Detecting resin adhesion defects

Abstract

To provide an apparatus and method for detecting slurry deposition and curing when manufacturing a component using an additive manufacturing process.SOLUTION: A method to monitor production of an additive manufacturing product collects images from a deposition process on a layer-by-layer basis, including a void image of a pattern left in a slurry layer after deposition of a layer and a displacement image formed by immersing a just-deposited layer in a renewed slurry layer. Image properties of the void image and displacement image are corrected, and then compared to a binary expected image from a computer generated model to identify defects in the just-deposited layer on a layer-by-layer basis. An additional method uses an output from a comparison to form a three-dimensional model corresponding to at least a portion of the additive manufacturing product. Constitutional components to control an additive manufacturing operation based on digital model data and to in-situ monitor successive layers for manufacturing defects can be embodied in a computer system or computer-aided machine, such as a computer-controlled additive manufacturing machine.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates generally to the fabrication of parts from slurries, such as curable resin or ceramic compositions, and to the detection of manufacturing defects in parts fabricated using slurries. In particular, an apparatus and method for detecting slurry adhesion and hardening when fabricating parts using additive manufacturing processes, such as when fabricating parts related to nuclear fission reactors using additive manufacturing processes, is disclosed. [Background technology]

[0002] In the following description, reference is made to certain structures and / or methods. However, the following references should not be construed as an admission that those structures and / or methods constitute prior art. Applicant expressly reserves the right to demonstrate that such structures and / or methods do not qualify as prior art against the present invention.

[0003] 3D printers offer the flexibility to manufacture a variety of components based on computer-input designs. However, manufacturing methods using 3D printers are plagued by adhesion failures caused by a variety of factors. These factors include insufficient curing, such as insufficient UV exposure; a lack of support structure, such as when a volume of cured material has no existing part of the manufactured product to adhere to; or improper resin formulation, which contributes to improper curing or insufficient mechanical properties. Each factor can prevent the additively manufactured material from adhering to the base part. Such failures can result in the manufactured part not meeting product specifications.

[0004] Inspection of manufactured parts can be challenging. For example, detecting and verifying as-manufactured features by direct imaging is often impractical due to the very uniformity of the optical properties of the manufactured material relative to the optical properties of the constituent materials. In another example, as-manufactured products may have features that include complex internal features that cannot be easily or accurately detected using ex-situ or post-construction measurements, especially optically-based measurements.

[0005] Furthermore, once a manufacturing defect, such as a poor adhesive, is identified, there is no known technique that can repair such a manufacturing defect. Summary of the Invention

[0006] In view of the above, it would be advantageous to have an apparatus and method for performing quality control during the production of resin-based manufactured parts. Both detection of resin build quality, such as resin adhesion, and detection of curing defects, such as resin delamination, improve the production of parts. In addition to and in conjunction with additive manufacturing, in-situ measurement and print verification can perform quality control on complex shapes and structures and optionally provide a means for repairing manufacturing defects, such as poor adhesion.

[0007] In general, the present disclosure relates to additive manufacturing methods for producing components, particularly components of nuclear fission reactor structures. In exemplary embodiments, additive manufacturing methods are based on deposition / curing techniques, which can be used to produce objects of nearly any shape or geometry using digital model data from another electronic data source, such as a 3D model or a computer-aided design (CAD) model, an additive manufacturing file (AMF) file, or a stereolithography contour (SLC) file (usually in successive layers). Curing of liquid-based materials, such as slurries, can use different techniques, each of which solidifies or forms the liquid-based material in a layer-by-layer manner to build the manufactured object. Examples include stereolithography (SL), which utilizes various light- or chemical-based curing processes (with associated light- or chemically-reactive materials). In exemplary embodiments, methods for in-situ monitoring of the production of additively manufactured products can be incorporated into the additive manufacturing methods. The in-situ monitoring method compares images of the slurry during and after production of layers of the component and compares regions of those images to corresponding images derived from the digital model data of the corresponding layers. Based on this comparison, threshold criteria are applied to determine the presence or absence of manufacturing defects, such as adhesion defects, delamination, or poor curing, where material from the slurry volume does not adhere to the deposition surface. Manufacturing defects can include, for example, adhesion defects, where material from the slurry volume does not adequately adhere to the deposition surface. Poor adhesion is typically caused by material adhering more strongly to the surface of the transport film than to the printed part, causing the part to break when lifted. Other manufacturing defects can include, for example, poor curing, where the resin did not cure (solidify) due to insufficient exposure or chemical issues, and delamination. The comparison is repeated for successively manufactured layers of the manufactured component.

[0008] Embodiments disclosed herein include methods for in-situ monitoring the production of additively manufactured products.

[0009] In a first embodiment, a method for in situ monitoring the production of an additively manufactured product includes acquiring a first baseline image of an additively manufactured slurry in a first volume of the slurry in a build zone of an additive manufacturing machine, forming a layer of the additively manufactured product by an additive manufacturing process, detecting one or more defects in the layer using void detection technology, acquiring a second baseline image of the slurry in a second volume of the slurry in the build zone of the additive manufacturing machine, detecting one or more defects in the layer using misalignment detection technology, and identifying the presence or absence of manufacturing defects in the additively manufactured product by analyzing output of the void detection technology and the misalignment detection technology. In an exemplary embodiment, the void detection technique includes: after forming a layer of the additively manufactured product by an additive manufacturing process, retracting the additively manufactured product from a first volume of slurry to a first retracted position, where the last formed layer includes an upper surface of the first volume of slurry and is spaced apart from a plane spanning the build zone; and capturing a void image of the first volume of slurry within the build zone, where in the void image the additively manufactured product is in the first retracted position and the first volume of slurry within the build zone is in a post-layer formation state including one or more voids in the first volume of slurry. In an exemplary embodiment, the misalignment detection technique includes immersing an additively manufactured product into a second volume of slurry in a build zone, where at the second immersion position the surface of the last formed layer is at a layer depth from the top surface of the second slurry volume, the layer depth being less than the thickness of the second volume of slurry in the build zone as supplied, and capturing a misalignment image of the second volume of slurry in the build zone, where in the misalignment image the additively manufactured product is at the second immersion position and the second volume of slurry in the build zone is in a pre-layer formation state including a reduced volume of slurry relative to the second volume of slurry in the as supplied state.

