3D printing adaptive powder spreading method

Abstract

The application discloses a 3D printing self-adaptive powder spreading method, comprising the following steps: (1) converting a three-dimensional model of a to-be-printed object into a two-dimensional model cross-section feature and transmitting the two-dimensional model cross-section feature to an upper computer; (2) judging whether the cross-section feature of the current printing layer changes; if not, keeping the powder feeding amount unchanged; if yes, adjusting the powder amount factor according to the region range where the current cross-section feature is located, and outputting a specific powder feeding amount; (3) after completing the powder spreading operation of the current layer printing, shooting a powder layer image of the current printing layer and transmitting the powder layer image to the upper computer, the upper computer performing powder bed feature identification and judging the uniformity of the current powder layer according to the powder layer image of the current printing layer, if the powder layer is uniform, completing the printing of the layer; if the powder layer is not uniform, adjusting the powder feeding amount and re-performing the powder spreading until the powder layer of the current printing layer is uniform; (4) after completing the printing of the current layer, lowering by one layer thickness, and repeating steps (1) to (3) to complete the powder spreading of the next layer. The method can save materials and reduce labor cost.

Description

Technical Field

[0001] This invention relates to the field of 3D printing technology, specifically to an adaptive powder spreading method for 3D printing. Background Technology

[0002] 3D printing technology, also known as additive manufacturing, is based on the principle of "discrete + stacking." It involves slicing a three-dimensional model, transferring the slice data into a printing device, and then using various techniques to deposit material layer by layer to form a three-dimensional solid part. As a type of additive manufacturing technology, laser powder bed selective melting (SPDM) technology features high forming precision, excellent surface quality, and high mechanical properties. It offers high processing flexibility in manufacturing complex structures and is increasingly being used in aerospace, shipbuilding, automotive, and biomedical fields.

[0003] In existing 3D printing processes, if powder bed defects are not detected in time, it will affect the continuity of part forming, leading to printing failures and reduced production efficiency. To address real-time monitoring of molten pool defects during powder bed printing, patent application number 202111505450.2 discloses a device and method for SLM powder bed defect identification and real-time molten pool status monitoring. This method combines a high-speed camera and image defect recognition technology for real-time monitoring of powder bed and molten pool processes. However, this method requires pausing the process and manually adjusting parameters when powder bed defects caused by molten pool problems are detected, which consumes a significant amount of manpower. Furthermore, existing technologies use the same amount of powder to cover the entire molding substrate and surrounding printing platform area in each powder bed layer, then use a scraper to remove excess powder. While this ensures the molding substrate is fully covered during printing, the cross-sectional features of some parts change continuously during printing. Using the same amount of powder throughout requires frequent powder replenishment, wasting labor costs and increasing powder costs. Summary of the Invention

[0004] Purpose of the invention: To address the above problems, this invention provides a 3D printing adaptive powder spreading method. Without manual parameter adjustment, the powder feeding amount is adjusted in a timely manner by identifying changes in the cross-sectional features of the model, ensuring that the molded substrate is evenly spread while saving powder costs and reducing labor costs.

[0005] Technical Solution: To solve the above problems, this invention discloses a 3D printing adaptive powder spreading method. The method uses a 3D printing device to complete the adaptive powder spreading operation. The 3D printing device includes a host computer, a powder feeding shaft mechanism, a powder spreading shaft mechanism, a forming cylinder mechanism, a powder collection cylinder, and a forming substrate. The 3D printing device is also equipped with a camera. The specific steps include:

[0006] (1) Convert the three-dimensional model of the item to be printed into the cross-sectional features of the two-dimensional model and transmit the cross-sectional features of each layer to the host computer;

[0007] (2) The host computer determines whether the cross-sectional features of the current printed layer have changed relative to the cross-sectional features of the previous printed layer; if the cross-sectional features have not changed, the amount of powder delivered to the molding substrate by the powder feeding shaft mechanism remains unchanged; if the cross-sectional features have changed, the powder quantity factor is adjusted according to the region where the current printed layer cross-sectional features are located, and the powder feeding shaft mechanism is controlled to output a specific amount of powder; the region specifically refers to several regions formed by dividing the molding substrate along the length direction of the molding substrate; the powder feeding amount output by the powder feeding shaft mechanism and the powder quantity factor satisfy:

[0008] Y = α·X

[0009] In the formula, Y represents the amount of powder fed, α is the powder quantity factor, and X is the amount of powder dispensed.

[0010] (3) After the host computer controls the powder spreading shaft mechanism to complete the powder spreading operation of the current layer printing on the molding substrate, the camera acquires the powder layer image of the current printing layer and transmits it to the host computer. The host computer performs powder bed feature recognition and judges the uniformity of the current powder layer based on the powder layer image of the current printing layer. If the powder layer is uniform, the printing of the current layer is completed; if the powder layer is not uniform, the amount of powder dropped by the powder falling shaft mechanism is adjusted, and the powder spreading shaft mechanism is used to spread the powder again until the powder layer of the current printing layer is uniform.

