Abnormal process parameter identification method and device of aluminum alloy material, electronic equipment and computer program product

Abstract

The invention discloses a method and a device for identifying abnormal process parameters of an aluminum alloy material, electronic equipment and a computer program product. The method comprises the following steps: obtaining to-be-analyzed material performance of an aluminum alloy material; the performance of the to-be-analyzed material is analyzed by using an abnormal process parameter identification model, abnormal process parameters causing the performance of the to-be-analyzed material not to conform to the material performance indexes are determined, and the abnormal process parameter identification model is trained by using a process parameter-material performance database in advance through a nearest neighbor classification machine learning algorithm; the training data for training the abnormal process parameter identification model is a plurality of training samples determined based on the process parameter-material performance database, the training samples comprise positive samples and negative samples, the preset material performance in the positive samples accords with the material performance indexes, and the preset material performance in the negative samples does not accord with the material performance indexes. The technical problem that unreasonable process parameters cannot be rapidly determined is solved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to a method, device, electronic equipment and computer program product for identifying abnormal process parameters of aluminum alloy materials. Background Art

[0002] 5xxx aluminum alloy automotive sheet materials refer to thin sheet materials made of alloys such as 5182, 5754 and 5052, and 6xxx aluminum alloy automotive sheet materials refer to thin sheet materials made of alloys such as 6016, 6014 and 6111, with a thickness of generally 0.5-3mm. One of its main uses is the manufacture of automotive inner panel parts. Its production process includes smelting, casting, homogenization, hot rolling, cold rolling, correction and annealing. Each production process has many process parameters, which will affect the performance of 5xxx aluminum alloy automotive sheets and 6xxx aluminum alloy automotive sheets.

[0003] Since the automotive industry has high requirements for stable product quality, it has placed high requirements on the performance stability of 5xxx aluminum alloy automotive sheets and 6xxx aluminum alloy automotive sheets. However, since the production process of 5xxx aluminum alloy automotive sheets and 6xxx aluminum alloy automotive sheets is long and has many process parameters, once there is an unreasonable change in the process parameters, it will cause abnormal material performance and scrap. At the same time, it also requires a lot of manpower and time to analyze massive production process data to find the causes and solutions to the problems.

[0004] With respect to the above-mentioned problem of being unable to quickly determine unreasonable process parameters, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present invention provide a method, device, electronic equipment and computer program product for identifying abnormal process parameters of aluminum alloy materials, so as to at least solve the technical problem that unreasonable process parameters cannot be quickly determined.

[0006] According to one aspect of an embodiment of the present invention, a method for identifying abnormal process parameters of an aluminum alloy material is provided, comprising: obtaining a material property to be analyzed of the aluminum alloy material, wherein the material property to be analyzed does not meet a preset material property index; analyzing the material property to be analyzed using an abnormal process parameter identification model, and determining abnormal process parameters that cause the material property to be analyzed to not meet the material property index, wherein the abnormal process parameter identification model is pre-trained using a process parameter-material property database through a nearest neighbor classification machine learning algorithm, the process parameter-material property database uses two-dimensional matrix data to represent a correspondence between preset process parameters and preset material properties, the training data for training the abnormal process parameter identification model is a plurality of training samples determined based on the process parameter-material property database, each of the training samples is used to represent a correspondence between the preset process parameters and the preset material property, and the training samples include: a positive sample and a negative sample, the preset material property in the positive sample meets the material property index, and the preset material property in the negative sample does not meet the material property index.

[0007] Optionally, the abnormal process parameter identification model includes: a first abnormal process parameter identification model for a first aluminum alloy material, using the abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determining the abnormal process parameters that cause the performance of the material to be analyzed not to meet the material performance indicators. The abnormal process parameters include: when the aluminum alloy material is the first aluminum alloy material, using the first abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determining the abnormal process parameters that cause the performance of the material to be analyzed not to meet the material performance indicators, wherein the preset process parameters used for training the first abnormal process parameter identification model include at least: intermediate annealing insulation temperature, intermediate annealing insulation time, continuous annealing insulation temperature and continuous annealing insulation time.

[0008] Optionally, the abnormal process parameter identification model includes: a second abnormal process parameter identification model for a second aluminum alloy material, using the abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determining the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators, including: when the aluminum alloy material is the second aluminum alloy material, using the second abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determining the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators, wherein the preset process parameters used for training the second abnormal process parameter identification model include at least: the ratio of the percentage of magnesium element content to the percentage of silicon element content, solution holding temperature, solution holding time, quenching cooling water temperature, pre-aging holding temperature, pre-aging holding time, aging holding temperature and aging holding time.

[0009] Optionally, the method also includes: obtaining at least one of the preset process parameters used to produce the aluminum alloy material, and at least one of the preset material properties, wherein the preset material properties include at least: tensile strength, yield strength and elongation; in the process parameter-material property database, using the two-dimensional matrix data to record the preset process parameters and the preset material properties, wherein each row of the two-dimensional matrix data is used to represent one of the preset process parameters for producing the aluminum alloy material, and at least one of the preset material properties corresponding to the preset process parameter, and each column of the two-dimensional matrix data represents one of the preset process parameters or one of the preset material properties.

[0010] Optionally, the method also includes: in the process parameter-material property database, determining the preset material property that meets the material performance index and the preset process parameters corresponding to the preset material property as the positive samples for training the abnormal process parameter identification model; based on the process parameter-material property database, determining the preset material property that does not meet the material performance index and the preset process parameters corresponding to the preset material property as the negative samples for training the abnormal process parameter identification model; using the positive samples and the negative samples to train the abnormal process parameter identification model through a nearest neighbor classification machine learning algorithm, wherein the abnormal process parameter identification model uses a grid search method to adjust the hyperparameters of the abnormal process parameter identification model.

[0011] Optionally, obtaining the material properties to be analyzed of the aluminum alloy material includes: when producing the aluminum alloy material, obtaining predicted material properties of the aluminum alloy material; detecting whether the predicted material properties meet the material performance indicators required by preset indicators; and when the predicted material properties do not meet the material performance indicators, determining the predicted material properties as the material properties to be analyzed.

[0012] Optionally, in the case of producing the aluminum alloy material, obtaining the predicted material properties of the aluminum alloy material includes: obtaining real-time process parameters for producing the aluminum alloy material; using a material abnormality process parameter identification model to analyze the real-time process parameters to determine the predicted material properties of the aluminum alloy material, wherein the material abnormality process parameter identification model is pre-trained using a process parameter-material property database through a random forest regression machine learning algorithm, and each group of training data for training the material abnormality process parameter identification model is the preset process parameter and the preset material property corresponding to the preset process parameter.

[0013] According to another aspect of an embodiment of the present invention, there is also provided an abnormal process parameter identification device for an aluminum alloy material, comprising: an acquisition module, used to acquire a material property to be analyzed of an aluminum alloy material, wherein the material property to be analyzed does not meet a preset material property index; a determination module, used to analyze the material property to be analyzed using an abnormal process parameter identification model, and determine the abnormal process parameters that cause the material property to be analyzed to not meet the material property index, wherein the abnormal process parameter identification model is pre-trained using a process parameter-material property database through a nearest neighbor classification machine learning algorithm, the process parameter-material property database uses two-dimensional matrix data to represent the correspondence between preset process parameters and preset material properties, the training data for training the abnormal process parameter identification model is a plurality of training samples determined based on the process parameter-material property database, each of the training samples is used to represent the correspondence between the preset process parameters and the preset material property, and the training samples include: positive samples and negative samples, the preset material property in the positive samples meets the material property index, and the preset material property in the negative samples does not meet the material performance index.

