Rapid prediction method for homogenization temperature of aluminum alloy cast ingot

Abstract

The invention belongs to the technical field of aluminum alloy material preparation, and particularly relates to a rapid prediction method for the homogenization temperature of an aluminum alloy cast ingot. According to the method, the contact thermocouple is arranged on the surface of the aluminum alloy cast ingot for actual temperature testing, simulation is carried out in combination with the finite element model, and accurate temperature data of the cast ingot in different homogenizing heat treatment processes can be obtained. Through the corrected finite element model, ingot homogenization process temperature simulation calculation is carried out on aluminum alloy ingots of different brands, and a temperature database between homogenization parameters and ingot core and surface temperatures can be efficiently established. A machine learning technology is utilized, a rapid prediction model between the homogenization parameters and the core and surface temperatures of the ingot is established according to a temperature database, the prediction time can be remarkably shortened, and the prediction efficiency is improved. By predicting and monitoring the temperature of the cast ingot in real time, the problems in the production process can be found and solved in time, and the production efficiency and the product quality are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of aluminum alloy material preparation, and particularly relates to a method for rapidly predicting the homogenization temperature of an aluminum alloy ingot. Background Art

[0002] In the production process of aluminum alloy plates, processes such as homogenization heat treatment, hot rolling, and cold rolling are required. Among them, homogenization heat treatment is an essential process before hot rolling. Due to the large size of the ingot, there is a large temperature difference between the surface and the core of the ingot during the heating process. A short heating time will cause the core of the ingot to not reach the set temperature, and a long heating time will increase the heating energy consumption and reduce the production efficiency. Therefore, in the actual operation of the factory, controlling the matching of temperature and time is the key to ensuring material performance and energy conservation and consumption reduction. However, there is no method in the prior art that can rapidly predict the homogenization temperature parameters of an aluminum alloy ingot. Summary of the Invention

[0003] In view of the problems existing in the prior art, the present invention provides a method for rapidly predicting the homogenization temperature of an aluminum alloy ingot. The specific content of the present invention is as follows:

[0004] A method for rapidly predicting the homogenization temperature of an aluminum alloy ingot includes the following steps:

[0005] S1, Temperature test of the aluminum alloy ingot: Set a contact thermocouple on the surface of aluminum alloy ingots of different grades, and detect the temperature T at a certain test position on the surface of the aluminum alloy ingot through the contact thermocouple 测 ;

[0006] S2, Establish a finite element model for the homogenization heat treatment process of the aluminum alloy ingot: Determine the geometric model according to the size of the aluminum alloy ingot, determine the physical property parameters of the model according to the grade or composition of the aluminum alloy, and determine the boundary conditions of the model according to the homogenization heating equipment; Use finite element simulation software to obtain the temperature data of the aluminum alloy ingot, and extract the temperature data T at the same position as the test position described in step S1 from the simulation results 仿 , and according to the test temperature T 测 and the simulation temperature data T 仿 correct the simulation model to obtain a corrected finite element model;

[0007] S3, Establish a temperature database: Use the corrected finite element model to perform temperature simulation calculations on the homogenization process of aluminum alloy ingots of different grades, and establish a temperature database between the homogenization parameters and the core and surface temperatures of the ingot;

[0008] S4, Establish a machine learning model: According to the temperature database of the ingot homogenization process, establish a rapid prediction model between the homogenization parameters and the core and surface temperatures of the ingot through machine learning;

[0009] S5. Based on the rapid prediction model, rapidly predict the temperature parameters of the aluminum alloy ingot during the homogenization process.

[0010] Further, in step S2, determine the boundary conditions of the model according to the structure of the homogenization heating equipment, the furnace gas temperature, the air flow direction, and the air flow velocity.

[0011] Further, the temperature data in the temperature database in step S3 includes: the core temperature of the aluminum alloy ingot, the surface temperature, the temperature difference between the core and the surface, and the time to reach temperature.

