Integrated management system and method for molten aluminum processing

Abstract

The application relates to the technical field of integrated management, and discloses a molten aluminum processing technology integrated management system and method. The system comprises an encoding module, a configuration module, a quality inspection module, an embedding module, a decoupling module and a regulation and control module, the modules are connected in series to form a data processing chain, the encoding module is further connected with the quality inspection module to provide deep identification information, and the regulation and control module is connected with the encoding module to provide feature chain data. The system realizes the whole-process management of molten aluminum processing through multi-level encoding, dynamic configuration, cross quality inspection, information embedding, parameter decoupling and linkage regulation and control. The application overcomes the problems of data fragmentation and poor information intercommunication in the existing molten aluminum processing technology management, and realizes the whole-process data integrated management from raw materials to finished products.

Description

Technical Field

[0001] This application relates to the field of integrated management, and in particular to an integrated management system and method for molten aluminum processing. Background Technology

[0002] Molten aluminum processing is a crucial step in aluminum production, and traditionally, it relied heavily on manual experience for control. Existing technologies have developed various process management methods for molten aluminum processing, including temperature control systems, quality inspection systems, and transportation monitoring systems. These systems can monitor and control parameters such as temperature, composition, and flowability of molten aluminum in real time, and record and analyze data through computer systems. Simultaneously, with the development of information technology, intelligent process management methods are gradually being applied to molten aluminum processing, such as using sensor networks for temperature field monitoring and employing digital equipment for quality inspection.

[0003] However, existing methods for managing molten aluminum processing suffer from data silos, with control systems for each process stage operating relatively independently and lacking effective data interaction and information sharing mechanisms. Particularly between different processes such as smelting, transportation, and casting, data transmission often relies on manual recording and transcription, easily leading to information distortion and delays. Furthermore, existing systems lack in-depth analysis of the correlation between raw material characteristics, process parameters, and quality indicators, making it difficult to achieve precise control of process parameters and stable quality control. In addition, during transportation, the dynamic changes in the temperature field and spatial location information are difficult to effectively integrate, affecting the maintenance of melt quality. Summary of the Invention

[0004] This application provides an integrated management system and method for molten aluminum processing, which overcomes the problems of data fragmentation and poor information exchange in the existing molten aluminum processing management, and realizes integrated data management of the entire process from raw materials to finished products.

[0005] In a first aspect, this application provides an integrated management system for molten aluminum processing, comprising: an encoding module, a configuration module, a quality inspection module, an embedding module, a decoupling module, and a control module; the output of the encoding module is connected to the input of the configuration module, the output of the configuration module is connected to the input of the quality inspection module, the output of the quality inspection module is connected to the input of the embedding module, the output of the embedding module is connected to the input of the decoupling module, and the output of the decoupling module is connected to the input of the control module; wherein, the encoding module is also connected to the quality inspection module to provide deep identification information to the quality inspection module, and the output of the control module is also connected to the encoding module to provide process feature chain data; the encoding module is used to obtain batch feature codes by performing multi-level cross-coding processing on aluminum raw materials, perform multi-dimensional mapping analysis of the batch feature codes and material inherent property data to obtain a process feature matrix, and perform hierarchical decomposition of material properties based on the process feature matrix to obtain deep identification information; the configuration module is used to configure the molten aluminum processing process according to the deep identification information. The system employs a multi-dimensional quantitative processing method to obtain a melting characteristic spectrum by nonlinearly and dynamically configuring melting parameters and coupling analysis of temperature, flow, and pressure field data during the melting process. A quality inspection module performs multi-variable cross-quality inspection on the molten aluminum based on the melting characteristic spectrum, performs hierarchical progressive analysis of the inspection data, and performs multi-dimensional cross-mapping of depth identification information, melting characteristic spectrum, and quality inspection data to obtain a quality characterization spectrum. An embedding module embeds hierarchical information into the ladle recording unit based on the quality characterization spectrum, dynamically tracks the temperature field, and performs four-dimensional correlation between spatial location data and the quality characterization spectrum to generate a transport control matrix. A decoupling module decouples multiple parameters of temperature changes based on the transport control matrix, performs hierarchical quantitative adjustment of the holding state, and adaptively fuses the position state and temperature field data to construct a casting characteristic field. A control module uses the casting characteristic field to perform multi-dimensional parameter linkage control of the holding furnace, adaptive compensation of the casting process, and multi-level progressive integration of process parameters to form a process characteristic chain.

[0006] Secondly, this application provides an integrated management method for molten aluminum processing technology. The method includes: obtaining batch feature codes by performing multi-level cross-coding on aluminum raw materials; performing multi-dimensional mapping analysis on the batch feature codes and material inherent property data to obtain a process feature matrix; performing hierarchical decomposition of material properties based on the process feature matrix to obtain depth identification information; performing nonlinear dynamic configuration of melting parameters based on the depth identification information; performing multi-dimensional quantification of the melting state by coupling analysis of temperature field, flow field, and pressure field data during the melting process to obtain a melting feature spectrum; and performing multi-variable cross-quality inspection on the molten aluminum based on the melting feature spectrum, and analyzing the detected data. The process involves hierarchical and progressive analysis, performing multi-dimensional cross-mapping of depth identification information, melting characteristic spectrum, and quality inspection data to obtain a quality characterization spectrum. Based on this spectrum, hierarchical information is embedded into the ladle recording unit. By dynamically tracking the temperature field, spatial location data is correlated with the quality characterization spectrum in four dimensions to generate a transport control matrix. According to this transport control matrix, multiple parameters are decoupled for temperature changes. By performing hierarchical quantitative adjustment of the heat preservation state, the position state and temperature field data are adaptively fused to construct a casting characteristic field. Through this casting characteristic field, multi-dimensional parameter linkage control of the heat preservation furnace is implemented, adaptive compensation is performed on the casting process, and process parameters are integrated in a multi-level progressive manner to form a process characteristic chain.

[0007] The technical solution provided in this application establishes a correspondence between raw material characteristics and process parameters through multi-level cross-coding and multi-dimensional mapping analysis via an encoding module, effectively improving the utilization efficiency of material characteristic data. The configuration module uses a non-linear dynamic configuration method to regulate melting parameters, achieving precise control of the melting state through coupled analysis of temperature, flow, and pressure fields, thus improving the control precision of the melting process. The quality inspection module achieves comprehensive collection and analysis of quality data through multi-variable cross-quality inspection and hierarchical progressive analysis, while the application of multi-dimensional cross-mapping technology ensures the accuracy of quality assessment. The embedding module uses hierarchical information embedding and four-dimensional correlation analysis technology to achieve real-time monitoring and data tracking during transportation, ensuring the stability of melt quality. The decoupling module achieves more precise temperature control through multi-parameter decoupling and hierarchical quantitative adjustment, while the adaptive fusion of position status and temperature field data improves the flexibility of temperature control. The control module uses multi-dimensional parameter linkage control and adaptive compensation technology to ensure the stable operation of the casting process, while the multi-level progressive integration of process parameters achieves the continuity of process control. The close collaboration between the modules forms a complete data processing chain, which significantly improves the automation level and quality control capabilities of the molten aluminum processing technology. Attached Figure Description

[0008] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic diagram of one embodiment of the integrated management system for molten aluminum processing in this application.

[0010] Figure 2 This is a schematic diagram of one embodiment of the integrated management method for molten aluminum processing in this application. Detailed Implementation

[0011] This application provides an integrated management system and method for molten aluminum processing. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0012] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the integrated management system for molten aluminum processing technology in this application includes:

[0013] The system comprises an encoding module 101, a configuration module 102, a quality inspection module 103, an embedding module 104, a decoupling module 105, and a control module 106. The output of the encoding module 101 is connected to the input of the configuration module 102, the output of the configuration module 102 is connected to the input of the quality inspection module 103, the output of the quality inspection module 103 is connected to the input of the embedding module 104, the output of the embedding module 104 is connected to the input of the decoupling module 105, and the output of the decoupling module 105 is connected to the input of the control module 106. The encoding module 101 is also connected to the quality inspection module 103 to provide deep identification information to the quality inspection module 103, and the output of the control module 106 is also connected to the encoding module 101 to provide process feature chain data.

[0014] The encoding module 101 is used to obtain batch feature codes by performing multi-level cross-encoding processing on aluminum raw materials, perform multi-dimensional mapping analysis between batch feature codes and material inherent property data to obtain process feature matrix, and perform hierarchical decomposition of material properties based on process feature matrix to obtain deep identification information.

