A laser powder bed fusion forming online monitoring device and method using OCT

Abstract

The application provides a laser powder bed melting forming online monitoring device and method using OCT. The method adds OCT imaging scanning technology in the powder bed melting forming process, forms a layer and scans a layer, so that the surface morphology of the forming process can be accurately characterized, thereby realizing the functions of accurately identifying defect morphology, classification and distribution, and further accurately studying the influence of process parameters on the forming quality. The method is convenient to operate, high in production and detection efficiency, and can further improve the production efficiency of the laser powder bed melting.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of laser powder scanning, and relates to an online monitoring device and method for laser powder bed melting forming by using OCT, in particular to a device and method for realizing defect monitoring and three-dimensional reconstruction of selective laser melting forming by using an optical coherence tomography method. BACKGROUND

[0002] L-PBF (laser powder bed fusion) technology, also known as laser powder bed melting technology, is a widely used technology in metal laser additive manufacturing. In this technology, a high-energy-density laser is used to scan the fine metal powder laid in a flat manner according to the specified path output by the digital design / manufacturing software, so as to realize the rapid melting and solidification of the metal powder in a specific point-specific path-specific single-layer area, and form the required thin-layer metal. By layer-by-layer scanning, the good combination between layers is further realized, and finally the forming of the entire component is completed. Due to the forming characteristics of point-line-surface-body of L-PBF, it is easier to complete the high-efficiency near-net forming of complex structural components. The rapid solidification characteristics also make the microstructure characteristics of the formed parts different from those of traditional forming methods, and the mechanical properties change significantly. However, this nearly free rapid solidification method of L-PBF may cause various defects in the solidification process, including surface unevenness, area unmelting, powder sticking, porosity, thermal stress deformation, etc., which will have a serious impact on the dimensional accuracy and mechanical properties of the formed components. Therefore, online monitoring of defects, real-time identification of defect occurrence and adjustment of process have a direct impact on the quality of component forming.

[0003] However, the existing various online technologies often use optical imaging technology to obtain two-dimensional images of the formed surface, and use image recognition, machine learning, etc. to identify and determine the defects. This requires a large amount of data to support, and belongs to indirect identification, which can only identify the classification, two-dimensional size and distribution of the defects, and cannot obtain the complete three-dimensional morphology of the defects, which may lead to misidentification of the defects, further causing missetting of the process parameters, equipment misoperation, causing serious forming errors, and even termination of the forming process, and the whole piece is scrapped. SUMMARY

[0004] The purpose of the present application is to overcome the shortcomings of the prior art and provide an online monitoring device and method for laser powder bed melting forming by using OCT to solve the problem of difficulty in finding defects in the L-PBF process.

[0005] To achieve the above purpose, the following technical solutions are adopted:

[0006] An online monitoring method for laser powder bed melting forming by using OCT, comprising the following steps:

[0007] Step 1, OCT imaging device and mechanical motion device initialization, OCT imaging device and powder bed fusion forming device without interference, mechanical motion device and powder bed fusion forming device without interference;

[0008] Step 2, OCT imaging device moves to the standard test block imaging position, and system calibration is performed;

[0009] Step 3, powder bed fusion forming device forms after powder laying;

[0010] Step 4, OCT imaging device scans forming to obtain a surface topography image;

[0011] Step 5, by the surface topography image, after noise reduction processing, coloring the pixel points, three-dimensional reconstruction is performed to obtain the surface topography three-dimensional morphology result, and the surface topography feature type, distribution position and defect size are obtained;

[0012] Step 6, steps 3-5 are repeated until the forming component is completed, and the online monitoring of the laser powder bed fusion forming component surface three-dimensional morphology and defect distribution is realized.

[0013] Further improvement of the application is that:

[0014] Preferably, in step 1, the OCT imaging device initialization process is: the camera imaging position and imaging quality are confirmed by the initial imaging of the CMOS camera, and the zero position and maximum stroke of the scanning camera are determined by the scanning light path unit.

