A method and equipment for preparing the microstructure of a high-strength and high-conductivity aluminum alloy

Abstract

The present invention belongs to the technical field of aluminum alloy heat treatment, and specifically discloses a method and equipment for preparing the microstructure of a high-strength and high-conductivity aluminum alloy. It comprehensively utilizes acoustic emission technology and conductivity detection to conduct on-line detection of an improved multi-stage solution-aging heat treatment process for aluminum alloy. A large amount of on-line detection data is obtained through experiments, and data processing and classification are carried out in combination with existing research to establish a database, and an on-line discrimination model for the microstructure state is established by comprehensively using an artificial neural network. The on-line monitoring of the microstructure state of the aluminum alloy during the heat treatment process is carried out by using this model, and the heat treatment process is adjusted in a timely manner according to the microstructure state and the purpose of each heat treatment process, so as to obtain a microstructure of high-strength and high-conductivity aluminum alloy with excellent comprehensive performance, and maximize the optimization effect of multi-stage solution-aging on the performance of aluminum alloy.

Description

Technical Field

[0001] The present invention belongs to the technical field of heat treatment of aluminum alloys, and more specifically, relates to a method and equipment for preparing the microstructure of a high-strength and high-conductivity aluminum alloy. Background Art

[0002] In recent years, in addition to the advantages of low density, high specific strength, and easy processing of conventional aluminum alloys, Al-Zn-Mg-Cu aluminum alloys also have ultra-high strength and are widely used in industries such as aerospace and vehicle equipment as one of the main structural materials. During the processing and production of Al-Zn-Mg-Cu aluminum alloy parts, the heat treatment process will have complex effects on the microstructure, thereby affecting the performance. There is a problem that it is difficult to simultaneously improve the strength, corrosion resistance, and toughness of Al-Zn-Mg-Cu aluminum alloys. When the strength is increased, problems such as a decrease in corrosion resistance and toughness, that is, low electrical conductivity, are likely to occur. In order to solve the problem of simultaneously improving the strength, corrosion resistance, and toughness of Al-Zn-Mg-Cu aluminum alloys and improve the application level, some heat treatment methods represented by three-stage aging have emerged. These heat treatment methods have improved the low corrosion resistance problem of the microstructure of high-strength aluminum alloys and ensured a certain strength. However, there are many process parameters in the three-stage aging process, and the process parameters have complex effects on the microstructure. It is difficult to obtain the optimal heat treatment process route, and the optimal heat treatment process routes for aluminum alloys of different grades and different sizes are not the same. This makes it very difficult to improve the comprehensive performance of Al-Zn-Mg-Cu aluminum alloys and obtain the microstructure of high-strength and high-conductivity aluminum alloys.

[0003] Performing cryogenic treatment before aging is considered to have a "pre-aging" effect, that is, it can promote the precipitation of precipitates during the aging process. A large number of studies have shown that the strength and conductivity of Al-Zn-Mg-Cu aluminum alloys are closely related to the state and distribution of precipitates in the microstructure, but it is difficult to control and characterize the state of precipitates. It is very difficult to obtain the optimal heat treatment parameters solely by adjusting the heat treatment parameters through experimental methods, and it is also very difficult to monitor the microstructure state through existing on-line detection technologies. There is no on-line detection technology specifically for detecting the microstructure state during the heat treatment process. During the aging process of Al-Zn-Mg-Cu aluminum alloys, microstructure evolutions such as the precipitation, growth, and re-dissolution of strengthening phases will occur. The behavior of strengthening phases will cause changes in the electrical conductivity of aluminum alloys, but there is no strong corresponding relationship between the electrical conductivity and the behavior of strengthening phases. When materials undergo microstructure evolutions such as phase changes, acoustic emissions will occur, and the acoustic signals can be detected through acoustic emission probes. There is a strong corresponding relationship between the generation of acoustic signals and the behavior of strengthening phases, but there are many acoustic signal parameters, and it is difficult to distinguish different strengthening phase behaviors based solely on the acoustic signal characteristics.

[0004] Based on this, how to improve and monitor the heat treatment process so as to effectively regulate the state of precipitates in the microstructure to prepare a high-strength and high-conductivity aluminum alloy microstructure has become an urgent problem to be solved. Summary of the Invention

[0005] In view of the above defects or improvement requirements of the prior art, the present invention provides a method and equipment for preparing a high-strength and high-conductivity aluminum alloy microstructure, aiming to monitor the microstructure state in real time during the heat treatment process, so as to timely adjust the heat treatment process to obtain a high-strength and high-conductivity aluminum alloy microstructure.

