Intelligent monitoring and autonomous regulation and control system and method for buckling deformation in melting additive manufacturing of laser powder bed

Abstract

The invention discloses a laser powder bed melting additive manufacturing buckling deformation intelligent monitoring and autonomous control system and method, the system comprises a powder platform, a powder laying mechanism, a first high-definition camera, a second high-definition camera and a user side remote monitoring module, the first high-definition camera is used for shooting block surface images; the second high-definition camera is used for shooting an image of the current powder layer; the user side remote monitoring module judges whether the surface warping degree of each formed block exceeds a preset range or not according to the received block surface image, and controls the printing working condition of the corresponding formed block according to the judgment result; and the user side remote monitoring module evaluates the powder laying quality of the current powder layer according to the received image of the current powder layer and determines the powder supply amount of the powder cylinder according to the evaluation result. Therefore, the influence of the warped block on the powder laying quality of the surrounding block can be automatically monitored, then the powder supply amount is adjusted in a self-adaptive mode, and the influence of the warped block on the forming quality of the surrounding block is reduced.

Description

Technical Field

[0001] The invention relates to an intelligent monitoring and autonomous control system for warping deformation in laser powder bed fusion additive manufacturing and a method thereof, and belongs to the field of laser powder bed fusion additive manufacturing. Background Art

[0002] When using high-throughput material testing and multi-material integrated printing to explore the printing process window of new materials under development, many block samples with different process parameters will be printed on a substrate, which can reduce the experimental cycle and quickly obtain suitable process parameters. However, blocks with inappropriate process parameters will warp. For example, difficult-to-process materials such as tungsten alloys are particularly sensitive to energy input. Excessive input energy will cause serious warping. The warping phenomenon is a cumulative process, from the initial mild warping to the final severe warping. The mildly warped blocks may reduce the subsequent warping degree by adjusting the amount of powder, but the severely warped blocks will block the movement of the powder wall, which will not only damage the powder laying device but also automatically stop the printing of all blocks if the load on the powder wall is too large. Blocks with suitable process parameters will not warp, but warped blocks will also affect the powder laying quality of the surrounding well-formed blocks, which may cause serious damage to their forming quality. Therefore, the warped blocks not only affect the powder laying quality of the surrounding blocks, causing damage to their forming quality, but also directly stop the printing of all blocks due to severe warping, which not only increases the printing cost but also affects the accurate acquisition of process parameters.

[0003] At present, it is difficult to establish the correlation between the cracks and warping caused by the heat distribution and heat accumulation of the formed blocks layer by layer based on the quality monitoring data collected during the printing process. In addition, during the printing process, manual supervision is usually required to observe the printing situation next to the printing equipment. When a warped block is encountered, the printing of the block is manually stopped, and other non-warped blocks continue to be printed. Sometimes the printing task takes too long, and it is difficult for the staff to observe the printing quality on site for a long time, or if the staff is on duty for a long time, they will inevitably ignore the small details of the printing, resulting in poor forming quality. Summary of the invention

[0004] The purpose of the present invention is to provide a laser powder bed fusion additive manufacturing warping deformation intelligent monitoring and autonomous control system and method, which integrates the precise identification of warped blocks, automatic selective shutdown of warped block printing, real-time detection of powder laying quality and adaptive compensation of powder supply, multi-directional data collection and user-end remote monitoring functions, which can not only meet the multi-block printing scenario, but also can be applied to the online monitoring scenario of simultaneous forming of multiple complex components, which has important practical significance for improving the degree of forming intelligence and printing stability.

[0005] In order to achieve the above technical objectives, the present invention will adopt the following technical solutions:

[0006] A laser powder bed fusion additive manufacturing warpage deformation intelligent monitoring and autonomous control system includes a powder platform, a forming cylinder, a powder cylinder and a powder spreading mechanism; it also includes a data acquisition device and a user-side remote monitoring module, the data acquisition device includes a first and a second high-definition camera, wherein:

[0007] The first high-definition camera is installed on the powder spreading mechanism, and is used to capture the block surface images including all formed blocks layer by layer, and can transmit the captured block surface images to the user-side remote monitoring module;

[0008] The second high-definition camera is mounted on the powder spreading mechanism, and is used to capture the image of the current powder layer spread by the powder spreading mechanism on the powder platform, and can transmit the captured image of the current powder layer to the user-side remote monitoring module;

[0009] The user-side remote monitoring module determines whether the surface warping degree of each formed block exceeds a preset range based on the block surface image fed back by the first high-definition camera, and controls the printing condition of the corresponding formed block based on the determination result;

[0010] The user-side remote monitoring module evaluates the powder spreading quality of the current powder layer according to the current powder layer image transmitted by the second high-definition camera, and determines the powder supply amount of the powder cylinder according to the evaluation result.

[0011] Preferably, the user-side remote monitoring module includes a block warping identification and shutdown module; the block warping identification and shutdown module includes a warping block image processing module and a warping block image evaluation module; wherein:

[0012] The warped block image processing module is used to receive the block surface image fed back by the first high-definition camera and filter and denoise the received block surface image frame by frame, and calculate the standard gray value when the number of deposition layers of each formed block reaches a preset number of layers, and calculate the real-time gray value of each formed block one by one in each deposition layer after the preset number of layers, and upload the percentage Y of the real-time gray value to the standard gray value to the warped block image evaluation module; wherein:

[0013] The calculation formula of standard gray value Gray0 is as follows:

[0014] Gray0=(0.299R0+0.587G0+0.144B0) / N

[0015] Among them, R0 is the red component of all pixels in the block surface image; G0 is the green component of all pixels in the block surface image; B0 is the blue component of all pixels in the block surface image; N is the number of formed blocks contained in the block surface image;

[0016] The real-time gray value of any formed block i in the block surface image is Gray i The calculation formula is as follows:

[0017] Gray i =0.299R i +0.587G i +0.144B i

[0018] Among them, R i G is the red component of the pixel points included in any formed block i in the block surface image; i B is the green component of the pixel points included in any formed block i in the block surface image; i is the blue component of the pixel points included in any formed block i in the block surface image;

[0019] The warped block image evaluation module first determines whether the real-time grayscale value of any formed block i is greater than the standard grayscale value. When the real-time grayscale value of the formed block i is less than the standard grayscale value, the formed block i continues to be printed. Otherwise, the percentage Y of the real-time grayscale value of the formed block i exceeding the standard grayscale value is calculated by the following formula: i :

[0020]

[0021] Among them, Gray0 is the standard gray value, Gray i is the real-time gray value of any formed block i;

[0022] The warped block image evaluation module calculates the percentage Y i , judge the severity of the surface warping of the corresponding formed block i, and determine the printing condition of the formed block i according to the judgment result.

