency management mechanisms; it is difficult to accumulate historical data during the cooling process and perform adaptive optimization, making it impossible to improve the level of intelligence during long-term operation. Summary of the Invention

[0004] This invention proposes an intelligent cooling system for rapidly cooling castings, which has the capabilities of structure recognition, dynamic control, energy efficiency optimization and self-learning, so as to achieve differentiated and precise cooling, closed-loop process control and continuous evolution of cooling strategies.

[0005] An intelligent cooling system for rapidly cooling castings includes:

[0006] Geometric sensing module: used to acquire the three-dimensional geometric features and material parameters of the casting;

[0007] Cooling template module: used to obtain the corresponding heat conduction and heat dissipation function model based on the three-dimensional geometric features of the casting, and generate an initial heat dissipation distribution map to construct the cooling template;

[0008] Cooling execution module: It contains three cooling units, namely air cooling unit, spray cooling unit and liquid cooling unit, which are used to implement differentiated cooling for different areas of the casting according to the cooling template;

[0009] Dynamic control module: Used to stabilize the model through cooling structure, perform differential analysis on the real-time temperature data of different areas of the casting surface and interior during the cooling process and the predicted value of the thermal function, and obtain the output parameters of each cooling unit for dynamic adjustment.

[0010] Resource recycling module: used to recover the medium used in the cooling process and reuse energy; and to comprehensively optimize the output parameters based on task intensity, cooling rate, energy consumption minimization and water resource reuse rate maximization;

[0011] Cooling template iteration module: Used to iterate and update the cooling template by collecting output parameters during the casting cooling process and adjusting the entire recording.

[0012] Data accumulation module: Used to store historical data relationships between cooling templates, output parameters and cooling effects, and to realize self-learning optimization of the cooling structure stability model.

[0013] As a preferred embodiment of the present invention, the method for obtaining the three-dimensional geometric features of the casting includes:

[0014] Direct analysis based on preset structural drawings or design model files; matching and identification based on structural parameters and standard casting models in the database; and geometric information acquisition through visual recognition equipment or 3D scanning equipment in the absence of drawings or parameters.

[0015] As a preferred embodiment of the present invention, obtaining the corresponding heat conduction and heat dissipation function model includes:

[0016] Based on the three-dimensional geometric features and material parameters of the casting, the structural features of each structural region of the casting are calculated, including heat transfer path, heat flux density and cooling time constant. According to the structural features of each structural region of the casting, a preset heat conduction and heat dissipation function model template is matched, including spherical structure, shell structure, thin-walled structure and multi-cavity structure. Combining historical cooling data and empirical models, the heat conduction and heat dissipation function model is corrected and fitted to optimize.

[0017] As a preferred technical solution of the present invention, the heat dissipation distribution map is constructed based on the heat conduction and heat dissipation function model, and is used to characterize the heat loss intensity of different regions of the casting per unit time. The heat dissipation distribution map has spatial partitioning attributes, which serve as the basis for the cooling execution module to select the cooling method and intensity.

[0018] As a preferred embodiment of the present invention, the structure of the cooling structure stabilization model includes:

[0019] The data acquisition layer is used to receive real-time temperature data, cooling rate, structural stress calculation parameters, and geometric feature information of different areas of the casting.

[0020] The thermal stress calculation layer, based on the finite element method or a simplified cooling template, predicts and analyzes the thermal stress distribution trend of the casting under the current cooling conditions, and obtains comprehensive thermal stress calculation results.

[0021] The temperature difference assessment layer is used to calculate the temperature gradient change between adjacent areas and compare it with a set stability threshold to obtain the temperature difference assessment result and identify the cooling path that causes warping, cracking or deformation.

[0022] The structural stability assessment layer integrates thermal stress calculation results and temperature difference assessment results to generate a structural stability risk level and formulate cooling control recommendations.

[0023] The regulation and intervention interface layer is used to send dynamic control commands to the cooling execution module to adjust the output parameters of the cooling unit.

