3D printer and exception handling method

Abstract

The invention provides a 3D printer and an exception handling method. The exception handling method comprises the following steps: acquiring continuous image data in a printing process; according to the continuous image data, whether the printed piece has abnormal displacement characteristics or not is determined; in response to the abnormal displacement characteristic of the printed piece, determining the spatial position of the printed piece according to the continuous image data, and generating a corresponding avoidance mask; and based on the avoidance mask, driving an execution mechanism to execute an exception handling operation. According to the method, the abnormal displacement characteristics of the printed piece are determined according to the continuous image data, and the abnormal processing operation is executed based on the avoidance mask, so that accurate positioning and automatic processing of the printing abnormality are realized, and the problems of printing failure and equipment damage caused by lack of an effective abnormal processing mechanism in related technologies are solved.

Description

Technical Field

[0001] This application relates to the field of 3D printing technology, specifically to a 3D printer and an anomaly handling method. Background Technology

[0002] With the popularization of 3D printing technology, users are paying increasing attention to the stability of the printing process. In related technologies, 3D printers often cannot quickly locate printing faults during the printing process and lack effective automated fault handling mechanisms. Therefore, the 3D printing process often leads to subsequent printing errors. These shortcomings not only cause the entire printing task to fail and waste printing materials, but also damage the hardware of the printing equipment. Summary of the Invention

[0003] This application primarily addresses the technical problem in related technologies where 3D printers are unable to quickly locate printing faults during the printing process, leading to printing failures and equipment damage.

[0004] The first aspect of this application provides an exception handling method applied to a 3D printer, the exception handling method including: Acquire continuous image data during the printing process; Based on continuous image data, determine whether the printed part has abnormal displacement characteristics; In response to abnormal displacement characteristics of the printed part, the spatial position of the printed part is determined based on continuous image data, and a corresponding avoidance mask is generated; Based on the avoidance mask, the actuator is driven to perform exception handling operations; Among them, the abnormal displacement feature is configured to characterize the unexpected physical position deviation of the printed part during the printing process; the avoidance mask is configured to characterize the projection position of the printed part with the abnormal displacement feature, so as to distinguish the abnormal area from the normal area.

[0005] In one embodiment, the 3D printer includes multiple image acquisition devices with different viewing angles; acquiring continuous image data during the printing process includes: acquiring continuous image data acquired by the multiple image acquisition devices; and determining the spatial features of the printed part based on the acquired continuous image data.

[0006] In one embodiment, determining whether a printed part has abnormal displacement features based on continuous image data includes: acquiring temporal features of the continuous image data; and determining that the printed part has abnormal displacement features in response to the temporal features characterizing abnormal movement.

[0007] In one embodiment, determining whether a printed part has abnormal displacement features based on continuous image data includes: calculating pixel motion vectors between adjacent frames based on continuous image data; and determining that the printed part has abnormal displacement features in response to the pixel motion vectors satisfying preset conditions.

[0008] In one embodiment, determining whether a printed part has abnormal displacement features based on continuous image data includes: extracting image segmentation masks of adjacent frames of the printed part in the continuous image data; and determining whether the printed part has abnormal displacement features based on the difference features of the image segmentation masks of adjacent frames.

[0009] In one embodiment, determining that the printed part has abnormal displacement features based on the difference characteristics of the image segmentation masks of adjacent frames includes: calculating the pixel motion vector between adjacent frames within the region corresponding to the image segmentation mask; and determining that the printed part has abnormal displacement features in response to the pixel motion vector within the mask region satisfying a preset condition.

[0010] In one embodiment, the 3D printer further includes an interactive interface; before driving the actuator to perform an exception handling operation based on the avoidance mask, the method further includes: outputting a visual prompt containing the avoidance mask in the interactive interface; obtaining an interactive instruction triggered based on the visual prompt; driving the actuator to perform an exception handling operation based on the avoidance mask, including: driving the actuator to perform an exception handling operation based on the avoidance mask and the interactive instruction; the interactive interface is configured to output exception prompt information and interact with the user.

[0011] In one embodiment, the printed component includes a main model and an auxiliary structure; the visual prompts include at least one of the main model anomaly prompts and the auxiliary structure anomaly prompts; the interactive instructions include avoidance instructions and termination instructions; the main model is used to represent a preset printed entity; the auxiliary structure is configured to provide physical support or a temporary structure for material transition during the printing of the main model.