[0023] Preferably, in the warped block image judging module, the judging criteria for the severity of the surface warping of the formed block i are: when the percentage is between 0% and 25%, it is slight warping; when the percentage is between 25% and 50%, it is moderate warping; when the percentage is between 50% and 100%, it is severe warping;

[0024] In the warping block image evaluation module, when the severity of the surface warping of the formed block i is mild or moderate, printing and monitoring are continued; when the severity of the surface warping of the formed block i is severe, printing is stopped.

[0025] Preferably, the user-side remote monitoring module includes a powder spreading quality monitoring and powder supply adaptive compensation module, which includes a powder spreading image processing module and a powder spreading image quality evaluation module, wherein:

[0026] The powder spreading image processing module is used to receive the current powder layer image transmitted by the second high-definition camera, filter and de-noise the received current powder layer image, and then use the threshold function to process it to obtain a binary gray-white image, and then count the number of black pixels in the gray-white image, and calculate the total area occupied by these black pixels based on the resolution of the gray-white image, so as to obtain the area S1 of the area with poor powder spreading quality in the current powder layer image, and transmit it to the powder spreading image quality evaluation module;

[0027] The powder spreading image quality judgment module is used to compare whether the area S1 of the area with poor powder spreading quality in the current powder layer image exceeds the powder spreading quality judgment set value U. When the judgment result shows that the area S1 of the area with poor powder spreading quality in the current powder layer image does not exceed the powder spreading quality judgment set value U, the current powder supply is maintained unchanged. Otherwise, the percentage G of the area S1 of the area with poor powder spreading quality in the current powder layer image exceeding the powder spreading quality judgment set value U is calculated, and the calculated percentage G is compared with the preset value. When the percentage is between 0% and 30%, 4 times the powder supply is adopted; when the percentage is between 30% and 60%, 5 times the powder supply is adopted; when the percentage is between 60% and 100%, 6 times the powder supply is adopted; wherein:

[0028]

[0029] In the formula: d1 is the diameter of the forming cylinder, d2 is the diameter of the powder cylinder, N is the number of forming blocks, and a and b correspond to the length and width of the forming block.

[0030] Preferably, the data acquisition device further comprises a thermal imager;

[0031] The thermal imager is mounted on the powder platform and is used to collect the surface heat distribution of each formed block layer by layer, and can transmit the collected surface heat distribution of each formed block to the control device;

[0032] The user remote monitoring module generates a heat distribution stereogram in real time according to the received surface heat distribution diagrams of each formed block.

[0033] Preferably, the local area network and wireless WIFI module in the user remote monitoring module are located in the computer host of the printing device, and the user-side remote monitoring module obtains the data collected by the first and second cameras and the thermal imager in real time through the local area network.

[0034] Preferably, the first high-definition camera is installed on the powder spreading arm on the side close to the powder cylinder, and the second high-definition camera is installed on the powder spreading arm on the side close to the forming cylinder, and the first and second high-definition cameras are arranged symmetrically.

[0035] Preferably, the powder spreading arm is symmetrically provided with a first and a second fixed bracket, the first high-definition camera is installed on the first fixed platform, and the first fixed platform is installed on the first fixed bracket in a telescopic manner through a first slider; the second high-definition camera is installed on the second fixed platform, and the second fixed platform is installed on the second fixed bracket in a telescopic manner through a second slider;

[0036] The thermal imager is mounted on the powder bed platform through a third fixed bracket.

[0037] Preferably, it also includes an on / off switch for controlling the first high-definition camera; the on / off switch includes an infrared emitting sensor and an infrared receiving device used in conjunction with each other; the infrared receiving device is arranged on the first fixed platform close to the first high-definition camera, and the infrared emitting sensor is installed on the side of the powder bed adjacent to the forming cylinder; when the powder spreading mechanism moves from one side of the powder cylinder to one side of the forming cylinder, when the infrared receiving device receives the infrared ray emitted by the infrared emitting device on one side of the powder bed, the first high-definition camera starts the shooting function.

[0038] Another technical purpose of the present invention is to provide a method for intelligent monitoring and autonomous control of warpage deformation in laser powder bed fusion additive manufacturing, including block warpage identification and shutdown control, powder laying quality monitoring and powder supply adaptive compensation control; wherein:

[0039] Block warping identification and shutdown control are achieved through the following steps:

[0040] Step 101, collecting block surface images of all formed blocks on the printing substrate layer by layer in real time;

[0041] Step 102, performing filtering and noise reduction processing and block edge enhancement processing on the block surface image collected in step 101;

[0042] Step 103, when the number of deposited layers of each formed block reaches a preset number of layers, based on the block surface image processed in step 102, the gray value of the block surface image is calculated as a standard gray value;

[0043] Step 104, starting from the next deposition layer of the preset number of layers, calculating the real-time gray value of each formed block layer by layer;

[0044] Step 105, judging whether the real-time gray value of each formed block is greater than the standard gray value, when the judgment result shows that the real-time gray value of the formed block i is less than the standard gray value, continue to print the formed block i, otherwise go to step 106;

[0045] Step 106, calculating the percentage Y of the real-time gray value of each formed block exceeding the standard gray value;

[0046] Step 107, based on the percentage Y of the real-time gray value calculated in step 106 exceeding the standard gray value, determine the surface warping degree of the corresponding formed block: when the percentage Y is between 0% and 25%, it is slight warping; when the percentage Y is between 25% and 50%, it is moderate warping; when the percentage Y is between 50% and 100%, it is severe warping;

[0047] Step 107, the formed blocks with slight warping and moderate warping continue to be printed and monitored, and the formed blocks with severe warping stop printing;

[0048] Powder spreading quality monitoring and powder supply adaptive compensation control are achieved through the following steps:

[0049] Step 201, collecting an image of the current powder layer spread by the powder spreading mechanism on the powder platform in real time;

[0050] Step 202, performing filtering, noise reduction and binarization processing on the image of the current powder layer collected in step 201;

[0051] Step 203, respectively calculate the area S1 of the area with poor powder spreading quality in the current powder layer image, the powder spreading quality judgment setting value U,

[0052]

[0053] Wherein: d1 is the diameter of the forming cylinder, d2 is the diameter of the powder cylinder, N is the number of forming blocks, a and b correspond to the length and width of the forming blocks; the area S1 of the area with poor powder spreading quality in the current powder layer image is obtained by: based on the binarized gray-white image, then counting the number of black pixels in the gray-white image, and calculating the total area occupied by these black pixels based on the resolution of the gray-white image, so as to obtain the area S1 of the area with poor powder spreading quality in the current powder layer image;

[0054] Step 204, judging whether the area S1 of the region with poor powder spreading quality in the current powder layer image exceeds the powder spreading quality judgment set value U. When the judgment result shows that the area S1 of the region with poor powder spreading quality in the current powder layer image is lower than the powder spreading quality judgment set value U, there is no need to adjust the powder supply amount. Otherwise, the process proceeds to step 205.

