A method for real-time regulation of laser additive manufacturing of alloys based on online ultrasonic monitoring

Abstract

The application discloses a kind of real-time regulation and control laser additive manufacturing alloy method based on online ultrasonic monitoring, which uses ultrasonic sensor to track the back wall reflection of laser forming sample in the process of laser additive manufacturing, and can accurately reflect the generation of crack and pore in the process of laser forming by processing ultrasonic signal data, and adjusts the laser forming process parameters in real time according to feedback data, to obtain the laser additive manufacturing alloy sample without crack and pore finally.The method provides a method for real-time monitoring and regulating alloy forming quality in the process of laser additive manufacturing, which significantly improves the laser forming quality of the formed alloy.

Description

Technical Field

[0001] This invention belongs to the field of laser additive manufacturing equipment and process innovation, and relates to a real-time control method for laser additive manufacturing of alloys based on online ultrasonic monitoring. Background Technology

[0002] Laser additive manufacturing is a process that uses metal powder or filament as raw materials. High-power lasers melt the raw materials, solidify them layer by layer, and finally deposit them into a uniform and dense three-dimensional solid part. It boasts advantages such as high forming speed, no structural limitations, and high material utilization, and is currently widely used in aerospace, defense, and medical fields. However, because the laser beams used in laser additive manufacturing are mostly Gaussian beams, although they have the characteristics of small spot diameter and high energy, they also have characteristics such as rapid focus movement and uneven energy distribution. During the forming process, this can cause problems such as excessively high local temperatures and large temperature gradients in the molten pool. This can easily lead to thermal stress concentration, uneven liquid phase spreading, and the formation of metallurgical defects such as hot cracks and porosity during solidification, ultimately affecting the quality and performance of the formed parts. Summary of the Invention

[0003] Purpose of the invention: The technical problem to be solved by the present invention is to provide a real-time controlled laser additive manufacturing alloy method based on online ultrasonic monitoring to eliminate thermal cracks and pores, form dense alloy samples without cracks and pores, and improve the forming quality and performance of laser additive manufacturing alloys, in order to address the shortcomings of existing alloy laser forming technology.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] Using a laser additive manufacturing system, the forming quality of alloy samples is monitored in real time during the laser forming process. If metallurgical defects, including cracks and pores, are detected, the laser forming process parameters are adjusted in real time to eliminate the metallurgical defects, and finally a dense three-dimensional alloy solid part without metallurgical defects is produced.

[0006] Furthermore, the laser additive manufacturing system includes a laser powder bed melting and forming equipment and an ultrasonic monitoring system. The ultrasonic monitoring system includes an ultrasonic sensor, a pulse transmitter and receiver, and a data collection and processing device. The laser powder bed melting and forming equipment includes a forming substrate and a computer control system. The ultrasonic sensor is installed below the forming substrate, and the alloy sample to be formed is placed above the forming substrate.

[0007] The pulse transmitter and receiver are respectively connected to the ultrasonic sensor and the data collection and processing device. They are used to control the ultrasonic sensor to emit sound signals to the alloy sample, collect reflected sound signals that are higher than the amplitude threshold, and send the reflected sound signals that are higher than the amplitude threshold to the data collection and processing device.

[0008] The data collection and processing device is connected to the computer control system and is used to process reflected sound signals that are higher than the amplitude threshold. If a metallurgical defect is detected, the computer control system is adjusted in real time to regulate the laser forming process parameters in order to eliminate the generation of metallurgical defects in the subsequent forming process.

[0009] Furthermore, the specific steps include the following:

[0010] Step 1: Use computer software to design and build a three-dimensional solid geometric model of the target part. Use Materialise Magics software to slice the three-dimensional model and plan the laser scanning path to discretize the three-dimensional solid into two-dimensional data.

[0011] Step 2: The two-dimensional data is imported into the computer control system of the laser powder bed melting and forming equipment. The alloy powder is melted and solidified layer by layer. During the laser forming process, the forming quality of the alloy sample is monitored in real time by an ultrasonic monitoring system. If the formation of metallurgical defects, including cracks and pores, is detected based on the reflected acoustic signal data, the laser forming process parameters are adjusted in real time to eliminate the metallurgical defects. Finally, a dense three-dimensional alloy solid part without cracks and pores is prepared.