[0012] In one embodiment, based on an avoidance mask and interactive instructions, the actuator is driven to perform an exception handling operation, including: in response to the avoidance instruction, the actuator is driven to skip the exception area and continue printing.

[0013] In one embodiment, based on an avoidance mask and interactive instructions, the actuator is driven to perform an exception handling operation, including: in response to a termination instruction, the actuator is driven to terminate the associated printing task of the exception region.

[0014] In one embodiment, before outputting a visual prompt containing an avoidance mask in the interactive interface, the process includes: extracting the shape components of the printed part based on the abnormal displacement features; determining the cause of the fault based on the shape components; and outputting a visual prompt containing an avoidance mask in the interactive interface, which includes: outputting a visual prompt containing an avoidance mask and the cause of the fault in the interactive interface based on the cause of the fault.

[0015] In one embodiment, the causes of failure include: first-layer adhesion failure, actuator physical interference, or structural strength collapse; first-layer adhesion failure is configured to characterize insufficient adhesion between the bottom of the print and the heated bed; actuator physical interference is configured to characterize an unexpected physical collision between the actuator and the print; structural strength collapse is configured to characterize fracture or deformation of the print due to internal stress or support failure.

[0016] The second aspect of this application provides a 3D printer, including: an image acquisition device, an actuator, a memory, and a processor; the memory stores a computer program, and the processor executes the computer program to implement the exception handling method provided in the first aspect or any embodiment of the first aspect.

[0017] A third aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the exception handling method provided in the first aspect or any embodiment of the first aspect.

[0018] A fourth aspect of this application provides an anomaly handling device for use in a 3D printer, comprising: a data acquisition module configured to acquire continuous image data during the printing process; an anomaly detection module configured to determine whether the printed part has abnormal displacement features based on the continuous image data; a positioning and mask generation module configured to determine the spatial position of the printed part based on the continuous image data and generate a corresponding avoidance mask in response to the presence of abnormal displacement features; and an execution control module configured to drive an actuator to perform anomaly handling operations based on the avoidance mask; the abnormal displacement features are configured to characterize the unexpected physical position deviation of the printed part during the printing process; and the avoidance mask is configured to characterize the projection position of the printed part with abnormal displacement features, so as to distinguish between abnormal areas and normal areas.

[0019] This application determines the abnormal displacement characteristics of the printed parts based on continuous image data and performs anomaly handling operations based on avoidance masks, thereby achieving accurate positioning and automated processing of printing anomalies. This solves the problem of printing failure and equipment damage caused by the lack of an effective anomaly handling mechanism in related technologies. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart illustrating an exception handling method provided in an embodiment of this application; Figure 2This is a schematic diagram of a process for acquiring continuous image data according to an embodiment of this application; Figure 3 A flowchart illustrating the process of determining abnormal displacement features based on temporal characteristics, provided as an embodiment of this application; Figure 4 A schematic diagram of a process for determining abnormal displacement features based on pixel motion vectors, provided in an embodiment of this application; Figure 5 A schematic diagram of the process for determining abnormal displacement features based on image segmentation mask according to an embodiment of this application; Figure 6 A flowchart illustrating the process of determining abnormal displacement features by combining pixel motion vectors, provided in an embodiment of this application; Figure 7 A flowchart illustrating an embodiment of this application of an interactive instruction-driven actuator; Figure 8 A flowchart illustrating the causes of output failures provided in an embodiment of this application; Figure 9 This is a structural block diagram of a 3D printer provided in one embodiment of this application; Figure 10 A structural block diagram of a computer-readable storage medium provided in an embodiment of this application; Figure 11 This is a structural block diagram of an exception handling device provided in an embodiment of this application.

[0022] Figure label: 100. 3D printer; 110. Image acquisition device; 120. Actuator; 130. Memory; 140. Processor; 200. Computer-readable storage medium; 300. Anomaly handling device; 310. Data acquisition module; 320. Anomaly detection module; 330. Positioning and mask generation module; 340. Execution control module. Detailed Implementation

[0023] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be noted that the following embodiments are for illustrative purposes only and do not limit the scope of the application. All other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this application.