[0006] To achieve the above object, according to one aspect of the present invention, a method for preparing a high-strength and high-conductivity aluminum alloy microstructure is proposed, which includes a model construction stage and a microstructure preparation stage, wherein:

[0007] The model construction stage includes:

[0008] Carry out aluminum alloy heat treatment experiments, collect online detection data during the heat treatment process to establish a database, where each piece of data includes heat treatment type, heat treatment time, conductivity value, acoustic signal, and strengthening phase behavior, and the acoustic signal includes acoustic signal amplitude record, acoustic signal frequency record, acoustic signal energy record, and acoustic signal source type;

[0009] Construct an online discrimination model for microstructure state, which includes a convolutional neural network model and a deep neural network model; use the acoustic signal amplitude record, acoustic signal frequency record, and acoustic signal energy record as inputs, and the acoustic signal source type as the output to train the convolutional neural network model; use the heat treatment type, heat treatment time, conductivity, and acoustic signal source type as inputs, and the strengthening phase behavior as the output to train the deep neural network model, so as to obtain a trained online discrimination model for microstructure state;

[0010] The microstructure preparation stage includes:

[0011] Carry out heat treatment on the aluminum alloy, collect the heat treatment type, heat treatment time, conductivity value, and acoustic signal in real time, and input them into the trained online discrimination model for microstructure state; among them, the convolutional neural network model obtains the acoustic signal source type according to the acoustic signal amplitude record, acoustic signal frequency record, and acoustic signal energy record, and then the deep neural network model obtains the strengthening phase behavior according to the acoustic signal source type obtained by the convolutional neural network model, the heat treatment type, the heat treatment time, and the conductivity value;

[0012] Adjust the heat treatment process in real time according to the strengthening phase behavior and conductivity value obtained by the online discrimination model for microstructure state until a satisfactory aluminum alloy microstructure is obtained.

[0013] As a further preference, the heat treatment process is multi-stage solution aging. Specifically, after multi-stage solution treatment, three-stage aging is carried out, and cryogenic treatment is carried out before the first-stage aging and the third-stage aging.

[0014] As a further preference, in the stage of preparing the microstructure, the real-time adjustment of the heat treatment process is specifically as follows:

[0015] In the first-stage aging, when the strengthening phase is fully precipitated and begins to grow, stop this aging process; in the second-stage aging, when the dissolution of the strengthening phase is basically completed, stop this aging process; in the third-stage aging, when the strengthening phase is fully precipitated and begins to grow, and the conductivity value reaches the ideal value, stop this aging process.

[0016] As a further preference, when performing heat treatment on the aluminum alloy, the temperature of the first-stage aging is 90-110°C, the temperature of the second-stage aging is 190-200°C, and the temperature of the third-stage aging is 110-130°C.

[0017] As a further preference, the cryogenic treatment is single or multiple times of liquid nitrogen immersion.

[0018] According to another aspect of the present invention, there is provided a device for realizing the method for preparing the microstructure of the high-strength and high-conductivity aluminum alloy as described above, including a calculation control module, a device cavity, and a heating and heat preservation module, a cryogenic module, and an on-line detection module arranged in the device cavity:

[0019] The heating and heat preservation module is used to form a uniform and stable temperature field in the device cavity;

[0020] The cryogenic module is used to form a stable liquid nitrogen liquid level during the cryogenic heat treatment process;

[0021] The on-line detection module includes an acoustic emission signal detection component and a conductivity detection component, which are respectively used to detect the acoustic signal and conductivity during the heat treatment process;

[0022] The calculation control module is used to receive the signals detected by the on-line detection module, and output control signals based on the strengthening phase behavior obtained by the on-line discrimination model of the microstructure state, and control the heating and heat preservation module and the cryogenic module to work, so as to autonomously and intelligently control the entire heat treatment process.

[0023] As a further preference, the cryogenic module includes a cryogenic tank, a bottom connecting frame, a liquid nitrogen pipeline, a liquid level sensor, and a self-pressurizing liquid nitrogen tank, wherein:

[0024] The cryogenic tank is connected to the bottom of the equipment cavity through a bottom connecting frame, which is a telescopic or detachable structure to realize the lifting or disassembly of the cryogenic tank; the liquid nitrogen pipeline is used to introduce liquid nitrogen from the self-pressurizing liquid nitrogen tank into the cryogenic tank; the liquid level sensor is installed on the upper part of the cryogenic tank and is higher than the upper surface of the aluminum alloy, and is used to detect the liquid level height so as to control the introduction of liquid nitrogen.

[0025] As a further preference, the heating and heat preservation module includes heating rods, fans and thermocouples, wherein:

[0026] Multiple heating rods are evenly distributed oppositely along the first direction on the side wall of the equipment cavity, and single or multiple fans are distributed along the second direction on the side wall of the equipment cavity. The first direction is perpendicular to the second direction, and the side wall where the heating rods and the fans are located is adjacent; the thermocouple is installed above the equipment cavity and close to the aluminum alloy.

[0027] As a further preference, the acoustic emission signal detection component includes high-temperature acoustic emission sensors and low-temperature acoustic emission sensors. Multiple high-temperature acoustic emission sensors are fixed on the top of the equipment cavity and are in direct contact with the aluminum alloy to be heat-treated during the aging heat treatment process; multiple low-temperature acoustic emission sensors are fixed on the bottom of the cryogenic tank and are in direct contact with the aluminum alloy to be heat-treated during the cryogenic heat treatment process; the conductivity detection component includes an eddy current conductivity meter and a conductivity probe supporting it, and the conductivity probe is fixed on the top of the equipment cavity.