[0012] Further, step 2 includes:

[0013] Step 2.1: Obtain acoustic signal characteristics from reflected acoustic signals that are above the amplitude threshold. The acoustic signal characteristics include rise time, peak amplitude, duration, kurtosis, number of counts, energy, and frequency.

[0014] Step 2.2: Perform cluster analysis on the acoustic signal features to obtain multiple clusters;

[0015] Step 2.3: Perform principal component analysis on multiple clusters to obtain the defect types of the acoustic emission source corresponding to each cluster, wherein the defect types include cracks and pores;

[0016] Step 2.4: If the formation of metallurgical defects, including cracks and pores, is detected, the laser forming process parameters are adjusted in real time to eliminate the metallurgical defects.

[0017] Further, in step 2.1, the peak amplitude of the acoustic signal characteristics is obtained by real-time measurement by the data collection and processing device (3), and the rise time, duration, kurtosis, number of counts, energy and frequency are calculated from the reflected acoustic signal above the amplitude threshold.

[0018] Further, step 2.2 includes: performing cluster analysis using the k-means clustering algorithm. The goal of the k-means clustering algorithm is to minimize the distance between all vectors in a cluster and the cluster center, and to maximize the distance between all cluster centers. By using the contour width to find the optimal number of clusters, the cluster with the largest average contour width divides the acoustic emission events into subgroups based on the number of reaction defects. The clustering algorithm ultimately yields two clusters, where acoustic events within the same cluster are similar to each other, and acoustic events in different clusters are different. The k-means algorithm can be described as follows:

[0019] Initialize each cluster center Ci, where 1 ≤ i ≤ k, k represents the number of clusters, and k ≥ 2. Calculate the Euclidean distance between the vector acoustic signal features and the cluster center Ci, and then assign each input vector to the nearest cluster. Recalculate the cluster center position based on the closest mean, and repeat the above steps until the cluster center positions no longer change. At this point, calculate the maximum average contour width. By repeating the above steps for all possible numbers of clusters, the average contour width is maximized when k = 2, indicating that the optimal number of clusters is 2.

[0020] Further, step 2.3 includes: using principal component analysis for dimensionality reduction, which uses orthogonal transformation to convert a set of correlated variables into a set of linearly uncorrelated values, called principal components. The seven features in the acoustic emission events are components of n input mode vectors Zj (j = 1, 2, ..., n). An n×7 feature matrix Z is obtained, where n is the number of acoustic emission events. First, the feature matrix Z is standardized by subtracting the mean of each column and then dividing by the standard deviation of the column. The covariance matrix E[ZZ] is calculated. T Calculate the eigenvectors and their corresponding eigenvalues. The two basis vectors with the largest eigenvalues ​​can be used as principal components.

[0021] Principal component analysis can be used to identify different defect types in the clusters obtained after cluster analysis. The defect types of the acoustic emission source include pores and cracks. The acoustic emission signals generated by pores have short decay times and small amplitudes; the acoustic emission signals generated by cracks have short durations and large amplitudes; the energy of acoustic emission events generated by pores is higher than that generated by cracks.

[0022] Furthermore, if a crack is detected in step 2.4, the laser forming process parameters are adjusted in real time to eliminate metallurgical defects, including reducing the laser scanning speed to eliminate the crack. Appropriately reducing the laser scanning speed and extending the solidification time of the molten pool is beneficial for liquid phase reflow filling, while reducing the temperature gradient of the molten pool, reducing the accumulation of thermal stress, and thus eliminating the crack.

[0023] Furthermore, if porosity is detected in step 2.4, the laser forming process parameters are adjusted in real time to eliminate metallurgical defects, including increasing laser power and decreasing laser scanning speed to eliminate porosity. Increasing laser power increases heat input and molten pool temperature, reduces melt viscosity, and facilitates complete melting and wetting of the powder, thereby reducing the formation of unmelted pores. Decreasing laser scanning speed prolongs molten pool solidification time, which facilitates gas escape from the liquid phase, reduces the formation of pores in the molten pool, and thus reduces porosity.

[0024] Furthermore, the transmission and reception bandwidth of the ultrasonic sensor is 100–1000 kHz.