[0014] According to another aspect of an embodiment of the present invention, there is also provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the above-mentioned abnormal process parameter identification method for aluminum alloy materials through the computer program.

[0015] According to another aspect of an embodiment of the present invention, there is also provided a computer program product, comprising computer instructions, which, when executed by a processor, implement the steps of the above-mentioned abnormal process parameter identification method for aluminum alloy materials.

[0016] In an embodiment of the present invention, the abnormal process parameter identification model is used to analyze the material properties of the aluminum alloy material to be analyzed. According to the influence of the preset process parameters on the material properties of the aluminum alloy material, the reason why the material properties to be analyzed do not meet the preset material performance indicators is quickly determined, thereby achieving the purpose of quickly determining the abnormal process parameters that cause the material properties to be analyzed to not meet the material performance indicators, realizing the technical effect of quickly determining the abnormal process parameters, and thus solving the technical problem of being unable to quickly determine unreasonable process parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0018] Figure 1is a flow chart of a method for identifying abnormal process parameters of an aluminum alloy material according to an embodiment of the present invention;

[0019] Figure 2 is a schematic diagram of a big data model for identifying abnormal process parameters of aluminum alloy automobile sheet materials according to an embodiment of the present invention;

[0020] Figure 3 is a schematic diagram of an abnormal process parameter identification device for aluminum alloy material according to an embodiment of the present invention;

[0021] Figure 4 It is a structural block diagram of a computer terminal according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0024] According to an embodiment of the present invention, an embodiment of a performance method for an aluminum alloy material is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0025] Figure 1 is a flow chart of a method for identifying abnormal process parameters of an aluminum alloy material according to an embodiment of the present invention. Figure 1 As shown, the method comprises the following steps:

[0026] Step S102, obtaining the material properties to be analyzed of the aluminum alloy material, wherein the material properties to be analyzed do not meet the preset material property index;

[0027] Step S104, using an abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determine the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators, wherein the abnormal process parameter identification model is pre-trained using a process parameter-material performance database through a nearest neighbor classification machine learning algorithm, the process parameter-material performance database uses two-dimensional matrix data to represent the correspondence between preset process parameters and preset material performance, and the training data for training the abnormal process parameter identification model is a plurality of training samples determined based on the process parameter-material performance database, each training sample is used to represent the correspondence between the preset process parameters and the preset material performance, and the training samples include: positive samples and negative samples, the preset material performance in the positive samples meets the material performance indicators, and the preset material performance in the negative samples does not meet the material performance indicators.

[0028] In an embodiment of the present invention, the abnormal process parameter identification model is used to analyze the material properties of the aluminum alloy material to be analyzed. According to the influence of the preset process parameters on the material properties of the aluminum alloy material, the reason why the material properties to be analyzed do not meet the preset material performance indicators is quickly determined, thereby achieving the purpose of quickly determining the abnormal process parameters that cause the material properties to be analyzed to not meet the material performance indicators, realizing the technical effect of quickly determining the abnormal process parameters, and thus solving the technical problem of being unable to quickly determine unreasonable process parameters.

[0029] In the above-mentioned embodiments of the present application, when the predicted material performance based on the real-time process parameters does not meet the material performance indicators, the abnormal process parameter identification model can be used to analyze the predicted material performance that does not meet the material performance indicators, and further determine the abnormal process parameters that cause the material performance indicators to not meet the requirements. Then, based on the abnormal process parameters identified by the abnormal process parameter identification model, the abnormal process parameters can be processed in the subsequent production of aluminum alloy materials to ensure that the aluminum alloy materials finally produced are qualified.

[0030] In the above step S102, the aluminum alloy material may be a 5xxx aluminum alloy automobile sheet material, such as thin sheet materials of alloy grades such as 5182, 5754 and 5052.

[0031] In the above step S102, the aluminum alloy material may also be a 6xxx aluminum alloy automobile sheet material, such as thin sheet materials of alloy grades such as 6016, 6014 and 6111.

[0032] In the above step S104, the preset process parameters represent the parameters of at least one process required for producing the aluminum alloy material; the real-time process parameters are any one of the preset process parameters.

[0033] In the above step S104, the preset material properties include: tensile strength (Mpa), yield strength (Mpa), and elongation (unit %).

[0034] Optionally, the material property to be analyzed is a preset material property that does not meet a preset material property index.

[0035] Optionally, in the case of producing aluminum alloy materials, corresponding process parameters may be selected in sequence according to the process flow of producing aluminum alloy materials, and the currently executed process parameters are real-time process parameters.

[0036] In the above step S104, different aluminum alloy materials are processed using different preset process parameters, and the material abnormal process parameter identification model is trained using the preset process parameters. Therefore, different aluminum alloy materials can be trained using different preset process parameters for different aluminum alloy materials.

[0037] For example, when the aluminum alloy material is a first aluminum alloy material, the preset process parameters of the first aluminum alloy material are used to train a first material abnormal process parameter identification model; when the aluminum alloy material is a second aluminum alloy material, the preset process parameters of the second aluminum alloy material are used to train a second material abnormal process parameter identification model.

[0038] In the above step S104, the nearest neighbor classification machine learning algorithm, namely the KNN classification machine learning algorithm, is a non-parametric supervised learning classification algorithm, which is mainly used for classification and regression tasks.

[0039] As an optional embodiment, after determining the abnormal process parameters, the abnormal process parameters may be processed, wherein the processing of the abnormal process parameters includes: generating early warning information for the abnormal process parameters, indicating that the subsequent process flow for producing aluminum alloy materials can compensate for the impact caused by the abnormal process parameters, or indicating that the abnormal process parameters should be improved when the aluminum alloy materials are generated next time; or when the impact caused by the abnormal process parameters cannot be compensated by the subsequent process flow, indicating that the aluminum alloy materials currently being produced should be scrapped / destroyed.

[0040] As an optional embodiment, before using the material property prediction model to analyze the real-time process parameters to determine the predicted material properties of the aluminum alloy material, the method also includes: identifying the material type of the aluminum alloy material, wherein the material type includes at least: a first aluminum alloy material and a second aluminum alloy material; selecting an abnormal process parameter identification model according to the material type, wherein the abnormal process parameter identification model includes at least: a first abnormal process parameter identification model for the first aluminum alloy material, and a second abnormal process parameter identification model for the second aluminum alloy material.

[0041] In the above-mentioned embodiments of the present application, a corresponding abnormal process parameter identification model is pre-set for each aluminum alloy material. Furthermore, when identifying abnormal process parameters of the aluminum alloy material, the material type of the aluminum alloy material can be first identified, and then, based on the identified material type, the corresponding abnormal process parameter identification model can be selected to identify the abnormal process parameters, thereby ensuring that the identified abnormal process parameters are sufficiently accurate.