[0024] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0025] The directional terms used in this description, such as "up," "down," "left," "right," "inner," and "outer," are only for reference to the directions shown in the attached diagram. These directional terms should be changed if the directions in the attached diagram change.

[0026] This application provides an exception handling method and a 3D printer 100. The exception handling method provided in this application is applied to the 3D printer 100. In the following detailed description of the embodiments, the main body executing each step of the exception handling method is the 3D printer 100, specifically the core control components such as the processor 140 inside the 3D printer 100.

[0027] Please see Figure 9 The 3D printer 100 includes at least an image acquisition device 110, an actuator 120, a memory 130, and a processor 140.

[0028] For example, the image acquisition device 110 is mounted on the internal column cavity of the 3D printer 100 to obtain a stable image. The lens of the image acquisition device 110 has a viewing angle facing the heated bed in the printing area, and is mainly used to acquire continuous image data in real time during the printing process. The image acquisition device 110 can be a visual sensing component such as a camera.

[0029] The actuator 120 includes at least one of a print head, a drive motor, a transmission assembly, and a heated bed. The actuator 120 is mainly used to perform the extrusion molding and printing action of a three-dimensional solid during operation, and to perform abnormal handling operations in response to control commands when a fault is detected.

[0030] The memory 130 is mainly used to store computer programs, control codes for the printing trajectory coordinate system, and visual data acquired by the image acquisition device 110.

[0031] The processor 140 is electrically or communicatively connected to the image acquisition device 110, the actuator 120, and the memory 130. Specifically, the processor 140 can be a central processing unit chip or a neural network processing chip built into the motherboard of the 3D printer 100. The processor 140 possesses image processing computing power and logic scheduling capabilities. By reading and executing the computer program in the memory 130, it performs visual analysis on the images transmitted from the image acquisition device 110 and issues instructions to other firmware to drive the actuator 120, thereby implementing the exception handling methods provided in the embodiments of this application.

[0032] Please see Figures 1 to 11 , Figure 1 This is a flowchart illustrating an exception handling method provided in an embodiment of this application. The exception handling method is applied to a 3D printer 100 and specifically includes the following steps: Step S100: Acquire continuous image data during the printing process.

[0033] When the 3D printer 100 performs a printing task, the image acquisition device 110 continuously captures images of the printing area where the heated bed is located, thereby acquiring continuous image data containing images of the printed part. This continuous image data is presented as a sequence of image frames or a video stream arranged in chronological order, objectively recording the physical morphological changes of the printed part on the heated bed. The continuous image data may include data from two or more images acquired sequentially in chronological order.

[0034] Step S200: Based on the continuous image data, determine whether there are abnormal displacement features in the printed part.

[0035] The processor 140 performs real-time analysis on the acquired continuous image data. In this application, abnormal displacement features are configured to characterize unexpected physical position deviations of the printed part during the printing process. Specifically, in a normal 3D printing process, the extruded printed part should be firmly attached to the heated bed and remain relatively stationary, or only move as expected as a whole following the mechanical transmission of the heated bed. If the processor 140 detects through image analysis that the printed part has tilted, detached from the heated bed, or partially broken and relatively slid on the heated bed, this movement state deviating from its original printed coordinate position, the processor 140 determines that the printed part has abnormal displacement features.

[0036] In step S300, in response to the presence of abnormal displacement features in the printed part, the spatial position of the printed part is determined based on continuous image data, and a corresponding avoidance mask is generated.

[0037] Once the processor 140 detects the aforementioned abnormal positional deviation of the printed part, it further combines continuous image data to extract the spatial position of the abnormal printed part in the current environment. Based on the calculated spatial position, the processor 140 generates a corresponding avoidance mask. In this application, the avoidance mask is configured to characterize the projected position of the printed part with abnormal displacement features, in order to distinguish between abnormal and normal areas. In a specific engineering implementation, the processor 140 maps the two-dimensional visual contour of the abnormal printed part into a mask data format with region identifiers. In this mask data, the area of ​​the printed part with abnormal displacement features is marked as an abnormal area, while the area where other unaffected parts being printed on the heated bed are located is marked as a normal area.

[0038] Step S500: Based on the avoidance mask, drive the actuator to perform an exception handling operation.