[0025] Beneficial effects:

[0026] 1. In the process of laser additive manufacturing of alloys, this invention can monitor the generation of cracks and pores during the laser additive manufacturing process by processing the acoustic signal of the back wall echo, and adjust the laser forming process parameters in real time to eliminate metallurgical defects such as cracks and pores, thus meeting the requirements of laser additive manufacturing of high forming quality alloy materials.

[0027] 2. This invention involves simple modifications to existing laser additive manufacturing equipment without damaging its original functions and structure. It is easy to operate and has low cost. Compared with the detection and analysis of alloy samples after laser forming, real-time monitoring can adjust the laser forming process parameters in real time based on cracks and pores generated during laser additive manufacturing, ensuring the forming quality and performance of the alloy samples and improving economic efficiency. Attached Figure Description

[0028] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, and the advantages of the present invention in the above and / or other aspects will become clearer.

[0029] Figure 1 This is a schematic diagram of the ultrasonic monitoring system of the present invention.

[0030] The markings in the diagram represent: 1-ultrasonic sensor, 2-pulse transmitter and receiver, 3-data collection and processing device, 4-sensor bracket, 5-bottom metal base, 6-bolt, 7-forming substrate, 8-alloy sample. Other unmodified parts of the laser powder bed melting forming equipment are not listed.

[0031] Figure 2 This is an optical image of an aluminum alloy sample formed by laser additive manufacturing in Example 1.

[0032] Figure 3 This is an optical image of a nickel-based alloy sample formed by laser additive manufacturing in Example 2.

[0033] Figure 4Optical image of the laser additive manufacturing aluminum alloy sample for Comparative Example 1.

[0034] Figure 5 Optical image of the laser additive manufacturing aluminum alloy sample for Comparative Example 2.

[0035] Figure 6 Optical image of the laser additive manufacturing aluminum alloy sample for Comparative Example 3. Detailed Implementation

[0036] The present invention can be better understood from the following embodiments.

[0037] This embodiment discloses a real-time controlled laser additive manufacturing alloy method based on online ultrasonic monitoring. Using a laser additive manufacturing system, the forming quality of the alloy sample is monitored in real time during the laser forming process. If metallurgical defects, including cracks and pores, are detected, the laser forming process parameters are adjusted in real time to eliminate the metallurgical defects, and finally a dense three-dimensional alloy solid part without metallurgical defects is prepared.

[0038] In this embodiment, the laser additive manufacturing system includes a laser powder bed melting and forming equipment and an ultrasonic monitoring system. Figure 1 The ultrasonic monitoring system of this invention includes an ultrasonic sensor 1, a pulse transmitter and receiver 2, and a data collection and processing device 3. The laser powder bed melting and forming equipment includes a forming substrate 7 and a computer control system. The ultrasonic sensor 1 is installed below the forming substrate 7, and the alloy sample 8 to be formed is placed on top of the forming substrate 7. In a specific implementation, the ultrasonic monitoring system also includes a sensor bracket 4. The ultrasonic sensor 1 is installed below the forming substrate 7 via the sensor bracket 4. The ultrasonic sensor 1 can be placed in a corresponding position according to the forming part. The sensor bracket 4 is provided with multiple sensor fixing cavities to facilitate the corresponding placement of the ultrasonic sensor 1 below the forming sample. The sensor bracket 4 is designed to be inserted into a bottom metal seat 5, which is fixed to the forming substrate 7 by bolts 6, so that the sensor is placed directly below the alloy sample 8 to be formed.

[0039] The laser powder bed melting and forming equipment also includes a laser, a laser forming chamber, an automatic powder spreading system, a protective atmosphere device, and a cooling circulation system, all of which are existing technologies and are not specifically limited in this embodiment.

[0040] The pulse transmitter receiver 2 is connected to the ultrasonic sensor 1 and the data collection and processing device 3 respectively. It is used to control the ultrasonic sensor 1 to emit sound signals to the alloy sample 8, collect reflected sound signals above the amplitude threshold, and send the reflected sound signals above the amplitude threshold to the data collection and processing device 3.

[0041] The data collection and processing device 3, connected to the computer control system, is used to process reflected sound signals exceeding the amplitude threshold, enabling the recording and analysis of sound signals from each layer. If a metallurgical defect is detected, the computer control system is adjusted in real time to regulate the laser forming process parameters to eliminate the generation of metallurgical defects during subsequent forming processes.

