Method for regulating and controlling type of iron-containing phase in secondary aluminum, medium and computer equipment

Abstract

The invention relates to the technical field of non-ferrous metal recycling, and discloses a method for regulating and controlling the type of an iron-containing phase in secondary aluminum, a medium and computer equipment. According to the regulation and control method, non-equilibrium solidification simulation (physical knowledge) and machine learning (data driving) are fused, and specifically, under the condition of limited experimental data, an associated data set of component-phase fraction-iron-containing phase types is constructed, a machine learning model is introduced based on the associated data set, iron-containing phase types corresponding to different alloy components are predicted, and the alloy components are subjected to non-equilibrium solidification simulation (physical knowledge) and machine learning (data driving). According to the method, the iron-containing phase type of any component point in the high-dimensional component space can be rapidly and accurately explored and predicted, the number of repeated experiments and trial and error times needed by a traditional method is greatly reduced, and the method is suitable for the component regulation and control of alloy design. And the research and development period is remarkably shortened, the research and development cost is reduced, and a new technical path is provided for high-value utilization of a complex secondary aluminum system.

Description

Technical Field

[0001] This invention relates to the field of non-ferrous metal recycling technology, and in particular to a method, medium, and computer equipment for controlling the type of iron phase in recycled aluminum. Background Technology

[0002] Aluminum and its alloys are highly favored for lightweight applications in aerospace, automotive, and other fields due to their excellent specific strength. However, during the casting and recycling of aluminum alloys, impurity elements such as iron (Fe) are often unavoidably introduced. Since Fe has extremely low solid solubility in the aluminum matrix (approximately 0.05 wt.%), excess Fe readily reacts with elements such as Al, Si, and Mg to form brittle iron-containing intermetallic compounds (iron-containing phases). Needle-like... The Fe phase severely disrupts the matrix, leading to a significant decrease in the mechanical properties of the alloy, especially ductility and fatigue resistance, which becomes a major obstacle to the high-value utilization of recycled aluminum.

[0003] Removing Fe directly from molten aluminum using existing technologies is technically challenging and economically impractical. Therefore, industry and academia generally employ phase morphology control strategies, which involve adding alloying elements (such as manganese and Mn) to promote the formation of harmful needle-like structures. Fe phase transformation is less harmful ( The Fe phase. However, existing methods for controlling the amount of Mn added mostly rely on empirical trial and error or limited experimental data, which are difficult to cope with the challenges of the complex and variable composition (high-dimensional composition space) of recycled aluminum raw materials. For example, excessive addition of Mn not only increases the total volume fraction of the iron-containing phase, but may also form coarse blocky slag phases, which in turn deteriorates the alloy properties.

[0004] Traditional alloy design methods typically rely on large-scale experimental sampling in high-dimensional compositional spaces to establish a mapping relationship between composition, microstructure, and properties. However, this purely experimental approach faces significant challenges in recycled aluminum systems: on the one hand, large compositional fluctuations and complex interactions between impurity elements lead to high experimental costs and lengthy cycles; on the other hand, under conditions of scarce data and small sample sizes, directly establishing a nonlinear mapping relationship between composition and iron-containing phase types is prone to overfitting, resulting in poor model generalization ability. How to uncover the microstructural features hidden behind the composition (such as the volume fraction of the iron-containing phase) under limited experimental data conditions, and use them as physical constraints to guide machine learning models, is a key scientific problem for achieving precise control of the iron-containing phase in recycled aluminum.

[0005] Therefore, it is necessary to provide a method for controlling the type of iron-containing phase in recycled aluminum with high prediction accuracy and strong generalization ability to solve the above-mentioned technical problems. Summary of the Invention

[0006] The main objective of this invention is to provide a method for controlling the type of iron-containing phase in recycled aluminum. This method integrates non-equilibrium solidification simulation (physical knowledge) and machine learning (data-driven) to quickly and accurately explore and predict the type of iron-containing phase at any composition point in a high-dimensional composition space. This greatly reduces the number of repeated experiments and trial and error required by traditional methods, significantly shortens the R&D cycle, and reduces R&D costs. It provides a new technical path for the high-value utilization of complex recycled aluminum systems.