[0039] After generating an avoidance mask that distinguishes between abnormal and normal areas, the processor 140 sends an execution command to the underlying control system based on the avoidance mask to drive the actuator 120 to perform an exception handling operation. Upon receiving the command, the actuator 120 can adjust the movement trajectory or extrusion action of the print head according to the area range marked by the avoidance mask to cope with the current fault state.

[0040] The anomaly handling method provided in this application, by real-time monitoring of continuous image data during the printing process, can accurately and promptly identify printed parts exhibiting abnormal displacement characteristics, and uses a generated avoidance mask to distinguish and isolate faulty areas from other normal areas. This method achieves accurate location and automated processing of printing anomalies, allowing for timely intervention in the early stages of a fault and effectively preventing partial printing faults from causing the overall printing task to fail.

[0041] In one embodiment, please refer to Figure 2 To better determine the spatial features of the printed part, the 3D printer 100 may include multiple image acquisition devices 110 with different viewpoints. In complex printing scenarios, a single-view image acquisition device 110 may be unable to fully observe the area where anomalies occur due to obstruction by a moving print head or the complex structure of the printed part itself. To solve the detection blind spot problem caused by viewpoint obstruction, the steps for acquiring continuous image data during the printing process may specifically include steps S110 and S120.

[0042] Step S110: Acquire continuous image data from multiple image acquisition devices.

[0043] Specifically, multiple image acquisition devices 110 with different perspectives can be installed at different diagonal or side wall positions within the internal cavity of the 3D printer 100. The processor 140 synchronously acquires continuous image data acquired by these multiple image acquisition devices 110.

[0044] Step S120: Determine the spatial features of the printed part based on the acquired continuous image data.

[0045] The processor 140 fuses images captured from different perspectives. By cross-referencing multi-view data, the processor 140 can determine the spatial characteristics of the printed part on the heated bed from all angles. This multi-view fusion method eliminates the occlusion interference of a single image, thus providing more comprehensive and accurate data support for subsequent determination of whether the printed part has undergone abnormal displacement.

[0046] Please refer to the following: Figures 3 to 6 There are multiple ways to determine whether a printed part has abnormal displacement characteristics based on continuous image data.

[0047] In one possible implementation, please refer to Figure 3 This application can capture abnormal displacement features based on an end-to-end video action classification algorithm.

[0048] Step S211: Obtain the temporal characteristics of continuous image data.

[0049] In the 3D printing process, the detachment of the printed part from the heated bed, the overall collapse, or the partial displacement of the material chute are essentially visually a continuous sequence of actions. In one embodiment, the processor 140 is equipped with an efficient video understanding model, such as a video action classification network like a time-shift module network model. The processor 140 inputs the acquired continuous image data as a video stream into this video understanding model. This video understanding model extracts the temporal change information between adjacent image frames, thereby obtaining deep temporal features of the continuous image data.

[0050] Step S212: In response to the time-series features characterizing abnormal movement, it is determined that the printed part has abnormal displacement features.

[0051] The video understanding model analyzes extracted temporal features to classify the motion occurring in the current video segment. If the action classification result output by the video understanding model is an abnormal object movement category, it is determined that the temporal feature represents abnormal movement, and the processor 140 then determines that the printed part has abnormal displacement features. Before actual model deployment, researchers can collect a large number of video clips containing real-world displacement and collapse scenarios as positive samples to train the video classification network. Using this end-to-end motion capture method, the processor 140 can capture the instantaneous action or relative motion trend of displacement in milliseconds. Compared to common detection methods such as head wrapping detection and fried noodle detection, the above embodiment can achieve immediate blocking in the early stages of a fault.

[0052] In one possible implementation, please refer to Figure 4 This application can capture minute movements of printed parts based on the principle of optical flow estimation.

[0053] Step S221: Calculate the pixel motion vector between adjacent frames based on continuous image data.

[0054] After acquiring continuous image data, the processor 140 invokes its internal lightweight optical flow network to perform pixel-by-pixel motion vector calculations on two adjacent frames on the time axis. The pixel motion vector reflects the direction and distance of movement of the same pixel in adjacent frames on a two-dimensional plane.

[0055] Step S222: In response to the pixel motion vector satisfying the preset conditions, it is determined that the printed part has abnormal displacement characteristics.