[0028] Generally speaking, compared with the prior art by the above technical solutions conceived by the present invention, the following technical advantages are mainly possessed:

[0029] 1. The present invention obtains online detection data through experiments and combines an artificial neural network to establish an online discrimination model for the microstructure state. Using this model, the precipitation, growth and dissolution processes of strengthening phases are automatically discriminated in real time during the heat treatment process of aluminum alloy, the microstructure state of aluminum alloy is monitored in real time, and the heat treatment process is adjusted in time according to the microstructure state and conductivity value, so as to obtain a high-strength and high-conductivity aluminum alloy microstructure with excellent comprehensive performance, maximizing the optimization effect of multi-stage solution aging on the performance of aluminum alloy.

[0030] 2. The present invention conducts online detection on the heat treatment process of aluminum alloy through acoustic emission technology and conductivity detection, can realize online monitoring of the microstructure state by detecting strengthening phases, uses the intelligent method of artificial neural network to process a large amount of complex acoustic emission detection data, which is faster, more efficient and accurate than manual discrimination. The data acquisition and analysis method makes the online detection of the microstructure state accurate and reliable and can make adjustments in time during the heat treatment process. The timely adjustment of heat treatment process parameters can make the improved multi-stage solution aging heat treatment process of aluminum alloy adapt to 7XXX series aluminum alloys of different grades and shapes.

[0031] 3. The present invention comprehensively applies a convolutional neural network model and a deep neural network model; since acoustic signals are two-dimensional data, it is more accurate and efficient to process them first through a convolutional neural network. Furthermore, due to the extremely complex correspondence relationship between heat treatment parameters, online detection data, and the behavior of strengthening phases, it is more efficient and accurate to process this multi-input single-output problem through a deep neural network, thereby achieving accurate prediction of the behavior of strengthening phases.

[0032] 4. The equipment for preparing the microstructure of high-strength and high-conductivity aluminum alloy designed by the present invention integrates functional modules such as an online detection module, a heating and insulation module, a cryogenic cooling module, and a calculation and control module. It can perform cryogenic cooling and aging treatment in the same equipment and obtain a large amount of online detection data during the heat treatment process. The arrangement of the online detection module ensures reliable and effective data collection and can adapt to aluminum alloys of different shapes. It can effectively ensure the continuity and monitorability of the heat treatment process and can promptly adjust process parameters during the heat treatment process, strongly supporting the realization of the method for preparing the microstructure of high-strength and high-conductivity aluminum alloy. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic structural diagram of the main part of the equipment for preparing the microstructure of high-strength and high-conductivity aluminum alloy according to an embodiment of the present invention;

[0034] Figure 2 It is a flowchart of the method for preparing the microstructure of high-strength and high-conductivity aluminum alloy according to an embodiment of the present invention;

[0035] Figure 3 It is a schematic structural diagram of the online discrimination model for the microstructure state according to an embodiment of the present invention.

[0036] In all the drawings, the same reference numerals are used to represent the same elements or structures, where: 1 - equipment cavity, 2 - first bracket, 3 - second bracket, 4 - heating rod, 5 - thermocouple, 6 - fan, 7 - cryogenic cooling tank, 8 - bottom connecting frame, 9 - liquid level sensor, 10 - high-temperature acoustic emission sensor, 11 - low-temperature acoustic emission sensor, 12 - conductivity probe, 13 - aluminum alloy to be heat-treated. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] In order to make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0038] A method for preparing the microstructure of a high-strength and high-conductivity aluminum alloy provided by an embodiment of the present invention performs heat treatment on the aluminum alloy, monitors the state of the microstructure of the aluminum alloy during the heat treatment process, and regulates the heat treatment process. Online detection data is obtained through experiments, and an online discrimination model for the microstructure state is established in combination with an artificial neural network. This model is used to automatically and real-time discriminate the precipitation, growth, and dissolution processes of strengthening phases during the heat treatment of the aluminum alloy, monitor the state of the microstructure of the aluminum alloy in real time, adjust the heat treatment process in a timely manner according to the microstructure state and conductivity value, and terminate the heat treatment process when the ideal microstructure state and conductivity value are obtained to obtain the microstructure of the high-strength and high-conductivity aluminum alloy.

[0039] Specifically, as Figure 2 shown, the method includes the following steps:

[0040] Step S1: Perform heat treatment experiments on the aluminum alloy using a device for preparing the microstructure of a high-strength and high-conductivity aluminum alloy, and collect online detection data during the heat treatment process.

[0041] In one embodiment, heat treatment experiments are performed on 7XXX series aluminum alloys that already have mature heat treatment research. The aluminum alloy specimens are subjected to multi-stage solution treatment and then placed in a device for preparing the microstructure of a high-strength and high-conductivity aluminum alloy for heat treatment experiments to record the online detection signals during the heat treatment process.