[0024] As a preferred embodiment of the present invention, the training of the cooling structure stability model includes:

[0025] Training samples are extracted from temperature change data, structural response data, cooling parameter adjustment records, and final casting quality inspection results collected during the actual cooling process of the casting. A mapping model between input features and cooling stability is established using supervised learning methods and casting structural stability evaluation results as training labels. New data is continuously collected during system operation, and the model is updated online to improve the accuracy of the model's prediction of cooling stress distribution, structural deformation trends, and anomaly risks. The model retraining process is automatically triggered according to the preset evaluation cycle or when a stability degradation trend is detected.

[0026] As a preferred embodiment of the present invention, the recycling and energy reuse include:

[0027] The cooling medium used in spray cooling and liquid cooling processes is collected and recycled for subsequent cooling processes after filtration, sedimentation or purification. The high-temperature airflow or hot water generated during the cooling process is guided to the heat exchange module for preheating the cooling medium, assisting in mold temperature control or providing heat source input for other energy utilization units. The cooling medium circulation efficiency and energy recovery rate are dynamically evaluated according to the system operating status and task load, and the recovery path and allocation priority are adjusted accordingly.

[0028] As a preferred embodiment of the present invention, the comprehensive optimization of the output parameters includes:

[0029] Based on the current task intensity of casting cooling, the priority of cooling efficiency is dynamically set, and the working frequency and intensity of each cooling unit are adjusted. Under the premise of meeting the target cooling rate, the power consumption and water usage during the cooling process are minimized. A comprehensive evaluation function is constructed based on a multi-objective optimization algorithm, and the optimal combination configuration of cooling execution parameters is dynamically generated by combining task intensity, cooling efficiency, energy consumption index and water resource reuse rate.

[0030] As a preferred technical solution of the present invention, the iteration and updating of the cooling template includes:

[0031] When performing cooling tasks for castings of the same type, a candidate cooling template with similar task intensity is selected from the historical template library as an initial reference; during the cooling process, the output parameter adjustment data and cooling effect evaluation results corresponding to the template are recorded; based on the performance indicators after cooling, the adaptability and effect of different templates under the task conditions are comprehensively evaluated, the optimal template is updated and marked, and it is replaced with the recommended template for this type of casting under similar task conditions.

[0032] As a preferred embodiment of the present invention, the self-learning optimization of the cooling structure stabilization model includes:

[0033] The cooling template used in each casting cooling process, the output parameters of each cooling unit, and the corresponding time series are recorded, and a mapping relationship is established with the final cooling effect evaluation results of the casting. Based on the above historical data relationship, a data sample set for model training is constructed, and the prediction accuracy of the cooling structure stability model is optimized by incremental update.

[0034] The present invention has the following advantages:

[0035] This invention improves the targeting and scientific nature of cooling strategies by accurately sensing the geometric features of castings and modeling heat conduction and heat dissipation functions. By setting up three cooling units—air cooling, spray cooling, and liquid cooling—and performing differentiated control based on the heat dissipation distribution map of spatial partitions, it achieves precise cooling control of multiple regions of complex castings.

[0036] This invention constructs a stable cooling structure model to evaluate the thermal stress distribution and structural stability of castings in real time during the cooling process, and combines temperature difference analysis for dynamic intervention, which significantly reduces the risk of deformation and cracking. By introducing a cooling template iteration mechanism, the template is automatically adjusted and optimized according to the task intensity and cooling effect, which improves the adaptability and reusability of the template.

[0037] This invention achieves automatic recovery of cooling medium and reuse of thermal energy by setting up a resource recycling module. Combined with a multi-objective optimization algorithm, it balances cooling speed and energy consumption control to meet the needs of green manufacturing.