[0042] The amplitude threshold is set based on a standard formed alloy sample, which is free of metallurgical defects. The amplitude threshold is determined by the maximum amplitude of the acoustic signal obtained from the standard formed alloy sample. The amplitude threshold is different for different alloys.

[0043] In addition, the ultrasonic sensor 1 can be an ultrasonic sensor with frequency filtering. In this embodiment, the transmission and reception bandwidth of the ultrasonic sensor 1 is 100 to 1000 kHz, and the frequency filtering reduces the influence of noise on the sound signal.

[0044] This embodiment discloses a real-time controlled laser additive manufacturing alloy method based on online ultrasonic monitoring, comprising the following steps:

[0045] Step 1: Use computer software to design and build a three-dimensional solid geometric model of the target part. Use Materialise Magics software to slice the three-dimensional model and plan the laser scanning path to discretize the three-dimensional solid into two-dimensional data.

[0046] Step 2 involves importing the two-dimensional data into the computer control system of the laser powder bed melting and forming equipment. The alloy powder is melted and solidified layer by layer. During the laser forming process, the forming quality of the alloy sample is monitored in real time using an ultrasonic monitoring system. Based on the reflected acoustic signal data, if metallurgical defects, including cracks and pores, are detected, the laser forming process parameters are adjusted in real time to eliminate these defects, ultimately producing a dense, crack-free, and pore-free three-dimensional alloy solid part. Specifically, this includes:

[0047] Step 2.1: Acoustic signal characteristics are acquired from reflected acoustic signals exceeding the amplitude threshold. Seven characteristics of the acoustic signal are collected by the data collection and processing device 3: rise time, peak amplitude, duration, kurtosis, number of counts, energy, and frequency. Among all characteristics, amplitude is measured in real time by the data collection and processing device 3, while the other descriptors are calculated from the reflected waveform of the shaped back wall, as they are highly dependent on the amplitude threshold used to detect the arrival and end times of the acoustic emission signal.

[0048] Other acoustic signal characteristics can be calculated as follows: Rise time is the difference between the time taken to reach the maximum amplitude of the acoustic signal and the time taken to start the signal. Duration is the difference between the end time and the start time of an acoustic signal. Kurtosis is a statistic describing the steepness or sharpness of the peak of a signal waveform. Mathematically, it is defined as the deviation K of the acoustic signal amplitude distribution from the normal distribution, usually expressed by the following formula: K = (1 / N)*Σ((xi-μ)^4) / σ^4. Where N is the number of acoustic signal samples, xi is the amplitude of the i-th sample, μ is the average amplitude of the entire acoustic signal, and σ is the standard deviation of the amplitude of the entire acoustic signal. The count can be obtained by selecting a time window on the waveform graph, recording the absolute values ​​of all acoustic signal amplitudes within a certain time range, and then taking the average of these amplitude values. Energy is the integral of the amplitude over the duration of an acoustic signal, which can be converted to decibels through calculation. Frequency can be obtained through Fourier transform.

[0049] All of these features are used in a multi-parameter statistical analysis and cluster analysis.

[0050] Step 2.2: Perform cluster analysis on the acoustic signal features to obtain multiple clusters;

[0051] Cluster analysis is performed using the k-means clustering algorithm. The goal of the k-means clustering algorithm is to minimize the distance between all vectors in a cluster and the cluster center, and to maximize the distance between all cluster centers. The optimal number of clusters is found by using the contour width; the cluster with the largest average contour width divides the acoustic emission events into subgroups based on the number of reaction defects. The clustering algorithm ultimately yields two clusters: acoustic events within the same cluster are similar to each other, while acoustic events in different clusters are different. The k-means algorithm can be described as follows:

[0052] Initialize each cluster center Ci, where 1 ≤ i ≤ k, k represents the number of clusters, and k ≥ 2. Calculate the Euclidean distance between the vector composed of acoustic signal features and the cluster center Ci, and then assign each input vector to the nearest cluster. Recalculate the position of the cluster center based on the closest mean, and repeat the above steps until the cluster center positions no longer change. At this point, calculate the maximum average contour width. By repeating the above steps for all possible numbers of clusters, the average contour width is maximized when k is 2, indicating that the optimal number of clusters is 2.