[0011] (4) After the current layer is printed, the host computer controls the forming cylinder mechanism to descend by one layer thickness, and repeats steps (1) to (3) to complete the next layer of powder spreading and printing, until all layers are printed.

[0012] Furthermore, in step (2), when the host computer determines that the cross-sectional characteristics have changed, it adjusts the powder quantity factor according to the area range where the current printed layer cross-sectional characteristics are located, and controls the powder feeding shaft mechanism to output a specific powder feeding amount. Specifically, this includes dividing the molding substrate into N regions along the length direction of the molding substrate, which are respectively denoted as the first region, the second region, ..., the Nth region, with the first region being close to the powder feeding shaft mechanism and the Nth region being close to the powder collection cylinder; the molding substrate is located between the powder feeding shaft mechanism and the powder collection cylinder.

[0013] If the cross-sectional features change and the area where the current printed layer cross-sectional features are located does not include the Nth region, then the powder quantity factor is adjusted to control the powder delivery amount output by the powder delivery shaft mechanism, Y1; Y1 is the amount of powder required to completely cover the area from the powder delivery shaft mechanism to the edge of the molding substrate and close to the powder collection cylinder.

[0014] If the cross-sectional features change and the area where the current printed layer cross-sectional features are located includes the Nth region, then adjust the toner quantity factor and control the toner delivery amount Y2 output by the toner drop shaft mechanism; Y2 is the amount of toner required to completely cover the area from the toner drop shaft mechanism to the toner collection cylinder.

[0015] Furthermore, in step (3), the host computer uses OpenCV image recognition to perform powder bed feature recognition on the powder layer image of the current printing layer and determines the uniformity of the current powder layer, specifically including:

[0016] First, the image is processed by Gaussian filtering and noise reduction, followed by grayscale processing. Sobel edge detection is used to detect the boundaries of the regions, and then the image is segmented using a fixed threshold method. The segmented image is then input into a trained neural network to identify the features of the current powder bed, and the uniformity of the current powder layer is determined based on the identified powder bed features.

[0017] Furthermore, the 3D printing device is also equipped with a human-computer interaction system, which communicates with the host computer through an RFID reader embedded in the host computer.

[0018] During the adaptive toner spreading printing operation, the host computer feeds back the toner spreading process parameters to the human-machine interaction system in real time through the RFID reader.

[0019] Furthermore, the range of values ​​for is [5,7].

[0020] Beneficial Effects: Compared with existing technologies, the adaptive powder spreading method for 3D printing described in this invention has the following significant advantages: 1. By identifying changes in the cross-sectional features of the 2D model through a host computer, the powder quantity factor is automatically adjusted to control the powder delivery shaft mechanism to output the corresponding powder quantity, ensuring uniform powder spreading on the molded substrate and saving powder and labor costs; 2. By performing algorithmic analysis on the powder bed features captured by the camera, the host computer drives the powder delivery shaft mechanism to adaptively adjust the powder delivery quantity, avoiding powder shortages and collapse in the powder layer, and ensuring a uniform and continuous powder layer. The powder spreading process described in this invention is highly automated, resulting in excellent printing quality and significantly improved production efficiency, meeting the requirements of digital production management in 3D printing workshops. Attached Figure Description

[0021] Figure 1 The diagram shown is a schematic diagram of the 3D printing device in an embodiment of the present invention.

[0022] Figure 2 The diagram shown is a flowchart of the adaptive powder feeding rate adjustment in an embodiment of the present invention.

[0023] Figure 3 The diagram shown is a flowchart of powder layer feature identification and uniformity judgment in an embodiment of the present invention;

[0024] Figure 4 The diagram shown illustrates the selection of powder feeding amount based on the cross-sectional characteristics of the model in an embodiment of the invention. Detailed Implementation

[0025] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0026] like Figure 1 and Figure 4 As shown, the 3D printing device specifically includes a host computer 1, a PLC controller 2, a powder feeding shaft mechanism 201, a powder spreading shaft mechanism 202, a forming cylinder mechanism 203, a human-machine interface system 4, an RFID reader 101, etc. The host computer 1 drives the operation of the powder feeding shaft mechanism 201, the powder spreading shaft mechanism 202, and the forming cylinder mechanism 203 through the PLC controller 2. The RFID reader 101 is embedded in the host computer 1, and the host computer 1 communicates with the human-machine interface system 4 through the RFID reader 101 to realize functions such as data storage, parameter setting, and fault alarm. In addition, the 3D printing device also includes a powder collection cylinder 2030 and a forming substrate 2031; the 3D printing device is also equipped with a high-speed camera 3.