[0007] To achieve the above objectives, the present invention provides a method for controlling the type of iron-containing phase in recycled aluminum, comprising the following steps: Step S1: Obtain different components Data on the iron-containing phase types of the cast alloy; and the phase fraction data of the iron-containing phases in the alloy were obtained through non-equilibrium solidification simulation; the iron-containing phase types include acicular phases. Fe phase and Chinese character form Fe phase; Step S2: Integrate the iron-containing phase type data and phase fraction data obtained in Step S1 to construct an associated dataset containing alloy composition, phase fraction, and iron-containing phase type; Step S3: Based on the associated dataset obtained in step S2, construct and train a machine learning model that takes the phase fraction as input and the iron-containing phase type as output. Step S4: Based on the machine learning model obtained in step S3, predict the types of iron-containing phases corresponding to different alloy compositions, and determine the composition control criteria used to distinguish different types of iron-containing phases accordingly. Step S5: Perform composition control for alloy design based on the composition control criteria obtained in step S4.

[0008] Preferably, the composition control criterion in step S4 is: when the Mn element content in the alloy... Needle-like structures are achieved when the following relationship is satisfied. Fe facing each other Chinese character shape The complete transformation of the Fe phase is as follows: ; in: This refers to the Fe element content; Fe element content The square of; This represents the Si element content.

[0009] Preferably, the different components in step S1 In the casting alloy: Si content is 6.0-12.0 wt.%; Mg content is 0-0.5 wt.%; Fe content is 0-2.0 wt.%; Mn content is 0-2.5 wt.%.

[0010] Preferred, different ingredients Casting alloys include at least one of the following alloys: , , , , as well as At least one of them.

[0011] Preferably, the machine learning model in step S3 adopts a logistic regression model.

[0012] The present invention also discloses a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method for controlling the type of iron-containing phase in recycled aluminum as described above.

[0013] The present invention also provides a computer device, comprising: a processor and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method for controlling the type of iron-containing phase in recycled aluminum as described above.

[0014] The effects of applying the technical solution of this invention are as follows: The core innovation of this invention's method for controlling the iron-containing phase type in recycled aluminum lies in using microstructural features (phase fraction) as input variables for a machine learning model. This constructs a correlation dataset of composition, phase fraction, and iron-containing phase type, enabling accurate prediction of the iron-containing phase type in a high-dimensional composition space even with only 63 sets of small sample data. This method fully utilizes the phase fraction generated by non-equilibrium solidification simulation as a physical constraint feature, significantly reducing the model's dependence on data volume and improving prediction accuracy and robustness under small sample conditions.