[0056] In a normal 3D printing process, the printed object should remain stationary relative to the heated bed. Therefore, reflected in the image, the pixel motion vectors on the surface of the printed part should be close to zero, or only exhibit a globally consistent vector that moves with the heated bed as a whole. This preset condition is configured to filter out abnormal physical movements. Specifically, if an unexpected pixel motion vector appears in a local area of ​​the image, and the processor 140 confirms through coordinate comparison that this local area is not the area of ​​the nozzle or other actuator performing the printing action, then the pixel motion vector is determined to meet the preset condition. By calculating the pixel motion vector, the processor 140 can promptly detect slippage or slight tilting of the printed part.

[0057] In one possible implementation, see [reference] Figure 5 This application can locate faults by comparing the differences in the shape and contour of printed parts.

[0058] Step S231: Extract the image segmentation mask of adjacent frames in the continuous image data of the printed part.

[0059] Processor 140 uses a semantic segmentation network to perform semantic segmentation on the current frame image and the previous frame image in continuous image data. Through the segmentation network, processor 140 extracts the pixel regions representing the printed part from the printer cavity, thereby extracting the image segmentation masks corresponding to each of the two adjacent frames. The image segmentation mask reflects the appearance outline of the printed part at the corresponding moment.

[0060] Step S232: Based on the difference features of the image segmentation masks of adjacent frames, determine that the printed part has abnormal displacement features.

[0061] After extracting the image segmentation masks at two time points, the processor 140 calculates the spatial difference between them. In specific implementations, the processor 140 can derive the difference features by calculating the intersection-over-union (IoU) ratio of the segmentation masks in adjacent frames, or by calculating the spatial geometric distance between the centroid coordinates of the two masks. Since 3D printing is a layer-by-layer stacking process, the contour of a normally printed part changes very little within a very short time interval between adjacent frames, its IoU should be close to its maximum value, and the centroid distance should be almost zero. If the processor 140 finds that the IoU value drops abruptly, or the centroid distance exceeds a set safety threshold, it determines that the mask position has changed abruptly, and thus determines that the printed part has abnormal displacement features.

[0062] In one possible implementation, see [reference] Figure 6 This application combines the technical advantages of image segmentation and pixel motion vectors to achieve more efficient fault detection.

[0063] Step S232a: Calculate the pixel motion vectors between adjacent frames within the region corresponding to the image segmentation mask.

[0064] To avoid computational resource consumption and interference from complex backgrounds caused by global optical flow calculation, processor 140 first executes step S231 to obtain an image segmentation mask containing the outline of the main body of the printed part. Subsequently, processor 140 uses the image segmentation mask as a visual filter to strictly limit the range of calculated pixel motion vectors to the area corresponding to the image segmentation mask. Because a 3D printer includes not only an image acquisition device but also at least a print head; some enclosed models also include components such as a printing cavity and a fan; implementing this application can eliminate interference from background areas such as the inner wall of the printing cavity, the fan, or the moving print head, significantly reducing the amount of invalid pixel calculations.

[0065] Step S232b: In response to the pixel motion vectors within the mask area satisfying preset conditions, it is determined that the printed part has abnormal displacement characteristics.

[0066] After locking onto the spatial region where the printed object is located, the processor 140 only performs conditional judgments on the pixel motion vectors extracted within the mask region. Similarly, if the pixels within the mask region exhibit abnormal non-stationary motion vectors, the processor 140 determines that the printed object has abnormal displacement characteristics. This embodiment not only eliminates interference caused by moving background components, making the detection results more accurate, but also reduces the computational load on the processor 140, ensuring the accuracy of anomaly detection results under high frame rate acquisition devices.

[0067] In one possible implementation, please refer to [the relevant documentation / reference]. Figure 7 After detecting abnormal displacement features and generating avoidance masks, this application introduces a human-computer interaction mechanism to achieve safer and more user-intended fault handling. In one embodiment, the 3D printer 100 also includes an interactive interface. The interactive interface can be a touch screen mounted on the printer body or a mobile terminal display screen communicatively connected to the printer. The interactive interface is configured to output abnormal prompts and interact with the user.

[0068] Specifically, before the exception handling operation is performed by the actuator driven by the avoidance mask, the exception handling method further includes steps S410 and S420. Correspondingly, the step of performing the exception handling operation can be specifically represented as step S510.

[0069] Step S410: Output a visual prompt containing the avoidance mask in the interactive interface.