[0027] like Figures 2 to 3 As shown, this invention provides a 3D printing adaptive powder spreading method, which uses a 3D printing device to complete the adaptive powder spreading operation. Specifically, it includes the following steps:

[0028] Step 1: Use slicing software to convert the 3D model of the item to be printed into 2D model cross-sectional features and transmit the obtained cross-sectional features of each layer to the host computer 1;

[0029] Step 2: The host computer 1 has a data processing module 102. The data processing module 102 determines and identifies whether the cross-sectional features of the current printing layer have changed relative to the cross-sectional features of the previous printing layer.

[0030] Among them, such as Figure 4 As shown, the powder feeding mechanism 201 specifically includes a powder feeding box 2010 and a powder feeding shaft 2011 with a cross-shaped cross section; the powder spreading mechanism 202 includes a flexible scraper 2020 with an inclined angle at its bottom; the forming substrate 2031 is located between the powder feeding mechanism 201 and the powder collection cylinder 2030. According to the actual printing conditions, the forming substrate 2031 can be divided into several regions. In this embodiment, the forming substrate 2031 is equally divided into region A, region B, and region C, with region A close to the powder feeding mechanism 201 and region C close to the powder collection cylinder 2030. The powder feeding amount and powder quantity factor output by the powder feeding mechanism 201 satisfy the following:

[0031] Y = α·X

[0032] In the formula, Y represents the powder feeding amount, which is the powder amount factor, and X represents the powder discharging amount, which is controlled by the powder dropping shaft mechanism 201; wherein, according to the model cross-sectional characteristics on the molding substrate 2031, it is set to 5-7, and in this embodiment, its value is 7.

[0033] (1) If the cross-sectional characteristics of the current printing layer have not changed relative to the previous printing layer, the output powder feeding amount of the powder feeding shaft mechanism 201 remains unchanged.

[0034] (2) If the cross-sectional characteristics of the current printing layer change relative to the previous printing layer and the cross-sectional characteristics of the current printing layer are in the range of A, B or A+B, then adjust the powder quantity factor and control the powder feeding amount Y1 output by the powder feeding shaft mechanism 201; Y1 is the amount of powder required to completely cover the edge of region C from the powder feeding shaft mechanism 201 to the molding substrate 2031 and close to the position of the powder collection cylinder 2030.

[0035] (3) If the cross-sectional characteristics of the current printing layer change relative to the previous printing layer and the cross-sectional characteristics of the current printing layer are in the range of C, B+C, or A+B+C, then adjust the toner quantity factor and control the output of the toner drop shaft mechanism 201 to deliver toner quantity Y2; Y2 is the amount of toner required to completely cover the area from the toner drop shaft mechanism 201 to the toner collection cylinder 203.

[0036] For each printed layer, existing technologies often involve directly covering the entire area from the bottom of the powder-feeding shaft 2011 to the opening of the powder collection cylinder 2030 on the printing platform to ensure that the molding substrate 2031 is fully covered with powder, and then scraping off the excess powder with a scraper 2020. However, according to this embodiment, adjusting the amount of powder to replenish according to the actual dimensions can save a certain amount of powder. For example, if the dimensions of the 300th layer of the item are in the range of A+B+C, and the dimensions of the 301st layer are in the range of A+B, then the dimensions of the 301st layer change. At this time, only Y1 powder needs to be replenished, and the corresponding amount of powder falls onto the molding substrate 2031. The powder-feeding shaft mechanism 202 completes the powder-feeding operation for this printed layer and scrapes the excess powder into the powder collection cylinder 2030, thus saving the amount of powder to be fed from Y2 to Y1.

[0037] Step 3: After the powder spreading mechanism 202 completes the powder spreading operation for the current layer printing on the molding substrate 2031, a high-speed camera 3 captures an image of the powder layer of the current printed layer and transmits it to the host computer 1. Based on the powder layer image of the current printed layer, the data processing module 102 in the host computer 1 uses OpenCV image recognition to identify the powder bed features of the printed layer and determine the uniformity of the current powder layer. Specifically, as follows... Figure 3 As shown,

[0038] First, the acquired image is subjected to Gaussian filtering and noise reduction, followed by grayscale processing. Sobel edge detection is used to detect the region boundaries to achieve gradient detection, and then the image is segmented using a fixed threshold method. The segmented image is input into a trained neural network, and the current layer's toner bed features are identified by matching it with the comparison image. If no defect features are identified in the current printing layer, the current toner layer is evenly spread, and the toner layer is melted to complete the printing. If defect features are identified in the current printing layer, i.e., the current printing layer has problems such as insufficient toner or collapse, it is determined that the current toner layer is not evenly spread. The amount of toner dispensed by the toner dispensing shaft mechanism 201 is adjusted, and the toner spreading shaft mechanism 202 is used to redistribute the toner until the current printing layer has a uniform toner layer.