[0042] As an optional embodiment, the abnormal process parameter identification model includes: a first abnormal process parameter identification model for a first aluminum alloy material, using the abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determining the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators. The abnormal process parameters include: when the aluminum alloy material is the first aluminum alloy material, using the first abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determining the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators, wherein the preset process parameters used to train the first abnormal process parameter identification model include at least: intermediate annealing insulation temperature, intermediate annealing insulation time, continuous annealing insulation temperature and continuous annealing insulation time.

[0043] In the above-mentioned embodiment of the present application, for the first aluminum alloy material, a corresponding first abnormal process parameter identification model is selected to predict the material properties, thereby ensuring that the predicted material properties predicted using the first abnormal process parameter identification model are sufficiently accurate.

[0044] Optionally, the first aluminum alloy material may be a 5xxx aluminum alloy automotive sheet material, such as a thin sheet material of 5182, 5754, 5052, or other grade alloys.

[0045] Optionally, when the aluminum alloy material is a 5xxx aluminum alloy automobile sheet material, the preset process parameters include: Si percentage (unit %), Fe percentage (unit %), Cu percentage (unit %), Mn percentage (unit %), Cr percentage (unit %), Mg percentage (unit %), online titanium wire addition amount (unit g), casting speed (unit mm / s), casting temperature (unit ° C), cooling water flow rate (mm3 / s), ingot thickness (unit mm), homogenization holding temperature (unit ° C), homogenization holding time (unit min), homogenization Heating rate (unit ℃ / min), hot rough rolling starting thickness (unit mm), hot rough rolling starting temperature (unit ℃), hot rough rolling relative reduction (unit %), hot rough rolling emulsion spraying amount (mm3 / s), hot final rolling thickness (unit mm), hot final rolling temperature (unit ℃), cold rolling intermediate billet thickness, intermediate annealing holding temperature (unit ℃), intermediate annealing holding time (unit min), cold rolling relative reduction (unit %), thickness after cold rolling (unit mm), continuous annealing holding temperature (unit ℃), continuous annealing holding time (unit min) and relative straightening amount (%).

[0046] As an optional embodiment, the abnormal process parameter identification model includes: a second abnormal process parameter identification model for a second aluminum alloy material, using the abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determining the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators. The abnormal process parameters include: when the aluminum alloy material is the second aluminum alloy material, using the second abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determining the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators, wherein the preset process parameters used to train the second abnormal process parameter identification model include at least: the ratio of the percentage of magnesium element content to the percentage of silicon element content, the solution holding temperature, the solution holding time, the quenching cooling water temperature, the pre-aging holding temperature, the pre-aging holding time, the aging holding temperature and the aging holding time.

[0047] In the above-mentioned embodiment of the present application, for the second aluminum alloy material, a corresponding second abnormal process parameter identification model is selected to predict the material properties, thereby ensuring that the predicted material properties predicted using the second abnormal process parameter identification model are sufficiently accurate.

[0048] Optionally, the second aluminum alloy material may be a 6xxx aluminum alloy automotive sheet material, such as a thin sheet material of 6016, 6014, 6111, or other alloy grades.

[0049] Optionally, when the aluminum alloy material is a 6xxx aluminum alloy automobile sheet material, the preset process parameters include: Si percentage (unit %), Fe percentage (unit %), Cu percentage (unit %), Mn percentage (unit %), Cr percentage (unit %), Mg percentage (unit %), ratio of Mg percentage to Si percentage (unit 1), casting speed (unit mm / s), casting temperature (unit °C), cooling water flow rate (mm3 / s), ingot thickness (unit mm), homogenization holding temperature (unit °C), homogenization holding time (unit min), homogenization heating rate (unit °C / m in), hot rough rolling starting thickness (unit mm), hot rough rolling starting temperature (unit ℃), hot rough rolling relative reduction (unit %), hot rough rolling emulsion spraying amount (mm3 / s), hot final rolling thickness (unit mm), hot final rolling temperature (unit ℃), cold rolling intermediate billet thickness, cold rolling relative reduction (unit %), thickness after cold rolling (unit mm), solution holding temperature (unit ℃), solution holding time (unit s), quenching cooling water temperature (unit ℃), pre-aging holding temperature (unit ℃), pre-aging holding time (unit min), aging holding temperature (unit ℃), aging holding time (unit min) and relative straightening amount (%).

[0050] As an optional embodiment, the method also includes: obtaining at least one preset process parameter used in the production of aluminum alloy materials, and at least one preset material property, wherein the preset material properties include at least: tensile strength, yield strength and elongation; in a process parameter-material property database, using two-dimensional matrix data to record the preset process parameters and preset material properties, wherein each row of the two-dimensional matrix data is used to represent a preset process parameter for the production of aluminum alloy materials, and at least one preset material property corresponding to the preset process parameter, and each column of the two-dimensional matrix data represents a preset process parameter or a preset material property.

[0051] In the above-mentioned embodiment of the present application, before producing the aluminum alloy material, at least one preset process parameter required for generating the aluminum alloy material and the influence of each preset process parameter on the material properties of the aluminum alloy material can be obtained in advance, and at least one preset material property corresponding to each preset process parameter is obtained. Then, the correspondence between the preset process parameters and the preset material properties is recorded in a process parameter-material property database in the form of two-dimensional matrix data. Based on the process parameter-material property database, the training data required for training the material property prediction model and the positive and negative samples required for training the abnormal process parameter recognition model can be constructed.

[0052] As an optional embodiment, the method also includes: in a process parameter-material performance database, determining preset material properties that meet the material performance indicators and preset process parameters corresponding to the preset material properties as positive samples for training an abnormal process parameter identification model; based on the process parameter-material performance database, determining preset material properties that do not meet the material performance indicators and preset process parameters corresponding to the preset material properties as negative samples for training an abnormal process parameter identification model; using positive samples and negative samples, through a nearest neighbor classification machine learning algorithm, to train an abnormal process parameter identification model, wherein the abnormal process parameter identification model uses a grid search method to adjust the hyperparameters of the abnormal process parameter identification model.

[0053] In the above-mentioned embodiments of the present application, the abnormal process parameter identification model is obtained by training the nearest neighbor classification machine learning algorithm. The positive samples and negative samples used for training the abnormal process parameter identification model can be determined according to the process parameter-material performance database. In the training stage of the abnormal process parameter identification model, the grid search method can be used to adjust the hyperparameters of the trained abnormal process parameter identification model to realize the training of the abnormal process parameter identification model.

[0054] As an optional embodiment, obtaining the material properties to be analyzed of the aluminum alloy material includes: when producing the aluminum alloy material, obtaining the predicted material properties of the aluminum alloy material; detecting whether the predicted material properties meet the material performance indicators required by preset indicators; when the predicted material properties do not meet the material performance indicators, determining the predicted material properties as the material properties to be analyzed.

[0055] In the above-mentioned embodiments of the present application, the material properties to be analyzed of the aluminum alloy material can be determined by performing performance testing on the aluminum alloy material. After obtaining the predicted material properties of the aluminum alloy material, the predicted material properties can be tested against the material performance indicators required by the preset indicators to determine whether the predicted material properties meet the material performance indicators. In the case that the predicted material properties do not meet the material performance indicators, the predicted material properties are determined as the material properties to be analyzed.