[0038] This invention establishes a correlation database between cooling templates, control parameters, and cooling effects through a data accumulation module, and optimizes the cooling structure stability model through an incremental learning mechanism, enabling the system to have intelligent evolution capabilities during long-term operation. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings in the following description are only schematic diagrams of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0040] Figure 1 This is a schematic diagram of the structure of an intelligent cooling system for rapidly cooling castings used in an embodiment of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0042] Example 1: An intelligent cooling system for rapidly cooling castings, specifically comprising:

[0043] 1. Geometric Sensing Module: Used to acquire the three-dimensional geometric features and material parameters of the casting;

[0044] The methods for obtaining the three-dimensional geometric features of the casting include:

[0045] Based on preset structural drawings or design model files, the volume, surface area, wall thickness distribution, key dimensions, and geometric feature points of the casting are directly analyzed. Based on the structural parameters, the casting models in the database are matched and identified. The geometric structure is quickly inferred by comparing the features of the parameter vectorization representation with the stored template. In the absence of drawings or parameters, geometric information is collected by visual recognition equipment or 3D scanning equipment, and a 3D model of the casting is generated by combining point cloud data reconstruction algorithms.

[0046] The aforementioned geometric features will serve as the input basis for the subsequent thermal modeling module, used to construct regional heat conduction paths, determine cooling-sensitive areas, and establish the spatial division basis for the cooling template.

[0047] 2. Cooling Template Module: Used to obtain the corresponding heat conduction and heat dissipation function model based on the three-dimensional geometric features of the casting, and generate an initial heat dissipation distribution map to construct the cooling template;

[0048] The process of obtaining the corresponding heat conduction and heat dissipation function model includes:

[0049] Based on the three-dimensional geometric features and material parameters of the casting, the structural characteristics of each structural region of the casting are calculated, including heat transfer path, heat flux density, and cooling time constant. A simplified heat conduction equation or finite difference method is used to initially model the heat distribution. According to the structural characteristics of the casting, a preset heat conduction and heat dissipation function model template is matched, including typical configurations such as spherical structure, shell structure, thin-walled structure, and multi-cavity structure. Approximate thermal behavior curves are automatically loaded through model library indexing. Combining historical cooling data and empirical models, the heat conduction and heat dissipation function model is corrected and optimized through fitting. Error minimization fitting or machine learning regression methods are used to improve the model prediction accuracy.

[0050] The heat dissipation distribution map is constructed based on the heat conduction and heat dissipation function model. It is used to characterize the heat loss intensity of different regions of the casting per unit time. The regional cooling load is calculated by combining factors such as regional thermal conductivity, heat dissipation area and boundary conditions. The heat dissipation distribution map has spatial partitioning attributes. The surface and interior of the casting are divided into regions by meshing or surface fitting methods to generate a spatial distribution map with cooling level labels, which serves as the basis for the cooling execution module to select the cooling method and intensity.

[0051] The empirical model refers to a mathematical or physical model established based on actual cooling data, engineering experience, and experimental statistical results from historical casting processes. It is used to characterize the heat conduction and heat dissipation characteristics of different casting structures under actual cooling conditions. Compared with purely theoretical models, empirical models realistically reflect the thermal behavior of castings under the influence of complex boundary conditions, material non-ideals, and process disturbances. The empirical model is classified and modeled based on various typical structural forms (such as shells, thin-walled structures, spheres, and multi-cavity structures). By combining it with real-time cooling data, the initial heat conduction or heat dissipation function model is corrected and optimized through fitting, thereby improving the model's accuracy and adaptability.

[0052] 3. Cooling execution module: It contains three cooling units, namely air cooling unit, spray cooling unit and liquid cooling unit, which are used to implement differentiated cooling for different areas of the casting according to the cooling template;

[0053] The air-cooling unit includes a variable frequency control fan, a flow channel, and a multi-angle air valve structure, which can adjust the wind speed, wind direction, and action time according to control commands for forced convection cooling of medium-thickness and symmetrical areas.

[0054] The spray cooling unit includes a high-precision nozzle assembly, an atomization control pump station, and a spray angle adjustment mechanism. It can achieve fine local cooling according to changes in regional heat load and surface shape, and is suitable for structurally weak areas or complex curved surfaces.

[0055] The liquid cooling unit is suitable for castings or molded structures with matching cooling channels. It includes sealed liquid flow pipelines, circulating pumps and temperature control devices, which can realize closed circulation of cooling medium and maintain dynamic control of liquid flow rate and temperature.