[0015] The configuration module 102 is used to perform nonlinear dynamic configuration of melting parameters based on depth identification information. By performing coupled analysis of temperature field, flow field and pressure field data during the melting process, the melting state is quantified in multiple dimensions to obtain melting characteristic spectrum.

[0016] The quality inspection module 103 is used to perform multivariate cross-quality inspection on molten aluminum based on the melting characteristic spectrum, perform hierarchical progressive analysis on the test data, and perform multidimensional cross-mapping of depth identification information, melting characteristic spectrum and quality inspection data to obtain quality characterization spectrum.

[0017] The embedding module 104 is used to embed hierarchical information into the ladle recording unit according to the quality characterization spectrum. By dynamically tracking the temperature field, the spatial location data is correlated with the quality characterization spectrum in four dimensions to generate a transportation control matrix.

[0018] The decoupling module 105 is used to decouple multiple parameters of temperature change based on the transport control matrix, and to construct the casting feature field by adaptively fusing the position state and temperature field data through hierarchical quantitative adjustment of the heat preservation state.

[0019] The control module 106 is used to control the holding furnace in multiple dimensions through the casting feature field, to perform adaptive compensation in the casting process, and to integrate the process parameters in a multi-level progressive manner to form a process feature chain.

[0020] Specifically, the integrated management system for molten aluminum processing manages the entire processing technology through the close cooperation of the coding module 101, configuration module 102, quality inspection module 103, embedding module 104, decoupling module 105, and control module 106. A tight data transmission chain is formed between the modules. The output of the coding module 101 is connected to the input of the configuration module 102, the output of the configuration module 102 is connected to the input of the quality inspection module 103, the output of the quality inspection module 103 is connected to the input of the embedding module 104, the output of the embedding module 104 is connected to the input of the decoupling module 105, and the output of the decoupling module 105 is connected to the input of the control module 106. In particular, the coding module 101 also establishes a direct connection with the quality inspection module 103 to transmit depth identification information. Simultaneously, the output of the control module 106 is connected to the coding module 101 via feedback to provide process feature chain data.

[0021] Encoding module 101 processes aluminum raw materials, employing multi-level cross-coding technology to extract material information. For a batch of aluminum raw materials, basic information is first digitized, such as standardizing parameters like Al-99.7% purity, melting point at 700℃, and density at 2.7 g / cm³. These values ​​are then used to generate a unique batch feature code, such as "AL97-D27-T700," through a batch coding algorithm. Subsequently, this batch feature code is mapped and analyzed with other inherent material properties (thermal conductivity, heat capacity, etc.) to construct a process feature matrix. This matrix contains a complete data structure of the material's physical, chemical, and process properties. Through a hierarchical deconstruction algorithm, the data in the matrix is ​​layered according to importance and correlation, ultimately forming deep identification information.

[0022] Upon receiving the depth identification information, the configuration module 102 initiates a nonlinear dynamic configuration process. Taking melting temperature control as an example, it establishes a temperature change curve based on the melting point parameters in the depth identification information, and simultaneously performs coupled analysis by combining flow field data (such as flow velocity and viscosity) and pressure field data (such as static pressure and dynamic pressure). By monitoring the changes in these three fields in real time, such as maintaining the temperature field within the range of 680-720℃, the flow velocity within 0.5-1.0 m / s, and the pressure within the range of ±5% of standard atmospheric pressure, these data undergo multi-dimensional quantification processing to ultimately generate a characteristic spectrum reflecting the melting state. After acquiring the melting characteristic spectrum, the quality inspection module 103 conducts multi-variable cross-quality inspection. It performs comprehensive quality inspection on the molten aluminum, including elemental content analysis, impurity detection, and temperature uniformity testing. The detection data undergoes layered progressive analysis, and each indicator is graded according to its importance. Subsequently, these data are cross-mapped with the previous depth identification information and melting characteristic spectrum in a multi-dimensional manner to generate a complete quality characterization spectrum.

[0023] The embedding module 104 is responsible for writing the information of the quality characterization spectrum into the ladle recording unit. Through hierarchical data processing, key information such as temperature, composition, and physical properties are allocated to different storage areas according to priority. Simultaneously, a dynamic temperature field tracking mechanism is established, using sensors placed at key locations in the ladle to collect temperature change data in real time. This spatial location data (including the three-dimensional coordinates and time information of the temperature acquisition points) is then subjected to four-dimensional correlation analysis with the quality characterization spectrum, ultimately generating a transport control matrix containing complete transport process monitoring information. After receiving the transport control matrix, the decoupling module 105 first decouples the temperature change data. By separating the time and spatial dimensions of the data, key influencing factors of temperature changes are identified. Layered quantitative adjustment of the insulation state is performed, and the optimal insulation parameters are calculated based on the temperature differences at different locations. The location status information and temperature field data are then fused to construct a casting characteristic field reflecting the temperature distribution characteristics of the entire casting process.

[0024] The control module 106 achieves precise control of the holding furnace based on data from the casting characteristic field. Through a multi-dimensional parameter linkage mechanism, it ensures coordinated adjustment of various process parameters. During the casting process, parameter compensation is performed in real time, dynamically adjusting for temperature deviations, fluidity changes, and other factors. Ultimately, the parameter changes and control records of the entire process are integrated in a multi-level progressive manner to form a complete process characteristic chain.

[0025] For example, when processing a batch of aluminum raw materials, the encoding module 101 first extracts its basic characteristics: purity 99.7%, density 2.7 g / cm³, and melting point 660℃. These data are then cross-coded at multiple levels to generate the batch feature code "AL97D27T660". Based on this information, the configuration module 102 sets the initial melting temperature to 700℃. When the actual temperature reaches 680℃, coupling analysis reveals that the flow rate has dropped to 0.3 m / s, lower than the expected value of 0.5 m / s, and the system immediately adjusts the heating power. The quality inspection module 103 continuously monitors the melting process. When a local temperature fluctuation exceeds ±10℃, it triggers the embedding module 104 to record the anomaly and writes this information to the ladle recording unit. The decoupling module 105 analyzes and finds a correlation between this temperature fluctuation and the transport speed, calculating that the optimal transport speed should be controlled between 15-20 km / h. Based on this data, the control module 106 adjusts the power output of the holding furnace in real time to ensure the temperature uniformity of the molten aluminum. The parameter changes throughout the process are recorded in the process feature chain.

[0026] In one specific embodiment, the encoding module 101 is specifically used for:

[0027] (1) Extract the composition, material specifications and production batch number data from aluminum raw materials, and group the data according to the preset arrangement order to obtain the raw material basic data group;

[0028] (2) Numericalize each set of data in the raw material basic data set, convert the composition into element content percentage, the specification data into standard size values, and the batch number data into time series values ​​to obtain a numerical data sequence.

[0029] (3) The numerical data sequence is weighted according to the importance of the elements, and the weighted data is cross-combined to generate batch feature codes;

[0030] (4) By extracting data on the purity, density, thermal conductivity and melting point of aluminum raw materials, a data set of inherent material properties is constructed. The batch feature code is matched with the data set of inherent material properties to establish a data mapping relationship and obtain the process feature matrix.

[0031] (5) Based on the process feature matrix, the material attribute data is processed in layers according to three dimensions: physical characteristics, chemical characteristics and process characteristics. The correlation degree of each layer of data is calculated to obtain attribute correlation data.

[0032] (6) Extract the key parameters from the attribute association data, quantify the degree of mutual influence between the parameters, and fuse the quantification results with the process feature matrix to obtain deep identification information.

[0033] Specifically, the encoding module 101 can convert the physical information of aluminum raw materials into calculable and processable digital information. During the information extraction stage, comprehensive parameter collection is performed on the raw materials, including basic information about the aluminum material (composition, material specifications, production batch number). Composition specifically refers to the types and contents of aluminum and its alloying elements; material specifications include dimensional specifications (length, width, thickness) and shape specifications (plates, bars, profiles, etc.); and the production batch number includes information such as the production date and production line number. The collected raw data is standardized to form a standardized data format. Specifically, the composition data is uniformly converted into percentages, such as the content of the main element Al and the percentage content of added elements Si, Cu, Mg, etc.; the specification data is standardized into metric units, such as the length unit being uniformly millimeters (mm); and the production batch number is converted into a time series value, using the format "year, month, day, hour, minute," facilitating subsequent data processing and traceability. Based on the numerical data series, weight values ​​are assigned according to the degree of influence of different elements on the melting process. The main element Al has the highest weight, followed by alloying elements that have a significant impact on melting performance (such as the effect of Si on fluidity), and then other trace elements. This hierarchical weighting method highlights the importance of key elements, and through a cross-combination algorithm, these weighted data are integrated into a unique batch feature code.