[0056] As an optional embodiment, in the case of producing aluminum alloy materials, obtaining predicted material properties of the aluminum alloy materials includes: obtaining real-time process parameters for producing aluminum alloy materials; using a material property prediction model to analyze the real-time process parameters to determine the predicted material properties of the aluminum alloy materials, wherein the material property prediction model is pre-trained using a process parameter-material property database through a random forest regression machine learning algorithm, and the process parameter-material property database uses two-dimensional matrix data to represent the correspondence between preset process parameters and preset material properties, and each group of training data for training the material property prediction model is the preset process parameter and the preset material property corresponding to the preset process parameter.

[0057] In the embodiment of the present invention, a material property prediction model is used to perform real-time analysis on real-time process parameters for producing aluminum alloy materials, determine the influence of each real-time process parameter on the material properties of the aluminum alloy materials, and obtain the predicted material properties of the aluminum alloy materials under the action of the real-time process parameters, thereby achieving the purpose of real-time supervision of the process parameters for producing aluminum alloy materials, and realizing the technical effect of real-time prediction of material properties of aluminum alloy materials based on the real-time process parameters for producing aluminum alloy materials, thereby solving the technical problem of being unable to quickly determine unreasonable process parameters.

[0058] As an optional embodiment, the method also includes: in a process parameter-material property database, selecting multiple preset process parameters and preset material properties corresponding to each preset process parameter to generate a training set, wherein each group of training data in the training set is a preset process parameter and a preset material property corresponding to the preset process parameter; using the training set to adopt a cross-validation method, and training a material property prediction model through a random forest regression machine learning algorithm, wherein the loss function used for training the material property prediction model is a mean square error function, and a grid search method is used to adjust the hyperparameters of the material property prediction model.

[0059] In the above-mentioned embodiments of the present application, the material property prediction model is obtained by training through a random forest regression machine learning algorithm. The training set used for training the material property prediction model can be selected from a process parameter-material property database, and the material property prediction model can use a mean square error function as a loss function. During the training stage of the material property prediction model, a grid search method can be used to adjust the hyperparameters to achieve the training of the material property prediction model.

[0060] As an optional embodiment, the method also includes obtaining the next real-time process parameter for producing the aluminum alloy material when the predicted material performance meets the material performance index.

[0061] In the above-mentioned embodiment of the present application, after determining the predicted material properties of the aluminum alloy material to be produced based on the real-time process parameters, the predicted material properties can be compared with the pre-set material performance indicators to determine whether the predicted material properties meet the material performance indicators required by the preset indicators. Then, if the predicted material properties based on the real-time process parameters meet the material performance indicators, the next real-time process parameters for producing the aluminum alloy material are obtained, and the material performance prediction model is continued to be used to predict the material properties.

[0062] As an optional embodiment, the method further includes: generating material performance warning information when the predicted material performance does not meet the material performance index, wherein the material performance warning information is at least used to indicate that the produced aluminum alloy material has abnormal material performance.

[0063] In the above-mentioned embodiment of the present application, when the predicted material performance based on the real-time process parameters does not meet the material performance indicators, material performance warning information can be generated, and the material performance warning information indicates that the aluminum alloy material being produced has abnormal material performance. Then, the production of the aluminum alloy material can be suspended based on the material performance warning information to avoid the losses caused by expanding the production of unqualified aluminum alloy materials; the subsequent process flow of producing the aluminum alloy material can also be adjusted based on the material performance warning information to compensate for the material performance defects of the aluminum alloy material caused by the abnormal real-time process parameters.

[0064] As an optional embodiment, the material property prediction model includes: a first material property prediction model for a first aluminum alloy material, using the material property prediction model to analyze real-time process parameters, and determining the predicted material properties of the aluminum alloy material includes: when the aluminum alloy material is the first aluminum alloy material, using the first material property prediction model to analyze real-time process parameters, and determining the predicted material properties of the first aluminum alloy material, wherein the preset process parameters used for training the first material property prediction model include at least: intermediate annealing insulation temperature, intermediate annealing insulation time, continuous annealing insulation temperature and continuous annealing insulation time.

[0065] In the above-mentioned embodiment of the present application, for the first aluminum alloy material, a corresponding first material property prediction model is selected to predict the material properties, thereby ensuring that the predicted material properties predicted using the first material property prediction model are sufficiently accurate.

[0066] As an optional embodiment, the material property prediction model includes: a second material property prediction model for a second aluminum alloy material, using the material property prediction model to analyze real-time process parameters, and determining the predicted material properties of the aluminum alloy material includes: when the aluminum alloy material is the second aluminum alloy material, using the second material property prediction model to analyze real-time process parameters to determine the predicted material properties of the second aluminum alloy material, wherein the preset process parameters used for training the second material property prediction model include at least: the ratio of the percentage of magnesium element content to the percentage of silicon element content, the solution holding temperature, the solution holding time, the quenching cooling water temperature, the pre-aging holding temperature, the pre-aging holding time, the aging holding temperature and the aging holding time.

[0067] In the above-mentioned embodiment of the present application, for the second aluminum alloy material, a corresponding second material property prediction model is selected to predict the material properties, thereby ensuring that the predicted material properties predicted using the second material property prediction model are sufficiently accurate.

[0068] The present invention also provides a preferred embodiment, which provides a big data model for identifying abnormal process parameters of aluminum alloy automobile plate materials, aiming to solve the problem that there are many process parameters in the production process of aluminum alloy automobile plates, and the influence mechanism and law of each process parameter on material properties are unclear.

[0069] As an optional example, a big data model for identifying abnormal process parameters of aluminum alloy automobile sheet materials is established. Through real-time calculation of the process parameters of the production process, the performance of aluminum alloy automobile sheet materials is predicted online, and timely warning and identification of the production process parameters that produce abnormal performance are achieved.

[0070] Figure 2 is a schematic diagram of a big data model for identifying abnormal process parameters of aluminum alloy automobile sheet materials according to an embodiment of the present invention. Figure 2 As shown, the steps include:

[0071] Step S201, building a "process parameter-material property" database for the production process of aluminum alloy automobile plates.

[0072] Step S202, using the database to train a random forest regression machine learning algorithm, establish a "process parameter-material performance" prediction model, and realize the prediction of material properties.

[0073] Step S203, using the database to train the K nearest neighbor classification machine learning algorithm, establish an abnormal process parameter recognition model, and realize the recognition of abnormal process parameters.

[0074] Step S204, predicting the material properties of the production process and processing abnormal process parameters.

[0075] The "process parameter-material property" prediction model trained in the above step S202 can analyze the real-time process parameters and determine the predicted material properties.

[0076] In the above step S204, the material properties predicted by the "process parameter-material property" prediction model based on the real-time process description can be compared with the material properties required by the pre-set index to determine whether the predicted material properties meet the index requirements. If they meet the index requirements, normal production will be carried out; if they do not meet the index requirements, an abnormal material property warning will be initiated, and the abnormal process parameter recognition model trained in the above step S203 will be used to analyze the predicted material properties and identify the abnormal process parameters. The material generation process can then be adjusted according to the identified abnormal process parameters to ensure that the produced materials meet the index requirements.