[0015] This control method integrates nonequilibrium solidification simulation (physical knowledge) and machine learning (data-driven). Specifically, under limited experimental data conditions, a correlation dataset of composition, phase fraction, and ferrous phase type is constructed. The key to this method's high-precision prediction under small sample conditions lies in its use of microstructure features (phase fraction) as intermediate variables. Compared to directly using alloy composition as input, phase fraction features have stronger physical meaning and lower dimensionality, effectively capturing the nonlinear influence of composition changes on phase transformation behavior. Through physical knowledge-guided feature engineering, the high-dimensional composition space is mapped to a low-dimensional, physically meaningful phase fraction space, thus providing high-quality input features for subsequent machine learning models. Based on the correlation dataset, a machine learning model is introduced to predict the ferrous phase type corresponding to different alloy compositions, obtaining composition control criteria to distinguish different ferrous phase types. Finally, composition control is performed on alloy design based on the obtained composition control criteria. This method can quickly and accurately explore and predict the iron-containing phase type at any composition point in a high-dimensional composition space, greatly reducing the number of repeated experiments and trial and error required by traditional methods, significantly shortening the research and development cycle, reducing research and development costs, and providing a brand-new technical path for the high-value utilization of complex recycled aluminum systems. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the overall process of the method for controlling the iron-containing phase type of recycled aluminum in an embodiment of the present invention; Figure 2 This is an alloy composition distribution map used to construct the dataset in an embodiment of the present invention; Figure 3 As described in the embodiments of the present invention Fe and Distribution diagram of the relationship between Fe phase fraction ratio and iron-containing phase type; Figure 4 Here is the confusion matrix of the machine learning model in this embodiment of the invention, where: (a) is the confusion matrix on the training set; and (b) is the confusion matrix on the test set. Figure 5 This is a comparison chart of the machine learning model prediction results and the actual results of the dataset in an embodiment of the present invention; Figure 6 This is a distribution diagram of the β-Fe phase fraction as a function of composition obtained through simulation in an embodiment of the present invention; Figure 7 This is a distribution diagram of the α-Fe phase fraction as a function of composition obtained through simulation in an embodiment of the present invention; Figure 8 The following are the prediction results for a fixed Si content in the embodiments of the present invention: (a) is a prediction result diagram of the iron-containing phase formation type as a function of Mg, Fe and Mn; (b) is a prediction result diagram of the iron-containing phase formation type as a function of Mn under different Fe contents. Figure 9 The following are prediction results for a fixed Mg content in the embodiments of the present invention: (a) is a prediction result for the type of iron-containing phase formation as a function of Si, Fe and Mn; (b) is a prediction result for the type of iron-containing phase formation as a function of Mn under different Fe contents. Figure 10 These are scanning electron micrographs of the alloy in the embodiments of the present invention, wherein: (a) is an alloy (a) Scanning electron micrograph; (b) is an alloy (c) is a scanning electron micrograph of the alloy. Scanning electron micrographs; (d) is an alloy Scanning electron micrographs of the alloy; (e) shows the alloy. Scanning electron micrographs; (f) shows the alloy. Scanning electron micrographs.

[0018] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0020] Example: A method for controlling the type of iron phase in recycled aluminum, the process of which is as follows: Figure 1 As shown, it includes the following steps: Step S1, data collection and non-equilibrium solidification simulation, specifically includes: obtaining data for different compositions. Data on the iron-containing phase types of the cast alloy; and the phase fraction data of the iron-containing phases in the alloy were obtained through non-equilibrium solidification simulation; the iron-containing phase types include acicular phases. Fe phase and Chinese character form Fe phase.

[0021] Data on the iron-containing phase types of Al-Si-Mg-Fe-Mn casting alloys with different compositions were collected from existing literature. Among these, 27 sets of data contained both acicular β-Fe phase and Chinese character-shaped α-Fe phase, while 36 sets contained only the Chinese character-shaped α-Fe phase. The collected data showed that different compositions... In the cast alloy: Si content is 6.0-12.0 wt.%; Mg content is 0-0.5 wt.%; Fe content is 0-2.0 wt.%; Mn content is 0-2.5 wt.%. For example... Figure 2 As shown.

[0022] The Scheil-Gulliver nonequilibrium solidification model was used to simulate the solidification process of the above 63 alloy combinations, and the volume fractions of β-Fe and α-Fe phases in each alloy combination were calculated. The simulated volume fractions of β-Fe phase ranged from 0.03% to 2.53%, and the volume fractions of α-Fe phase ranged from 0.27% to 8.36%.

[0023] Step S2: Integrate the iron-containing phase type data and phase fraction data obtained in Step S1 to construct a related dataset containing alloy composition, phase fraction, and iron-containing phase type, as follows: The iron-containing phase type data and phase fraction data obtained in step S1 are integrated to construct a related dataset containing alloy composition, phase fraction and iron-containing phase type as shown in Table 1.