[0042] In one embodiment, the heat treatment experiment process is mainly three-stage aging. Multi-stage solution treatment is performed before aging, and cryogenic treatment with a single liquid nitrogen immersion is performed before the first-stage aging and the third-stage aging. Preferably, during the experiment, the temperature of the first-stage aging is 90-110°C, the time is 8-12 h, the temperature of the second-stage aging is 190-200°C, the time is 1-15 min, the temperature of the third-stage aging is 110-130°C, the time is 8-24 h; the liquid nitrogen immersion time is 1-6 h.

[0043] In one embodiment, the environmental noise is detected when the device for preparing the microstructure of a high-strength and high-conductivity aluminum alloy is working but the aluminum alloy to be heat-treated is not placed in it, so as to set the threshold values under different heat treatment parameters. The setting of the threshold value should be higher than the noise during the no-load operation of the device.

[0044] In one embodiment, the online detection data collected during the heat treatment experiment process includes conductivity value records, sound signal amplitude records, sound signal frequency records, and sound signal energy records, and at the same time, the heat treatment type and heat treatment time are recorded.

[0045] Step S2: Manually classify the online detection data obtained from the experiment, correspond the online monitoring data to the behavior of the strengthening phase, and form a database.

[0046] Specifically, in combination with relevant research, the changes in the on-line detection signals are corresponded to the strengthening phase behaviors to manually classify the on-line detection data. After manual classification, the strengthening phase behaviors are corresponded to the heat treatment type, heat treatment time, conductivity value, acoustic signal amplitude record, acoustic signal frequency record, and acoustic signal energy record.

[0047] In one embodiment, the strengthening phase behaviors are obtained through direct observation during the test process.

[0048] In one embodiment, the increase in the conductivity value is corresponded to the precipitation and growth of the strengthening phase, and the decrease in the conductivity value is corresponded to the re-solution of the strengthening phase. The precipitation and growth behaviors are separated by the magnitude of the conductivity value change rate and the acoustic signal characteristics, and the degree of growth and re-solution is judged.

[0049] In one embodiment, when an acoustic signal exceeding the threshold value is detected during the heat treatment process of the aluminum alloy, it is considered that the microstructure of the heat-treated aluminum alloy has changed. In combination with relevant research, the conductivity value record, and the acoustic signal category, different parts (divided by time nodes) of the acoustic signal amplitude record, acoustic signal frequency record, and acoustic signal energy record are corresponded to the strengthening phase behaviors.

[0050] Step S3: Extract the acoustic signal data in the database for further classification, correspond the acoustic signal data to the acoustic signal source type to form an acoustic signal database, and train a convolutional neural network model.

[0051] In one embodiment, different parts of the acoustic signal amplitude record, acoustic signal frequency record, and acoustic signal energy record corresponding to the strengthening phase behaviors are further classified. According to the different acoustic signal types, these acoustic signals are divided into two major categories: burst type and continuous type. Combining the precipitation phase behaviors and acoustic signal characteristics, the burst type acoustic signals are subdivided into three levels, and the continuous type acoustic signals are subdivided into five levels.

[0052] In one embodiment, the data form of the acoustic signal record is a picture directly output by the acoustic signal processing software. The content of the picture is a two-dimensional scatter plot, with the horizontal axis being time and the vertical axis being the acoustic signal amplitude or acoustic signal frequency or acoustic signal energy. That is, the input data of the convolutional neural network model is the scatter plot of the acoustic signal amplitude record, acoustic signal frequency record, and acoustic signal energy record, and the output of the neural network model is the acoustic signal source type.

[0053] In one embodiment, there are 13 convolutional layers and 5 pooling layers in the convolutional neural network model. The size of all convolutional kernels is 3, and the activation function is ReLU. A large number of iterative trainings are carried out using 90% of the acoustic signal database, and the model accuracy is verified using the other 10% of the data. When the accuracy rate reaches 95%, the optimal model is retained.

[0054] Step S4: Train the deep neural network model in the online discrimination model of the microstructure state by calling the trained convolutional neural network model using the database, so as to establish the online discrimination model of the microstructure state.

[0055] Build the framework of the online discrimination model of the microstructure state, as Figure 3 shown, that is, a multi-input single-output program with a deep neural network model as the main body and capable of calling the convolutional neural network model established in step S3. The inputs of the program include at least the heat treatment type, heat treatment time, conductivity value, conductivity change rate, acoustic signal amplitude record, acoustic signal frequency record, and acoustic signal energy record, and the output is the strengthening phase behavior. Among the data input into the program, the heat treatment type, heat treatment time, and conductivity value are directly used as the inputs of the deep neural network, and the acoustic signal amplitude record, acoustic signal frequency record, and acoustic signal energy record are used as the inputs of the convolutional neural network model established in step S3. The type of acoustic signal source is obtained through the convolutional neural network model and input into the deep neural network. The inputs of the deep neural network model are the heat treatment type, heat treatment time, conductivity value, and acoustic signal source type, and the output is the strengthening phase behavior.