[0077] In the above step S201, the aluminum alloy automobile sheet may be a 5xxx aluminum alloy automobile sheet, or a 6xxx aluminum alloy automobile sheet.

[0078] In the above step S201, when the aluminum alloy automobile plate is a 5xxx aluminum alloy automobile plate, the production process "process parameters-material properties" database is a two-dimensional matrix data, one row of the matrix data is the process parameters and corresponding material properties of the 5xxx aluminum alloy automobile plate material in a certain production process, and one column of the matrix data is a certain process parameter or material property of the 5xxx aluminum alloy automobile plate material.

[0079] Optionally, the process parameters of 5xxx aluminum alloy automobile sheet include: Si percentage (unit %), Fe percentage (unit %), Cu percentage (unit %), Mn percentage (unit %), Cr percentage (unit %), Mg percentage (unit %), online titanium wire addition amount (unit g), casting speed (unit mm / s), casting temperature (unit ° C), cooling water flow (mm3 / s), ingot thickness (unit mm), homogenization holding temperature (unit ° C), homogenization holding time (unit min), homogenization heating rate (unit ℃ / min), hot rough rolling starting thickness (unit mm), hot rough rolling starting temperature (unit ℃), hot rough rolling relative reduction (unit %), hot rough rolling emulsion spraying amount (mm3 / s), hot final rolling thickness (unit mm), hot final rolling temperature (unit ℃), cold rolling intermediate billet thickness, intermediate annealing holding temperature (unit ℃), intermediate annealing holding time (unit min), cold rolling relative reduction (unit %), thickness after cold rolling (unit mm), continuous annealing holding temperature (unit ℃), continuous annealing holding time (unit min) and relative straightening amount (%).

[0080] Optionally, the material properties of 5xxx aluminum alloy automobile plates include: tensile strength (Mpa), yield strength (Mpa), and elongation (unit %).

[0081] In the above step S201, when the aluminum alloy automobile plate is a 6xxx aluminum alloy automobile plate, the production process "process parameters-material properties" database is a two-dimensional matrix data, one row of the matrix data is the process parameters and corresponding material properties of the 6xxx aluminum alloy automobile plate material in a certain production process, and one column of the matrix data is a certain process parameter or material property of the 6xxx aluminum alloy automobile plate material.

[0082] Optionally, the process parameters of 6xxx aluminum alloy automobile sheet include: Si percentage (unit %), Fe percentage (unit %), Cu percentage (unit %), Mn percentage (unit %), Cr percentage (unit %), Mg percentage (unit %), ratio of Mg percentage to Si percentage (unit 1), casting speed (unit mm / s), casting temperature (unit °C), cooling water flow (mm3 / s), ingot thickness (unit mm), homogenization holding temperature (unit °C), homogenization holding time (unit min), homogenization heating rate (unit °C / min), hot rough rolling opening Rolling thickness (unit: mm), hot rough rolling start temperature (unit: °C), hot rough rolling relative reduction (unit: %), hot rough rolling emulsion spraying amount (mm3 / s), hot final rolling thickness (unit: mm), hot final rolling temperature (unit: °C), cold rolling intermediate billet thickness, cold rolling relative reduction (unit: %), thickness after cold rolling (unit: mm), solution holding temperature (unit: °C), solution holding time (unit: s), quenching cooling water temperature (unit: °C), pre-aging holding temperature (unit: °C), pre-aging holding time (unit: min), aging holding temperature (unit: °C), aging holding time (unit: min) and relative straightening amount (%).

[0083] Optionally, the material properties of 6xxx aluminum alloy automobile plates include: tensile strength (Mpa), yield strength (Mpa), and elongation (unit %).

[0084] In the above step S202, for the establishment of the "process parameters-material properties" prediction model, a random forest regression machine learning algorithm is adopted, the training set data volume is 70%-80%, the test data volume is 20%-30%, the cross-validation method is adopted during training, the loss function is the mean square error, and the grid search method is used to perform hyperparameter tuning of the random forest regression machine learning algorithm.

[0085] In the above step S203, for the establishment of the abnormal process parameter identification model, the KNN classification machine learning algorithm is adopted, the classification dimensions are qualified and unqualified material performance, the training set data volume is 70%-80%, and the test data volume is 20%-30%. The data in the database is normalized before training, and the grid search method is used to perform hyperparameter tuning on the KNN classification machine learning algorithm.

[0086] In the above step S204, for the material property prediction and processing method of the production process, the material property is predicted using the established "process parameter-material property" prediction model, and the predicted material property is compared to see whether it meets the material property required by the index (i.e., material property index). If it meets the requirements, no processing is performed; if it does not meet the requirements, an early warning is issued, and the abnormal process parameters are identified using the established abnormal process parameter identification model.

[0087] The technical solution provided in this application solves the problems that the production process mechanism of 5xxx aluminum alloy automobile plates is complex and it is difficult to establish an accurate mechanism model, and that traditional process optimization methods and management and control technologies are unable to effectively optimize and control the complex process parameters of each process in the whole process. It makes full use of the massive production data of 5xxx aluminum alloy automobile plates, establishes intelligent forecasting and processing technology through database technology and machine learning algorithms, realizes rapid early warning and problem location of the production process, and avoids a large amount of material and time waste.

[0088] According to an embodiment of the present invention, an embodiment of an abnormal process parameter identification device for aluminum alloy materials is also provided. It should be noted that the abnormal process parameter identification device for aluminum alloy materials can be used to execute the abnormal process parameter identification method for aluminum alloy materials in the embodiment of the present invention, and the abnormal process parameter identification method for aluminum alloy materials in the embodiment of the present invention can be executed in the abnormal process parameter identification device for aluminum alloy materials.

[0089] Figure 3 is a schematic diagram of an abnormal process parameter identification device for aluminum alloy material according to an embodiment of the present invention. Figure 3 As shown, the device may include: an acquisition module 32, used to acquire the material properties to be analyzed of the aluminum alloy material, wherein the material properties to be analyzed do not meet the preset material performance indicators; a determination module 34, used to use an abnormal process parameter identification model to analyze the material properties to be analyzed, and determine the abnormal process parameters that cause the material properties to be analyzed to not meet the material performance indicators, wherein the abnormal process parameter identification model is pre-trained using a process parameter-material performance database through a nearest neighbor classification machine learning algorithm, the process parameter-material performance database uses two-dimensional matrix data to represent the correspondence between preset process parameters and preset material properties, and the training data for training the abnormal process parameter identification model is a plurality of training samples determined based on the process parameter-material performance database, each training sample is used to represent the correspondence between the preset process parameters and the preset material properties, and the training samples include: positive samples and negative samples, the preset material properties in the positive samples meet the material performance indicators, and the preset material properties in the negative samples do not meet the material performance indicators.

[0090] It should be noted that the acquisition module 32 in this embodiment can be used to execute step S102 in the embodiment of the present application, and the determination module 34 in this embodiment can be used to execute step S104 in the embodiment of the present application. The examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the contents disclosed in the above embodiments.