[0024] Table 1. Association Data Set of Alloy Composition-Phase Fraction-Iron-Containing Phase Type

[0025] Based on the data in Table 1, the volume fraction ratio of α-Fe to β-Fe phases (f(α-Fe) / f(β-Fe)) for each alloy combination was calculated, and a correlation analysis was performed with the measured results of the iron-containing phase type. The results are as follows: Figure 3 As shown. From Figure 3 It is clear that when the ratio of f(α-Fe) / f(β-Fe) is less than 4.81, both acicular and Chinese character-shaped iron-containing phases usually coexist in the alloy; while when the ratio is greater than or equal to 4.81, only the Chinese character-shaped iron-containing phase is observed in the alloy. This correlation indicates that the phase fractions of α-Fe and β-Fe are the key factors determining the type of iron-containing phase. Therefore, the phase fraction was subsequently selected as the input feature for the machine learning model.

[0026] Step S3: Based on the associated dataset obtained in step S2, construct and train a machine learning model that takes the phase fraction as input and the iron-containing phase type as output, as follows: Using the 63 sets of β-Fe and α-Fe phase fraction data obtained in step S1 as input features and the corresponding 63 sets of iron-containing phase type data as output labels, a machine learning model is constructed. This embodiment uses the logistic regression algorithm, mainly considering its good interpretability and robustness under small sample conditions. Since the sample size is only 63 sets, using complex models (such as support vector machines or random forests) is prone to overfitting or insufficient generalization ability. Logistic regression can effectively avoid the risk of overfitting by constructing a linear decision boundary. The model is implemented based on the Python Scikit-learn library. The 63 sets of samples are randomly divided into a training set (50 sets) and a test set (13 sets) in a ratio of 50:13. The model parameters are set as follows: L2 regularization parameter C=10, maximum number of iterations is 10000, the solver is 'lbfgs', and the random state is set to 29.

[0027] The model performance was evaluated using the confusion matrix, and the results are as follows: Figure 4 As shown. On the training set, the model achieved a prediction accuracy of 96.8% for acicular phases only and 90.9% for mixed phases; on the test set, the corresponding accuracies were 87.5% and 100%, respectively. These results indicate that the trained logistic regression model can stably predict the iron-bearing phase type based on the input phase fraction.

[0028] To further verify the advantages of logistic regression in tasks with few samples, this embodiment also compares it with two common classifiers, support vector machines and random forests. The results show that on the same training and test sets, logistic regression achieves the highest F1 scores (0.94 and 0.93 respectively), and is insensitive to sample distribution, demonstrating the best robustness and generalization ability. This further verifies that, based on feature engineering (phase score) guided by physical knowledge, simple linear models can achieve excellent performance under few sample conditions.

[0029] Subsequently, the model was used to back-judge the iron-bearing phase types of all 63 sets of data, and the predicted results were compared with the actual iron-bearing phase types in the dataset. The results were then plotted on a phase diagram with Fe and Mn contents as coordinates, as shown below. Figure 5 As shown in the figure. The results show that the model's prediction boundary is in high agreement with the distribution of data points, proving the effectiveness of the model.

[0030] Step S4: Based on the machine learning model obtained in step S3, predict the types of iron-containing phases corresponding to different alloy compositions, and determine the composition control criteria used to distinguish different types of iron-containing phases, as follows: Large-scale simulations were performed using the Scheil-Gulliver nonequilibrium solidification model within a broader, high-dimensional compositional space. The alloy composition ranges were set as follows: Si 6.0-12.0 wt.% (step 0.25 wt.%), Mg 0-0.5 wt.% (step 0.025 wt.%), Fe 0-2.0 wt.% (step 0.1 wt.%), and Mn 0-2.0 wt.% (step 0.1 wt.%). A total of 231,525 virtual alloy compositions and their corresponding β-Fe and α-Fe phase fractions were obtained, as shown below. Figure 6 , Figure 7 As shown.