[0056] In one embodiment, the inputs of the deep neural network model are the heat treatment type, heat treatment time, conductivity value, and acoustic signal source type, and the output is the strengthening phase behavior. 80% of the database data is used for model training. During the training process, the optimal number of hidden layers is found to achieve a balance between accuracy and efficiency. The number of hidden layers is not less than 3. After training, the other 20% of the data is used for model verification. When the prediction accuracy of the online discrimination model of the microstructure state reaches 95%, the deep neural network model is considered to be accurate and reliable.

[0057] Step S5: Perform aluminum alloy heat treatment using the high-strength and high-conductivity aluminum alloy microstructure preparation equipment, and monitor the heat treatment process in real time through the online discrimination model of the microstructure state, and adjust the heat treatment process in a timely manner to obtain an ideal aluminum alloy microstructure.

[0058] In one embodiment, the aluminum alloy to be heat-treated is heat-treated in the high-strength and high-conductivity aluminum alloy microstructure preparation equipment, and the heat treatment process route is multi-stage solution treatment + single-time liquid nitrogen immersion + first-stage aging + second-stage aging + three-time liquid nitrogen immersion + third-stage aging. The time for single-time liquid nitrogen immersion is 2 - 6h, and the time for each of the three-time liquid nitrogen immersions is 1 - 2h. After each liquid nitrogen immersion, it stays at room temperature for 30min.

[0059] In one embodiment, the heat treatment process is adjusted specifically by adjusting the time of each stage of aging. The aging time is adjusted according to the purpose of each stage of aging and the judgment of the microstructure state by the on-line discrimination model of the microstructure state. In the first-stage aging, the aging process is stopped when the strengthening phase has fully precipitated and begins to grow; in the second-stage aging, the aging process is stopped when the dissolution of the strengthening phase is basically completed; in the third-stage aging, the aging process is stopped when the strengthening phase has fully precipitated and begins to grow and the conductivity value reaches the ideal value.

[0060] The main part of the preparation equipment for the microstructure of high-strength and high-conductive aluminum alloy for implementing the above method is as Figure 1 shown. The equipment includes an equipment cavity 1, a heating and heat preservation module, a cryogenic module, an on-line detection module, and a calculation and control module, where:

[0061] The equipment cavity 1 is used for heat preservation during the heat treatment process, fixing and supporting the functional modules, the aluminum alloy to be heat treated, and heat insulation. Specifically, in the equipment cavity 1, a first bracket 2 and a second bracket 3 are respectively provided at its lower part and upper part. The second bracket 3 is used for fixing and supporting the high-temperature acoustic emission sensor and the conductivity probe in the on-line detection module, and the first bracket 2 is used for fixing the aluminum alloy 13 to be heat treated; an interface is provided at the bottom of the equipment cavity 1 to be connected to the bottom connecting frame in the cryogenic module, and pre-installed holes or grooves are provided on the side wall for installing the heating rods, thermocouples, and fans in the heating and heat preservation module.

[0062] The heating and heat preservation module is used to form a uniform and stable temperature field in the cavity. It includes heating rods 4, thermocouples 5, and fans 6. Among them, multiple heating rods 4 are uniformly distributed oppositely along the first direction on the side wall of the equipment cavity 1, and single or multiple fans 6 are distributed along the second direction on the side wall of the equipment cavity 1. The first direction is perpendicular to the second direction, and multiple fans can be located on the same side wall or distributed oppositely; the thermocouple 5 is located at the upper part of the equipment cavity 1 near the aluminum alloy 13 to be heat treated.

[0063] The cryogenic module is used to form a stable liquid nitrogen liquid level during the cryogenic heat treatment process. It includes a cryogenic tank 7, a liquid nitrogen pipeline, a bottom connecting frame 8, a liquid level sensor 9, and a self-pressurizing liquid nitrogen tank. Among them, multiple brackets are provided in the middle of the side wall of the cryogenic tank 7 for fixing and supporting the low-temperature acoustic emission sensor, and multiple brackets at different heights are provided on the upper part of the side wall of the cryogenic tank 7 for fixing and supporting the liquid level sensor 9. The bottom connecting frame 8 is installed and fixed on the outside of the bottom of the cryogenic tank 7 and connected to the bottom of the equipment cavity 1; the self-pressurizing liquid nitrogen tank passes liquid nitrogen into the cryogenic tank 7 from the bottom through the liquid nitrogen pipeline; the liquid level sensor 9 is located higher than the aluminum alloy to be heat treated and is used to detect the liquid level height in the cryogenic tank 7, so as to control the liquid nitrogen input situation and ensure that the aluminum alloy is immersed in liquid nitrogen.

[0064] The on-line detection module is used to detect acoustic emission signals and conductivity. The detection of acoustic emission signals is realized by a high-temperature acoustic emission sensor 10, a low-temperature acoustic emission sensor 11, a preamplifier and a multi-channel acoustic emission system, and the detection of conductivity is realized by an eddy current conductivity meter and a conductivity probe 12 supporting it. The high-temperature acoustic emission sensor 10 is fixed on the top of the equipment cavity and is in direct contact with the aluminum alloy to be heat-treated during the aging heat treatment process; a plurality of low-temperature acoustic emission sensors 11 are fixed on the bottom of the cryogenic tank and are in direct contact with the aluminum alloy to be heat-treated during the cryogenic heat treatment process; the conductivity probe 12 is fixed on the top of the equipment cavity and is close to the aluminum alloy to be heat-treated during the heat treatment process.