[0091] In an embodiment of the present invention, the abnormal process parameter identification model is used to analyze the material properties of the aluminum alloy material to be analyzed. According to the influence of the preset process parameters on the material properties of the aluminum alloy material, the reason why the material properties to be analyzed do not meet the preset material performance indicators is quickly determined, thereby achieving the purpose of quickly determining the abnormal process parameters that cause the material properties to be analyzed to not meet the material performance indicators, realizing the technical effect of quickly determining the abnormal process parameters, and thus solving the technical problem of being unable to quickly determine unreasonable process parameters.

[0092] As an optional embodiment, the abnormal process parameter identification model includes: a first abnormal process parameter identification model for a first aluminum alloy material, the determination module includes: a first determination sub-module, which is used to use the first abnormal process parameter identification model to analyze the performance of the material to be analyzed when the aluminum alloy material is the first aluminum alloy material, and determine the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators, wherein the preset process parameters used to train the first abnormal process parameter identification model include at least: intermediate annealing insulation temperature, intermediate annealing insulation time, continuous annealing insulation temperature and continuous annealing insulation time.

[0093] As an optional embodiment, the abnormal process parameter identification model includes: a second abnormal process parameter identification model for a second aluminum alloy material, using the abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determining the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators, including: a second determination submodule, used to, when the aluminum alloy material is the second aluminum alloy material, use the second abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determine the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators, wherein the preset process parameters used for training the second abnormal process parameter identification model include at least: the ratio of the percentage of magnesium element and the percentage of silicon element, solution holding temperature, solution holding time, quenching cooling water temperature, pre-aging holding temperature, pre-aging holding time, aging holding temperature and aging holding time.

[0094] As an optional embodiment, the device also includes: a first acquisition submodule, used to acquire at least one of the preset process parameters used to produce the aluminum alloy material, and at least one of the preset material properties, wherein the preset material properties include at least: tensile strength, yield strength and elongation; a storage submodule, used to record the preset process parameters and the preset material properties in the process parameter-material property database using the two-dimensional matrix data, wherein each row of the two-dimensional matrix data is used to represent one of the preset process parameters for producing the aluminum alloy material, and at least one of the preset material properties corresponding to the preset process parameter, and each column of the two-dimensional matrix data represents one of the preset process parameters or one of the preset material properties.

[0095] As an optional embodiment, the device also includes: a third determination submodule, used to determine, in the process parameter-material property database, the preset material property that meets the material performance index and the preset process parameters corresponding to the preset material property as the positive samples for training the abnormal process parameter identification model; a fourth determination submodule, used to determine, based on the process parameter-material property database, the preset material property that does not meet the material performance index and the preset process parameters corresponding to the preset material property as the negative samples for training the abnormal process parameter identification model; a training submodule, used to use the positive samples and the negative samples to train the abnormal process parameter identification model through a nearest neighbor classification machine learning algorithm, wherein the abnormal process parameter identification model uses a grid search method to adjust the hyperparameters of the abnormal process parameter identification model.

[0096] As an optional embodiment, the acquisition module includes: an acquisition unit, used to obtain the predicted material properties of the aluminum alloy material when producing the aluminum alloy material; a detection unit, used to detect whether the predicted material properties meet the material performance indicators required by preset indicators; and a determination unit, used to determine the predicted material properties as the material properties to be analyzed when the predicted material properties do not meet the material performance indicators.

[0097] As an optional embodiment, the acquisition unit includes: an acquisition subunit, used to acquire real-time process parameters for producing the aluminum alloy material; and a determination subunit, used to analyze the real-time process parameters using a material property prediction model to determine the predicted material properties of the aluminum alloy material, wherein the material property prediction model is pre-trained using a process parameter-material property database through a random forest regression machine learning algorithm, and each set of training data for training the material property prediction model is the preset process parameter and the preset material property corresponding to the preset process parameter.

[0098] An embodiment of the present invention may provide an electronic device, which may be a computer terminal, and the computer terminal may be any computer terminal device in a computer terminal group. Optionally, in this embodiment, the computer terminal may also be replaced by a terminal device such as a mobile terminal.

[0099] Optionally, in this embodiment, the computer terminal may be located in at least one network device among a plurality of network devices of the computer network.

[0100] In this embodiment, the above-mentioned computer terminal can execute the program code of the following steps in the abnormal process parameter identification method for aluminum alloy materials: obtaining the material properties to be analyzed of the aluminum alloy material, wherein the material properties to be analyzed do not meet the preset material performance indicators; using the abnormal process parameter identification model to analyze the material properties to be analyzed, and determine the abnormal process parameters that cause the material properties to be analyzed to not meet the material performance indicators, wherein the abnormal process parameter identification model is pre-trained using the process parameter-material performance database through the nearest neighbor classification machine learning algorithm, the process parameter-material performance database uses two-dimensional matrix data to represent the correspondence between preset process parameters and preset material properties, and the training data for training the abnormal process parameter identification model is a plurality of training samples determined based on the process parameter-material performance database, each training sample is used to represent the correspondence between the preset process parameters and the preset material properties, and the training samples include: positive samples and negative samples, the preset material properties in the positive samples meet the material performance indicators, and the preset material properties in the negative samples do not meet the material performance indicators.

[0101] Figure 4 is a structural block diagram of a computer terminal according to an embodiment of the present invention. Figure 4 As shown, the computer terminal 40 may include: one or more (only one is shown in the figure) processors 42 , and a memory 44 .

[0102] Among them, the memory can be used to store software programs and modules, such as the program instructions / modules corresponding to the abnormal process parameter identification method and device of the aluminum alloy material in the embodiment of the present invention. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, that is, the abnormal process parameter identification method of the aluminum alloy material mentioned above is realized. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal 40 via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0103] The processor can call the information and application programs stored in the memory through the transmission device to perform the following steps: obtaining the material properties to be analyzed of the aluminum alloy material, wherein the material properties to be analyzed do not meet the preset material performance indicators; using the abnormal process parameter identification model to analyze the material properties to be analyzed, and determine the abnormal process parameters that cause the material properties to be analyzed to not meet the material performance indicators, wherein the abnormal process parameter identification model is pre-trained using the process parameter-material performance database through the nearest neighbor classification machine learning algorithm, the process parameter-material performance database uses two-dimensional matrix data to represent the correspondence between preset process parameters and preset material properties, and the training data for training the abnormal process parameter identification model is a plurality of training samples determined based on the process parameter-material performance database, each training sample is used to represent the correspondence between the preset process parameters and the preset material performance, and the training samples include: positive samples and negative samples, the preset material performance in the positive samples meets the material performance indicators, and the preset material performance in the negative samples does not meet the material performance indicators.

[0104] Optionally, the abnormal process parameter identification model includes: a first abnormal process parameter identification model for a first aluminum alloy material, the above-mentioned processor can also execute the program code of the following steps: when the aluminum alloy material is the first aluminum alloy material, use the first abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determine the abnormal process parameters that cause the performance of the material to be analyzed to not meet the material performance indicators, wherein the preset process parameters used to train the first abnormal process parameter identification model include at least: intermediate annealing insulation temperature, intermediate annealing insulation time, continuous annealing insulation temperature and continuous annealing insulation time.