[0056] The cooling execution module receives multi-zone control commands from the dynamic control module and the cooling template, supports the combined and collaborative work of multiple cooling methods, and realizes customized cooling control for different parts of the casting through time segmentation, spatial selection or cooling intensity adjustment.

[0057] 4. Dynamic control module: It is used to stabilize the model through the cooling structure, and perform differential analysis on the real-time temperature data of the surface and interior of different areas of the casting during the cooling process and the predicted value of the thermal function to obtain the output parameters of each cooling unit for dynamic adjustment.

[0058] The structure of the cooling structure stabilization model includes:

[0059] The data acquisition layer is used to receive temperature data, cooling rate, structural stress estimation parameters and geometric feature information of different areas of the casting in real time, and supports multi-sensor fusion, including infrared thermal imaging, embedded thermocouple or fiber optic temperature sensor system.

[0060] The thermal stress calculation layer, based on the finite element method or a simplified thermo-mechanical coupling model combined with a cooling template, dynamically predicts the thermal stress evolution trend and critical risk area distribution in each region, and obtains the thermal stress calculation results.

[0061] The temperature difference assessment layer is used to calculate the temperature gradient between adjacent areas and compare the results with the stability threshold set by the system in real time to obtain the temperature difference assessment results, which are used to identify areas that may cause warping, thermal cracking, and internal stress concentration.

[0062] The structural stability assessment layer integrates thermal stress and temperature difference assessment results to form a structural risk level evaluation model, and optimizes the accuracy of risk identification based on historical data;

[0063] The control and intervention interface layer generates corresponding control strategies based on the judgment results and sends them to the cooling execution module in the form of instructions. It adjusts parameters such as wind speed, spray flow rate or liquid cooling rate of each cooling unit in real time to realize dynamic reconstruction and rhythm control of the cooling path, so as to reduce local thermal stress and maintain the stability of the overall structure.

[0064] The training of the cooling structure stability model includes:

[0065] Training samples are extracted from temperature change data, structural response data, cooling parameter adjustment records, and final casting quality inspection results collected during the actual cooling process of the casting. A supervised learning method is used, with the casting structural stability evaluation results as training labels, to establish a mapping model between input features and cooling stability. New data is continuously collected during system operation, and the model is updated online to improve the accuracy of its predictions on cooling stress distribution, structural deformation trends, and anomaly risks. The model retraining process is automatically triggered according to a preset evaluation cycle or when a stability degradation trend is detected, enabling the model to continuously iterate and maintain long-term adaptability.

[0066] 5. Resource recycling module: used to recover the medium used in the cooling process and reuse energy; and to comprehensively optimize the output parameters based on task intensity, cooling speed, energy consumption minimization and water resource reuse rate maximization.

[0067] The recycling and energy reuse include:

[0068] The cooling medium used in spray cooling and liquid cooling processes is recovered and collected using a closed-loop liquid collection device combined with multi-stage filtration, sedimentation, and ultraviolet sterilization. The recovered liquid is then processed and recycled for subsequent cooling tasks. The high-temperature airflow or hot water generated during the cooling process is guided to the heat exchange module, where heat energy is recovered through plate or shell-and-tube heat exchangers. This heat energy is used to preheat the coolant for the next cycle or to provide a heat source for the mold constant temperature system. The heat recovery path and redistribution process are dynamically allocated by the central scheduling controller according to the task plan and the current heat load, thereby maximizing the overall thermal energy utilization efficiency of the system.

[0069] The comprehensive optimization of the output parameters includes:

[0070] Based on the current task intensity of casting cooling, and taking into account delivery cycle and defect risk, the priority weight of cooling efficiency is dynamically set; the working frequency, intensity and duration of each cooling unit are adjusted, and the objective constraint method is used to balance cooling speed, system power consumption and coolant consumption; a multi-objective comprehensive evaluation function is constructed, with function variables including task intensity, cooling efficiency index, unit energy consumption cost, water resource recovery rate, etc., and the optimal control parameter combination configuration is generated through evolutionary algorithm, fuzzy control or adaptive weighted strategy as the output execution scheme under the current working condition.