[0070] Processor 140 converts the generated avoidance mask into a user-friendly visual image and presents it in the interactive interface. This visualization can be a two-dimensional image or a three-dimensional model rendering. Processor 140 displays the fault area covered by the avoidance mask in the interactive interface as a highlighted bounding box or a special marker color, allowing the user to intuitively and clearly understand which specific area on the hotbed has an anomaly.

[0071] In this application, to enable different processing of different situations of the printed parts, the printed parts are subdivided into a main model and an auxiliary structure. The main model is used to represent the final printed entity that the user presets and expects to obtain; the auxiliary structure is configured as a temporary structure that provides physical support or material transition during the printing of the main model, such as a tree-like support to prevent the collapse of suspended parts, or a wiping tower for cleaning residual impurities during multi-color printing color change.

[0072] Correspondingly, the visual prompts include at least one of the main model anomaly prompts and auxiliary structure anomaly prompts. The processor 140 will clearly display text or graphic alerts in the interactive interface to inform the user whether the abnormal displacement is occurring in the main model of the printed part or the auxiliary structure.

[0073] Step S420: Obtain the interactive instructions triggered by the visual prompts.

[0074] After reviewing the visual prompts on the interactive interface, users can make decisions based on the actual situation and issue intervention strategies through touch clicks or other means. Interactive commands must include at least avoidance commands and termination commands.

[0075] Step S510: Based on the avoidance mask and interactive instructions, drive the actuator to perform exception handling operations.

[0076] The system provides different processing logic for different types of printout anomalies. This application illustrates the following two cases: In response to abnormal displacement characteristics of the main model and the interaction command received by the processor 140 being an avoidance command, the processor 140 drives the actuator 120 to skip the abnormal area and continue printing. In scenarios involving batch printing of multiple entities, if a single main model collapses, the user can choose to issue an avoidance command. In this case, the processor 140 will modify the subsequent printing trajectory according to the avoidance mask, causing the print head to avoid the abnormal area where the damaged main model is located in subsequent jobs, and continue printing other normal main models on the heated bed. In addition, the user can also actively issue a termination command to stop the printing task as needed.

[0077] In response to abnormal displacement of the auxiliary structure and the interaction command being a termination command, the processor 140 drives the actuator 120 to terminate the associated printing task in the abnormal area. Since auxiliary structures such as the wiping tower are involved in operations such as cleaning up impurities during multi-color printing color changes, if the wiping tower shifts or collapses, executing an avoidance command would cause subsequent main model printing to result in color mixing or material shortages. Therefore, in response to the abnormality of the auxiliary structure, the processor 140 cannot execute the avoidance command; instead, the processor 140 needs to guide the user to issue a termination command, driving the actuator 120 to stop printing, thereby avoiding the unnecessary production of defective products.

[0078] Please refer to the following: Figure 8 To further assist users in troubleshooting printing failures, this application introduces a fault analysis mechanism in addition to displaying printing anomaly prompts. Specifically, before outputting a visual prompt containing an avoidance mask in the interactive interface, the anomaly handling method also includes steps S401 and S402. Correspondingly, step S410 can be specifically manifested as step S411 in scenarios involving fault causes.

[0079] Step S401: Extract the morphological components of the printed part based on the abnormal displacement features.

[0080] After confirming that the printed part has deviated from its position, the processor 140 further analyzes the specific visual manifestations of this deviation. The processor 140 extracts morphological components reflecting the displacement state of the printed part by performing feature comparison on continuous image data. These morphological components record specific deformation and movement details, such as whether the printed part has undergone overall translation, tilting and collapsing, or merely a local structural breakage.

[0081] Step S402: Determine the cause of the fault based on the morphological components.

[0082] Different physical faults often exhibit different displacement patterns. The processor 140 has a built-in classification algorithm that can deduce the root cause of the printing failure based on the extracted morphological components.

[0083] In the embodiments of this application, the causes of failure mainly include three situations: failure of the first layer adhesion, physical interference of the actuator, and collapse of structural strength.

[0084] Specifically, first-layer adhesion failure indicates insufficient adhesion between the bottom of the printed part and the heated bed. When the morphological component extracted by the processor 140 shows that the entire printed part is sliding and translating on the heated bed, and the bottom of the printed part is intact and without residue, the processor 140 determines that the cause of the failure is first-layer adhesion failure. In this case, there may be oil stains or no solid adhesive tape applied to the surface of the heated bed.