[0105] Optionally, the abnormal process parameter identification model includes: a second abnormal process parameter identification model for a second aluminum alloy material, the above-mentioned processor can also execute the program code of the following steps: when the aluminum alloy material is the second aluminum alloy material, use the second abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determine the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators, wherein the preset process parameters used to train the second abnormal process parameter identification model include at least: the ratio of the percentage of magnesium element content to the percentage of silicon element content, the solution holding temperature, the solution holding time, the quenching cooling water temperature, the pre-aging holding temperature, the pre-aging holding time, the aging holding temperature and the aging holding time.

[0106] Optionally, the processor may also execute the program code of the following steps: obtaining at least one preset process parameter and at least one preset material property used in producing aluminum alloy materials, wherein the preset material properties include at least tensile strength, yield strength and elongation; in a process parameter-material property database, using two-dimensional matrix data to record the preset process parameters and preset material properties, wherein each row of the two-dimensional matrix data is used to represent a preset process parameter for producing aluminum alloy materials, and at least one preset material property corresponding to the preset process parameter, and each column of the two-dimensional matrix data represents a preset process parameter or a preset material property.

[0107] Optionally, the processor may also execute the program code of the following steps: in the process parameter-material performance database, the preset material performance that meets the material performance index and the preset process parameters corresponding to the preset material performance are determined as positive samples for training the abnormal process parameter recognition model; based on the process parameter-material performance database, the preset material performance that does not meet the material performance index and the preset process parameters corresponding to the preset material performance are determined as negative samples for training the abnormal process parameter recognition model; using the positive samples and the negative samples, the abnormal process parameter recognition model is trained through the nearest neighbor classification machine learning algorithm, wherein the abnormal process parameter recognition model uses a grid search method to adjust the hyperparameters of the abnormal process parameter recognition model.

[0108] Optionally, the processor may also execute the program code of the following steps: when producing aluminum alloy materials, obtaining predicted material properties of the aluminum alloy materials; detecting whether the predicted material properties meet the material performance indicators required by preset indicators; if the predicted material properties do not meet the material performance indicators, determining the predicted material properties as the material properties to be analyzed.

[0109] Optionally, the processor may also execute the program code of the following steps: obtaining real-time process parameters for producing aluminum alloy materials; analyzing the real-time process parameters using a material property prediction model to determine predicted material properties of the aluminum alloy materials, wherein the material property prediction model is pre-trained using a process parameter-material property database through a random forest regression machine learning algorithm, and each set of training data for training the material property prediction model is a preset process parameter and a preset material property corresponding to the preset process parameter.

[0110] By adopting the embodiment of the present invention, a scheme for identifying abnormal process parameters of aluminum alloy materials is provided. In the embodiment of the present invention, the abnormal process parameter identification model can be used to analyze the material properties of the aluminum alloy materials to be analyzed, and according to the influence of the preset process parameters on the material properties of the aluminum alloy materials, the reasons causing the material properties to be analyzed not to meet the preset material performance indicators are quickly determined, so as to achieve the purpose of quickly determining the abnormal process parameters causing the material properties to be analyzed not to meet the material performance indicators, and achieve the technical effect of quickly determining the abnormal process parameters, thereby solving the technical problem of being unable to quickly determine unreasonable process parameters.

[0111] It can be understood by those skilled in the art that Figure 4 The structure shown is for illustration only, and the computer terminal may also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a handheld computer, a mobile Internet device (Mobile Internet Devices, MID), a PAD, or other terminal devices. Figure 4 The structure of the electronic device is not limited. For example, the computer terminal 40 may also include Figure 4 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 4 Different configurations are shown.

[0112] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a computer program. The computer program can be stored in a non-volatile medium. The non-volatile storage medium may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0113] The embodiment of the present invention further provides a non-volatile storage medium. Optionally, in this embodiment, the non-volatile storage medium can be used to store the program code executed by the abnormal process parameter identification method for aluminum alloy material provided by the above embodiment.

[0114] Optionally, in this embodiment, the non-volatile storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0115] Optionally, in this embodiment, the non-volatile storage medium is configured to store program codes for executing the following steps: obtaining material properties to be analyzed of an aluminum alloy material, wherein the material properties to be analyzed do not meet preset material performance indicators; analyzing the material properties to be analyzed using an abnormal process parameter identification model, and determining abnormal process parameters that cause the material properties to be analyzed to not meet the material performance indicators, wherein the abnormal process parameter identification model is pre-trained using a process parameter-material performance database through a nearest neighbor classification machine learning algorithm, the process parameter-material performance database uses two-dimensional matrix data to represent the correspondence between preset process parameters and preset material properties, the training data for training the abnormal process parameter identification model is a plurality of training samples determined based on the process parameter-material performance database, each training sample is used to represent the correspondence between a preset process parameter and a preset material performance, and the training samples include: positive samples and negative samples, the preset material performance in the positive samples meets the material performance indicators, and the preset material performance in the negative samples does not meet the material performance indicators.

[0116] Optionally, in this embodiment, the abnormal process parameter identification model includes: a first abnormal process parameter identification model for a first aluminum alloy material, the non-volatile storage medium is configured to store a program code for executing the following steps: when the aluminum alloy material is a first aluminum alloy material, using the first abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determining the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators, wherein the preset process parameters used to train the first abnormal process parameter identification model include at least: intermediate annealing insulation temperature, intermediate annealing insulation time, continuous annealing insulation temperature and continuous annealing insulation time.

[0117] Optionally, in this embodiment, the abnormal process parameter identification model includes: a second abnormal process parameter identification model for a second aluminum alloy material, the non-volatile storage medium is configured to store a program code for executing the following steps: when the aluminum alloy material is the second aluminum alloy material, the second abnormal process parameter identification model is used to analyze the performance of the material to be analyzed, and determine the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators, wherein the preset process parameters used to train the second abnormal process parameter identification model include at least: the ratio of the percentage of magnesium element content to the percentage of silicon element content, the solution holding temperature, the solution holding time, the quenching cooling water temperature, the pre-aging holding temperature, the pre-aging holding time, the aging holding temperature and the aging holding time.

[0118] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: obtaining at least one preset process parameter used in producing aluminum alloy materials, and at least one preset material property, wherein the preset material properties include at least: tensile strength, yield strength and elongation; in a process parameter-material property database, using two-dimensional matrix data to record the preset process parameters and preset material properties, wherein each row of the two-dimensional matrix data is used to represent a preset process parameter for producing aluminum alloy materials, and at least one preset material property corresponding to the preset process parameter, and each column of the two-dimensional matrix data represents a preset process parameter or a preset material property.

[0119] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: in a process parameter-material property database, preset material properties that meet the material performance indicators and preset process parameters corresponding to the preset material properties are determined as positive samples for training an abnormal process parameter identification model; based on the process parameter-material property database, preset material properties that do not meet the material performance indicators and preset process parameters corresponding to the preset material properties are determined as negative samples for training an abnormal process parameter identification model; using positive samples and negative samples, the abnormal process parameter identification model is trained through a nearest neighbor classification machine learning algorithm, wherein the abnormal process parameter identification model uses a grid search method to adjust the hyperparameters of the abnormal process parameter identification model.