[0015] Preferably, in step 2, the process of system calibration of the OCT imaging device is: the scattered point depth signal of the scanning area is obtained by scanning the scanning galvanometer group in the OCT imaging device, the scattered point depth signal is returned to the upper computer, after data processing, the processing result is compared with the standard result, the error value is obtained, whether the error value meets the requirements is judged, if it does not meet the requirements, the imaging parameters are adjusted until the final error value meets the requirements.

[0016] Preferably, the specific process of step 3 is that the powder scraper lays the powder on the substrate or the deposition layer, after the powder scraper returns to the home position, the L-PBF forming system works to complete the single-layer laser scanning and solidification forming process.

[0017] Preferably, in step 5, after noise reduction processing, the actual height of the sample surface scanning point and the reference line is calculated according to the image scale, and the pixel points are colored according to different heights.

[0018] Preferably, in step 5, the forming layer is three-dimensionally reconstructed according to the actual height of the sample surface scanning point and the reference line.

[0019] Preferably, in step 6, before each repetition of steps 3-5, the three-dimensional morphology result is analyzed by machine learning to obtain an optimized forming control strategy, and then the next laser powder bed melting forming is carried out.

[0020] An online monitoring device for laser powder bed melting forming using OCT is used to implement the above-mentioned online monitoring method of laser powder bed melting forming, comprising a mechanical movement device arranged in the laser powder bed melting forming device; the mechanical movement device is connected with an OCT imaging device, and the mechanical movement device drives the OCT imaging device to move; the mechanical movement device and the OCT imaging device are jointly connected with an upper computer.

[0021] Preferably, the mechanical movement device is an external support frame arranged in the laser powder bed melting forming device, a movement mechanism is slidably connected to the external support frame, an optical sensor, a position encoder and an OCT imaging device are installed at the lower end of the movement mechanism; a locking mechanism is arranged at the connection between the movement mechanism and the external support frame; the optical sensor and the position encoder are electrically connected with the OCT imaging device; the optical sensor, the position encoder and the OCT imaging device are connected with the upper computer.

[0022] Preferably, the mechanical movement device is a mechanical arm.

[0023] Compared with the prior art, the present application has the following beneficial effects:

[0024] The present application provides a laser powder bed melting forming online monitoring method using OCT technology. The method adds OCT imaging scanning technology in the powder bed melting forming process, one layer is formed and one layer is scanned, so that the surface morphology of the forming process can be accurately characterized, thereby realizing the functions of accurately identifying defect morphology, classification and distribution, and further accurately studying the influence of process parameters on forming quality. The method is convenient to operate, has high production and detection efficiency, and can further improve the production efficiency of laser powder bed melting.

[0025] Further, based on the discovery of printing defects in each layer, the method can timely feedback and adjust the printing parameters, timely discover forming problems, eliminate the cumulative error formed by the surface slight height difference with the increase of the number of melting layers in the forming process, reduce the probability of defects of the formed parts, improve the forming quality, improve the production efficiency and reduce the cost.

[0026] The present application also discloses a laser powder bed melting forming online monitoring device using OCT, which is simple in equipment and can be directly improved on the basis of the existing laser powder bed melting forming device, and has low improvement cost. BRIEF DESCRIPTION OF DRAWINGS

[0027] Figure 1An online monitoring method for laser powder bed melting forming by using OCT;

[0028] Figure 2 An online monitoring method for laser powder bed melting forming by using OCT;

[0029] Figure 3 A structural schematic diagram of an online monitoring device for laser powder bed melting forming by using OCT technology;

[0030] Figure 4 A structural schematic diagram of a mechanical movement device in the online monitoring device;

[0031] Figure 5 A structural schematic diagram of another mechanical movement device in the online monitoring device.

[0032] Wherein: 1-OCT imaging device; 2-external support frame; 3-movement mechanism; 4-locking mechanism; 5-servo control motor; 6-optical sensor; 7-position encoder. DETAILED DESCRIPTION

[0033] In the description of the present application, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application; the terms "first", "second", "third" are only for the purpose of description, and cannot be understood as indicating or implying relative importance; in addition, unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connecting" should be understood in a broad sense, for example, it can be fixed connection, or detachable connection; it can be directly connected, or indirectly connected through an intermediate medium; it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0034] Reference Figure 1 The present application discloses an online monitoring method for laser powder bed melting forming by using OCT, which comprises the following steps:

[0035] S1, the OCT imaging device 1 and the mechanical movement device are initialized: the system is powered on, the central control system is self-checked, the host computer issues an instruction, the position of the OCT imaging device 1 is zeroed, and there is no interference with the powder bed melting forming device.