[0053] Clustering analysis is used to group the obtained acoustic signals, thereby creating clusters. In this embodiment, the clustering analysis yields two clusters. The acoustic signals within each cluster are similar to those within the clusters, while the acoustic signals between different clusters are different.

[0054] Step 2.3: Perform principal component analysis on multiple clusters to obtain the defect types of the acoustic emission source corresponding to each cluster, wherein the defect types include cracks and pores;

[0055] Principal component analysis (PCA) is used for dimensionality reduction. It utilizes orthogonal transformations to convert a set of correlated variables into a set of linearly uncorrelated values, called principal components. The seven features of acoustic emission events are components of the n input mode vectors Zj(). This yields an n×7 feature matrix, where n is the number of acoustic emission events. First, the feature matrix Z is standardized by subtracting the mean of each column and then dividing by the standard deviation of the column. The covariance matrix E[ZZ] is then calculated. T Calculate the eigenvectors and their corresponding eigenvalues. The two basis vectors with the largest eigenvalues ​​can be used as principal components.

[0056] Principal component analysis (PCA) can be applied to the clusters obtained after cluster analysis to identify different defect types. PCA can also be used for dimensionality reduction to visualize the cluster analysis results. In the two clusters, most cluster features show significant differences in mean, i.e., larger F-values, and a larger F-value indicates greater feature importance. The differences in event characteristics between the two clusters lead to different acoustic emission source mechanisms, i.e., different defect types in the acoustic emission sources. Acoustic emission signals generated by pores have short decay times and small amplitudes, while acoustic emission signals from cracks have short durations and large amplitudes. Compared to cracks, acoustic emission events generated by pores have higher energy.

[0057] Step 2.4: If the formation of metallurgical defects, including cracks and pores, is detected, the laser forming process parameters are adjusted in real time to eliminate the metallurgical defects.

[0058] The generation of cracks and pores can be monitored by visualizing the cluster analysis results during the above process. The data collection and processing device 3 is connected to the computer control system. When the generation of cracks and pores is detected, the computer control system adjusts the laser powder bed melting and forming equipment in real time to change the forming process parameters, thereby eliminating cracks and pores.

[0059] If a crack is detected, it can be eliminated by reducing the scanning speed. Appropriately reducing the laser scanning speed and extending the solidification time of the molten pool facilitates liquid phase reflow and filling, while also reducing the temperature gradient of the molten pool, decreasing the accumulation of thermal stress, and thus eliminating the crack.

[0060] If pores are detected, they can be eliminated by increasing laser power and decreasing laser scanning speed. Increasing laser power increases heat input and molten pool temperature, reduces melt viscosity, and promotes complete melting and wetting of the powder, thereby reducing the formation of unmelted pores. Decreasing laser scanning speed prolongs molten pool solidification time, which facilitates the escape of gas from the liquid phase, reduces the formation of pores in the molten pool, and thus reduces porosity.

[0061] In the following embodiments, the relevant parameters of the aluminum and nickel alloy compositions remain unchanged. The aluminum alloy composition contains 3.90-4.80 wt.% copper, 1.10-1.90 wt.% Mg, and 0.40-1.10 wt.% Mn, with the balance being aluminum. The aluminum alloy powder has a particle size of 19-52 μm. The nickel alloy composition contains 21.1-22.8 wt.% Cr, 17.5-18.9 wt.% Fe, 8.2-9.4 wt.% Mo, 0.5-0.8 wt.% W, and 1.1-1.8 wt.% Co, with the balance being Ni. The powder particle size is 15-46 μm.

[0062] Example 1

[0063] (1) An ultrasonic monitoring system is installed in the laser additive manufacturing system. The data collection and processing device of the ultrasonic monitoring system is connected to the computer control system of the laser powder bed melting forming equipment. The computer control system is used to adjust the laser forming process parameters of the laser powder bed melting forming equipment. When cracks and pores are detected, the forming process parameters can be adjusted in real time.

[0064] (2) Use computer software to design and build a three-dimensional solid geometric model of the target part, and then use Materialise Magics software to perform layer slicing processing on the three-dimensional model and plan the laser scanning path.