[0055] Step 205, calculate the percentage G of the area S1 of the area with poor powder spreading quality in the current powder layer image exceeding the powder spreading quality judgment set value U, and compare the calculated percentage G with the preset value, when the percentage is between 0% and 30%, use 4 times the powder supply; when the percentage is between 30% and 60%, use 5 times the powder supply; when the percentage is between 60% and 100%, use 6 times the powder supply;

[0056] The percentage G of the area S1 of the region with poor powder spreading quality in the current powder layer image exceeding the powder spreading quality judgment set value U is calculated by the following formula:

[0057]

[0058] Based on the above technical objectives, the present invention has the following advantages over the prior art:

[0059] 1. The intelligent monitoring and autonomous control system for warping deformation in laser powder bed fusion additive manufacturing provided by the present invention can solve the problem that when exploring the process parameters of new materials, during the multi-block printing process, it is impossible to accurately identify the warped blocks and automatically shut down the subsequent printing of the warped blocks, while continuing to print the blocks with better forming quality. The system can quickly identify blocks with slightly warped surfaces, and on this basis, it can also automatically monitor the impact of the warped blocks on the powder spreading quality of the surrounding blocks, and then adaptively adjust the powder supply to reduce the impact of the warped blocks on the forming quality of the surrounding blocks.

[0060] 2. The system also adds a user remote monitoring module and a block layer-by-layer heat distribution map acquisition module. During the printing process, multi-directional data is collected and uploaded to the data gimbal. Researchers do not need to be on duty at the printing site for a long time, which can reduce the operator's irregular operations. The heat distribution on the block surface is collected layer by layer and a heat accumulation and defect formation model is established, which provides an important reference for process parameter optimization and defect formation mechanism research.

[0061] In general, the functions involved in this system are highly practical, especially when researching new materials and new processes, where multi-block printing is often performed. Moreover, compared with existing monitoring technologies, the installation of this forming quality monitoring system requires less modification of the internal structure of the printing equipment, and can be easily installed and disassembled. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is the application scenario of the present invention: random warping phenomenon occurs when printing multiple blocks;

[0063] Figure 2 Provides the workflow of intelligent monitoring and autonomous control system for warpage deformation in laser powder bed fusion additive manufacturing;

[0064] Figure 3 Flow chart for judging warped blocks;

[0065] Figure 4 This is a flow chart for judging the quality of powder spreading;

[0066] Figure 5 This is a schematic diagram of the structure of the intelligent monitoring and autonomous control device for warping deformation in laser powder bed fusion additive manufacturing;

[0067] Figure 6 This is a schematic diagram of the installation of the warped block recognition module and the powder laying quality detection module;

[0068] Figure 7 It is a schematic diagram of the composition of the warped block recognition module;

[0069] Figure 8 Schematic diagram of the block layer-by-layer thermal distribution image acquisition module

[0070] In the figure: 1. powder cylinder; 2. powder spreading arm; 3. flexible scraper; 4. forming cylinder; 5. infrared receiving device; 6. first fixed bracket; 7. second fixed bracket; 8. thermal imager; 9. third fixed bracket; 10. infrared emitting device; 11. powder bed platform; 12. first fixed table; 13. first high-definition camera; 14. first slider; 15. second slider; 16. second high-definition camera; 17. second fixed table; 18. fill light; 19. first locking bolt; 20. second locking bolt; 21. third fixed table; 22. third slider. DETAILED DESCRIPTION

[0071] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and is by no means any limitation to the present invention and its application or use. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Unless otherwise specified, the relative arrangement, expressions and numerical values ​​of the components and steps described in these embodiments do not limit the scope of the present invention. The techniques, methods and equipment known to ordinary technicians in the relevant fields may not be discussed in detail, but in appropriate cases, the techniques, methods and equipment should be regarded as part of the specification. In all examples shown and discussed here, any specific value should be interpreted as merely exemplary, rather than as a limitation. Therefore, other examples of the exemplary embodiments may have different values.

[0072] For ease of description, spatially relative terms such as "above", "above", "on the upper surface of", "above", etc. may be used here to describe the spatial positional relationship between a device or feature and other devices or features as shown in the figure. It should be understood that spatially relative terms are intended to include different orientations of the device in use or operation in addition to the orientation described in the figure. For example, if the device in the accompanying drawings is inverted, the device described as "above other devices or structures" or "above other devices or structures" will be positioned as "below other devices or structures" or "below other devices or structures". Thus, the exemplary term "above" can include both "above" and "below". The device can also be positioned in other different ways (rotated 90 degrees or in other orientations).

[0073] Combined with the picture Figure 1-Figure 8 The workflow of a laser powder bed fusion additive manufacturing warpage intelligent monitoring and autonomous control system of the present invention includes the following steps:

[0074] Step S1: Establish a multi-block model in the slicing software, assign different process parameters to each block, such as laser power, scanning speed, scanning strategy, etc., and then import the slicing data of the block into the printing device, such as Figure 1 shown.

[0075] Step S2: Install the new substrate and perform manual leveling, pour metal powder into the powder tank, replace the new scraper and debug the powder spreading device to achieve a good powder spreading effect, and then introduce argon gas into the closed forming cavity for gas washing. The contents involved in this step are included in the following. Figure 2 The workflow diagram is shown in the figure.

[0076] Step S3: After completing the above preparations, the block starts printing, and the block warping identification and shutdown module, powder laying quality monitoring and adaptive compensation module, block layer-by-layer thermal distribution data acquisition module and remote monitoring module start working. The workflow of each module is as follows: Figure 2 shown.