[0034] In the extraction of inherent material property data, the focus is on key physical parameters affecting the melting process: purity (reflecting the material's purity), density (determining the material's mass-volume relationship), thermal conductivity (affecting heating efficiency), and melting point temperature (determining process temperature settings). These parameters, acquired using specialized testing equipment, are used to construct a multidimensional dataset, which is then mapped to the previously generated batch feature codes to form a process feature matrix. Layered processing of the process feature matrix is ​​a crucial step, classifying material property data according to three dimensions: physical properties (e.g., thermal conductivity, density), chemical properties (e.g., composition purity, chemical activity), and process properties (e.g., plasticity, flowability). By calculating the correlation coefficients between different properties, the mutual influence relationships between parameters are evaluated, generating attribute correlation data.

[0035] In the deep identification information generation stage, key parameters that have the most significant impact on the melting process, such as the melting characteristics and flow properties of the material, are extracted from attribute association data. By establishing a parameter influence assessment model, the interactions between these parameters are quantitatively analyzed. Finally, these quantitative results are integrated with the process feature matrix to form complete deep identification information.

[0036] For example, the complete workflow of coding module 101 is as follows: A batch of aluminum alloy material to be processed undergoes initial testing to obtain basic information such as its composition (Al-96.8%, Si-2.5%, Cu-0.5%, other-0.2%), specifications (plate size 1200mm×600mm×10mm), and production batch number (20231225-A01). This data is first converted into a standardized form: composition data is retained as a percentage to two decimal places, size data is standardized to millimeters, and the batch number is converted into a time series value (2023122508, where 08 represents the production period). In the grading and weighting stage, the weight of the main element Al is set to 1.0, Si (significantly affected by the process) is set to 0.8, Cu to 0.6, and other elements to 0.4. These weight values ​​are weighted and calculated with the content of each element, combined with the specifications and time series value, to generate a unique batch feature code. Meanwhile, the purity (99.3%), density (2.71 g / cm³), thermal conductivity (235 W / (m·K)), and melting point (660℃) of this batch of materials were measured using professional testing equipment. These data constitute a dataset of the inherent properties of the materials.

[0037] In the three-dimensional hierarchical processing of physical, chemical, and technological properties, each dimension contains multiple parameters. By calculating the correlation between these parameters, such as the relationship between thermal conductivity and melting rate, or purity and flowability, a complete correlation matrix is ​​obtained. All processing results are then integrated to form in-depth identification information reflecting the complete characteristics of the batch of materials.

[0038] In one specific embodiment, the configuration module 102 is specifically used for:

[0039] (1) Extract the physical property parameters, chemical property parameters, and process property parameters from the depth identification information, divide the parameters into numerical ranges, and obtain the melt control parameters;

[0040] (2) Calculate the data correlation degree of the melting control parameters, and group together the parameters with correlation coefficients higher than the set threshold to obtain parameter correlation groups;

[0041] (3) Prioritize the parameter association groups according to the importance of the process, and perform cross-validation on each parameter association group to obtain nonlinear dynamic configuration parameters;

[0042] (4) By collecting temperature, flow rate and pressure change data in the melting furnace, the collected data is analyzed for time series to obtain the field data sequence;

[0043] (5) Perform correlation calculations on the temperature field, flow field, and pressure field data in the field data sequence, and obtain coupling characteristic values ​​by quantifying the interaction strength of physical quantities;

[0044] (6) Perform multidimensional spatial transformation on the coupled eigenvalues, cross-map the transformed data with the nonlinear dynamic configuration parameters, quantify and evaluate the mapping results, and obtain the melting feature spectrum.

[0045] Specifically, after receiving the depth identification information transmitted by the encoding module 101, the configuration module 102 performs parameter classification and extraction. The depth identification information contains three main categories of parameters: physical property parameters (such as thermal conductivity, melting point temperature, specific heat capacity, etc.), chemical property parameters (such as element content, purity value, activity level, etc.), and process property parameters (such as fluidity, formability, etc.). Each type of parameter needs to be divided into numerical ranges to determine the control range during the melting process. For example, the melting point temperature in the physical property parameters is divided into preheating range, melting range, and superheating range; the element content in the chemical property parameters is divided into standard range and allowable fluctuation range; and the fluidity in the process property parameters is divided into high, medium, and low levels. This range division forms the initial melting control parameters. For the extracted melting control parameters, data correlation calculation is required to determine the mutual influence relationship between parameters. First, the correlation coefficient between each parameter is calculated, and the Pearson correlation coefficient is calculated using standardized data. When the correlation coefficient exceeds a preset threshold (usually set to 0.7 or higher), these highly correlated parameters are grouped together to form a parameter association group. These correlation groups reflect the inherent relationships between parameters, providing a basis for subsequent process control.

[0046] The priority ranking of parameter association groups is determined based on their importance in the process. In molten aluminum processing, temperature-related parameters have the highest priority because temperature directly affects the melting quality; followed by flowability-related parameters, which affect the material's filling effect; and then pressure-related parameters, which affect the stability of the melt. Cross-validation of each parameter association group is conducted to examine their adaptability under different operating conditions, ultimately forming a nonlinear dynamic configuration parameter set. Field data acquisition is a crucial aspect of melting process control. Multiple sensors are deployed within the melting furnace to collect real-time temperature values ​​(including temperature distribution at different locations), flow velocities (including surface and internal velocities), and pressure change data (including static and dynamic pressure). These raw data undergo time-series analysis to examine the changing patterns of each parameter over time. Through noise reduction and smoothing, an effective field data sequence is obtained.

[0047] Complex interactions exist among the temperature, flow, and pressure fields in the field data sequence. Changes in the temperature field affect the material's fluidity, thus influencing the flow field distribution; changes in the flow field, in turn, affect the pressure field distribution, and vice versa, affecting temperature conduction. By establishing a model of the interactions between physical quantities, the strength of these interactions is quantitatively described, yielding coupling characteristic values ​​reflecting the degree of coupling between fields. Performing a multidimensional spatial transformation on these coupling characteristic values ​​essentially processes the data from the three fields within a unified mathematical space. Coordinate transformation maps the data from each field to the same reference system, cross-mapping with nonlinear dynamic configuration parameters. This process requires consideration of the time-varying and spatial distribution characteristics of the parameters, and the mapping results are quantitatively evaluated using appropriate evaluation metrics, ultimately forming a complete melting characteristic spectrum.

[0048] For example, after receiving the deep identification information of a batch of aluminum alloy materials, the configuration module 102 first extracts key parameters: melting point temperature (660℃) and thermal conductivity (237 W / m·K) in physical properties; main element content (Al-96.8%) and impurity content (0.2%) in chemical properties; and flowability index in process properties. These parameters are divided into different control ranges: temperature control range is 660℃±40℃, thermal conductivity allows fluctuation range ±5%, and impurity content upper limit is 0.3%. Through data correlation analysis, it is found that the correlation coefficient between temperature and flowability reaches 0.85, indicating that they have a strong correlation, so they are grouped into the same parameter correlation group. At the same time, it is found that the correlation coefficient between pressure change and flowability is 0.65, which is below the threshold, so it is treated as an independent parameter. After sorting the parameter correlation groups according to process importance, the temperature-flowability correlation group is given the highest priority.

[0049] During the actual melting process, temperature data collected by multi-point temperature sensors showed that the temperature difference at different locations fluctuated between 5-15℃, and flow velocity detection revealed a 2-3 fold difference between the surface flow velocity and the internal flow velocity. These data underwent time-series analysis to eliminate abnormal fluctuations, forming a field data sequence reflecting changes in the melt state. Correlation analysis between the temperature field data and the flow field data revealed that when the temperature increased by 10℃, the flow velocity increased by approximately 0.1 m / s; this correlation was quantified as a coupling characteristic value. Finally, through spatial transformation and cross-mapping, a melting characteristic spectrum containing complete information on temperature distribution, flow characteristics, and pressure changes was generated.

[0050] In one specific embodiment, the quality inspection module 103 is specifically used for:

[0051] (1) The melting characteristic spectrum is segmented according to element content, impurity content and temperature distribution, and the segmented data is normalized to obtain the quality inspection benchmark data.

[0052] (2) The density, hardness, thermal conductivity and fluidity of molten aluminum are tested, and the test results are compared with the quality inspection benchmark data to obtain the quality inspection parameter set;

[0053] (3) Combine and classify the relevant parameters in the quality inspection parameter group, and perform progressive analysis on the classified data to obtain the quality inspection characteristic values;

[0054] (4) Extract material attribute data from the depth identification information, perform correlation calculation between the material attribute data and the quality inspection feature values, and obtain the quality correlation matrix;

[0055] (5) The quality correlation matrix and the fused feature spectrum are fused together, and the quality feature group is obtained by cross-validating the fused data;

[0056] (6) The quality feature groups are classified according to the importance of parameters, the classified data are combined in multiple dimensions, and numerical mapping is performed according to the quality evaluation standard to obtain the quality characterization spectrum.