[0012] Further, the temperature simulation calculation of the ingot homogenization process in step S3 includes: carrying out the temperature calculation of the ingot homogenization process within the ranges of the homogenization furnace gas temperature T ± 10%, the homogenization time t ± 20%, the length L of the aluminum alloy ingot ± 50%, the width W of the aluminum alloy ingot ± 50%, and the thickness H of the aluminum alloy ingot ± 50%. Specifically, the homogenization furnace gas temperature T ± 10% ranges from 300 - 600 °C, the homogenization time t ± 20% means the time is 0 - 50 h, the length L of the aluminum alloy ingot ± 50% means the length is 1 - 10 m, the width W of the aluminum alloy ingot ± 50% means the width is 1 - 3 m, and the thickness H of the aluminum alloy ingot ± 50% means the thickness is 100 - 800 mm.

[0013] Further, the homogenization parameters in steps S3 and S4 include the homogenization furnace gas temperature T, the homogenization time t, the length L of the aluminum alloy ingot, the width W of the aluminum alloy ingot, and the thickness H of the aluminum alloy ingot.

[0014] Further, the temperature parameters of the aluminum alloy ingot during the homogenization process in step S5 include the relationship curve between the ingot surface temperature and the homogenization time, the relationship curve between the ingot core temperature and the homogenization time, and the ingot time to reach temperature.

[0015] Further, the method for rapidly predicting the temperature parameters of the aluminum alloy ingot during the homogenization process in step S5 is: input the homogenization furnace gas temperature T, the homogenization time t, the length L of the aluminum alloy ingot, the width W of the aluminum alloy ingot, and the thickness H of the aluminum alloy ingot into the rapid prediction model, and the model rapidly predicts the temperature parameters of the aluminum alloy ingot through calculation.

[0016] Advantages of the present invention:

[0017] (1) The rapid prediction method for the homogenization temperature of the aluminum alloy ingot disclosed by the present invention can improve the accuracy of prediction. By setting contact thermocouples on the surface of the aluminum alloy ingot for actual temperature measurement (S1) and combining with a finite element model for simulation (S2), the present invention can obtain accurate temperature data of the ingot during different homogenization heat treatment processes. This method combines the advantages of experimental testing and numerical simulation, effectively improving the accuracy and reliability of temperature prediction.

[0018] (2) The rapid prediction method for the homogenization temperature of aluminum alloy ingots disclosed by the present invention can optimize the model adaptability. During the process of establishing the finite element model (S2), personalized settings are made according to the size, grade or composition of the aluminum alloy ingot and the boundary conditions of the homogenization heating equipment, and the model is corrected by comparing the measured temperature with the simulated temperature, making the model more in line with the actual production situation and improving the adaptability and prediction accuracy of the model.

[0019] (3) The rapid prediction method for the homogenization temperature of aluminum alloy ingots disclosed by the present invention can improve the prediction efficiency. By simulating the temperature during the homogenization process of aluminum alloy ingots of different grades through the corrected finite element model (S3), a temperature database between the homogenization parameters and the core and surface temperatures of the ingot can be efficiently established. This database provides rich data support for the establishment of subsequent machine learning models and helps to achieve more accurate predictions. Using machine learning technology (S4), a rapid prediction model between the homogenization parameters and the core and surface temperatures of the ingot is established based on the temperature database, which can significantly shorten the prediction time and improve the prediction efficiency. This is of great significance for quality control and process optimization in the production process of aluminum alloy ingots.

[0020] (4) The method described in the present invention realizes the rapid prediction of the homogenization temperature of aluminum alloy ingots (S5) by integrating experimental testing, numerical simulation and machine learning technologies, providing strong support for the intelligent production of aluminum alloy ingots. By predicting and monitoring the ingot temperature in real time, problems in the production process can be discovered and solved in time, improving production efficiency and product quality. Description of the Drawings

[0021] Figure 1 is a schematic diagram of the method disclosed by the present invention;

[0022] Figure 2 is the change curve of the ingot temperature with time obtained in Example 1. Detailed Embodiments

[0023] The present invention will be described in detail below with reference to the drawings and specific embodiments. The following embodiments do not limit the content of the invention described in the claims in any way. In addition, all the contents shown in the following embodiments are not limited to those necessary for the solution of the invention described in the claims.