[0077] Step S4: The block warping identification and shutdown module starts to monitor the surface morphology of each block in real time. The four fill lights 18 located under the first fixed platform 12 are turned on to make the image collected by the first high-definition camera 13 clearer. If the image recognition program identifies a warped block, it will combine the slice models of each block to accurately locate the position of the block on the substrate. Then the program determines whether to continue printing the block based on whether the warping degree exceeds the threshold. If it exceeds the threshold, the printing of the block stops. If it does not exceed the threshold, the block continues to print, and the warping identification system continues to monitor the block. Other well-formed blocks continue to print. The detailed workflow is as follows: Figure 3As shown. The first high-definition camera is installed perpendicular to the powder bed, which can avoid the inaccurate defect recognition caused by the installation of the first high-definition camera at a certain tilt angle to the powder bed, resulting in erroneous decisions made by the program. Because during the printing process, the metal luster of the horizontal cross-section of the block is inconsistent at different angles and in different areas, which will cause errors in visual dynamic recognition. The first high-definition camera is installed on the first fixed bracket above the powder wall through the first fixed platform. The camera can be adjusted to a suitable height according to the size of the printed block to accurately obtain the surface forming quality of the block. That is, the first fixed platform is assembled to the first fixed bracket through the first slider to realize the lifting and lowering adjustment of the first high-definition camera.

[0078] When the powder-laying wall moves to the left to a specific position, the first high-definition camera turns on the camera function and continuously photographs the surface of the block. During the camera process, it cooperates with the image recognition algorithm to judge in real time whether the surface of each formed block is warped. If the warping is serious, the recognition program will combine the position of the warped block in the real-time printing screen with the slice model of all blocks to accurately locate the position of the warped block on the substrate and automatically close the printing of the warped block, while other well-formed blocks continue to be printed.

[0079] It can be seen that in the present invention, the first high-definition camera is mainly used to capture images of all formed blocks on the forming substrate. In other words, when the first high-definition camera turns on the shooting function, its field of view is the forming substrate, and it can clearly capture images of all formed blocks on the forming substrate and transmit the captured images to the data center of the remote monitoring module on the user side.

[0080] In the present invention, the position of the first high-definition camera when the camera function is turned on is controlled by an on-off switch, and the on-off switch includes an infrared emitting sensor and an infrared receiving device used in conjunction; the infrared receiving device is arranged on the first fixed platform close to the first high-definition camera, and the infrared emitting sensor is installed on the side of the powder bed adjacent to the forming cylinder; when the powder spreading mechanism moves from one side of the powder cylinder to one side of the forming cylinder, the infrared receiving device receives the infrared rays emitted by the infrared emitting device on the side of the powder bed, and the first high-definition camera turns on the shooting function. The central axis of the installation position of the infrared emitting sensor on one side of the powder bed is tangent to the edge of the circular or square forming cylinder. Taking the circular forming cylinder as an example, when the powder spreading wall moves to the left to spread powder, the infrared receiving device located below the camera installation platform receives the infrared rays emitted by the infrared emitting device on the side of the powder bed, and the camera turns on the camera function.

[0081] The user-side remote monitoring module analyzes the block surface images taken frame by frame by the first high-definition camera. Since the surface flatness and edge contour deformation degree of different warped blocks are different, the grayscale values ​​calculated from the collected images will be different. Therefore, the degree of block warping can be judged by the grayscale value. When the block is printed to 20 layers (20 layers are based on experience, and in fact, this value may not be selected), the grayscale value calculated from the collected block surface image after filtering and denoising to remove the powder bed background is used as the standard grayscale value. The percentage of the real-time grayscale value exceeding the standard grayscale value is calculated. When the percentage is between 0% and 25%, it is slight warping, when the percentage is between 25% and 50%, it is moderate warping, and when the percentage is between 50% and 100%, it is severe warping. For blocks with slight and moderate warping, continue printing and monitoring, and stop printing for blocks with severe warping.

[0082] In other words, the user-side remote monitoring module of the present invention determines whether the surface warping degree of each formed block exceeds a preset range based on the block surface image fed back by the first high-definition camera, and controls the printing condition of the corresponding formed block according to the judgment result; it includes a block warping identification and shutdown module; the block warping identification and shutdown module includes a warping block image processing module and a warping block image evaluation module; wherein:

[0083] The warped block image processing module is used to receive the block surface image fed back by the first high-definition camera and filter and denoise the received block surface image frame by frame, and calculate the standard gray value when the number of deposition layers of each formed block reaches a preset number of layers, and calculate the real-time gray value of each formed block one by one in each deposition layer after the preset number of layers, and upload the percentage Y of the real-time gray value to the standard gray value to the warped block image evaluation module; wherein:

[0084] The calculation formula of standard gray value Gray0 is as follows:

[0085] Gray0=(0.299R0+0.587G0+0.144B0) / N

[0086] Among them, R0 is the red component of all pixels in the block surface image; G0 is the green component of all pixels in the block surface image; B0 is the blue component of all pixels in the block surface image; N is the number of formed blocks contained in the block surface image;

[0087] The real-time gray value of any formed block i in the block surface image is Gray i The calculation formula is as follows:

[0088] Gray i =0.299R i+0.587G i +0.144B i

[0089] Among them, R i G is the red component of the pixel points included in any formed block i in the block surface image; i B is the green component of the pixel points included in any formed block i in the block surface image; i is the blue component of the pixel points included in any formed block i in the block surface image;

[0090] The warped block image evaluation module first determines whether the real-time grayscale value of any formed block i is greater than the standard grayscale value. When the real-time grayscale value of the formed block i is less than the standard grayscale value, the formed block i continues to be printed. Otherwise, the percentage Y of the real-time grayscale value of the formed block i exceeding the standard grayscale value is calculated by the following formula: i :

[0091]

[0092] Among them, Gray0 is the standard gray value, Gray i is the real-time gray value of any formed block i;

[0093] The warped block image evaluation module calculates the percentage Y i , judge the severity of the surface warping of the corresponding formed block i, and determine the printing condition of the formed block i according to the judgment result.

[0094] In the warping block image judgment module, the judgment criteria for the severity of surface warping of the formed block i are: when the percentage is between 0% and 25%, it is slight warping, when the percentage is between 25% and 50%, it is moderate warping, and when the percentage is between 50% and 100%, it is severe warping; in the warping block image judgment module, when the severity of surface warping of the formed block i is slight or moderate warping, continue printing and monitoring; when the severity of surface warping of the formed block i is severe warping, stop printing.