[0036] OCT imaging device 1 initialization: OCT imaging device 1 moves to the standard test block imaging position. The CMOS camera initial imaging, confirm the camera imaging position and imaging quality, the scanning optical path unit in the OCT imaging device 1 determines the zero position and the maximum stroke.

[0037] S2, OCT online monitoring device imaging subsystem calibration: image the standard test block, the scanning galvanometer group moves to the scanning starting position, the OCT light source emits the initial light beam, which is divided into scanning beam and reference beam by laser beam splitter module, the scanning beam is projected to the standard test block surface through the scanning optical path system, and the reflected back to the equipment becomes the signal reflected beam, which enters the laser beam splitter again through the reflected light path system, at the same time, the reference beam forms a signal reference beam through the reference light path, enters the laser beam splitter module, the signal reflected beam and the signal reference beam interfere to form an interference signal beam into the signal acquisition system, which converts the interference signal beam into a single point depth signal. The scanning galvanometer points the scanning beam to the next scanning point, and the above process is repeated, and finally the scatter point depth signal of the entire scanning area is obtained, the scatter point depth signal, i.e. the imaging result is returned to the host computer for data point processing, the result is compared with the pre-stored standard result, the error value is calculated, and the imaging parameter is adjusted, the imaging is repeated, the error value is confirmed, and the above result is repeated until the error value meets the initial set value or the default value;

[0038] S3: the powder scraper lays the powder to be printed on the powder laying substrate or the deposited layer, and the powder scraper returns to the home position after the powder laying is completed; the L-PBF forming system works, and the single layer laser scanning and solidification forming process is completed, and the laser stops;

[0039] S4: capture the morphology image: OCT imaging device 1 to the set observation area, image the whole area to form the initial data, and transmit to the host computer. The OCT imaging device 1 moves to the zero position.

[0040] L-PBF single layer forming image capture: the L-PBF device automatically runs the scanning and solidification program, and after the solidification is completed, the surface morphology image after the solidification forming is obtained.

[0041] S5: Component digital morphology three-dimensional reconstruction and defect online monitoring: based on the surface tomography height information obtained in step 4, the massive two-dimensional picture data obtained by the OCT imaging device, all OCT scanning two-dimensional jpeg images are denoising processed and the actual height of the sample surface scanning point from the reference line is calculated according to the image scale, and the pixel points are colored according to different heights. The two-dimensional jpeg images after processing are three-dimensionally reconstructed to obtain the surface topography three-dimensional morphology result of the formed layer. The machine learning algorithm is used to automatically analyze and measure the surface three-dimensional topography of the formed part, and the information such as the surface topography feature type, distribution position, size and the like of the formed part is recorded, so as to realize the online monitoring of the actual surface three-dimensional morphology and defect distribution of the laser powder bed fusion formed component.

[0042] S6, according to the specific program setting, steps 3-5 are repeated until the L-PBF forming process is completed, and the actual three-dimensional morphology and defect distribution of the formed component are output.

[0043] Referring to Figure 2 Further, the application discloses a laser powder bed fusion forming closed-loop online monitoring method using OCT, comprising the following steps:

[0044] S1, setting the laser powder bed fusion forming process parameters, the powder scraper lays the powder to be printed on the first powder laying substrate or the deposited layer, and the powder scraper is returned to the home position after the powder laying is completed; the L-PBF forming system works, and the single-layer laser scanning fusion forming process is completed, and the laser is stopped;

[0045] S2, moving the OCT imaging device to the set observation area by the motion mechanism, performing full-area tomographic scanning imaging, and transmitting the scanning data to the upper computer;

[0046] S3, based on the surface tomography height information obtained in S2, all two-dimensional jpeg images obtained by the OCT imaging device are denoising processed, and the actual height of the sample surface scanning point from the reference line is calculated according to the image scale, and the pixel points are colored according to different heights. The two-dimensional jpeg images after processing are three-dimensionally reconstructed to obtain the surface topography three-dimensional morphology result of the formed layer;