[0065] The calculation and control module includes an industrial control computer and a PLC, which establish signal connections with each module through data cables and receive on-line detection signals during the heat treatment process. The on-line discrimination model of the microstructure state is stored in the industrial control computer, which can automatically analyze and process on-line detection data and output control signals according to the microstructure state to control the cryogenic and aging parameters and then control the heat treatment process.

[0066] In an embodiment, the number of high-temperature acoustic emission sensors is 3, the number of conductivity probes is 2, and the number of low-temperature acoustic emission sensors is 4. The number of heating rods is 8, and 4 heating rods are evenly distributed on each of the two side walls of the equipment cavity along the first direction. The number of fans is 2, and 1 fan is on each of the two side walls of the equipment cavity along the second direction. The first direction is perpendicular to the second direction. The number of brackets for fixing and supporting the liquid level sensor on the upper part of the side wall of the cryogenic tank is 3, which are located at different heights to adapt to aluminum alloy samples to be processed at different heights.

[0067] In the above-mentioned equipment for preparing the microstructure of high-strength and high-conductive aluminum alloy, the specific number and distribution of high-temperature acoustic emission sensors, conductivity probes, low-temperature acoustic emission sensors, heating rods, fans, etc. can provide a uniform and accurate temperature field during the heat treatment process of aluminum alloy, collect a large amount of effective on-line detection data, and have strong adaptability to the shape and size of the aluminum alloy to be heat-treated, and can realize two heat treatments of cryogenic and aging in the same equipment.

[0068] The following is further illustrated by specific embodiments.

[0069] The heat-treated aluminum alloy grade is 7085, the target conductivity is 40% IACS, and the aluminum alloy grade for establishing the online discrimination model of the microstructure state is 7050. The reference alloy grades are: 7055, 7075, and the reference heat treatment methods are: three-stage aging, two-stage aging. First, use the preparation equipment for the microstructure of high-strength and high-conductivity aluminum alloy to conduct heat treatment experiments on 7050 aluminum alloy, mainly including three-stage aging and cryogenic treatment after multi-stage solution treatment. During the experiment, online detection data is collected, and then the online detection data is manually classified to form a database, and the database is used to train the acoustic signal processing model and the online discrimination model of the microstructure state. After the online discrimination model of the microstructure state is established, heat treatment of 7085 aluminum alloy is carried out, and the model is used to automatically analyze and process the online detection signal to monitor the heat treatment process in real time, judge the behavior of the precipitation phase, and terminate the heat treatment process when the ideal aluminum alloy microstructure is obtained, and finally obtain the high-strength and high-conductivity aluminum alloy microstructure.

[0070] The high-strength and high-conductivity aluminum alloy microstructure is prepared according to the following method in this embodiment:

[0071] (1) Use a 7050 aluminum alloy block with a suitable size for heat treatment experiments and collect online detection data during the heat treatment process. First, detect the ambient noise when the preparation equipment for the microstructure of high-strength and high-conductivity aluminum alloy is working but without putting the aluminum alloy to be heat-treated. The threshold value during aging is set to 45 dB, which is higher than the noise during the no-load operation of the equipment. Install the online detection module after putting the 7050 aluminum alloy block into the preparation equipment, and test whether the conductivity detection is normal, and use the "broken lead" experiment to test whether the acoustic emission detection is normal. After the preparation equipment is debugged, a heat treatment experiment with a process route of soaking in liquid nitrogen for 6 h, first-stage aging at 100 °C for 12 h, second-stage aging at 200 °C for 15 min, soaking in liquid nitrogen for 6 h, and third-stage aging at 120 °C for 24 h is carried out, and online detection data is collected. Among them, the data sampling interval is 1 min except for the second-stage aging, and the data sampling interval for the second-stage aging is 1 s.

[0072] (2) Manually classify the online detection data obtained from the experiment, correspond the online monitoring data with the behavior of the strengthening phase, and form a database. Combine relevant research to classify the online detection data into four types: no change in the strengthening phase, precipitation, growth, and re-dissolution of the strengthening phase. Each piece of data after classification includes heat treatment type, heat treatment time, conductivity value, acoustic signal amplitude record, acoustic signal frequency record, acoustic signal energy record, and strengthening phase behavior. All data entries constitute a database, with a total of 3500 valid data entries.