[0065] (3) Import the slice file obtained in step (2) into the computer control system of the laser powder bed melting and forming equipment, and connect the ultrasonic monitoring system simultaneously. Place aluminum alloy powder in the equipment for forming the sample. The transmission and reception bandwidth of the ultrasonic sensor 1 is 100~1000kHz. During the forming process, the ultrasonic sensor 1 transmits acoustic signals to the forming sample under the control of the pulse transmitter receiver 2, and receives the acoustic signals reflected from the back wall of the forming sample in real time, records and analyzes the characteristic acoustic signals of each layer of the sample during the forming process. In addition, only characteristic acoustic signals above the threshold are collected. The ultrasonic sensor 1 reduces the influence of noise by using frequency filtering. The characteristic acoustic signals obtained in the data collection and processing device 3 are processed and divided into two clusters by cluster analysis. The acoustic signal types of different clusters are different, reflecting different types of defects.

[0066] After the data collection and processing device 3 identifies cracks or pores generated during the forming process through cluster analysis, it controls the computer control system in the laser powder bed melting equipment in real time to adjust the laser forming process parameters to eliminate the generation of cracks or pores in the subsequent forming process, ensure the forming quality of each layer of the aluminum alloy sample, and finally obtain a uniform, dense aluminum alloy sample without cracks or pores.

[0067] (4) After forming, the part is separated from the forming substrate by wire cutting to obtain an aluminum alloy block sample. The aluminum alloy block sample is ground and polished according to the standard metallographic sample preparation method. The optical microscope image is shown below. Figure 2 As shown, no obvious cracks were observed in the aluminum alloy sample, only a few micropores, indicating good forming quality. The density reached 99.6%.

[0068] Example 2

[0069] (1) The ultrasonic monitoring system is installed in the laser additive manufacturing system. The data collection and processing device 3 of the ultrasonic monitoring system is connected to the computer control system. When cracks and pores are detected, the forming process parameters of the laser powder bed melting forming equipment can be changed in real time by the computer control system.

[0070] (2) Use computer software to design and build a three-dimensional solid geometric model of the target part, and then use Materialise Magics software to perform layer slicing processing on the three-dimensional model and plan the laser scanning path.

[0071] (3) Import the slice file obtained in step (2) into the computer control system of the laser powder bed melting and forming equipment, and connect the ultrasonic monitoring system at the same time. Place the nickel alloy powder in the equipment for forming the sample. The ultrasonic sensor 1 has a transmission and reception bandwidth of 100~1000kHz. During the forming process, the ultrasonic sensor 1 transmits acoustic signals to the forming sample under the control of the pulse transmitter receiver 2, and receives the acoustic signals reflected from the back wall of the forming sample in real time. Record and analyze the characteristic acoustic signals of each layer of the sample during the forming process. In addition, only characteristic acoustic signals above the threshold are collected. The ultrasonic sensor 1 reduces the influence of noise by using frequency filtering. The characteristic acoustic signals obtained in the data collection and processing device 3 are processed and divided into two clusters by cluster analysis. The acoustic signal types of different clusters are different, reflecting different types of defects.

[0072] After the data collection and processing device 3 identifies cracks or pores generated during the forming process through cluster analysis, it controls the forming process parameters in the computer control system of the laser powder bed melting equipment in real time to eliminate cracks or pores in the subsequent forming process, ensure the forming quality of each layer of the sample, and finally obtain a uniform, dense nickel-based alloy sample without cracks or pores.

[0073] (4) After forming, the part is separated from the forming substrate by wire cutting to obtain a nickel alloy block sample. The nickel alloy block sample is ground and polished according to the standard metallographic sample preparation method. The optical microscope image is shown below. Figure 3 As shown, the nickel alloy sample exhibits good forming quality, with no cracks or obvious porosity observed. The density reaches 99.8%, demonstrating good laser forming quality.

[0074] Comparative Example 1

[0075] (1) Unlike Example 1, Comparative Example 1 did not have an ultrasonic monitoring system installed in its laser additive manufacturing system. Therefore, it was impossible to monitor the generated cracks and pores in real time during the laser additive manufacturing process, making it difficult to control the laser forming process parameters in real time. Obvious cracks and pores were observed in the formed aluminum alloy sample. Figure 4 The forming quality was poor, with a density of only 96.4%.