[0095] Step S5: The powder spreading quality monitoring and adaptive compensation module monitors the powder spreading quality of the blocks around the warped block in real time, and reduces the influence of the warped block on the powder spreading quality of the surrounding blocks by automatically adjusting the powder supply amount. When the warped block is identified in step S4, the four fill lights located below the second fixed platform 17 are turned on, which also makes the image captured by the second high-definition camera 16 clearer. The second high-definition camera moves from right to left with the powder spreading wall 2. The image recognition program in the user-side remote monitoring module monitors the surface flatness of the powder bed and the powder spreading quality of the warped block and its surrounding blocks in real time, and independently selects the appropriate powder supply amount according to the powder spreading quality to perform powder supply compensation. If the warped block and its surrounding blocks are evenly powdered in subsequent monitoring, the powder supply amount returns to normal. The detailed work flow is as follows: Figure 4 shown.

[0096] During the filming process, the second high-definition camera cooperates with the image recognition algorithm to judge in real time the defects remaining after the scraper scrapes the surface powder of the powder cylinder and the size of the exposed metal area on the surface of the block during the scraper spreading the powder. If the exposed area is too large, the powder spreading compensation program will be started. Generally, 3 times the powder supply is used during the printing process, that is, the forming cylinder drops 20μm and the powder cylinder rises 60μm. In order to improve the powder spreading quality, the powder supply in the powder spreading compensation program is set to 4, 5, and 6 times the powder supply. The program will automatically set different powder supply amounts for compensation according to the size of the exposed metal surface area of ​​the block. If it is recognized that the powder spreading quality has improved, the compensation program will gradually reduce the powder supply until the initial 3 times the powder supply is restored to prevent the remaining powder from being too little and the overall printing cannot be completed.

[0097] The user-side remote monitoring module analyzes the frame-by-frame powder spreading images taken by the second camera to obtain the uneven powder spreading area within the powder spreading image, including the exposed surface of the formed block after powder spreading and the flatness of the powder bed. When the uneven powder spreading area is lower than the set value, there is no need to adjust the powder supply; when the uneven powder spreading area is higher than or equal to the set value, the powder supply needs to be adjusted, and the percentage of the uneven powder spreading area exceeding the set value is calculated. When the percentage is between 0% and 30%, 4 times the powder supply is used; when the percentage is between 30% and 60%, 5 times the powder supply is used; when the percentage is between 60% and 100%, 6 times the powder supply is used. Specifically, the powder spreading image is first denoised by filtering and denoising, and the acquired powder spreading image is processed by the threshold function to obtain a binary gray-white image, and then the pixels in the image are calculated. The area of ​​the black pixel region in the image is the uneven powder spreading area. The image is binarized and converted into a gray-white image containing only black and white colors. The number of black pixels in the image is counted and the total area occupied by these black pixels is calculated based on the image resolution. This area represents the size of the unevenly powdered area in the image.

[0098] In other words, in the present invention, the user-side remote monitoring module evaluates the powder spreading quality of the current powder layer according to the current powder layer image transmitted by the second high-definition camera, and determines the powder supply amount of the powder cylinder according to the evaluation result. Specifically, the user-side remote monitoring module includes a powder spreading quality monitoring and powder supply amount adaptive compensation module, and the powder spreading quality monitoring and powder supply amount adaptive compensation module includes a powder spreading image processing module and a powder spreading image quality evaluation module, wherein:

[0099] The powder spreading image processing module is used to receive the current powder layer image transmitted by the second high-definition camera, filter and de-noise the received current powder layer image, and then use the threshold function to process it to obtain a binary gray-white image, and then count the number of black pixels in the gray-white image, and calculate the total area occupied by these black pixels based on the resolution of the gray-white image, so as to obtain the area S1 of the area with poor powder spreading quality in the current powder layer image, and transmit it to the powder spreading image quality evaluation module;

[0100] The powder spreading image quality judgment module is used to compare whether the area S1 of the area with poor powder spreading quality in the current powder layer image exceeds the powder spreading quality judgment set value U. When the judgment result shows that the area S1 of the area with poor powder spreading quality in the current powder layer image does not exceed the powder spreading quality judgment set value U, the current powder supply is maintained unchanged. Otherwise, the percentage G of the area S1 of the area with poor powder spreading quality in the current powder layer image exceeding the powder spreading quality judgment set value U is calculated, and the calculated percentage G is compared with the preset value. When the percentage is between 0% and 30%, 4 times the powder supply is adopted; when the percentage is between 30% and 60%, 5 times the powder supply is adopted; when the percentage is between 60% and 100%, 6 times the powder supply is adopted; wherein:

[0101]

[0102] In the formula: d1 is the diameter of the forming cylinder, d2 is the diameter of the powder cylinder, N is the number of forming blocks, and a and b correspond to the length and width of the forming block.

[0103] It can be seen that in the present invention, the warping identification and shutdown module is linked with the powder spreading quality monitoring and adaptive compensation module. When the former module identifies a warped block, if the degree of warping is too large, the printing of the warped block will be stopped immediately. If the degree of warping is relatively light, the block can continue to be printed because the degree of warping may be reduced through subsequent improvement of the powder spreading quality. After identifying the warped block, the powder spreading quality monitoring and adaptive compensation module identifies the powder spreading quality of the blocks around the warped block and starts the powder spreading compensation program to increase the powder supply until it is monitored that the surface of the surrounding blocks is uniform, and then the initial powder supply is restored.

[0104] Step S6: The layer-by-layer heat distribution data acquisition module collects the heat distribution image of the block surface in real time. When printing starts, the thermal imager 8 starts working and collects the heat distribution image of each block surface layer by layer. Different process parameters have different heat distribution. The heat distribution image has different color depths, indicating the heat level. The darker the color, the more serious the heat accumulation. The temperature value is displayed in the local area with the same color. Then the background program builds a three-dimensional heat distribution model of the block based on the collected data, which is conducive to researchers studying the defect formation mechanism and providing a reference for process parameter optimization. The core component is the thermal imager, which is installed above the bracket on the left side of the forming cylinder. The third platform that fixes the thermal imager is adjusted to a suitable inclination angle by rotating up and down, and the height of the thermal imager can also be adjusted by moving the slider embedded in the third fixed bracket up and down.

[0105] The image data collected by warpage identification and powder laying quality monitoring, as well as the heat distribution data collected layer by layer by the thermal imager, will be synchronized to the data processing center. Users only need to log in to the program remotely to observe the block printing quality and equipment status in real time.

[0106] The LAN and wireless WIFI modules in the user remote monitoring module are located in the computer host of the printing device. The function of the LAN and wireless WIFI modules is to transmit a large number of real-time video images, heat distribution data and printing device operating status to the data processing center. After analysis and evaluation, the key data is remotely transmitted in small batches to mobile terminals such as mobile phones via wireless WIFI.