[0057] Specifically, after receiving the melting characteristic spectrum from the configuration module 102, the data needs to be segmented. The melting characteristic spectrum contains data such as elemental content (content data of main elements, alloying elements, and impurity elements), impurity content (non-metallic inclusions, gas content, etc.), and temperature distribution (temperature values ​​at various points in space), which need to be segmented according to different standards. Elemental content is segmented according to chemical composition standards, impurity content is segmented according to purity levels, and temperature distribution is segmented according to spatial location. These segmented data are then normalized to convert data of different dimensions into a unified standard range, forming comparable quality inspection benchmark data. Performing multiple physical property tests on molten aluminum is a key step in quality control. Density testing uses the water displacement method or X-ray method to obtain the density value of the melt; hardness testing uses a dedicated melt hardness tester to measure the resistance value of the melt; thermal conductivity testing uses a thermal diffusivity measuring device to obtain the thermal conductivity; and flowability testing uses a dedicated flowability tester to record the flow distance and time of the melt. These test data are compared and analyzed with the previously obtained quality inspection benchmark data to examine the degree of deviation between the actual measured values ​​and the standard values, thus forming a set of quality inspection parameters.

[0058] The quality inspection parameter set contains a large amount of test data, which needs to be organized and classified in a reasonable manner. First, the data is grouped according to the correlation of physical properties. For example, parameters related to flow characteristics, such as density and fluidity, are grouped together, while parameters related to intrinsic material properties, such as hardness and thermal conductivity, are grouped together. A progressive analysis method is then used on the classified data, gradually analyzing the relationships between parameters from basic physical quantities to composite physical quantities, ultimately obtaining quality inspection characteristic values ​​that reflect the overall performance of the material.

[0059] The deep identification information contains the material's original property data, which are the material's characteristic parameters at room temperature. By extracting this material property data and performing correlation analysis with the quality inspection characteristic values ​​obtained in the molten state, a correspondence between the material's performance parameters in the room temperature and molten states is established. Through methods such as calculating correlation coefficients and establishing regression equations, a quality correlation matrix reflecting the material's performance change patterns is formed. The quality correlation matrix contains data on the material's performance changes from the solid to the liquid state and needs to be fused with the process parameters in the melting characteristic spectrum. The fusion process employs multi-source data fusion technology, comprehensively processing data from different sources and verifying the consistency and reliability of the data through cross-validation, ultimately obtaining a quality characteristic set that comprehensively reflects the material's performance.

[0060] A quality characteristic group is a collection of multidimensional data that needs to be graded according to the degree of influence of each parameter on product quality. For example, parameters that directly affect product performance (such as component uniformity) are classified as primary indicators, while parameters that indirectly affect product performance (such as temperature uniformity) are classified as secondary indicators. The graded data are then combined and numerically mapped according to quality evaluation standards to ultimately form a complete quality characterization spectrum.

[0061] For example, when a batch of aluminum alloy material enters the melting stage, the quality inspection module 103 first receives the melting characteristic spectrum of that batch. The elemental content data in the characteristic spectrum shows the distribution of the main element Al, the alloying element Si, and other trace elements. This data is first divided into different composition intervals. Simultaneously, the temperature distribution data shows the temperature values ​​of the melt at different locations, and this data is segmented according to spatial coordinates. All segmented data is normalized and converted into values ​​within a standard 0-1 range, forming the initial quality inspection benchmark. Subsequently, the melt is monitored in real time using various testing devices: a density meter measures the actual density of the melt and compares it with the theoretical density; a hardness tester measures the apparent hardness of the melt; a thermal conductivity meter records the heat transfer capacity; and a flowability tester determines the flow characteristics of the melt. Comparing these test data with the quality inspection benchmark data reveals a significant correlation between density and flowability, therefore these two sets of parameters are grouped together.

[0062] Through progressive analysis, it was discovered that a density change of 0.1 g / cm³ corresponds to a change in flowability; this relationship was recorded as a quality inspection characteristic value. Simultaneously, the material's performance parameters at room temperature were extracted from the depth identification information and compared with the characteristic values ​​in the molten state to establish a temperature-performance correlation, which was recorded in the quality correlation matrix. The quality correlation matrix was then fused with the original melting characteristic spectrum to obtain a complete dataset reflecting the material's performance changes throughout the melting process. These data were then graded according to their impact on the final product quality, forming a multidimensional quality characterization system.

[0063] In one specific embodiment, the embedded module 104 is specifically used for:

[0064] (1) The temperature data, composition data and physical property data in the quality characterization spectrum are classified according to their importance, and the classified data are encoded and converted to obtain the classified data group;

[0065] (2) Divide the storage capacity of the ladle recording unit into blocks, and write the hierarchical data groups into the corresponding blocks according to the data priority order to obtain the recording data sequence;

[0066] (3) Temperature data is collected by temperature sensing points arranged inside the ladle, and time series analysis is performed on the collected data to obtain the temperature change curve;

[0067] (4) Combine the longitude, latitude, altitude and time data of the steel ladle during transportation to obtain a spatial location sequence;

[0068] (5) Perform data association between temperature change curves and spatial location sequences, and cross-validate the association results with the recorded data sequences to obtain multidimensional association data;

[0069] (6) By performing numerical transformation on multidimensional related data, the transformation results are reorganized and integrated according to the spatiotemporal correspondence to generate a transportation control matrix.

[0070] Specifically, the embedded module 104 effectively records and tracks the transportation process using the quality characterization spectrum data generated by the quality inspection module 103. First, three types of key data are extracted from the quality characterization spectrum: temperature data (including melt temperature, ambient temperature, and their distribution), composition data (including main element content, alloy element distribution, and impurity levels), and physical property data (including fluidity, thermal conductivity, density, and other physical properties). These data are classified according to their importance based on their impact on the quality of the molten aluminum. Temperature data, because it directly affects the melt state, is classified as Level 1 important data; composition data, because it affects the performance of the final product, is classified as Level 2 important data; and physical property data, as a condition monitoring parameter, is classified as Level 3 important data. The classified data is converted into standardized digital sequences using specific coding rules, forming standardized graded data groups. The ladle recording unit is a data storage device installed on the transport ladle and needs to be divided into blocks according to the importance and storage characteristics of the data. The storage space is divided into multiple functional blocks: a real-time data area (for storing frequently updated data such as temperature), a static data area (for storing relatively stable data such as composition), and a status data area (for storing periodically updated data such as physical properties). Data in the hierarchical data groups is written to corresponding storage blocks according to priority. High-priority temperature data is written to the real-time data area to ensure fast access and updates; composition data is written to the static data area to ensure data integrity; and physical property data is written to the status data area for easy monitoring and analysis. This partitioned storage method forms an ordered sequence of recorded data.

[0071] Temperature monitoring is one of the most critical aspects of the transportation process. Multiple temperature sensors are installed inside the steel ladle. These sensors are arranged according to a "three-dimensional uniform distribution" principle: temperature sensors are placed at the top, middle, and bottom, while also maintaining a uniform radial distribution. Each sensor collects temperature data at a preset sampling frequency (e.g., once per minute). These data points are arranged chronologically, and abnormal fluctuations are removed through data smoothing to ultimately form a temperature change curve reflecting temperature variations over time. The geographical location information of the steel ladle, including longitude, latitude, and altitude, is recorded in real time using a GPS positioning system. This location data, along with time information (accurate to the second), is recorded to form a spatial location sequence describing the transportation trajectory. The introduction of the time dimension gives the location information a temporal characteristic, facilitating subsequent correlation analysis with temperature data.

[0072] Correlation analysis between temperature change curves and spatial location sequences is a crucial method for discovering patterns in temperature variations. By pairing temperature data at each time point with corresponding location data, a three-dimensional correlation between temperature, location, and time is established. This correlated data then needs to be cross-validated with other parameters (such as composition and physical properties) stored in the recorded data sequence to ensure data consistency and reliability, ultimately forming multidimensional correlated data containing information across multiple dimensions. This multidimensional correlated data is then converted into a standardized transportation control matrix. This process includes data normalization, enabling comparisons of different data types within the same numerical range; and reorganizing the data structure according to the temporal and spatial correspondences, integrating discrete data points into a continuous control sequence. The resulting transportation control matrix contains complete transportation process monitoring data, providing a basis for subsequent process control.