[0076] Comparative Example 2

[0077] (1) The difference between this comparative example and Example 1 is that the data collection and processing module of the ultrasonic monitoring system is not connected to the laser powder bed melting and forming equipment (the equipment for controlling the laser forming process parameters). In this case, the ultrasonic monitoring system can only collect the acoustic signal reflected from the back wall of the sample during forming and cannot control the laser forming process parameters in real time. Although the generation of pores and cracks is monitored by processing the acoustic signal and visualizing the cluster analysis results during the forming process, due to the inability to control the forming process parameters in real time to avoid the formation of cracks and pores, obvious cracks and pores are ultimately observed in the formed aluminum alloy sample. Figure 5 The forming quality was poor, with a density of only 97.1%.

[0078] Comparative Example 3

[0079] (1) This comparative example differs from Example 1 in that, in the first half of the formed sample, the data collection and processing device 3 of the ultrasonic monitoring system is not connected to the laser powder bed melting forming equipment (the equipment for controlling laser forming process parameters). At this time, the ultrasonic monitoring system can only collect the acoustic signal reflected from the back wall of the sample during forming and cannot control the forming process parameters in real time. When the formed sample reaches a certain stage, the data collection and processing device 3 of the ultrasonic monitoring system is connected to the laser powder bed melting forming equipment (the equipment for controlling laser forming process parameters in real time). In the second half of the formed sample, the laser forming process parameters can be controlled in real time to eliminate metallurgical defects such as cracks and pores. Figure 6 As shown, obvious cracks and pores were observed in the early forming part of the final laser-formed aluminum alloy sample, indicating poor forming quality. However, the later forming sample had better quality and no obvious cracks or pores.

[0080] In its specific implementation, this application provides a computer storage medium and a corresponding data processing unit. The computer storage medium is capable of storing a computer program, which, when executed by the data processing unit, can run the invention's content regarding a real-time controlled laser additive manufacturing alloy method based on online ultrasonic monitoring, as well as some or all of the steps in various embodiments. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0081] Those skilled in the art will clearly understand that the technical solutions in the embodiments of the present invention can be implemented using computer programs and their corresponding general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of computer programs, i.e., software products. These computer program software products can be stored in a storage medium and include several instructions to cause a device containing a data processing unit (which may be a personal computer, server, microcontroller, MUU, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

[0082] This invention provides a real-time controlled laser additive manufacturing method for alloys based on online ultrasonic monitoring. Many methods and approaches exist for implementing this technical solution; the above description is merely a specific embodiment of this invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications should also be considered within the scope of protection of this invention. All components not explicitly stated in this embodiment can be implemented using existing technologies.

Claims

1. A real-time controlled laser additive manufacturing alloy method based on online ultrasonic monitoring, characterized in that, Using a laser additive manufacturing system, the forming quality of alloy samples during the laser forming process is monitored in real time. The laser additive manufacturing system includes a laser powder bed melting and forming equipment and an ultrasonic monitoring system. The ultrasonic monitoring system includes an ultrasonic sensor (1), a sensor bracket (4), a pulse transmitter receiver (2), and a data collection and processing device (3). The laser powder bed melting and forming equipment includes a forming substrate (7) and a computer control system. The alloy sample (8) to be formed is placed on top of the forming substrate (7). The ultrasonic sensor (1) is installed below the forming substrate (7) through the sensor bracket (4). The sensor bracket (4) is provided with multiple sensor fixing cavities, which can be used to place the ultrasonic sensor (1) below the alloy sample (8). The pulse transmitter receiver (2) is connected to the ultrasonic sensor (1) and the data collection and processing device (3) respectively. It is used to control the ultrasonic sensor (1) to emit sound signals to the alloy sample (8), collect reflected sound signals above the amplitude threshold, and send the reflected sound signals above the amplitude threshold to the data collection and processing device (3). The data collection and processing device (3) is connected to the computer control system and is used to process reflected sound signals that are higher than the amplitude threshold. If a metallurgical defect is detected, the computer control system is adjusted in real time to adjust the laser forming process parameters to eliminate the generation of metallurgical defects in the subsequent forming process. The method includes the following steps: Step 1: Use computer software to design and build a three-dimensional solid geometric model of the target part. Use Materialise Magics software to slice the three-dimensional model and plan the laser scanning path to discretize the three-dimensional solid into two-dimensional data. Step 2: Import the two-dimensional data into the computer control system of the laser powder bed melting and forming equipment, melt and solidify the alloy powder layer by layer, and monitor the forming quality of the alloy sample in real time through the ultrasonic monitoring system during the laser forming process. If the formation of metallurgical defects, including cracks and pores, is detected based on the reflected acoustic signal data, the laser forming process parameters are adjusted in real time to eliminate the metallurgical defects, and finally a dense three-dimensional alloy solid part without cracks and pores is prepared. Step 2 includes: Step 2.1, acquiring acoustic signal characteristics in reflected acoustic signals that are above the amplitude threshold, wherein the acoustic signal characteristics include rise time, peak amplitude, duration, kurtosis, number of counts, energy, and frequency; Step 2.2: The acoustic signal features are subjected to cluster analysis using the k-means clustering algorithm to obtain multiple clusters; Step 2.3: Perform principal component analysis on multiple clusters to obtain the defect types of the acoustic emission source corresponding to each cluster, wherein the defect types include cracks and pores; Step 2.4: If the formation of metallurgical defects, including cracks and pores, is detected, the laser forming process parameters are adjusted in real time to eliminate the metallurgical defects.