[0107] Step S7: The user remote monitoring module starts working when printing begins. During the printing process, the images taken by the high-definition camera and the thermal imager are uploaded to the data processing center in real time. Researchers can log in to the user-side program to remotely observe the printing process. If a fault alarm signal occurs during the printing process, it will be immediately uploaded to the user-side program, which can solve the problem of long-term on-site duty.

[0108] Step S8: After printing is completed, wait for the sample to slowly cool down to room temperature inside the forming chamber, take out the sample, and perform subsequent processing.

[0109] The laser powder bed fusion additive manufacturing warping deformation intelligent monitoring and autonomous control system of the present invention can monitor the warped blocks in real time and automatically close the blocks with a larger degree of warping, while continuing to print the well-formed blocks. Through the powder spreading adaptive compensation program, the influence of the warped blocks on the surrounding blocks is reduced. The user can remotely monitor the printing process without having to be on duty on site for a long time. The block surface thermal distribution data collected layer by layer can provide a reference for the next printing parameter adjustment and the study of the defect formation mechanism caused by heat accumulation.

[0110] The present invention can effectively solve the practical problems faced in performing block printing with multiple process parameters during the development of new material processes.

[0111] The present invention is described in further detail below in conjunction with embodiments:

[0112] Embodiment 1

[0113] Taking the printing of 98wt.% W-1.5wt.% TiC-0.5wt.% Y2O3 block by single laser equipment as an example, the process parameters of the material are studied.

[0114] Step S1: Establish a block model in the slicing software and assign corresponding process parameters to each block. The block size is 8mm×8mm×6mm, the laser power range is 200W~450W, the scanning speed range is 200mm / s~600mm / s, the powder layer thickness is 25μm, the interlayer rotation is 67°, the checkerboard scanning strategy, and then the slicing data is imported into the printing device.

[0115] Step S2: Install a new substrate and perform manual leveling, pour metal powder into the powder cylinder, replace a new scraper and debug the powder spreading device to achieve a good powder spreading effect, then introduce argon gas into the closed forming chamber for gas washing until the oxygen content inside the forming chamber is less than 500ppm.

[0116] Step S3: After completing the above preparations, the substrate is first scanned for 3 layers, and then the block printing is started. The intelligent monitoring and autonomous control system for warping deformation also starts to run. The block warping identification and shutdown module starts to monitor the surface morphology of each block in real time. During the printing of multiple blocks, the real-time acquired block surface image is analyzed, and the standard gray value Gray1 of each block is calculated to be 100, and the real-time gray value Gray2 and the percentage Y are calculated to determine whether Gray2 is greater than Gray1. If it is less than, continue to print the block. If it is greater than, choose to continue printing or stop printing the warped block according to the size of the percentage Y. The powder spreading quality monitoring and adaptive compensation module starts to monitor the powder spreading quality of the blocks around the warped block in real time and automatically adjusts the powder supply to reduce the influence of the warped block on the powder spreading quality of the surrounding blocks. The specific process is to analyze the real-time acquired powder spreading quality image and calculate the powder spreading quality assessment value U to be 42.25mm 2 And calculate the real-time uneven powder spreading area S1 and percentage G, and determine whether the uneven powder spreading area exceeds the set value. If it exceeds, select the corresponding powder supply according to the excess range. If it does not exceed, maintain the initial powder supply. The layer-by-layer thermal distribution data acquisition module collects the heat distribution image of the block surface in real time, and the data collected during the printing process is uploaded to the data processing center in real time.

[0117] Step S4: Check whether the network connected to the printing device is stable and whether the real-time image of the printing process can be viewed by logging into the client program. After completing the above steps, the researcher leaves the laboratory to remotely monitor the printing process.

[0118] Step S5: When printing is completed, the device will send a printing completion instruction to the user end. The researchers only need to wait for the sample to slowly cool to room temperature inside the forming chamber, take out the sample, and perform subsequent processing.

[0119] The intelligent monitoring and autonomous control system for warping deformation in laser powder bed fusion additive manufacturing operated smoothly during the single laser forming process of 98wt.% W-1.5wt.% TiC-0.5wt.% Y2O3 blocks. It was monitored that 8 blocks had severe warping, 2 blocks had moderate warping, 4 blocks had slight warping, and the remaining blocks did not show any warping. The subsequent printing of the blocks with severe warping was directly stopped. The warping degree of the blocks with moderate and slight warping was reduced after autonomous control of the powder supply, and the influence of the warped blocks on the surrounding blocks was reduced. The heat accumulation distribution map collected layer by layer provided a certain reference for the study of the warping and crack formation mechanism of the tungsten alloy blocks formed by single laser. The entire printing process does not require manual supervision, which significantly improves the intelligence of the additive manufacturing process.

[0120] Embodiment 2

[0121] Taking the dual laser follow-up printing equipment to form 98wt.% W-1.5wt.% TiC-0.5wt.% Y2O3 blocks, the process parameters of the new material were studied as an example.

[0122] Step S1: The researchers informed the company engineers of the printing plan. The engineers established a block model in the slicing software according to the printing plan and assigned corresponding process parameters to each block. The block size was 10mm×10mm×8mm, the laser power range was 100W~250W, the scanning speed range was 200mm / s~400mm / s, the powder layer thickness was 25μm, the inter-layer rotation was 67°, and the checkerboard scanning strategy was used. The slicing data was then imported into the printing device.

[0123] Step S2: Install a new substrate and perform manual leveling, pour metal powder into the powder cylinder, replace a new scraper and debug the powder spreading device to achieve a good powder spreading effect, then introduce argon gas into the closed forming chamber for gas washing until the oxygen content inside the forming chamber is less than 500ppm.

[0124] Step S3: After completing the above preparations, the substrate is first scanned for 3 layers, and then the block printing is started. The intelligent monitoring and autonomous control system for warping deformation also starts to run. The block warping identification and shutdown module starts to monitor the surface morphology of each block in real time. During the printing of multiple blocks, the real-time acquired block surface image is analyzed, and the standard gray value Gray1 of each block is calculated to be 120, and the real-time gray value Gray2 and percentage Y are calculated to determine whether Gray2 is greater than Gray1. If it is less than, continue to print the block. If it is greater than, choose to continue printing or stop printing the warped block according to the size of the percentage Y. The powder spreading quality monitoring and adaptive compensation module starts to monitor the powder spreading quality of the blocks around the warped block in real time and automatically adjusts the powder supply to reduce the influence of the warped block on the powder spreading quality of the surrounding blocks. The specific process is to analyze the real-time acquired powder spreading quality image and calculate the powder spreading quality assessment value U to be 72.25mm 2 And calculate the real-time uneven powder spreading area S1 and percentage G, and determine whether the uneven powder spreading area exceeds the set value. If it exceeds, select the corresponding powder supply according to the excess range. If it does not exceed, maintain the initial powder supply. The layer-by-layer thermal distribution data acquisition module collects the heat distribution image of the block surface in real time, and the data collected during the printing process is uploaded to the data processing center in real time.