[0085] The second scenario is physical interference of the actuator, mainly manifested as an unexpected physical collision between the actuator 120 and the printed part. When the shape component manifests as the printed part suddenly tilting as the print head passes by, the processor 140 determines it as physical interference of the actuator. This is usually caused by the print head scraping against the edge of the printed part that is raised.

[0086] The third scenario is structural strength collapse, primarily manifested as breakage or deformation of the printed part due to internal stress or support failure. When the morphological component shows that the bottom of the printed part is firmly adhered to the heated bed, but its upper structure breaks off or undergoes severe deformation, the processor 140 determines it to be structural strength collapse. This suggests to the user that the current support structure design of the printed part may have mechanical defects.

[0087] Step S411: Based on the cause of the fault, output a visual prompt containing the avoidance mask and the cause of the fault in the interactive interface.

[0088] After completing the fault analysis, the processor 140 will show which printout is damaged in the interactive interface by using an avoidance mask. It will also simultaneously present the determined fault cause in the interactive interface in the form of text description or icon warning, so that users can take targeted measures before reprinting, such as cleaning the heated bed or adjusting the printing parameters.

[0089] Please refer to the following: Figure 9 This application provides a 3D printer 100. The 3D printer 100 includes an image acquisition device 110, an actuator 120, a memory 130, and a processor 140. The memory 130 stores a computer program, and the processor 140 executes the computer program to implement the exception handling methods provided in the above embodiments of this application. The specific hardware configuration, communication relationship, and respective steps performed by the image acquisition device 110, actuator 120, memory 130, and processor 140 have been described in detail above and will not be repeated here.

[0090] Please see Figure 10 This application also provides a computer-readable storage medium 200. The computer-readable storage medium 200 stores a computer program, which, when called and executed by the processor 140, implements the exception handling methods provided in the above embodiments of this application. The computer-readable storage medium 200 may specifically include data storage media such as flash memory, solid-state drive, or read-only memory.

[0091] In addition, please see Figure 11 This application also provides an anomaly handling device 300, applied to a 3D printer 100. The anomaly handling device 300 includes a data acquisition module 310, an anomaly detection module 320, a positioning and mask generation module 330, and an execution control module 340.

[0092] Specifically, the data acquisition module 310 is configured to acquire continuous image data during the printing process. The anomaly detection module 320 is configured to determine whether the printed part has abnormal displacement features based on the continuous image data. The positioning and mask generation module 330 is configured to, in response to the presence of abnormal displacement features in the printed part, determine the spatial position of the printed part based on the continuous image data and generate a corresponding avoidance mask. The execution control module 340 is configured to drive the actuator 120 to perform anomaly handling operations based on the avoidance mask.

[0093] Among them, the abnormal displacement feature is used to characterize the unexpected physical position deviation of the printed part during the printing process; the avoidance mask is used to characterize the projection position of the printed part with abnormal displacement features, so as to distinguish the abnormal area from the normal area.

[0094] The division of the functional modules in the exception handling device 300 corresponds one-to-one with the steps in the aforementioned exception handling method. The specific working principle and execution details of each module are exactly the same as those in the aforementioned method embodiment, and will not be repeated here.

[0095] It should be noted that the terms "first," "second," and "third" in this application are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first," "second," or "third" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0096] The above description is only a part of the embodiments of this application and does not limit the scope of protection of this application. Any equivalent device or equivalent process transformation made based on the content of this application specification and drawings, or direct or indirect application in other related technical fields, are similarly included in the patent protection scope of this application.

Claims

1. An anomaly handling method for 3D printing, applied to a 3D printer, characterized in that, The exception handling method includes: Acquire continuous image data during the printing process; Based on the continuous image data, determine whether the printed part has abnormal displacement characteristics; In response to the presence of abnormal displacement characteristics in the printed part, the spatial position of the printed part is determined based on the continuous image data, and a corresponding avoidance mask is generated; Based on the avoidance mask, the actuator is driven to perform an exception handling operation; The abnormal displacement feature is configured to characterize the unexpected physical position deviation of the printed part during the printing process; the avoidance mask is configured to characterize the projection position of the printed part with the abnormal displacement feature, so as to distinguish the abnormal area from the normal area.