[0073] (3) Extract the acoustic signal data from the database for further classification, correspond the acoustic signal data with the acoustic signal source type to form an acoustic signal database and train the convolutional neural network model. Extract all the acoustic signal data from the database, manually judge the acoustic signal source type at each time node, and the types are divided into background noise, first-level burst type, second-level burst type, third-level burst type, first-level continuous type, second-level continuous type, third-level continuous type, fourth-level continuous type and fifth-level continuous type. Establish an acoustic signal database. Each data includes the acoustic signal amplitude record, acoustic signal frequency record, acoustic signal energy record and acoustic signal source type, totaling 2500 valid data. Establish a convolutional neural network model, which has 13 convolutional layers and 5 pooling layers. The size of all convolution kernels is 3, and the activation function is ReLU. Take the acoustic signal data as input and the acoustic signal source type as output to train the convolutional neural network model. Use 90% of the acoustic signal database for iterative training, and use the other 10% of the data to verify the model accuracy. When the number of iterations exceeds 3000 times, the prediction accuracy reaches 96.3%, and the model is retained.

[0074] (4) By calling the trained convolutional neural network model, the deep neural network model in the online discrimination model of microstructure state is trained using the database to establish the online discrimination model of microstructure state. The input of the online discrimination model of microstructure state is the heat treatment type, heat treatment time, conductivity value, acoustic signal amplitude record, acoustic signal frequency record, and acoustic signal energy record, and the output is the enhanced phase behavior. Among the input data, the heat treatment type, heat treatment time, and conductivity value are directly used as the input of the deep neural network, and the acoustic signal amplitude record, acoustic signal frequency record, and acoustic signal energy record are used as the input of the convolutional neural network model established in step (3). The acoustic signal source type is obtained through the convolutional neural network and input into the deep neural network. After the deep neural network obtains all the input quantities, it outputs the enhanced phase behavior. The input of the deep neural network model is the heat treatment type, heat treatment time, conductivity value, and acoustic signal source type, and the output is the enhanced phase behavior. The deep neural network model is trained using the data in the database during the operation of the online discrimination model program of microstructure state. 80% of the database data was used for model training. After adjusting the hidden layer and the number of hidden layer neurons, the network prediction output prediction accuracy of the test sample reached 97.1% when the number of hidden layers was 5 and the number of hidden layer neurons was 13. It was considered that the deep neural network model was established. At this point, the online discrimination model of microstructure state was established.

[0075] (5) The heat treatment of 7085 aluminum alloy blocks is carried out using a microstructural preparation equipment for high-strength and high-conductivity aluminum alloy. The heat treatment process is monitored in real time through an on-line discrimination model of the microstructural state, and the heat treatment process is adjusted in a timely manner to obtain an ideal aluminum alloy microstructure. Place the 7085 aluminum alloy block in the preparation equipment and adjust the installation of the on-line detection module. Repeat the steps before the heat treatment test of the 7050 aluminum alloy block. The heat treatment process route parameters are single-time liquid nitrogen immersion for 2 h, primary aging at 100 °C, secondary aging at 200 °C, three-time liquid nitrogen immersion for 1 h, and tertiary aging at 120 °C. The specific aging time is adjusted according to the real-time monitoring results. During the heat treatment process, the precipitation of strengthening phases starts at 10 h during the primary aging, and the precipitation basically ends at 11.5 h and the growth of strengthening phases starts. At this time, the primary aging is stopped; the dissolution of strengthening phases starts at 1 min during the secondary aging, and the dissolution basically ends at 8 min and the growth of undissolved strengthening phases starts. At this time, the secondary aging is stopped; the precipitation of strengthening phases starts at 13 h during the tertiary aging. At this time, the conductivity is 37.2%, and the precipitation behavior basically ends at 15 h and the growth of strengthening phases starts. At this time, the conductivity is 39.6%. The tertiary aging process continues. When the strengthening phases further grow for 0.5 h, the conductivity reaches 40%. At this time, the tertiary aging is stopped and the heat treatment ends, obtaining a 7085 aluminum alloy block with a high-strength and high-conductivity microstructure.

[0076] Through the above preparation method, a 7085 aluminum alloy block with a high-strength and high-conductivity microstructure is obtained. The best heat treatment process route obtained is single-time liquid nitrogen immersion for 2 h, primary aging at 100 °C for 11.5 h, secondary aging at 200 °C for 8 min, three-time liquid nitrogen immersion for 1 h, and tertiary aging at 120 °C for 15.5 h.

[0077] It is easy for those skilled in the art to understand that the above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for preparing the microstructure of a high-strength and high-conductivity aluminum alloy, characterized in that, it includes a model construction stage and a microstructure preparation stage, wherein: The model construction stage includes: Conducting heat treatment tests on the aluminum alloy, collecting online detection data during the heat treatment process to establish a database, where each piece of data includes heat treatment type, heat treatment time, conductivity value, acoustic signal, and strengthening phase behavior, and the acoustic signal includes acoustic signal amplitude record, acoustic signal frequency record, acoustic signal energy record, and acoustic signal source type; Constructing an online discrimination model for the microstructure state, which includes a convolutional neural network model and a deep neural network model; using the acoustic signal amplitude record, acoustic signal frequency record, and acoustic signal energy record as inputs, and the acoustic signal source type as the output to train the convolutional neural network model; using the heat treatment type, heat treatment time, conductivity, and acoustic signal source type as inputs, and the strengthening phase behavior as the output to train the deep neural network model, so as to obtain a trained online discrimination model for the microstructure state; The microstructure preparation stage includes: Conducting heat treatment on the aluminum alloy, and real-time collecting the heat treatment type, heat treatment time, conductivity value, and acoustic signal, and inputting them into the trained online discrimination model for the microstructure state; among them, the convolutional neural network model obtains the acoustic signal source type based on the acoustic signal amplitude record, acoustic signal frequency record, and acoustic signal energy record, and then the deep neural network model obtains the strengthening phase behavior based on the acoustic signal source type obtained by the convolutional neural network model, as well as the heat treatment type, heat treatment time, and conductivity value; According to the strengthening phase behavior and conductivity value obtained by the online discrimination model for the microstructure state, the heat treatment process is adjusted in real time until the aluminum alloy microstructure that meets the requirements is obtained.