[0047] S4, using the machine learning algorithm to automatically measure and analyze the surface three-dimensional topography of the formed part, and recording the information such as the surface topography feature type, distribution position, size and the like of the formed part. According to the analysis result of the surface topography of the formed part, the machine learning algorithm optimizes the forming control strategy, and adjusts the process parameters such as laser power and scanning speed in the next layer fusion forming process;

[0048] S5, transmitting the optimized next layer fusion forming process parameters to the L-PBF forming system;

[0049] S6, repeating S1-S5 until all printing tasks are completed.

[0050] Referring to Figure 3 The application discloses an online monitoring device for laser powder bed melting forming by using OCT technology, which comprises an OCT imaging device 1, a mechanical motion device and an upper computer.

[0051] In the embodiment of the application, the OCT imaging device 1 comprises: an OCT light source, which internally contains a laser generator and outputs a laser beam with low coherence characteristics; a laser beam splitter, which is used for separating the scanning light beam output by the light source and the reference light beam, so that the signal reflected light beam and the signal reference light beam interfere with each other; a reflected light path, which provides the reference signal light beam; a scanning light path, which provides accurate pointing for the scanning light beam and outputs the final scanning light beam to the surface to be scanned; a signal acquisition device, which acquires the interference signal of the signal reflected light beam and the signal reference light beam; a CMOS camera, which is used for synchronously capturing a two-dimensional visible light image of the imaging area and positioning the imaging area; and a central control device, which is used for processing the interference signal to convert it into a depth signal of the scanned surface, uploading the depth information of each scanning point to the upper computer and receiving the control of the scanning light path to point to the scanning point by the upper computer.

[0052] In the embodiment of the application, the mechanical motion device is fixedly installed in the laser powder bed melting forming device and is used for supporting and moving the OCT imaging device 1.

[0053] In the embodiment of the application, the upper computer comprises a man-machine interface, which is used for displaying the imaging result and the real-time state of the system to the operator, setting the automatic operation logic of the system by the operator, inputting various imaging or motion parameters, and directly operating the OCT imaging device 1 or the mechanical motion device beyond the system logic; an image processing card, which is used for collecting the original scanning point depth data output by the OCT imaging device 1, performing three-dimensional reconstruction and storing the data; and a personal computer, which is used for providing a basic operation platform for the man-machine interface and the image processing card.

[0054] Referring to Figure 4 The application discloses an online monitoring device for laser powder bed melting forming by using OCT technology, which comprises a mechanical motion device, an external support frame 2, which is used for supporting the whole device, and Figure 3 The external support frame is installed in the forming cabin of the L-PBF device and realizes vibration isolation and electrical connection of the system and the L-PBF device.

[0055] Motion servo mechanism: including motion mechanism 3, locking mechanism 4, servo control motor 5, optical sensor 6 and position encoder 7; for realizing the three-dimensional motion of OCT imaging device in L-PBF equipment forming cabin, avoiding high-power laser scanning area and scraper motion area in forming process, positioning imaging equipment at set scanning position before OCT imaging device starts imaging, positioning and keeping OCT imaging device during imaging, and withdrawing OCT imaging device after imaging is completed; motion mechanism 3 is provided with two, which can move in the whole laser powder bed melting forming system along the track of external support frame 2, and locking mechanism 4 is arranged at the connection of motion mechanism 3 and track, so that after motion mechanism 3 drives OCT imaging device 1 to the set position, positioning can be carried out for subsequent operation, one end of the track is provided with servo control motor 5 for driving motion mechanism 3 to move along the track. The lower part of motion mechanism 3 is fixedly connected with optical sensor 6, position encoder 7 and OCT imaging device 1, optical sensor 6 and position encoder 7 are connected with OCT imaging device 1, and optical sensor 6 and position encoder 7 are connected with upper computer.

[0056] Control system, receiving upper computer motion instruction for outputting motion instruction to motion servo mechanism, collecting motion servo mechanism sensor signal to realize motion servo mechanism closed loop control, uploading motion state to upper computer.