2. The real-time controlled laser additive manufacturing alloy method based on online ultrasonic monitoring according to claim 1, characterized in that, In step 2.1, the peak amplitude of the acoustic signal characteristics is obtained by real-time measurement by the data collection and processing device (3), and the rise time, duration, kurtosis, number of counts, energy and frequency are calculated from the reflected acoustic signal above the amplitude threshold.

3. The real-time controlled laser additive manufacturing alloy method based on online ultrasonic monitoring according to claim 2, characterized in that, Step 2.2 includes: initializing each cluster center Ci, where 1 ≤ i ≤ k, k represents the number of clusters, k ≥ 2; calculating the Euclidean distance between the vector composed of acoustic signal features and the cluster center Ci, and then assigning each input vector to the nearest cluster; recalculating the position of the cluster center based on the closest mean, and repeating the above steps until the cluster center position does not change, at which point the maximum average contour width is calculated; by repeating the above steps for all clusters, the average contour width is maximized when k is 2, indicating that the optimal number of clusters is 2, the acoustic events within the same cluster are similar to each other, and the acoustic events within different clusters are different.

4. The real-time controlled laser additive manufacturing alloy method based on online ultrasonic monitoring according to claim 3, characterized in that, Step 2.3 includes: dimensionality reduction using principal component analysis (PCA). The seven features of the acoustic emission events are components of the n input mode vectors Zj, j=1, 2, ..., n, resulting in an n×7 feature matrix Z, where n is the number of acoustic emission events. First, the feature matrix Z is standardized by subtracting the mean of each column and then dividing by the standard deviation of the column. Then, the covariance matrix E[ZZ] is calculated. T Calculate the eigenvectors and their corresponding eigenvalues; the two basis vectors with the largest eigenvalues ​​are taken as principal components; Principal component analysis can be used to identify different types of defects in the clusters obtained after cluster analysis. The types of defects in the acoustic emission sources include pores and cracks. The acoustic emission signals generated by pores have short decay time and small amplitude; the acoustic emission signals generated by cracks have short duration and large amplitude; the acoustic emission events generated by pores have higher energy than those generated by cracks.

5. The real-time controlled laser additive manufacturing alloy method based on online ultrasonic monitoring according to claim 4, characterized in that, If a crack is detected in step 2.4, the laser forming process parameters are adjusted in real time to eliminate the metallurgical defect, including reducing the laser scanning speed to eliminate the crack.

6. The real-time controlled laser additive manufacturing alloy method based on online ultrasonic monitoring according to claim 4, characterized in that, If porosity is detected in step 2.4, the laser forming process parameters are adjusted in real time to eliminate metallurgical defects, including increasing laser power and reducing laser scanning speed to eliminate porosity.

7. A real-time controlled laser additive manufacturing alloy method based on online ultrasonic monitoring according to claim 6, characterized in that, The transmission and reception bandwidth of the ultrasonic sensor (1) is 100~1000 kHz.

Images