[0039] Step 4: After the current layer is printed, the host computer 1 controls the forming cylinder mechanism 203 to descend by one layer thickness, repeating steps 1 to 3 to complete the powder application and printing of the next layer, until all layers are printed. The layer thickness range of the forming cylinder mechanism 203 descending is 20-40μm; in this embodiment, 20μm is selected.

Claims

1. A 3D printing adaptive powder spreading method, comprising a 3D printing device to perform adaptive powder spreading operation, the 3D printing device including a host computer, a powder dispensing shaft mechanism, a powder spreading shaft mechanism, a forming cylinder mechanism, a powder collection cylinder, and a forming substrate, and the 3D printing device is also equipped with a camera; characterized in that, Specifically, the following steps are included: (1) Convert the three-dimensional model of the item to be printed into the cross-sectional features of the two-dimensional model and transmit the cross-sectional features of each layer to the host computer; (2) The host computer determines whether the cross-sectional features of the current printing layer have changed relative to the cross-sectional features of the previous printing layer; If the cross-sectional characteristics remain unchanged, the amount of powder delivered to the molding substrate by the powder-feeding shaft mechanism remains constant; if the cross-sectional characteristics change, the powder quantity factor is adjusted according to the region where the current printed layer cross-sectional characteristics are located, controlling the powder-feeding shaft mechanism to output a specific amount of powder; the region specifically refers to several regions formed by dividing the molding substrate along the length direction of the molding substrate; specifically including: The molding substrate is divided into N regions along its length, denoted as the first region, the second region, ..., the Nth region, with the first region being closer to the powder dispensing shaft mechanism and the Nth region being closer to the powder collection cylinder; the molding substrate is located between the powder dispensing shaft mechanism and the powder collection cylinder. If the cross-sectional features change and the area where the current printed layer cross-sectional features are located does not include the Nth region, then adjust the powder quantity factor and control the powder delivery amount output Y1 of the powder delivery shaft mechanism; Y1 is the amount of powder required to completely cover the area from the powder delivery shaft mechanism to the edge of the molding substrate and close to the powder collection cylinder. If the cross-sectional features change and the area where the current printed layer cross-sectional features are located includes the Nth area, then adjust the toner quantity factor and control the toner delivery amount Y2 output by the toner drop shaft mechanism; Y2 is the amount of toner required to completely cover the area from the toner drop shaft mechanism to the toner collection cylinder. The powder feeding amount and powder quantity factor output by the powder feeding shaft mechanism satisfy the following: In the formula, Y represents the amount of powder fed, α is the powder quantity factor, and X is the amount of powder dispensed. (3) After the host computer controls the powder spreading shaft mechanism to complete the powder spreading operation of the current printing layer on the molding substrate, the camera acquires the powder layer image of the current printing layer and transmits it to the host computer. The host computer performs powder bed feature recognition and judges the uniformity of the current powder layer based on the powder layer image of the current printing layer. If no defect features are identified, the current powder layer is uniform and the printing of the layer is completed. If defect features are identified, the current powder layer is not uniform. The amount of powder dropped by the powder falling shaft mechanism is adjusted, and the powder spreading shaft mechanism is used to spread the powder again until the powder layer of the current printing layer is uniform. (4) After the current layer is printed, the host computer controls the forming cylinder mechanism to descend by one layer thickness, and repeats steps (1) to (3) to complete the next layer of powder spreading and printing until all layers are printed.

2. The 3D printing adaptive powder spreading method according to claim 1, characterized in that, In step (3), the host computer uses OpenCV image recognition to perform powder bed feature recognition on the powder layer image of the current printing layer and determines the uniformity of the current powder layer. Specifically, this includes: First, the image is processed by Gaussian filtering and noise reduction, followed by grayscale processing. Sobel edge detection is used to detect the boundaries of the regions, and then the image is segmented using a fixed threshold method. The segmented image is then input into a trained neural network to identify the features of the current powder bed, and the uniformity of the current powder layer is determined based on the identified powder bed features.

3. The 3D printing adaptive powder spreading method according to claim 1, characterized in that, The 3D printing device is also equipped with a human-computer interaction system, which communicates with the host computer through an RFID reader embedded in the host computer. During the adaptive toner spreading printing operation, the host computer feeds back the toner spreading process parameters to the human-machine interaction system in real time through the RFID reader.

4. The 3D printing adaptive powder spreading method according to claim 1, characterized in that, The range of α is [5,7].

Images