[0073] For example, during the transport of a batch of molten aluminum from the smelting workshop to the casting workshop, the embedded module 104 first receives the batch's quality characterization spectrum. The temperature data shows the current melt temperature is 720℃, the composition data shows the main element Al content is 99.7%, and the physical property data shows a thermal conductivity of 237 W / (m·K). After these data are classified, the temperature data is marked as the highest priority and updated in real time; the composition data is stored as a fixed basic parameter; and the physical property data is updated and stored periodically. Temperature sensors arranged inside the ladle continuously collect data during transportation. For example, at a certain point in time, the top sensor records a temperature distribution of 718℃, the middle 721℃, and the bottom 719℃. These data points form a temperature curve over time, while the GPS system records the location information of the transport vehicle. When the transport vehicle turns or climbs a slope, the temperature sensors may detect local temperature fluctuations. These changes correspond to the location data, helping to analyze the impact of the external environment on the temperature distribution. Finally, all this data is integrated into the transportation control matrix, providing a basis for real-time adjustment of insulation measures.

[0074] In one specific embodiment, the decoupling module 105 is specifically used for:

[0075] (1) Group the temperature data, location data and time data in the transportation control matrix, perform correlation analysis on the grouped data, and obtain the parameter combination sequence;

[0076] (2) Decompose the temperature change data in the parameter combination sequence into time dimension, classify the decomposed data according to the change trend, and obtain the temperature change feature value;

[0077] (3) Perform corresponding analysis on the temperature change characteristic values ​​and the location change data, sort the analysis results, and obtain the heat preservation control parameters;

[0078] (4) By performing layered processing on the thermal insulation control parameters, the processing results are quantified according to the thermal insulation requirements to obtain thermal insulation adjustment data;

[0079] (5) Perform numerical matching between the thermal insulation adjustment data and the location status data, and combine the matching results according to the spatiotemporal correspondence to obtain the temperature fusion sequence;

[0080] (6) Extract the key parameters from the temperature fusion sequence, perform spatial distribution analysis on the extracted data, reconstruct the data according to the process requirements, and construct the casting feature field.

[0081] Specifically, the decoupling module 105 is responsible for processing the transport control matrix transmitted from the embedded module 104. First, it classifies the data in the matrix. Temperature data includes real-time temperature values ​​and trends at various points in the melt; location data includes spatial coordinates and speed during transport; and time data records the time points of the entire transport process. When grouping this data, a time window method is used, segmenting the continuous data stream according to fixed time intervals (e.g., 5 minutes), with each time window's data serving as an analysis unit. By calculating the correlation coefficients between parameters, significantly correlated parameter combinations are identified, such as the relationship between temperature change and transport speed, or the relationship between temperature distribution and terrain changes, forming a parameter combination sequence reflecting the correlation between parameters. The time dimension decomposition of temperature change data is for a deeper understanding of temperature change patterns. First, the temperature data is expanded according to a time series, and then Fourier transform is used to decompose temperature changes into three parts: a trend term, a periodic term, and a random term. The trend term reflects the overall direction of temperature change, the periodic term reflects the regular fluctuations in temperature, and the random term represents unpredictable fluctuations. Based on these characteristics, temperature changes are divided into stable periods, fluctuating periods, and transition periods. Characteristic parameters for each period, such as the rate of temperature change and the amplitude of fluctuation, are calculated to obtain characteristic values ​​of temperature change that describe the characteristics of temperature change.

[0082] When analyzing the correspondence between temperature change characteristic values ​​and location change data, a spatiotemporal correspondence is first established. Each temperature characteristic value corresponds to a specific location and time point. By analyzing the mapping relationship between temperature change and location change, key location points affecting temperature change are identified, such as turning points and slope change points. The data of these key points are prioritized, and the insulation control requirements for different location points are determined based on the sensitivity of temperature changes, forming insulation control parameters to guide insulation control. Layered processing of insulation control parameters is used to achieve precise temperature control. The insulation control parameters are divided into an emergency control layer, a normal control layer, and a preventative control layer. The emergency control layer handles situations of rapid temperature changes, the normal control layer maintains a normal temperature fluctuation range, and the preventative control layer addresses potential temperature change risks. The control parameters at each level are quantified to determine specific control indicators, such as heating power and insulation time, ultimately forming executable insulation adjustment data.

[0083] When matching insulation regulation data with location status data, the actual conditions during transportation are taken into account. For example, insulation intensity needs to be increased on uphill sections, constant temperature needs to be maintained in waiting areas, and rapid temperature drop needs to be prevented on downhill sections. By establishing a correspondence between insulation regulation and location status, appropriate insulation measures can be taken at different locations. These matching results are organized according to the correspondence between time and space to form a temperature fusion sequence reflecting the overall temperature control strategy. Constructing a casting characteristic field is a crucial step in preparing for subsequent casting processes. Key control parameters, such as temperature uniformity indicators and temperature gradient distribution, are extracted from the temperature fusion sequence. Spatial distribution analysis is performed on these parameters to understand the overall distribution characteristics of the temperature field. Based on the specific requirements of the casting process, this data is reorganized to construct a characteristic field reflecting the casting conditions, providing a basis for precise control of the casting process.

[0084] For example, during the transport of molten aluminum, the transport control matrix received by the decoupling module 105 shows that at the start of transport, the temperature distribution inside the ladle is relatively uniform, with temperature differences between the top, middle, and bottom within 5°C. As the transport vehicle begins to move, the temperature data shows a correlation with speed changes; temperature fluctuations increase during acceleration, while the temperature tends to stabilize at a constant speed. Time-dimensional decomposition reveals a periodic fluctuation in temperature changes of approximately 30 minutes, which coincides with the terrain change cycle of the transport route. Corresponding analysis shows that the temperature drop rate increases at specific locations (such as long uphill sections), requiring increased insulation intensity in advance. These key locations are given higher control priority, and corresponding insulation control parameters are adjusted. For example, insulation power is increased before uphill sections, and basic insulation levels are maintained on flat sections to ensure that temperature fluctuations are controlled within allowable limits throughout the process. The constructed casting characteristic field contains complete temperature distribution information, showing the spatial distribution characteristics of the temperature inside the ladle and the evolution of the temperature field over time. This information directly guides the setting of subsequent casting process parameters, ensuring the smooth progress of the casting process.

[0085] In one specific embodiment, the control module 106 includes:

[0086] (1) Association unit, used to classify and combine temperature distribution data, spatial location data and temporal change data in the casting characteristic field, perform association analysis on the combined data, and obtain furnace temperature control parameters;

[0087] (2) Divide into units, which are used to subdivide the furnace temperature control parameters. The subdivision results are divided into intervals according to the temperature fluctuation trend to obtain the heat preservation control sequence;

[0088] (3) Extraction unit, used to extract data from key control points in the heat preservation control sequence, perform cross-validation on the extracted data, and obtain compensation control parameters;

[0089] (4) Comparison unit, used to compare the compensation control parameters with the casting process requirements, and to classify the comparison results according to the degree of deviation to obtain the compensation data set;

[0090] (5) Arrangement unit, used to arrange the process parameters in the compensation data group according to the process sequence, perform progressive analysis on the arranged data, and obtain the process parameter sequence;

[0091] (6) Combination unit, used to combine related parameters in the process parameter sequence in multiple levels, and reconstruct the data of the combination result according to the process flow to form a process feature chain.

[0092] Specifically, the control module 106 converts the casting characteristic field transmitted from the decoupling module 105 into specific process control parameters. First, the correlation unit receives the casting characteristic field data, which includes three types of key information: temperature distribution data (reflecting the temperature values ​​and distribution patterns of the melt at various points in space), spatial location data (describing the spatial coordinates of temperature measurement points and their relative relationships), and temporal change data (recording the temperature change trend over time). These three types of data are categorized and combined to form data groups. Then, the inherent relationships between the data are identified through correlation analysis methods, such as the correlation between temperatures at different locations and the correspondence between temperature changes and time, ultimately yielding the control parameters used to control the furnace temperature. After receiving the furnace temperature control parameters, the division unit first subdivides the parameters. The temperature change range is divided into multiple control intervals, such as preheating intervals, holding intervals, and temperature adjustment intervals. Within each interval, specific control requirements are determined based on the characteristics of temperature fluctuations (such as fluctuation amplitude, fluctuation frequency, and rate of change). These subdivided control requirements are integrated into a complete holding control sequence to guide the operation of the holding furnace.