[0031] By inputting the phase fraction data obtained from the above simulation into the machine learning model trained in step S3, the predicted iron-containing phase types corresponding to these more than 230,000 sets of virtual alloy compositions can be quickly obtained. By analyzing these prediction results, the influence of each element on the phase transformation can be systematically studied.

[0032] First, with a fixed Si content of 7 wt.%, the relationship between the amount of Mn required to achieve the β-Fe→α-Fe phase transformation and the Fe and Mg contents was studied under different Mg contents. The results are as follows: Figure 8 As shown in the figure. The results indicate that the required amount of Mn increases with increasing Fe content, but changes in Mg content have virtually no effect on the required amount of Mn.

[0033] Secondly, with the Mg content fixed at 0.4 wt.%, the results were studied under different Si contents, as follows: Figure 9 As shown in the figure. The results indicate that the required amount of Mn added increases with increasing Fe content, but the amount of Mn added required to achieve complete transformation decreases significantly with increasing Si content.

[0034] based on Figure 9 The prediction results shown are used to mathematically fit the transformation boundary lines under different Si contents, and finally construct a composition control criterion for distinguishing different iron-containing phase types, as follows: When the Mn content in the alloy Needle-like structures are achieved when the following relationship is satisfied. Fe facing each other Chinese character shape The complete transformation of the Fe phase, that is, the minimum Mn content required to achieve the complete transformation of β-Fe to α-Fe, is expressed as a function of Fe and Si contents, as follows: ; in: This refers to the Fe element content; Fe element content The square of; This represents the Si element content.

[0035] This mathematical expression has a clear physical meaning and engineering operability. It is the most direct output of the method of this invention to achieve high-precision prediction under small sample conditions, and provides quantitative guidance for the composition design of recycled aluminum.

[0036] Step S5: Perform composition control for alloy design based on the composition control criteria obtained in step S4.

[0037] The experimental verification process is as follows: To verify the accuracy of the above criteria, this embodiment designed and prepared six key alloys, the compositions of which are shown in Table 2. The alloys used industrial pure Al, Al-20Si, pure Mg, Al-10Fe, and Al-10Mn master alloys as raw materials, and were melted at 740℃ in a ZG-25 type vacuum induction medium-frequency melting furnace. After electromagnetic stirring and settling, they were cast into cylindrical ingots with a diameter of 20mm. After grinding and polishing, the types of iron-containing phases in each alloy were observed using a FEG Quanta 250 field emission scanning electron microscope. The results are as follows: Figure 10 As shown.

[0038] Based on the composition control criteria in step S4, the iron-containing phase types of the six alloys were predicted and compared with the experimental results, which are summarized in Table 2.

[0039] Table 2 Experimental verification results

[0040] As can be seen from Table 2, the predicted results for all six alloys are completely consistent with the experimental results, which fully demonstrates that the composition control criterion proposed in this invention has extremely high accuracy and reliability.

[0041] In summary, this invention systematically solves the challenge of precisely controlling the type of iron-containing phase in recycled aluminum within a high-dimensional composition space by integrating non-equilibrium solidification simulation and machine learning. The established quantitative control criterion for composition provides direct and efficient theoretical guidance for the grade preservation and composition design of recycled aluminum, demonstrating significant industrial application value and economic benefits.

[0042] This embodiment also provides a computer device, including: a processor and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method for controlling the type of iron-containing phase in recycled aluminum as described above.

[0043] In addition, this embodiment also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method for controlling the type of iron-containing phase in recycled aluminum as described above.