[0125] Step S4: The engineer checks whether the network connected to the printing device is stable, and informs the researchers whether they can view the real-time images during the printing process by logging into the user-side program. After the adjustment is completed, the researchers can remotely monitor the printing process. If an unexpected situation occurs during the printing process, such as the printing device detecting insufficient chamber protective gas, the printing device will immediately alarm and upload the problem instruction to the user side. After receiving the instruction, the researchers will immediately contact the engineer to handle it on site.

[0126] The intelligent monitoring and autonomous control system for warping deformation in laser powder bed fusion additive manufacturing operated smoothly during the dual laser forming process of 98wt.%W-1.5wt.%TiC-0.5wt.%Y2O3 blocks. It was detected that 5 blocks were severely warped, 4 blocks were moderately warped, 6 blocks were slightly warped, and the remaining blocks did not show any warping. The subsequent printing of the severely warped blocks was directly stopped. The warping degree of the moderately and slightly warped blocks was reduced after autonomous control of the powder supply, and the influence of the warped blocks on the surrounding blocks was reduced. The heat accumulation distribution map collected layer by layer provided a certain reference for the study of the warping and crack formation mechanism of the tungsten alloy blocks formed by dual lasers. The use of this system to monitor the printing process in real time can capture all-round unexpected situations in the printing process and avoid the omission of details by manual supervision.

[0127] Step S5: When printing is completed, the device will send a printing completion instruction to the user end. The researchers only need to wait for the sample to slowly cool to room temperature inside the forming chamber, and then contact the engineer to take out the sample for subsequent processing.

[0128] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A laser powder bed fusion additive manufacturing warpage deformation intelligent monitoring and autonomous control system, comprising a powder platform, a forming cylinder, a powder cylinder and a powder spreading mechanism; characterized in that: It also includes a data acquisition device and a user-side remote monitoring module, wherein the data acquisition device includes a first and a second high-definition camera, wherein: The first high-definition camera is installed on the powder spreading mechanism, and is used to capture the block surface images including all formed blocks layer by layer, and can transmit the captured block surface images to the user-side remote monitoring module; The second high-definition camera is mounted on the powder spreading mechanism, and is used to capture the image of the current powder layer spread by the powder spreading mechanism on the powder platform, and can transmit the captured image of the current powder layer to the user-side remote monitoring module; The user-side remote monitoring module determines whether the surface warping degree of each formed block exceeds a preset range based on the block surface image fed back by the first high-definition camera, and controls the printing condition of the corresponding formed block based on the determination result; The user-side remote monitoring module evaluates the powder spreading quality of the current powder layer according to the current powder layer image transmitted by the second high-definition camera, and determines the powder supply amount of the powder cylinder according to the evaluation result.

2. The laser powder bed fusion additive manufacturing warpage deformation intelligent monitoring and autonomous control system according to claim 1 is characterized in that: The user-side remote monitoring module includes a block warping identification and shutdown module; the block warping identification and shutdown module includes a warping block image processing module and a warping block image evaluation module; wherein: The warped block image processing module is used to receive the block surface image fed back by the first high-definition camera and filter and denoise the received block surface image frame by frame, and calculate the standard gray value when the number of deposition layers of each formed block reaches a preset number of layers, and calculate the real-time gray value of each formed block one by one in each deposition layer after the preset number of layers, and upload the percentage Y of the real-time gray value to the standard gray value to the warped block image evaluation module; wherein: The calculation formula of standard gray value Gray0 is as follows: Gray0=(0.299R0+0.587G0+0.144B0) / N Among them, R0 is the red component of all pixels in the block surface image; G0 is the green component of all pixels in the block surface image; B0 is the blue component of all pixels in the block surface image; N is the number of formed blocks contained in the block surface image; The real-time gray value Gray of any formed block i in the block surface image i The calculation formula is as follows: Gray i =0.299R i +0.587G i +0.144B i Among them, R i G is the red component of the pixel points included in any formed block i in the block surface image; i B is the green component of the pixel points included in any formed block i in the block surface image; i is the blue component of the pixel points included in any formed block i in the block surface image; The warped block image evaluation module first determines whether the real-time grayscale value of any formed block i is greater than the standard grayscale value. When the real-time grayscale value of the formed block i is less than the standard grayscale value, the formed block i continues to be printed. Otherwise, the percentage Y of the real-time grayscale value of the formed block i exceeding the standard grayscale value is calculated by the following formula: i : Among them, Gray0 is the standard gray value, Gray i is the real-time gray value of any formed block i; The warped block image evaluation module calculates the percentage Y i , judge the severity of the surface warping of the corresponding formed block i, and determine the printing condition of the formed block i according to the judgment result.

3. The laser powder bed fusion additive manufacturing warpage deformation intelligent monitoring and autonomous control system according to claim 2 is characterized in that: In the warped block image evaluation module, the evaluation criteria for the severity of the surface warping of the formed block i are: when the percentage is between 0% and 25%, it is slight warping; when the percentage is between 25% and 50%, it is moderate warping; when the percentage is between 50% and 100%, it is severe warping; In the warping block image evaluation module, when the severity of the surface warping of the formed block i is mild or moderate, printing and monitoring are continued; when the severity of the surface warping of the formed block i is severe, printing is stopped.

4. The laser powder bed fusion additive manufacturing warpage deformation intelligent monitoring and autonomous control system according to claim 2 is characterized in that: The user-side remote monitoring module includes a powder spreading quality monitoring and powder supply adaptive compensation module, which includes a powder spreading image processing module and a powder spreading image quality evaluation module, wherein: The powder spreading image processing module is used to receive the current powder layer image transmitted by the second high-definition camera, filter and de-noise the received current powder layer image, and then use the threshold function to process it to obtain a binary gray-white image, and then count the number of black pixels in the gray-white image, and calculate the total area occupied by these black pixels based on the resolution of the gray-white image, so as to obtain the area S1 of the area with poor powder spreading quality in the current powder layer image, and transmit it to the powder spreading image quality evaluation module; The powder spreading image quality judgment module is used to compare whether the area S1 of the area with poor powder spreading quality in the current powder layer image exceeds the powder spreading quality judgment set value U. When the judgment result shows that the area S1 of the area with poor powder spreading quality in the current powder layer image does not exceed the powder spreading quality judgment set value U, the current powder supply is maintained unchanged. Otherwise, the percentage G of the area S1 of the area with poor powder spreading quality in the current powder layer image exceeding the powder spreading quality judgment set value U is calculated, and the calculated percentage G is compared with the preset value. When the percentage is between 0% and 30%, 4 times the powder supply is adopted; when the percentage is between 30% and 60%, 5 times the powder supply is adopted; when the percentage is between 60% and 100%, 6 times the powder supply is adopted; wherein: In the formula: d1 is the diameter of the forming cylinder, d2 is the diameter of the powder cylinder, N is the number of forming blocks, and a and b correspond to the length and width of the forming block.