[0057] The detection process of the device is:

[0058] S1, OCT imaging device 1 and mechanical motion device initialization: system power on, central control system self-checking, upper computer issuing instruction,

[0059] Servo control motor 5 starts, locking mechanism 4 is unlocked, motion mechanism 3 moves to zero position, optical sensor 6 and position encoder 7 are zeroed, and the normality of zero position is determined without interference with the forming system; the upper computer instructs motion mechanism 3 to move to the maximum stroke, and the optical sensor 6 and position encoder 7 are reset to the maximum stroke. The expected scanning area and area division parameters are input to the upper computer, and the upper computer automatically divides the expected scanning area into a plurality of scanning subareas, scanning sequence and scanning points of each subarea. In each scanning subarea, OCT imaging device scans and images, and between each subarea, motion mechanism 3 moves OCT imaging device 1 to position;

[0060] OCT online monitoring device imaging subsystem initialization: OCT imaging device 1 moves to standard test block imaging position. CMOS camera initial imaging, confirming camera imaging position and imaging quality. The scanning light path system in the OCT imaging device determines zero position and maximum stroke;

[0061] S2, OCT online monitoring device imaging subsystem calibration: image the standard test block, move the scanning galvanometer group to the scanning starting position, the OCT light source emits an initial light beam, which is divided into a scanning beam and a reference beam by a laser beam splitter module, the scanning beam is projected onto the surface of the standard test block through the scanning light path system, and the reflected light beam becomes a signal reflected light beam when it is reflected back to the device, then enters the laser beam splitter through the reflected light path system, at the same time, the reference beam forms a signal reference light beam through the reference light path, enters the laser beam splitter module, the signal reflected light beam and the signal reference light beam interfere to form an interference signal light beam into the signal acquisition system, which converts the interference signal light beam into a single-point depth signal. The scanning galvanometer points the scanning beam to the next scanning point, and the above process is repeated to finally obtain the scatter depth signal of the entire scanning area, the imaging result is returned to the host computer, the data points are processed, the result is compared with the pre-stored standard result, the error value is calculated, and the imaging parameters are adjusted for imaging again, the error value is confirmed, and the above result is repeated until the error value meets the initial set value or the default value;

[0062] S3, the powder scraper lays the powder to be printed on the first powder-laid substrate or the deposited layer, and the powder scraper returns to the home position after the powder laying is completed; the L-PBF forming system works to complete the single-layer laser scanning and solidification forming process, and the laser stops;

[0063] S4, capture the initial morphology image: the servo control motor 5 is started, the locking mechanism 4 is unlocked, the motion mechanism 3 moves the OCT imaging device 1 to the set observation area, the servo control motor 5 is stopped, the locking mechanism 4 is locked, and step 3 is repeated to form the initial data by imaging the whole area and transmitting it to the host computer. The locking mechanism 4 is unlocked, the servo control motor 5 is started, and the motion mechanism 3 is moved to the zero position and locked;

[0064] S5, capture the image after L-PBF single-layer forming: the L-PBF device automatically runs the first-layer scanning and solidification program, and after the solidification is completed, the surface topography image after the solidification is obtained,

[0065] S6, three-dimensional reconstruction of digital morphology and online monitoring of defects of the component: based on the scanning area surface tomography height information obtained in step 5, all OCT scanning two-dimensional jpeg images are processed by noise reduction, and the actual height of the sample surface scanning point from the reference line is calculated according to the image scale, and the pixel points are colored according to different heights. According to the actual height, the OCT scanning two-dimensional jpeg image after processing is three-dimensionally reconstructed to obtain the three-dimensional morphology result of the surface topography of the formed part. The machine learning algorithm is used to automatically analyze and measure the surface three-dimensional topography of the formed part, and the information such as the surface topography feature type, distribution position, and size of the formed part is recorded to realize the online monitoring of the actual surface three-dimensional morphology and defect distribution of the laser powder bed melting formed component.

[0066] S7, repeating S6-S7 until the L-PBF forming process is completed, and outputting the actual three-dimensional shape and defect distribution of the formed component.