[0071] 6. Cooling Template Iteration Module: This module is used to iterate and update the cooling template by collecting output parameters during the casting cooling process and adjusting the entire recording.

[0072] The iteration and updating of the cooling template includes:

[0073] When performing cooling tasks on castings of the same type, candidate cooling templates with similar task intensity, structural characteristics, and material types are selected from the historical template library as initial references. The matching process combines task tag retrieval and thermal behavior curve similarity comparison algorithms for screening. During the cooling process, the output parameter adjustment data corresponding to the template is recorded, including the real-time adjustment trajectory, execution frequency, medium consumption, and regional temperature response of each cooling unit, and is archived synchronously with the risk assessment results generated by the structural stability model. Based on the performance indicators after cooling, including cooling time, surface / internal temperature distribution consistency, defect risk score, and energy efficiency, the adaptability and control performance of different templates under the task conditions are comprehensively evaluated. After multiple task runs, the template with the best performance is marked and updated through a template performance ranking mechanism, and set as the recommended default template for this type of casting under similar task conditions, while reserve alternative templates for future special working condition matching.

[0074] 7. Data Accumulation Module: Used to store historical data relationships between cooling templates, output parameters and cooling effects, and to realize self-learning optimization of the cooling structure stability model.

[0075] The self-learning optimization of the cooling structure stabilization model includes:

[0076] The cooling template used in each casting cooling process, the output parameters of each cooling unit, and the corresponding time series are recorded and mapped to the final cooling effect evaluation results of the casting (including temperature distribution consistency, structural deformation degree, surface defect detection results, etc.), forming a multi-dimensional and multi-level data sample set. The data accumulation module uses timestamp marking and task number indexing to classify and store historical samples according to casting type, cooling strategy, anomaly marking, etc., forming a structured data warehouse to support subsequent training calls and backtracking analysis. Based on the above historical data relationships, a dataset for model training is constructed, and the prediction accuracy and generalization ability of the cooling structure stability model are optimized through supervised learning, incremental training, or adaptive fine-tuning strategies. During system operation, the latest cooling task data is introduced in real time for online model updates, and a dynamic evaluation mechanism is set. When the prediction deviation exceeds the set threshold or the model performance deteriorates, the retraining process is automatically triggered to ensure the continuous evolution and long-term adaptability of the model.

[0077] Example 2: An intelligent cooling system for rapidly cooling castings, see [link to example]. Figure 1 As shown, it includes the following modules and units:

[0078] The system includes a geometric sensing module, a cooling template module, a cooling execution module, a dynamic control module, a resource circulation module, a cooling template iteration module, and a data accumulation module. Each module consists of several functional units that work together to complete the intelligent and rapid cooling process of the casting.

[0079] 1. Geometric perception module, including:

[0080] Structural diagram analysis unit: used to extract the three-dimensional geometric information of castings from CAD drawings or design models;

[0081] Parameter matching unit: compares and matches the extracted parameters with standard casting models in the database;

[0082] 3D scanning and recognition unit: When drawings or parameters are missing, it calls visual recognition or laser scanning equipment to automatically obtain the shape and size data of the casting.

[0083] 2. Cooling template module, including:

[0084] Function modeling unit: Substitutes the input geometric information and material parameters into the heat transfer formula to construct multi-region heat conduction and heat dissipation functions;

[0085] Structure recognition and matching unit: Identifies the structure type of the casting (such as shell, cavity, thin wall) and matches the corresponding function template;

[0086] Distribution map generation unit: Maps the function output to a spatial heat dissipation distribution map to form an initial cooling template.

[0087] 3. Cooling execution module, including:

[0088] Air-cooled unit: Equipped with a variable frequency fan and airflow guide device, suitable for rapid cooling of medium to thick areas;

[0089] Spray cooling unit: Equipped with micro-atomizing nozzles with controllable pressure, suitable for thin-walled or irregularly shaped parts;

[0090] Liquid cooling unit: Used in molds or structures for castings with cooling channels, supporting closed-loop liquid circulation;

[0091] Zoned control interface: Different cooling units and parameters are allocated to different zones according to the heat dissipation diagram.