2. The anomaly handling method according to claim 1, wherein the 3D printer includes multiple image acquisition devices with different viewing angles, characterized in that, The acquisition of continuous image data during the printing process includes: Acquire continuous image data from multiple image acquisition devices; Based on the acquired continuous image data, the spatial features of the printed part are determined.

3. The anomaly handling method according to claim 1, characterized in that, The step of determining whether the printed part has abnormal displacement features based on the continuous image data includes: Obtain the temporal characteristics of the continuous image data; In response to the timing feature indicating abnormal movement, it is determined that the printed part has the abnormal displacement feature.

4. The anomaly handling method according to claim 1, characterized in that, The step of determining whether the printed part has abnormal displacement features based on the continuous image data includes: Based on the continuous image data, calculate the pixel motion vectors between adjacent frames; In response to the pixel motion vector satisfying a preset condition, it is determined that the printed part has the abnormal displacement feature.

5. The anomaly handling method according to claim 1, characterized in that, The step of determining whether the printed part has abnormal displacement features based on the continuous image data includes: Extract the image segmentation mask of the printout from adjacent frames in the continuous image data; Based on the difference features of the image segmentation masks of the adjacent frames, it is determined that the printed part has the abnormal displacement feature.

6. The anomaly handling method according to claim 5, characterized in that, The step of determining the presence of the abnormal displacement feature in the printed part based on the difference features of the image segmentation masks of the adjacent frames includes: Within the region corresponding to the image segmentation mask, calculate the pixel motion vectors between adjacent frames; In response to the pixel motion vector within the mask area satisfying a preset condition, it is determined that the printed part has the abnormal displacement feature.

7. The anomaly handling method according to claim 1, wherein the 3D printer further includes an interactive interface; characterized in that, Before driving the actuator to perform exception handling operations based on the avoidance mask, the method further includes: The interactive interface outputs a visual prompt containing the avoidance mask; Obtain interactive commands triggered based on the visual prompts; The step of driving the actuator to perform exception handling operations based on the avoidance mask includes: Based on the avoidance mask and the interaction instructions, the actuator is driven to perform an exception handling operation; The interactive interface is configured to output error messages and interact with the user via commands.

8. The anomaly handling method according to claim 7, characterized in that, The printed part includes a main model and auxiliary structures; The visual prompts include prompts for anomalies in the main model and / or prompts for anomalies in the auxiliary structures; The main model is used to represent a pre-defined printable entity; the auxiliary structure is configured to provide physical support or a temporary structure for material transition during the printing of the main model.

9. The anomaly handling method according to claim 8, characterized in that, The interactive instructions include avoidance instructions; the step of driving the actuator to perform exception handling operations based on the avoidance mask and the interactive instructions includes: In response to the avoidance command, the actuator is driven to skip the abnormal area and continue printing.

10. The anomaly handling method according to claim 8, characterized in that, The interactive instructions include a termination instruction; the step of driving the actuator to perform exception handling operations based on the avoidance mask and the interactive instructions includes: In response to the termination command, the actuator is driven to terminate the printing task associated with the abnormal region.

11. The anomaly handling method according to claim 7, characterized in that, Before outputting a visual prompt containing the avoidance mask in the interactive interface, the following steps are included: The morphological components of the printed part are extracted based on the abnormal displacement features; The cause of the fault is determined based on the morphological components; The step of outputting a visual prompt containing the avoidance mask in the interactive interface includes: Based on the cause of the fault, a visual prompt containing the avoidance mask and the cause of the fault is output in the interactive interface.

12. The anomaly handling method according to claim 11, characterized in that, The causes of the failure include: failure of the first layer adhesion, physical interference of the actuator, or collapse of the structural strength. Wherein, the first-layer adhesion failure is configured to characterize insufficient adhesion between the bottom of the printed part and the heated bed; the actuator physical interference is configured to characterize an unexpected physical collision between the actuator and the printed part; and the structural strength collapse is configured to characterize the fracture or deformation of the printed part due to internal stress or support failure.

13. A 3D printer, characterized in that, include: Image acquisition device, actuator, memory, and processor; The memory stores a computer program, and when the processor executes the computer program, it implements the exception handling method according to any one of claims 1 to 12.

14. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the exception handling method according to any one of claims 1 to 12.

Images