[0024] Refer to the attached Figure 1 , a rapid prediction method for the homogenization temperature of aluminum alloy ingots, comprising the following steps:

[0025] S1, Testing the temperature of the aluminum alloy ingot: Contact thermocouples are arranged on the surface of aluminum alloy ingots of different grades, and the temperature T at a certain test position on the surface of the aluminum alloy ingot is detected through the contact thermocouples测 ;

[0026] S2. Establish a finite element model for the homogenization heat treatment process of aluminum alloy ingots: Determine the geometric model according to the size of the aluminum alloy ingot, determine the physical properties parameters of the model according to the aluminum alloy grade or composition, and determine the boundary conditions of the model according to the structure of the homogenization heating equipment, furnace gas temperature, air flow direction, and air flow velocity. The temperature data T at the same position as the test position 仿 , according to the test temperature T 测 and the simulated temperature data T 仿 to correct the simulation model to obtain the corrected finite element model;

[0027] S3. Establish a temperature database: Use the corrected finite element model to calculate the temperature of the ingot homogenization process within the ranges of homogenization furnace gas temperature T±10%, homogenization time t±20%, aluminum alloy ingot length L±50%, aluminum alloy ingot width W±50%, and aluminum alloy ingot thickness H±50% for different grades of aluminum alloy ingots. Specifically, the homogenization furnace gas temperature T±10% ranges from 300 to 600 °C, the homogenization time t±20% refers to a time of 0 to 50 h, the aluminum alloy ingot length L±50% refers to a length of 1 to 10 m, the aluminum alloy ingot width W±50% refers to a width of 1 to 3 m, and the aluminum alloy ingot thickness H±50% refers to a thickness of 100 to 800 mm. Use the simulation calculation results to establish a temperature database between the homogenization parameters and the core and surface temperatures of the ingot; The temperature data in the temperature database includes: the core temperature, surface temperature, core and surface temperature difference, time to reach temperature, etc. of the aluminum alloy ingot;

[0028] S4. Establish a machine learning model: According to the temperature database of the ingot homogenization process, establish a rapid prediction model between the homogenization parameters and the core and surface temperatures of the ingot through machine learning;

[0029] S5. Input the homogenization furnace gas temperature T, homogenization time t, aluminum alloy ingot length L, aluminum alloy ingot width W, and aluminum alloy ingot thickness H into the rapid prediction model. The model quickly predicts through calculation and outputs the relationship curve between the ingot surface temperature and the homogenization time, the relationship curve between the ingot core temperature and the homogenization time, and the ingot time to reach temperature.

[0030] Example 1

[0031] A rapid prediction method for the homogenization temperature of aluminum alloy ingots, comprising the following steps:

[0032] S1. The heat treatment equipment used is a vertical homogenization heat treatment furnace. The temperature measurement area is divided into 6 zones, and 5 ingots can be placed in each zone. The temperature of each zone can be controlled independently. The heat treatment material is 3104 aluminum alloy. The ingot has a thickness of 400 mm, a width of 1200 mm, a length of 5000 mm, a thermal conductivity of 126 W / m℃, a specific heat capacity of 998 J / kg℃, and a density of 2670 kg / m 3 . According to the structure and heating characteristics of the vertical pusher furnace, the ingots are placed vertically in the soaking furnace. The heating rate of the bottom surface of the ingot is the fastest, and the heating temperature of the top surface is the slowest. The heat transfer coefficient of the bottom surface is set to 50 W / m 2 ℃, and the heat transfer coefficient of the top surface is set to 30 W / m 2 ℃. The middle area decreases linearly. The furnace gas temperature is set to 550℃, and the heating time is 20 hours. The geometric model size is the same as that of the ingot.

[0033] S2. Based on the finite element model established in step S1, a series of temperature calculations are carried out. The furnace gas temperatures are set to 500, 510, 520, 530, 540, 550, 560℃ respectively, the heating times are set to 16, 18, 20, 22, 24 hours respectively, the ingot lengths are set to 4000, 4500, 5000, 5500, 6000 mm respectively, the ingot widths are set to 1000, 1200, 1400, 1600, 1800, 2000 mm respectively, and the ingot thicknesses are set to 400, 450, 500, 550, 600 mm respectively. A total of 28 groups of ingot temperature calculations are carried out to obtain the surface temperature, core temperature, surface-core temperature difference, and temperature arrival time data of the ingot under the corresponding parameters, and a database is established.