[0092] 4. Dynamic control module, including:

[0093] Temperature acquisition unit: Deploys multiple infrared or embedded sensors to acquire surface and core temperatures;

[0094] Thermal function comparison and analysis unit: Performs differential analysis between real-time temperature and predicted temperature;

[0095] Cooling structure stability model: including thermal stress calculation, temperature difference assessment and structural judgment mechanism;

[0096] Parameter adjustment unit: Outputs control commands to each cooling unit to dynamically adjust air volume, spray volume, or flow rate.

[0097] 5. Resource recycling module, including:

[0098] Media recovery unit: collects cooling water or liquid, and filters and purifies it;

[0099] Energy recovery unit: introduces high-temperature airflow or waste heat liquid into the heat exchange module for preheating;

[0100] Optimize decision-making units: Adjust the frequency and path of data collection based on task urgency and resource utilization.

[0101] 6. Cooling template iteration module, including:

[0102] Template selection unit: Selects initial solutions from historical templates based on casting type and task intensity;

[0103] Data recording unit: Records execution parameters and temperature feedback information throughout the entire process;

[0104] Evaluation and Update Unit: Based on the results of the performance evaluation, the template is optimized, replaced, or recommended for further evaluation.

[0105] 7. Data sedimentation module, including:

[0106] Historical data storage unit: stores templates, control parameters, and test results during the cooling process;

[0107] Sample extraction unit: Constructs a sample set for training;

[0108] Model learning unit: Incremental learning method is used to optimize the cooling structure and stabilize the model;

[0109] Results Feedback Unit: Feeds the learning results back to the heat function prediction and control strategy.

[0110] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An intelligent cooling system for rapidly cooling castings, characterized in that, include: Geometric sensing module: used to acquire the three-dimensional geometric features and material parameters of the casting; Cooling template module: used to obtain the corresponding heat conduction and heat dissipation function model based on the three-dimensional geometric features of the casting, and generate an initial heat dissipation distribution map to construct the cooling template; The process of obtaining the corresponding heat conduction and heat dissipation function model includes: Based on the three-dimensional geometric features and material parameters of the casting, the structural features of each structural region of the casting are calculated, including heat transfer path, heat flux density and cooling time constant; according to the structural features of each structural region of the casting, a preset heat conduction and heat dissipation function model template is matched, including spherical structure, shell structure, thin-walled structure and multi-cavity structure; combined with historical cooling data and empirical models, the heat conduction and heat dissipation function model is corrected and fitted and optimized. Cooling execution module: It contains three cooling units, namely air cooling unit, spray cooling unit and liquid cooling unit, which are used to implement differentiated cooling for different areas of the casting according to the cooling template; Dynamic control module: Used to stabilize the model through cooling structure, perform differential analysis on the real-time temperature data of different areas of the casting surface and interior during the cooling process and the predicted value of the thermal function, and obtain the output parameters of each cooling unit for dynamic adjustment. The structure of the cooling structure stabilization model includes: The data acquisition layer is used to receive real-time temperature data, cooling rate, structural stress calculation parameters, and geometric feature information of different areas of the casting. The thermal stress calculation layer, based on the finite element method or a simplified cooling template, predicts and analyzes the thermal stress distribution trend of the casting under the current cooling conditions, and obtains comprehensive thermal stress calculation results. The temperature difference assessment layer is used to calculate the temperature gradient change between adjacent areas and compare it with a set stability threshold to obtain the temperature difference assessment result and identify the cooling path that causes warping, cracking or deformation. The structural stability assessment layer integrates thermal stress calculation results and temperature difference assessment results to generate a structural stability risk level and formulate cooling control recommendations. The regulation and intervention interface layer is used to send dynamic control commands to the cooling execution module to adjust the output parameters of the cooling unit. Resource recycling module: used to recover the medium used in the cooling process and reuse energy; and to comprehensively optimize the output parameters based on task intensity, cooling rate, energy consumption minimization and water resource reuse rate maximization; Cooling template iteration module: Used to iterate and update the cooling template by collecting output parameters during the casting cooling process and adjusting the entire recording. Data accumulation module: Used to store historical data relationships between cooling templates, output parameters and cooling effects, and to realize self-learning optimization of the cooling structure stability model.