[0044] It should be noted that the division of the various modules in the above system is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. These modules can be implemented entirely in software via processing element calls; they can be fully implemented in hardware; or some modules can be implemented by processing element calls to software, while others are implemented in hardware. Each module can be a separate processing element, or it can be integrated into a chip in the aforementioned device. Alternatively, it can be stored as program code in the memory of the aforementioned device, and its functions can be called and executed by a processing element of the device. Furthermore, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each of the above modules can be completed through the integrated logic circuits in the hardware of the processor element or through software instructions.

[0045] It should be understood that the above-described device embodiments are merely illustrative, and the device of this application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units, modules, or components may be combined, or integrated into another system, or some features may be ignored or not executed.

[0046] Furthermore, unless otherwise specified, the functional units / modules in the various embodiments of this application can be integrated into one unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated together. The integrated units / modules described above can be implemented in hardware or as software program modules.

[0047] When an integrated unit / module is implemented in hardware, the hardware can be digital circuits, analog circuits, etc. The physical implementation of the hardware structure includes, but is not limited to, transistors, memristors, etc. Unless otherwise specified, the processor can be any suitable hardware processor, such as a Central Processing Unit (CPU), Graphics Processing Unit (GPU), Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor, or other electronic component. Unless otherwise specified, memory can be implemented from any type of volatile or non-volatile storage device or a combination thereof, such as a USB flash drive, Random-Access Memory (RAM), Static Random-Access Memory (SRAM), Dynamic Random-Access Memory (DRAM), Enhanced Dynamic Random-Access Memory (EDRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), Resistive Random Access Memory (RRAM), High-Bandwidth Memory (HBM), and Hybrid Memory Cube (HMC). Cube, magnetic storage, flash memory, disk, optical disk, portable hard drive or magnetic disk, and other media that can store program code.

[0048] If the integrated unit / module is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application.

[0049] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for controlling the type of iron-containing phase in recycled aluminum, characterized in that, Includes the following steps: Step S1: Obtain different components Data on the iron-containing phase types of the cast alloy; and the phase fraction data of the iron-containing phases in the alloy were obtained through non-equilibrium solidification simulation; the iron-containing phase types include acicular phases. Fe phase and Chinese character form Fe phase; Step S2: Integrate the iron-containing phase type data and phase fraction data obtained in Step S1 to construct an associated dataset containing alloy composition, phase fraction, and iron-containing phase type; Step S3: Based on the associated dataset obtained in step S2, construct and train a machine learning model that takes the phase fraction as input and the iron-containing phase type as output. Step S4: Based on the machine learning model obtained in step S3, predict the types of iron-containing phases corresponding to different alloy compositions, and determine the composition control criteria used to distinguish different types of iron-containing phases accordingly. Step S5: Perform composition control for alloy design based on the composition control criteria obtained in step S4.

2. The method for controlling the type of iron-containing phase in recycled aluminum as described in claim 1, characterized in that, The composition control criterion in step S4 is: when the Mn element content in the alloy... Needle-like structures are achieved when the following relationship is satisfied. Fe facing each other Chinese character shape The complete transformation of the Fe phase is as follows: ； in: This refers to the Fe element content; Fe element content The square of; This represents the Si element content.

3. The method for controlling the type of iron-containing phase in recycled aluminum as described in claim 1, characterized in that, Different components in step S1 In the casting alloy: Si content is 6.0-12.0 wt.%; Mg content is 0-0.5 wt.%; Fe content is 0-2.0 wt.%; Mn content is 0-2.5 wt.%.

4. The method for controlling the type of iron-containing phase in recycled aluminum as described in claim 3, characterized in that, Different ingredients Casting alloys include at least one of the following alloys: , , , , as well as At least one of them.

5. The method for controlling the type of iron-containing phase in recycled aluminum as described in claim 1, characterized in that, In step S3, the machine learning model uses a logistic regression model.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method for controlling the type of iron-containing phase in recycled aluminum as described in any one of claims 1 to 5.

7. A computer device, characterized in that, include: A processor and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method for controlling the type of iron-containing phase in recycled aluminum as described in any one of claims 1 to 5.

Images