[0093] The main task of the extraction unit is to identify and extract key control points from the heat preservation control sequence. These control points include temperature abrupt change points, steady-state transition points, and critical control points. The data for each control point needs to undergo cross-validation, i.e., comparison with data from adjacent time points and adjacent spatial points to ensure the accuracy and representativeness of the data. The validated control point data is compiled into compensation control parameters for subsequent temperature compensation control. The comparison unit performs a detailed comparison of the compensation control parameters with the preset casting process requirements. These process requirements include multiple aspects such as temperature control accuracy, uniformity requirements, and heat preservation time requirements. By calculating the deviation between the parameters and the requirements, and classifying the deviation according to its magnitude (e.g., severe deviation, moderate deviation, slight deviation), corresponding compensation strategies are formulated for different levels of deviation, ultimately forming a compensation data set.

[0094] The arrangement unit is responsible for rearranging the process parameters in the compensation data set according to the actual casting process sequence. The casting process includes multiple stages such as preheating, heating, temperature holding, temperature adjustment, and pouring, each with its specific process requirements. By performing progressive analysis on the rearranged data, the connection and continuity of parameters between each process are ensured, ultimately generating a complete process parameter sequence. The combination unit is the final processing step, responsible for multi-level combination of related parameters in the process parameter sequence. This combination considers the mutual influence and constraints between parameters, ensuring that all parameters can coordinate and cooperate during execution. The final combination result is reconstructed according to the process flow, forming a complete process feature chain. This feature chain includes all process parameters and control requirements from raw materials to finished products.

[0095] For example, when a batch of molten aluminum enters the casting stage, the control module 106 first processes the casting characteristic field data through the correlation unit. Temperature distribution data shows the temperature values ​​at different locations within the furnace, spatial location data records the specific locations of these temperature measuring points, and temporal variation data reflects the temperature change trend. After correlation analysis, the influence relationships between various temperature control points are determined. The division unit divides the temperature control range into different intervals, such as the working interval of 720-740℃ and the warning interval of 710-720℃. Each interval has a corresponding control strategy; for example, a conventional heating mode is used in the working interval, while a rapid heating mode is activated in the warning interval. These control strategies are integrated into a heat preservation control sequence. The extraction unit identifies key points from the control sequence, such as the inflection point where the temperature begins to decrease and the abrupt change point where temperature fluctuations increase. After cross-validation with surrounding data to confirm their reliability, compensation control parameters are formed. The comparison unit compares these parameters with process standards, and when a deviation in temperature uniformity is found, a corresponding compensation scheme is immediately generated. The arrangement unit sorts the compensation schemes according to the casting process to ensure that the temperature parameters of each process meet the process requirements. Finally, the combination unit integrates these sorted parameters into a complete process feature chain to guide the temperature control of the entire casting process.

[0096] In one specific embodiment, the dividing unit is specifically used for:

[0097] (1) Group the furnace temperature control parameters according to temperature range, rate of change and fluctuation amplitude, and perform numerical conversion on the grouped data to obtain temperature characteristic data;

[0098] (2) Divide the temperature feature data into time dimensions, sort the results according to their numerical values, and obtain the time series data group;

[0099] (3) Combine the correlation parameters in the time series data group, perform correlation calculation on the combined data, and obtain the fluctuation trend value;

[0100] (4) Define the interval of the fluctuation trend value according to the change law, and normalize the value of the definition to obtain the interval parameter;

[0101] (5) Match the interval parameters with the temperature control requirements, and perform hierarchical processing on the matching results to obtain the control data set;

[0102] (6) The control data group is reconstructed according to the insulation requirements, and the reconstruction results are integrated to obtain the insulation control sequence.

[0103] Specifically, the segmentation unit first groups the received furnace temperature control parameters in multiple dimensions, including temperature range (the temperature range of the melt in different regions), rate of change (how quickly the temperature changes over time), and fluctuation amplitude (the degree to which the temperature deviates from the average value). These grouped data are then standardized and converted into numerical forms with uniform dimensions, forming temperature characteristic data. Time dimension segmentation is a crucial step in performing time series analysis on the temperature characteristic data. Continuous temperature data is segmented according to fixed time intervals, with each time segment serving as an independent analysis unit. These segmented data are then sorted according to their temperature values, forming ordered time series data sets, which facilitates the discovery of regular patterns in temperature changes.

[0104] Time-series data sets contain multiple related parameters, such as average temperature, rate of temperature change, and temperature fluctuation amplitude over a certain period. These intrinsically linked parameters are combined, and the correlation coefficients between them are calculated to assess the degree of association. This correlation analysis helps identify the main trends in temperature change, obtaining fluctuation trend values ​​that reflect the characteristics of temperature changes. Defining the intervals of these fluctuation trend values ​​is to divide continuous data into discrete control intervals. Based on the patterns of temperature change, stable intervals, fluctuating intervals, and transitional intervals can be identified. The results of these interval definitions are then normalized to unify data at different scales within a standard range, resulting in standardized interval parameters.

[0105] The interval parameters need to be matched with specific temperature control requirements. Temperature control requirements include allowable temperature fluctuation range, temperature uniformity requirements, and limits on heating and cooling rates. By comparing the degree of compliance between the interval parameters and the control requirements, the matching results are classified and hierarchical control data sets are formed.

[0106] Parameter reconstruction involves reorganizing the control data set according to the requirements of the actual insulation process. This process considers multiple requirements such as insulation time, insulation temperature, and temperature uniformity, integrating discrete control parameters into a complete insulation control sequence.

[0107] For example, when a batch of molten aluminum requires temperature control, the furnace temperature regulation parameters are received first. These parameters show that the temperature distribution at different locations ranges from 710 to 730°C, with a temperature change rate of approximately 2°C / minute and a temperature fluctuation range of ±5°C. This data is standardized and converted to relative values. Time-division is then applied, dividing the temperature data into 10-minute time windows. The temperature data within each window is sorted, revealing the periodicity of temperature changes. Correlation analysis reveals a positive correlation between the rate of temperature change and the fluctuation amplitude; that is, the faster the temperature changes, the greater the fluctuation amplitude.

[0108] Based on these analysis results, the temperature control range is divided into: a stable range (temperature change rate < 1℃ / min), an adjustment range (temperature change rate 1-3℃ / min), and a transition range (temperature change rate > 3℃ / min). Each range has a corresponding control strategy. For example, the stable range maintains the existing heating power, the adjustment range requires dynamic adjustment of the heating power, and the transition range requires special control measures.

[0109] In one specific embodiment, the comparison unit is specifically used for:

[0110] (1) Extract the temperature, time and flow values ​​from the compensation control parameters, standardize the extracted data, and obtain the compensation benchmark value;

[0111] (2) Numericalize the parameters in the casting process requirements, sort the results according to priority, and obtain the process standard values;

[0112] (3) Calculate the difference between the compensation benchmark value and the process standard value, and filter the calculation results to obtain the deviation data;

[0113] (4) The deviation data are classified according to their numerical values, and the classification results are defined by thresholds to obtain a deviation level table.

[0114] (5) Reorganize the parameters in the deviation level table according to the compensation requirements, optimize the reorganization results, and obtain the compensation parameter set;

[0115] (6) Integrate the compensation parameter group, and calibrate the parameters according to the process requirements to obtain the compensation data group.

[0116] Specifically, key data from the compensation control parameters are extracted, including temperature values ​​(reflecting the current temperature state of the melt), time values ​​(representing the duration of the process), and fluidity values ​​(characterizing the flow characteristics of the melt). These extracted raw data undergo standardization, converting data of different dimensions into a unified numerical range to form comparable compensation benchmark values. Casting process requirements typically include multiple levels of process indicators, requiring the conversion of these qualitative and quantitative requirements into specific numerical parameters. For example, temperature uniformity requirements are converted into allowable temperature fluctuation ranges, fluidity requirements into flow rate ranges, and time requirements into the duration of each process step. These converted values ​​are prioritized according to their impact on product quality, ultimately forming standardized process standard values.

[0117] The difference between the compensation benchmark value and the process standard value reflects the degree of deviation between the actual process parameters and the standard requirements. The difference calculation uses a relative error approach, considering both the magnitude of the absolute error and the proportion of deviation relative to the standard value. The calculated differences are then filtered to remove insignificant deviations, retaining only those requiring compensation adjustments. The tiered processing of deviation data is used to develop different levels of compensation strategies. Based on the magnitude of the deviation, multiple threshold ranges are set, classifying deviations into different levels. For example, they can be divided into minor deviations (requiring fine-tuning), moderate deviations (requiring routine adjustments), and severe deviations (requiring urgent handling), forming a complete deviation level table.