[0120] Optionally, in this embodiment, the non-volatile storage medium is configured to store program codes for executing the following steps: in the case of producing aluminum alloy materials, obtaining predicted material properties of the aluminum alloy materials; detecting whether the predicted material properties meet the material performance indicators required by preset indicators; in the case where the predicted material properties do not meet the material performance indicators, determining the predicted material properties as the material properties to be analyzed.

[0121] Optionally, in this embodiment, the non-volatile storage medium is configured to store program code for executing the following steps: obtaining real-time process parameters for producing aluminum alloy materials; analyzing the real-time process parameters using a material property prediction model to determine predicted material properties of the aluminum alloy materials, wherein the material property prediction model is pre-trained using a process parameter-material property database through a random forest regression machine learning algorithm, and each set of training data for training the material property prediction model is a preset process parameter and a preset material property corresponding to the preset process parameter.

[0122] The embodiment of the present invention further provides a computer program product, including a computer program. Optionally, in this embodiment, when the computer program is executed by a processor, the steps of the abnormal process parameter identification method for aluminum alloy material provided in the above embodiment are implemented.

[0123] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0124] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0125] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0126] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0127] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0128] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a non-volatile storage medium. Based on this understanding, the technical solution of the present invention, 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. The computer software product is stored in a non-volatile storage medium, including several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned non-volatile storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0129] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for identifying abnormal process parameters of aluminum alloy materials, characterized in that: include: Obtaining the material properties to be analyzed of the aluminum alloy material, wherein the material properties to be analyzed do not meet the preset material property index; The material properties to be analyzed are analyzed using an abnormal process parameter identification model to determine the abnormal process parameters that cause the material properties to be analyzed to fail to meet the material performance index, wherein the abnormal process parameter identification model is pre-trained using a process parameter-material performance database through a nearest neighbor classification machine learning algorithm, the process parameter-material performance database uses two-dimensional matrix data to represent the correspondence between preset process parameters and preset material properties, the training data for training the abnormal process parameter identification model is a plurality of training samples determined based on the process parameter-material performance database, each of the training samples is used to represent the correspondence between the preset process parameters and the preset material properties, and the training samples include: positive samples and negative samples, the preset material properties in the positive samples meet the material performance index, and the preset material properties in the negative samples do not meet the material performance index.

2. The method according to claim 1, characterized in that: The abnormal process parameter identification model includes: a first abnormal process parameter identification model for a first aluminum alloy material, using the abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determining that the abnormal process parameters that cause the performance of the material to be analyzed to not meet the material performance index include: In the case where the aluminum alloy material is the first aluminum alloy material, the first abnormal process parameter identification model is used to analyze the performance of the material to be analyzed to determine the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators, wherein the preset process parameters used to train the first abnormal process parameter identification model include at least: intermediate annealing insulation temperature, intermediate annealing insulation time, continuous annealing insulation temperature and continuous annealing insulation time.

3. The method according to claim 1, characterized in that The abnormal process parameter identification model includes: a second abnormal process parameter identification model for a second aluminum alloy material, using the abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determining that the abnormal process parameters that cause the performance of the material to be analyzed to not meet the material performance index include: In the case where the aluminum alloy material is the second aluminum alloy material, the second abnormal process parameter identification model is used to analyze the performance of the material to be analyzed to determine the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance indicators, wherein the preset process parameters used for training the second abnormal process parameter identification model include at least: the ratio of the percentage of magnesium element content to the percentage of silicon element content, the solution holding temperature, the solution holding time, the quenching cooling water temperature, the pre-aging holding temperature, the pre-aging holding time, the aging holding temperature and the aging holding time.

4. The method according to claim 1, characterized in that: The method further comprises: Acquire at least one of the preset process parameters and at least one of the preset material properties used in producing the aluminum alloy material, wherein the preset material properties at least include: tensile strength, yield strength and elongation; In the process parameter-material property database, the two-dimensional matrix data is used to record the preset process parameters and the preset material properties, wherein each row of the two-dimensional matrix data is used to represent one of the preset process parameters for producing the aluminum alloy material, and at least one of the preset material properties corresponding to the preset process parameter, and each column of the two-dimensional matrix data represents one of the preset process parameters or one of the preset material properties.

5. The method according to claim 1, characterized in that: The method further comprises: In the process parameter-material property database, the preset material properties that meet the material property index and the preset process parameters corresponding to the preset material properties are determined as the positive samples for training the abnormal process parameter recognition model; Based on the process parameter-material property database, the preset material properties that do not meet the material property index and the preset process parameters corresponding to the preset material properties are determined as the negative samples for training the abnormal process parameter recognition model; The positive samples and the negative samples are used to train the abnormal process parameter identification model through a nearest neighbor classification machine learning algorithm, wherein the abnormal process parameter identification model uses a grid search method to adjust the hyperparameters of the abnormal process parameter identification model.

6. The method according to claim 1, characterized in that Obtaining the material properties to be analyzed for aluminum alloy materials includes: In the case of producing the aluminum alloy material, obtaining predicted material properties of the aluminum alloy material; Detecting whether the predicted material performance meets the material performance index required by the preset index; In the case that the predicted material performance does not meet the material performance index, the predicted material performance is determined as the material performance to be analyzed.

7. The method according to claim 6, characterized in that In the case of producing the aluminum alloy material, obtaining the predicted material properties of the aluminum alloy material includes: Acquire real-time process parameters for producing the aluminum alloy material; The real-time process parameters are analyzed using a material abnormality process parameter identification model to determine the predicted material properties of the aluminum alloy material, wherein the material abnormality process parameter identification model is pre-trained using a process parameter-material property database through a random forest regression machine learning algorithm, and each group of training data for training the material abnormality process parameter identification model is the preset process parameter and the preset material property corresponding to the preset process parameter.

8. An abnormal process parameter identification device for aluminum alloy material, characterized in that: include: An acquisition module, used for acquiring the material properties to be analyzed of the aluminum alloy material, wherein the material properties to be analyzed do not meet the preset material property index; A determination module is used to use an abnormal process parameter identification model to analyze the performance of the material to be analyzed, and determine the abnormal process parameters that cause the performance of the material to be analyzed to fail to meet the material performance index, wherein the abnormal process parameter identification model is pre-trained using a process parameter-material performance database through a nearest neighbor classification machine learning algorithm, and the process parameter-material performance database uses two-dimensional matrix data to represent the correspondence between preset process parameters and preset material performance. The training data for training the abnormal process parameter identification model is a plurality of training samples determined based on the process parameter-material performance database, each of the training samples is used to represent the correspondence between the preset process parameters and the preset material performance, and the training samples include: positive samples and negative samples, the preset material performance in the positive samples meets the material performance index, and the preset material performance in the negative samples does not meet the material performance index.

9. An electronic device, comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the abnormal process parameter identification method for the aluminum alloy material according to any one of claims 1 to 7 through the computer program.

10. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by a processor, the steps of the method for identifying abnormal process parameters of aluminum alloy materials described in any one of claims 1 to 7 are implemented.

Images