2. The method for preparing the microstructure of a high-strength and high-conductivity aluminum alloy according to claim 1, characterized in that, The heat treatment process is multi-stage solution aging, specifically, three-stage aging is carried out after multi-stage solution, and cryogenic treatment is carried out before the first-stage aging and the third-stage aging.

3. The method for preparing the microstructure of a high-strength and high-conductivity aluminum alloy according to claim 2, characterized in that, In the microstructure preparation stage, the real-time adjustment of the heat treatment process is specifically: In the first-stage aging, when the strengthening phase has fully precipitated and begins to grow, stop this aging process; in the second-stage aging, when the re-dissolution of the strengthening phase is basically completed, stop this aging process; in the third-stage aging, when the strengthening phase has fully precipitated and begins to grow, and the conductivity value reaches the ideal value, stop this aging process.

4. The method for preparing the microstructure of a high-strength and high-conductivity aluminum alloy according to claim 2, characterized in that, When conducting heat treatment on the aluminum alloy, the temperature of the first-stage aging is 90 - 110 °C, the temperature of the second-stage aging is 190 - 200 °C, and the temperature of the third-stage aging is 110 - 130 °C.

5. The method for preparing the microstructure of a high-strength and high-conductivity aluminum alloy according to claim 2, characterized in that, The cryogenic treatment is single or multiple immersions in liquid nitrogen.

6. A device for implementing the method for preparing the microstructure of a high-strength and high-conductivity aluminum alloy according to any one of claims 1 - 5, characterized in that, It includes a calculation and control module, a device cavity (1), and a heating and heat preservation module, a cryogenic module, and an on-line detection module arranged in the device cavity (1): The heating and heat preservation module is used to form a uniform and stable temperature field in the device cavity (1); The cryogenic module is used to form a stable liquid nitrogen liquid level during cryogenic heat treatment; The on-line detection module includes an acoustic emission signal detection component and a conductivity detection component, which are respectively used to detect acoustic signals and conductivity during heat treatment; The calculation and control module is used to receive the signals detected by the on-line detection module, and output control signals based on the strengthening phase behavior obtained from the microscopic structure state on-line discrimination model, and control the operation of the heating and heat preservation module and the cryogenic module, so as to autonomously and intelligently control the entire heat treatment process.

7. The device according to claim 6, wherein, the cryogenic module includes a cryogenic tank (7), a bottom connecting frame (8), a liquid nitrogen pipeline, a liquid level sensor (9), and a self-pressurizing liquid nitrogen tank, wherein: The cryogenic tank (7) is connected to the bottom of the device cavity (1) through the bottom connecting frame (8), and the bottom connecting frame (8) is a telescopic or detachable structure to realize the lifting or disassembly of the cryogenic tank (7); the liquid nitrogen pipeline is used to introduce liquid nitrogen from the self-pressurizing liquid nitrogen tank into the cryogenic tank (7); the liquid level sensor (9) is installed on the upper part of the cryogenic tank (7) and is higher than the upper surface of the aluminum alloy, and is used to detect the liquid level height, so as to control the introduction of liquid nitrogen.

8. The device according to claim 6, wherein, the heating and heat preservation module includes heating rods (4), fans (6), and thermocouples (5), wherein: Multiple heating rods (4) are uniformly distributed oppositely along the first direction on the side wall of the device cavity, and single or multiple fans (6) are distributed along the second direction on the side wall of the device cavity. The first direction is perpendicular to the second direction, and the side wall where the heating rods (4) and the fans (6) are located is adjacent; the thermocouple (5) is installed above the device cavity and close to the aluminum alloy.

9. The device according to any one of claims 6-8, wherein, the acoustic emission signal detection component includes a high-temperature acoustic emission sensor (10) and a low-temperature acoustic emission sensor (11). Multiple high-temperature acoustic emission sensors (10) are fixed on the top of the device cavity and are in direct contact with the aluminum alloy to be heat treated during aging heat treatment; multiple low-temperature acoustic emission sensors (11) are fixed on the bottom of the cryogenic tank and are in direct contact with the aluminum alloy to be heat treated during cryogenic heat treatment; the conductivity detection component includes an eddy current conductivity meter and a conductivity probe (12) supporting it, and the conductivity probe (12) is fixed on the top of the device cavity.

Images