[0118] The parameters in the deviation level table need to be reorganized according to the compensation requirements. Different compensation strategies are used for different deviation levels; for example, slight deviations use gradual compensation, moderate deviations use stepwise compensation, and severe deviations require immediate mandatory compensation measures. These compensation strategies are optimized to form an executable set of compensation parameters. The compensation parameter set is then integrated and calibrated. The integration process needs to consider the mutual influence between parameters to ensure that there are no conflicts between the various compensation measures. The calibration process adjusts the compensation parameters to the range most suitable for the actual process conditions, ultimately forming a complete set of compensation data.

[0119] For example, when a set of compensation control parameters is received, data such as the current temperature (725℃), process duration (45 minutes), and flowability indicators are first extracted. This data is standardized and converted to a uniform numerical range of 0-1, forming a compensation baseline value. Simultaneously, the process requirements specify a standard temperature of 720±5℃, a process time of 40-50 minutes, and flowability within a specific range. These requirements are converted into specific numerical standards, with temperature control given the highest priority because it directly affects product quality.

[0120] By calculating the difference between the compensation benchmark value and the process standard value, it was found that the temperature exceeded the standard by 5℃, the time was within the allowable range, and the fluidity had a slight deviation. These deviations were classified into different levels: temperature deviation was considered moderate and required timely adjustment; fluidity deviation was considered slight and could be resolved through fine-tuning. Compensation strategies were developed based on the deviation levels. For temperature deviation, a step-by-step cooling method was adopted, with each temperature drop of 1℃ followed by a 2-minute interval to observe the effect; for fluidity deviation, indirect adjustment was achieved through fine-tuning the temperature. These compensation measures were integrated and calibrated to form a complete compensation scheme to guide subsequent process adjustments.

[0121] The integrated management system for molten aluminum processing in the embodiments of this application has been described above. The integrated management method for molten aluminum processing in the embodiments of this application is described below. Please refer to [link / reference]. Figure 2One embodiment of the integrated management method for molten aluminum processing technology in this application includes:

[0122] S201. By performing multi-level cross-coding on aluminum raw materials, batch feature codes are obtained. The batch feature codes are then mapped and analyzed with the inherent material property data in multiple dimensions to obtain a process feature matrix. Based on the process feature matrix, the material properties are deconstructed hierarchically to obtain deep identification information.

[0123] S202. Based on the depth identification information, nonlinear dynamic configuration of melting parameters is performed. By coupling analysis of temperature field, flow field and pressure field data during the melting process, multi-dimensional quantitative processing of the melting state is performed to obtain the melting characteristic spectrum.

[0124] S203. Based on the melting characteristic spectrum, multivariate cross-quality inspection is performed on molten aluminum. The test data is analyzed in a hierarchical and progressive manner. The depth identification information, melting characteristic spectrum and quality inspection data are cross-mapped in multiple dimensions to obtain the quality characterization spectrum.

[0125] S204. Based on the quality characterization spectrum, hierarchical information is embedded into the ladle recording unit. By dynamically tracking the temperature field, the spatial location data is correlated with the quality characterization spectrum in four dimensions to generate a transportation control matrix.

[0126] S205. Based on the transportation control matrix, multiple parameters are decoupled for temperature changes. By performing layered quantitative adjustment of the heat preservation state, the position state and temperature field data are adaptively fused to construct the casting characteristic field.

[0127] S206. Through the casting characteristic field, the holding furnace is controlled by multi-dimensional parameters, the casting process is adaptively compensated, and the process parameters are integrated in a multi-level progressive manner to form a process characteristic chain.

[0128] In this embodiment, a deep deconstruction of raw material characteristics is achieved through a combination of multi-level cross-coding and multi-dimensional mapping analysis, providing an accurate data foundation for subsequent process control. Nonlinear dynamic configuration technology is used to regulate melting parameters, and coupled analysis of temperature, flow, and pressure fields makes the control of the melting process more precise. Multivariate cross-quality inspection and hierarchical progressive analysis enable comprehensive monitoring of quality data, and a complete quality characterization system is established through multi-dimensional cross-mapping. In the transportation stage, hierarchical information embedding technology and four-dimensional correlation analysis are used to achieve precise tracking and control of the transportation process. The application of multi-parameter decoupling technology makes temperature change control more flexible, and hierarchical quantitative adjustment and adaptive fusion ensure the stability of the melt temperature. Finally, multi-dimensional parameter linkage control and adaptive compensation technology are used to achieve precise control of the casting process, while multi-level progressive integration ensures the continuity and stability of process parameters. This integrated management method for the entire process significantly improves the automation level and process control accuracy of molten aluminum processing, effectively reduces product quality fluctuations, and provides reliable data support for process optimization through full data traceability.

[0129] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An integrated management system for molten aluminum processing, characterized in that, The integrated management system for molten aluminum processing includes: an encoding module, a configuration module, a quality inspection module, an embedding module, a decoupling module, and a control module. The output of the encoding module is connected to the input of the configuration module, the output of the configuration module is connected to the input of the quality inspection module, the output of the quality inspection module is connected to the input of the embedding module, the output of the embedding module is connected to the input of the decoupling module, and the output of the decoupling module is connected to the input of the control module. The encoding module is also connected to the quality inspection module to provide deep identification information to the quality inspection module, and the output of the control module is also connected to the encoding module to provide process feature chain data. The encoding module is used to obtain batch feature codes by performing multi-level cross-encoding processing on aluminum raw materials, perform multi-dimensional mapping analysis on the batch feature codes and material inherent property data to obtain a process feature matrix, and perform hierarchical decomposition of material properties based on the process feature matrix to obtain deep identification information. The configuration module is used to perform nonlinear dynamic configuration of melting parameters based on the depth identification information. It performs multi-dimensional quantization of the melting state by coupling analysis of temperature, flow, and pressure field data during the melting process to obtain a melting characteristic spectrum. Specifically, it is used for: The physical, chemical, and process characteristic parameters are extracted from the deep identification information, and the parameters are divided into numerical ranges to obtain melting control parameters. Data correlation is calculated for these melting control parameters, and parameters with correlation coefficients higher than a set threshold are grouped together to obtain parameter correlation groups. These parameter correlation groups are prioritized according to process importance, and cross-validation is performed on each group to obtain nonlinear dynamic configuration parameters. Temperature, flow velocity, and pressure change data are collected from the melting furnace, and time-series analysis is performed on the collected data to obtain a field data sequence. The temperature, flow, and pressure field data in the field data sequence are correlated, and the interaction strength of physical quantities is quantified to obtain coupling characteristic values. These coupling characteristic values ​​undergo multidimensional spatial transformation, and the transformed data is cross-mapped with the nonlinear dynamic configuration parameters. The mapping results are quantitatively evaluated to obtain a melting characteristic spectrum. The quality inspection module is used to perform multivariate cross-quality inspection on molten aluminum based on the melting characteristic spectrum, perform hierarchical progressive analysis on the test data, and perform multidimensional cross-mapping of depth identification information, melting characteristic spectrum and quality inspection data to obtain quality characterization spectrum. An embedding module is used to embed hierarchical information into the ladle recording unit according to the quality characterization spectrum. By dynamically tracking the temperature field, it establishes a four-dimensional correlation between spatial location data and the quality characterization spectrum to generate a transportation control matrix. Specifically, it is used to: classify the temperature data, composition data, and physical property data in the quality characterization spectrum according to their importance; encode and convert the classified data to obtain classified data groups; divide the storage capacity of the ladle recording unit into blocks; write the classified data groups into corresponding blocks according to data priority to obtain a recorded data sequence; collect temperature data by temperature sensing points arranged inside the ladle; perform time series analysis on the collected data to obtain a temperature change curve; combine the longitude, latitude, altitude, and time data of the ladle during transportation to obtain a spatial location sequence; correlate the temperature change curve with the spatial location sequence; cross-validate the correlation result with the recorded data sequence to obtain multi-dimensional correlated data; and perform numerical transformation on the multi-dimensional correlated data, reorganize the transformation results, and integrate the data according to the spatiotemporal correspondence to generate a transportation control matrix. The decoupling module is used to decouple multiple parameters of temperature change based on the transport control matrix, and to construct a casting feature field by adaptively fusing the position state and temperature field data through hierarchical quantitative adjustment of the heat preservation state and adaptive fusion of the position state and temperature field data. The control module is used to perform multi-dimensional parameter linkage control of the holding furnace through the casting feature field, to perform adaptive compensation of the casting process, and to integrate the process parameters in a multi-level progressive manner to form a process feature chain.