[0034] S3. Data analysis is carried out on the database established in step S2, and a machine learning model for temperature prediction is established. According to the input information such as furnace gas temperature, ingot length, ingot width, ingot thickness, and ingot grade, data such as the surface temperature, core temperature, and temperature arrival time of the ingot are calculated. The change curve of the ingot temperature over time is as Figure 2 shown (furnace gas temperature 560℃).

[0035] The method of the present invention realizes the rapid prediction of the homogenization temperature of aluminum alloy ingots by integrating experimental testing, numerical simulation, and machine learning technologies, providing strong support for the intelligent production of aluminum alloy ingots. By predicting and monitoring the ingot temperature in real time, problems in the production process can be discovered and solved in a timely manner, improving production efficiency and product quality.

[0036] The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for quickly predicting the homogenization temperature of an aluminum alloy ingot, characterized in that: The following steps are involved: S1, aluminum alloy ingot temperature test: contact thermocouples are set on the surface of aluminum alloy ingots of different grades, and the temperature T of a certain test position on the surface of the aluminum alloy ingot is detected by the contact thermocouples. 测 ; S2, establish a finite element model of the homogenization heat treatment process of aluminum alloy ingots: determine the geometric model according to the size of the aluminum alloy ingots, determine the physical parameters of the model according to the aluminum alloy grade or composition, and determine the boundary conditions of the model according to the homogenization heating equipment; use finite element simulation software to obtain the temperature data of the aluminum alloy ingots, and extract the temperature data T at the same position as the test position described in step S1 in the simulation results. 仿 , according to the test temperature T 测 And the simulation temperature data T 仿 Modifying the simulation model to obtain a modified finite element model; S3, establishing a temperature database: using the modified finite element model, performing temperature simulation calculations on the homogenization process of aluminum alloy ingots of different grades, and establishing a temperature database between homogenization parameters and the core and surface temperatures of the ingots; S4, establishing a machine learning model: according to the ingot homogenization process temperature database, establishing a rapid prediction model between homogenization parameters and the ingot core and surface temperatures through machine learning; S5, quickly predicting the temperature parameters of the aluminum alloy ingot during the homogenization process according to the rapid prediction model.

2. The method for rapidly predicting the homogenization temperature of an aluminum alloy ingot according to claim 1, characterized in that: Step S2 determines the boundary conditions of the model according to the structure of the homogenizing heating equipment, furnace gas temperature, air flow direction, and air flow velocity.

3. The method for rapidly predicting the homogenization temperature of an aluminum alloy ingot according to claim 1, characterized in that: The temperature data in the temperature database in step S3 include: the core temperature, the surface temperature, the temperature difference between the core and the surface, and the time to reach the temperature of the aluminum alloy ingot.

4. The method for rapidly predicting the homogenization temperature of an aluminum alloy ingot according to claim 1, characterized in that: The temperature simulation calculation of the ingot homogenization process described in step S3 includes: carrying out the ingot homogenization process temperature calculation within the range of homogenization furnace gas temperature T±10%, homogenization time t±20%, aluminum alloy ingot length L±50%, aluminum alloy ingot width W±50%, and aluminum alloy ingot thickness H±50%.

5. The method for rapidly predicting the homogenization temperature of an aluminum alloy ingot according to claim 4, characterized in that: The homogenization parameters in steps S3 and S4 include the homogenization furnace gas temperature T, the homogenization time t, the length L of the aluminum alloy ingot, the width W of the aluminum alloy ingot, and the thickness H of the aluminum alloy ingot.

6. The method for rapidly predicting the homogenization temperature of an aluminum alloy ingot according to any one of claims 1 to 5, characterized in that: The temperature parameters of the aluminum alloy ingot during the homogenization process in step S5 include a curve of the relationship between the surface temperature of the ingot and the homogenization time, a curve of the relationship between the core temperature of the ingot and the homogenization time, and the time it takes for the ingot to reach temperature.

7. The method for rapidly predicting the homogenization temperature of an aluminum alloy ingot according to claim 6, characterized in that: Step S5 is a method for quickly predicting the temperature parameters of the aluminum alloy ingot during the homogenization process: inputting the homogenization furnace gas temperature T, the homogenization time t, the aluminum alloy ingot length L, the aluminum alloy ingot width W, and the aluminum alloy ingot thickness H into the rapid prediction model, and the model quickly predicts the temperature parameters of the aluminum alloy ingot through calculation.

Images