2. The intelligent cooling system for rapidly cooling castings according to claim 1, characterized in that, The methods for obtaining the three-dimensional geometric features of the casting include: Direct analysis based on preset structural drawings or design model files; matching and identification based on structural parameters and standard casting models in the database; and geometric information acquisition through visual recognition equipment or 3D scanning equipment in the absence of drawings or parameters.

3. The intelligent cooling system for rapidly cooling castings according to claim 1, characterized in that, The heat dissipation distribution map is constructed based on the heat conduction and heat dissipation function model and is used to characterize the heat loss intensity of different areas of the casting per unit time. The heat dissipation distribution map has spatial partitioning attributes and serves as the basis for the cooling execution module to select the cooling method and intensity.

4. The intelligent cooling system for rapidly cooling castings according to claim 1, characterized in that, The training of the cooling structure stability model includes: Training samples are extracted from temperature change data, structural response data, cooling parameter adjustment records, and final casting quality inspection results collected during the actual cooling process of the casting. A mapping model between input features and cooling stability is established using supervised learning methods and casting structural stability evaluation results as training labels. New data is continuously collected during system operation, and the model is updated online to improve the accuracy of the model's prediction of cooling stress distribution, structural deformation trends, and anomaly risks. The model retraining process is automatically triggered according to the preset evaluation cycle or when a stability degradation trend is detected.

5. The intelligent cooling system for rapidly cooling castings according to claim 1, characterized in that, The recycling and energy reuse include: The cooling medium used in spray cooling and liquid cooling processes is collected and recycled for subsequent cooling processes after filtration, sedimentation or purification. The high-temperature airflow or hot water generated during the cooling process is guided to the heat exchange module for preheating the cooling medium, assisting in mold temperature control or providing heat source input for other energy utilization units. The cooling medium circulation efficiency and energy recovery rate are dynamically evaluated according to the system operating status and task load, and the recovery path and allocation priority are adjusted accordingly.

6. The intelligent cooling system for rapidly cooling castings according to claim 1, characterized in that, The comprehensive optimization of the output parameters includes: Based on the current task intensity of casting cooling, the priority of cooling efficiency is dynamically set, and the working frequency and intensity of each cooling unit are adjusted. Under the premise of meeting the target cooling rate, the power consumption and water usage during the cooling process are minimized. A comprehensive evaluation function is constructed based on a multi-objective optimization algorithm, and the optimal combination configuration of cooling execution parameters is dynamically generated by combining task intensity, cooling efficiency, energy consumption index and water resource reuse rate.

7. The intelligent cooling system for rapidly cooling castings according to claim 1, characterized in that, The iteration and updating of the cooling template includes: When performing cooling tasks for castings of the same type, a candidate cooling template with similar task intensity is selected from the historical template library as an initial reference; during the cooling process, the output parameter adjustment data and cooling effect evaluation results corresponding to the template are recorded; based on the performance indicators after cooling, the adaptability and effect of different templates under the task conditions are comprehensively evaluated, the optimal template is updated and marked, and it is replaced with the recommended template for this type of casting under similar task conditions.

8. The intelligent cooling system for rapidly cooling castings according to claim 1, characterized in that, The self-learning optimization of the cooling structure stabilization model includes: The cooling template used in each casting cooling process, the output parameters of each cooling unit, and the corresponding time series are recorded, and a mapping relationship is established with the final cooling effect evaluation results of the casting. Based on the above historical data relationship, a data sample set for model training is constructed, and the prediction accuracy of the cooling structure stability model is optimized by incremental update.

Images