2. The integrated management system for molten aluminum processing according to claim 1, characterized in that, The encoding module is specifically used for: The composition, material specifications, and production batch number data are extracted from the aluminum raw materials and grouped according to a preset arrangement order to obtain the raw material basic data group. Each set of data in the raw material basic data group is numerically processed, converting the composition into element content percentage, the specification data into standard size values, and the batch number data into time series values ​​to obtain a numerical data sequence. The numerical data sequence is weighted according to the importance of its elements, and the weighted data is cross-combined to generate batch feature codes. By extracting data on the purity, density, thermal conductivity, and melting point temperature of the aluminum raw materials, a data set of inherent material properties is constructed. The batch feature code is matched with the data set of inherent material properties to establish a data mapping relationship and obtain a process feature matrix. Based on the process feature matrix, the material property data is processed in layers according to three dimensions: physical properties, chemical properties, and process properties. The correlation degree of each layer of data is calculated to obtain the attribute correlation data. Key parameters are extracted from the attribute association data, the degree of mutual influence between parameters is quantified, and the quantification results are fused with the process feature matrix to obtain deep identification information.

3. The integrated management system for molten aluminum processing according to claim 1, characterized in that, The quality inspection module is specifically used for: The melting characteristic spectrum is segmented according to element content, impurity content, and temperature distribution, and the segmented data is normalized to obtain quality inspection benchmark data. The molten aluminum is tested for density, hardness, thermal conductivity, and fluidity. The test results are compared and analyzed with the quality inspection benchmark data to obtain a set of quality inspection parameters. The relevant parameters in the quality inspection parameter group are combined and classified, and the classified data are analyzed progressively to obtain quality inspection feature values. Material attribute data is extracted from the depth identification information, and the material attribute data is correlated with the quality inspection feature values ​​to obtain a quality correlation matrix; The quality correlation matrix and the fused feature spectrum are fused together, and the quality feature group is obtained by cross-validation of the fused data. The quality feature groups are classified according to the importance of parameters, and the classified data are combined in multiple dimensions and numerically mapped according to the quality evaluation criteria to obtain the quality characterization spectrum.

4. The integrated management system for molten aluminum processing according to claim 1, characterized in that, The decoupling module is specifically used for: The temperature data, location data, and time data in the transportation control matrix are grouped, and correlation analysis is performed on the grouped data to obtain a parameter combination sequence. The temperature change data in the parameter combination sequence is decomposed into time dimension, and the decomposed data is classified according to the change trend to obtain temperature change feature values. The temperature change characteristic values ​​are correlated with the location change data, and the analysis results are sorted to obtain the heat preservation control parameters. By performing layered processing on the thermal insulation control parameters, the processing results are quantified according to the thermal insulation requirements to obtain thermal insulation adjustment data; The thermal insulation adjustment data and the position status data are numerically matched, and the matching results are combined according to the spatiotemporal correspondence to obtain a temperature fusion sequence; Key parameters are extracted from the temperature fusion sequence, spatial distribution analysis is performed on the extracted data, and the data is reconstructed according to process requirements to construct a casting feature field.

5. The integrated management system for molten aluminum processing according to claim 1, characterized in that, The control module includes: The correlation unit is used to classify and combine the temperature distribution data, spatial location data, and temporal change data in the casting characteristic field, perform correlation analysis on the combined data, and obtain furnace temperature control parameters. A division unit is used to subdivide the furnace temperature control parameters, and the subdivision results are divided into intervals according to the temperature fluctuation trend to obtain the heat preservation control sequence. The extraction unit is used to extract data from the key control points in the heat preservation control sequence, perform cross-validation on the extracted data, and obtain compensation control parameters. The comparison unit is used to compare the compensation control parameters with the casting process requirements, and to classify the comparison results according to the degree of deviation to obtain a compensation data set. The arrangement unit is used to arrange the process parameters in the compensation data group according to the process sequence, perform progressive analysis on the arranged data, and obtain the process parameter sequence. The combination unit is used to perform multi-level combination of relevant parameters in the process parameter sequence, and reconstruct the data of the combination result according to the process flow to form a process feature chain.

6. The integrated management system for molten aluminum processing according to claim 5, characterized in that, The partitioning unit is specifically used for: The furnace temperature control parameters are grouped according to temperature range, rate of change, and fluctuation amplitude. The grouped data are then numerically converted to obtain temperature characteristic data. The temperature feature data is segmented along the time dimension, and the segmentation results are sorted according to their numerical values ​​to obtain a time-series data set. The correlation parameters in the time series data group are combined and processed, and the correlation of the combined data is calculated to obtain the fluctuation trend value. The fluctuation trend value is defined into intervals according to the changing pattern, and the defined results are numerically normalized to obtain the interval parameters. The interval parameters are matched with the temperature control requirements, and the matching results are graded to obtain a control data set. The control data set is reconstructed according to the insulation requirements, and the reconstruction results are integrated to obtain the insulation control sequence.

7. The integrated management system for molten aluminum processing according to claim 6, characterized in that, The comparison unit is specifically used for: The temperature, time, and flowability values ​​in the compensation control parameters are extracted, and the extracted data are standardized to obtain the compensation benchmark value. The parameters in the casting process requirements are numerically processed, and the processing results are sorted according to priority to obtain the process standard values. The difference between the compensation benchmark value and the process standard value is calculated, and the calculation results are filtered to obtain the deviation data. The deviation data is classified according to its numerical value, and the classification results are defined by a threshold to obtain a deviation level table. The parameters in the deviation level table are reorganized according to the compensation requirements, and the reorganization results are optimized to obtain a compensation parameter set. The compensation parameter set is integrated, and the integration results are calibrated according to process requirements to obtain the compensation data set.

8. An integrated management method for molten aluminum processing, characterized in that, The integrated management method for molten aluminum processing includes: By performing multi-level cross-coding on aluminum raw materials, batch feature codes are obtained. The batch feature codes are then mapped and analyzed with the inherent material property data in a multi-dimensional manner to obtain a process feature matrix. Based on the process feature matrix, the material properties are deconstructed hierarchically to obtain deep identification information. Based on the depth identification information, nonlinear dynamic configuration of melting parameters is performed. Through coupled analysis of temperature, flow, and pressure field data during the melting process, multi-dimensional quantification of the melting state is conducted to obtain a melting characteristic spectrum. This includes: extracting physical, chemical, and process characteristic parameters from the depth identification information; dividing the parameters into numerical ranges to obtain melting control parameters; calculating the data correlation of the melting control parameters; grouping parameters with correlation coefficients higher than a set threshold to obtain parameter correlation groups; prioritizing the parameter correlation groups according to process importance; cross-validating each parameter correlation group to obtain nonlinear dynamic configuration parameters; collecting temperature, flow velocity, and pressure change data within the melting furnace; performing time-series analysis on the collected data to obtain a field data sequence; performing correlation calculations on the temperature, flow, and pressure field data in the field data sequence; quantifying the interaction strength of physical quantities to obtain coupling characteristic values; performing multi-dimensional spatial transformation on the coupling characteristic values; cross-mapping the transformed data with the nonlinear dynamic configuration parameters; and quantifying the mapping results to obtain a melting characteristic spectrum. Based on the melting characteristic spectrum, multivariate cross-quality inspection is performed on molten aluminum, and the detection data is analyzed in a hierarchical and progressive manner. The depth identification information, melting characteristic spectrum and quality inspection data are cross-mapped in multiple dimensions to obtain the quality characterization spectrum. Based on the quality characterization spectrum, hierarchical information is embedded into the ladle recording unit. By dynamically tracking the temperature field, spatial location data is correlated with the quality characterization spectrum in four dimensions to generate a transportation control matrix. Specifically, this involves: classifying the temperature, composition, and physical property data in the quality characterization spectrum according to their importance; encoding and converting the classified data to obtain classified data groups; dividing the storage capacity of the ladle recording unit into blocks; writing the classified data groups into corresponding blocks according to data priority to obtain a recorded data sequence; collecting temperature data through temperature sensors arranged inside the ladle; performing time series analysis on the collected data to obtain a temperature change curve; combining the longitude, latitude, altitude, and time data of the ladle during transportation to obtain a spatial location sequence; correlating the temperature change curve with the spatial location sequence; cross-validating the correlation results with the recorded data sequence to obtain multidimensional correlation data; and performing numerical transformation on the multidimensional correlation data, reorganizing the transformation results, and integrating the data according to spatiotemporal correspondence to generate a transportation control matrix. Based on the transport control matrix, multiple parameters are decoupled for temperature changes. By performing layered quantitative adjustment of the heat preservation state, the position state and temperature field data are adaptively fused to construct a casting feature field. Through the casting characteristic field, the holding furnace is controlled by multi-dimensional parameters, the casting process is adaptively compensated, and the process parameters are integrated in a multi-level progressive manner to form a process characteristic chain.

Images