5. The laser powder bed fusion additive manufacturing warpage deformation intelligent monitoring and autonomous control system according to claim 2 is characterized in that: The data acquisition device also includes a thermal imager; The thermal imager is mounted on the powder platform and is used to collect the surface heat distribution of each formed block layer by layer, and can transmit the collected surface heat distribution of each formed block to the control device; The user remote monitoring module generates a heat distribution stereogram in real time according to the received surface heat distribution diagrams of each formed block.

6. The laser powder bed fusion additive manufacturing warpage deformation intelligent monitoring and autonomous control system according to claim 5 is characterized in that: The local area network and wireless WIFI module in the user remote monitoring module are located in the computer host of the printing device. The user-side remote monitoring module obtains the data collected by the first and second cameras and the thermal imager in real time through the local area network.

7. The laser powder bed fusion additive manufacturing warpage deformation intelligent monitoring and autonomous control system according to claim 6 is characterized in that: The first high-definition camera is installed on the powder spreading arm at the side close to the powder cylinder, while the second high-definition camera is installed on the powder spreading arm at the side close to the forming cylinder, and the first and second high-definition cameras are arranged symmetrically.

8. The laser powder bed fusion additive manufacturing warpage deformation intelligent monitoring and autonomous control system according to claim 7 is characterized in that: The first and second fixing brackets are symmetrically mounted on the powder spreading arm, the first high-definition camera is mounted on the first fixing platform, and the first fixing platform is mounted on the first fixing bracket in a telescopic manner through a first slider; the second high-definition camera is mounted on the second fixing platform, and the second fixing platform is mounted on the second fixing bracket in a telescopic manner through a second slider; The thermal imager is mounted on the powder bed platform through a third fixed bracket.

9. The laser powder bed fusion additive manufacturing warpage deformation intelligent monitoring and autonomous control system according to claim 8, characterized in that: It also includes an on / off switch for controlling the first high-definition camera; the on / off switch includes an infrared emitting sensor and an infrared receiving device used in conjunction with each other; the infrared receiving device is arranged on the first fixed platform close to the first high-definition camera, and the infrared emitting sensor is installed on the side of the powder bed adjacent to the forming cylinder; when the powder spreading mechanism moves from one side of the powder cylinder to one side of the forming cylinder, when the infrared receiving device receives the infrared ray emitted by the infrared emitting device on one side of the powder bed, the first high-definition camera starts the shooting function.

10. A method for intelligent monitoring and autonomous control of warpage deformation in laser powder bed fusion additive manufacturing, characterized in that: It includes block warping identification and shutdown control, powder spreading quality monitoring and powder supply adaptive compensation control; among which: Block warping identification and shutdown control are achieved through the following steps: Step 101, collecting block surface images of all formed blocks on the printing substrate layer by layer in real time; Step 102, filtering and denoising the block surface image collected in step 101 and performing block edge enhancement processing; Step 103, when the number of deposition layers of each formed block reaches a preset number of layers, based on the block surface image processed in step 102, calculating the grayscale value of the block surface image as the standard grayscale value; Step 104, starting from the next deposition layer of the preset number of layers, calculating the real-time gray value of each formed block layer by layer; Step 105, judging whether the real-time gray value of each formed block is greater than the standard gray value, when the judgment result shows that the real-time gray value of the formed block i is less than the standard gray value, continue to print the formed block i, otherwise go to step 106; Step 106, calculating the percentage Y of the real-time gray value of each formed block exceeding the standard gray value; Step 107, based on the percentage Y of the real-time gray value calculated in step 106 exceeding the standard gray value, determine the surface warping degree of the corresponding formed block: when the percentage Y is between 0% and 25%, it is slight warping; when the percentage Y is between 25% and 50%, it is moderate warping; when the percentage Y is between 50% and 100%, it is severe warping; Step 107, the slightly warped and moderately warped formed blocks continue to be printed and monitored, and the severely warped formed blocks stop printing; Powder spreading quality monitoring and powder supply adaptive compensation control are achieved through the following steps: Step 201, collecting an image of the current powder layer spread by the powder spreading mechanism on the powder platform in real time; Step 202, performing filtering, noise reduction and binarization processing on the image of the current powder layer collected in step 201; Step 203, respectively calculate the area S1 of the area with poor powder spreading quality in the current powder layer image, the powder spreading quality judgment setting value U, Wherein: d1 is the diameter of the forming cylinder, d2 is the diameter of the powder cylinder, N is the number of forming blocks, a and b correspond to the length and width of the forming blocks; the area S1 of the area with poor powder spreading quality in the current powder layer image is obtained by: based on the binarized gray-white image, then counting the number of black pixels in the gray-white image, and calculating the total area occupied by these black pixels based on the resolution of the gray-white image, so as to obtain the area S1 of the area with poor powder spreading quality in the current powder layer image; Step 204, judging whether the area S1 of the region with poor powder spreading quality in the current powder layer image exceeds the powder spreading quality judgment set value U. When the judgment result shows that the area S1 of the region with poor powder spreading quality in the current powder layer image is lower than the powder spreading quality judgment set value U, there is no need to adjust the powder supply amount. Otherwise, the process proceeds to step 205. Step 205, calculate the percentage G of the area S1 of the area with poor powder spreading quality in the current powder layer image exceeding the powder spreading quality judgment set value U, and compare the calculated percentage G with the preset value, when the percentage is between 0% and 30%, use 4 times the powder supply; when the percentage is between 30% and 60%, use 5 times the powder supply; when the percentage is between 60% and 100%, use 6 times the powder supply; The percentage G of the area S1 of the region with poor powder spreading quality in the current powder layer image exceeding the powder spreading quality judgment set value U is calculated by the following formula:

Images