[0067] Referring to Figure 5 Optionally, the external support frame in the mechanical movement device is replaced by a mechanical arm, the front end of the mechanical arm has an optical sensor and a position encoder. The online monitoring device is no longer installed in the L-PBF forming cabin through the external support frame, but is fixed with the L-PBF forming cabin wall through the mechanical arm, and the movement servo device no longer has a separate mechanical structure, and the movement of the OCT imaging device is realized through the movement of the mechanical arm.

[0068] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for online monitoring of laser powder bed fusion forming using OCT, characterized in that, Comprising the following steps: Step 1, the OCT imaging device (1) and the mechanical motion device are initialized, and the OCT imaging device (1) and the powder bed fusion forming device are not interfered, and the mechanical motion device and the powder bed fusion forming device are not interfered; Step 2, the OCT imaging device (1) is moved to the standard test block imaging position, and the system calibration is carried out; In step 2, the process of the OCT imaging device (1) for system calibration is that the scanning galvanometer group in the OCT imaging device (1) scans to obtain the scatter point depth signal of the scanning area, returns the scatter point depth signal to the upper computer, after data processing, compares the processing result with the standard result, obtains the error value, judges whether the error value meets the requirements, if not, adjusts the imaging parameters, until the final error value meets the requirements; Step 3, the powder bed fusion forming device is formed after powder laying; Step 4, the OCT imaging device (1) scans the forming to obtain the surface topography image; Step 5, through the surface topography image, after noise reduction processing, the pixel points are colored, three-dimensional reconstruction is carried out to obtain the surface topography three-dimensional morphology result, and the surface topography feature type, distribution position and defect size are obtained; In step 5, after noise reduction processing, according to the image scale, the actual height of the sample surface scanning point and the reference line is calculated, and the pixel points are colored according to different heights; Step 6, repeat steps 3-5 until the forming component is completed, and the online monitoring of the three-dimensional morphology and defect distribution of the laser powder bed fusion forming component surface is realized; In step 6, before each time repeating steps 3-5, the three-dimensional morphology result is analyzed through machine learning to obtain an optimized forming control strategy, and then laser powder bed fusion forming is carried out.

2. The method of claim 1, wherein the method is an online monitoring method of laser powder bed fusion forming using OCT. In step 1, the initialization process of the OCT imaging device (1) is that the camera imaging position and imaging quality are confirmed through the initial imaging of the CMOS camera, and the zero position and maximum stroke of the scanning camera are determined through the scanning light path unit.

3. The method of claim 1, wherein the method is an online monitoring method of laser powder bed fusion forming using OCT. The specific process of step 3 is that the powder scraper lays the powder on the substrate or the deposition layer, the powder scraper returns to the home position, the L-PBF forming system works, and the single-layer laser scanning and solidification forming process is completed.

4. The method of claim 1, wherein the method is an online monitoring method of laser powder bed fusion forming using OCT. In step 5, according to the actual height of the sample surface scanning point and the reference line, the three-dimensional reconstruction of the forming layer is carried out.

5. An application OCT laser powder bed fusion forming online monitoring device for implementing the online monitoring method of claim 1, characterized in that, Comprising a mechanical motion device arranged in the laser powder bed fusion forming device; the mechanical motion device is connected with an OCT imaging device (1), and the mechanical motion device drives the OCT imaging device (1) to move; the mechanical motion device and the OCT imaging device (1) are jointly connected with an upper computer.

6. An on-line monitoring device according to claim 5, characterised in that, The mechanical movement device is an external support frame (2) arranged in a laser powder bed fusion forming device, a movement mechanism (3) is slidably connected to the external support frame (2), an optical sensor (6), a position encoder (7) and an OCT imaging device (1) are installed at the lower end of the movement mechanism (3); a locking mechanism (4) is arranged at the connection between the movement mechanism (3) and the external support frame (2); the optical sensor (6) and the position encoder (7) are electrically connected to the OCT imaging device (1); the optical sensor (6), the position encoder (7) and the OCT imaging device (1) are connected to an upper computer.

7. The on-line monitoring device of claim 5, wherein, The mechanical movement device is a mechanical arm.

Images