[0010] In a second embodiment, a method for in situ monitoring production of an additively manufactured product includes: forming a first portion of an additively manufactured product in a first deposition step, the additively manufactured product being attached to a build stage of an additive manufacturing machine; providing a first volume of a slurry for additive manufacturing to a build zone of an additive manufacturing machine; capturing a first image of the first slurry volume within the build zone, wherein in the first image the first slurry volume within the build zone is as-dispensed and has a thickness between a top surface oriented toward a build stage of the additive manufacturing machine and a bottom surface oriented toward a curing radiation source of the additive manufacturing machine; dipping the additively manufactured product into a first slurry volume within the build zone, wherein at the first immersion position, a first portion of the additively manufactured product is positioned to a first layer depth (D Ln ) and the first layer depth (D Ln ) is less than the thickness of the first slurry volume in the build zone as delivered; A first layer (L) is deposited on the deposition surface of the first portion from at least a first portion of the first slurry volume located between the deposition surface and the bottom surface of the first slurry volume. n ), wherein the distance between the deposition surface and the bottom surface of the first slurry volume is n ) layer thickness (TL n ) defining, forming, Retracting the additively manufactured product from the first slurry volume, at the first retracted position, n ) is retracted away from a plane that includes an upper surface of the first slurry volume and spans the build zone; capturing a second image of the first slurry volume within the build zone, where in the second image the additively manufactured product is in a first retracted position and the first slurry volume within the build zone is in a post-layer state including one or more voids within the first slurry volume; providing a second volume of the additive manufacturing slurry to a build zone of the additive manufacturing machine; capturing a first image of the second slurry volume within the build zone, where in the second image the second slurry volume within the build zone is as-dispensed and has a thickness between a top surface oriented toward a build stage of the additive manufacturing machine and a bottom surface oriented toward a curing radiation source of the additive manufacturing machine; dipping the additively manufactured product into a second slurry volume within the build zone, at the second immersion location, n+1 ) will be deposited on the first layer (L n ) surface is at a second layer depth (D Ln+1 ) and the second layer depth (D Ln+1 ) is less than the thickness of the second slurry volume in the build zone as delivered; capturing a second image of the second slurry volume within the build zone, where in the second image the additively manufactured product is in a second immersion position and the second slurry volume within the build zone is in a pre-layered state including a reduced volume of the second slurry relative to the as-dispensed second slurry volume; correcting image characteristics of the second image of the first slurry volume based on the first image of the first slurry volume to form a corrected void image; correcting image characteristics of the second image of the second slurry volume based on the first image of the second slurry volume to form a corrected misregistration image; and (a) comparing the corrected void image with a binary predicted image from a computer-generated model; and (b) comparing the corrected misalignment image with a binary predicted image from a computer-generated model, the binary predicted image from the computer-generated model being a first layer (L n ) is of a layer in the additive manufacturing product corresponding to the layer; and and identifying the presence or absence of defects in the additively manufactured product based on the comparing step.

[0011] The disclosed methods may be embodied as instructions in a non-transitory computer-readable storage medium that stores instructions for execution by a process.

[0012] The disclosed methods are applicable to different types of additive manufacturing processes. For example, both the void detection and misalignment detection techniques disclosed herein are applicable to additive manufacturing processes that use delivery systems for slurries. Also, for example, the misalignment detection techniques disclosed herein are applicable to additive manufacturing processes that use vat-based deposition systems for slurries or laser-curing radiation, such as SLA and DLP vat-based deposition systems.

[0013] The disclosed method allows for repeatable, in-situ defect detection, thereby gaining efficiencies (time and material) in manufacturing, for example, in that production of a component can be stopped and parts scrapped as necessary. Furthermore, defects detected in one manufactured layer can be compensated for, corrected, or "fixed" by adjustments in the manufacturing process of subsequent layers (and such repairs can be confirmed in-situ before further production of the component).

[0014] The foregoing summary, as well as the following detailed description of the embodiments, can be better understood when read in conjunction with the appended drawings. It is understood that the illustrated embodiments are not limited to the precise arrangements and instrumentalities shown. [Brief explanation of the drawings]

[0015] [Figure 1] FIG. 1 is a schematic block diagram of an additive manufacturing machine according to some embodiments. [Figure 2A] FIG. 2 is a schematic block diagram of the various sub-units of the additive manufacturing machine shown in FIG. 1. [Figure 2B] FIG. 2 is a schematic block diagram of the various sub-units of the additive manufacturing machine shown in FIG. 1. [Figure 2C] FIG. 2 is a schematic block diagram of the various sub-units of the additive manufacturing machine shown in FIG. 1. [Figure 2D] FIG. 2 is a schematic block diagram of the various sub-units of the additive manufacturing machine shown in FIG. 1. [Figure 2E] FIG. 2 is a schematic block diagram of the various sub-units of the additive manufacturing machine shown in FIG. 1. [Figure 3A] 10A-10C show a schematic of slurry delivery to a transport film of an additive manufacturing machine. [Figure 3B] 10 is a corresponding exemplary image of a layer of liquid-based slurry material on a transfer film as applied, viewed from below the transfer film. [Figure 4A] 10A-10C schematically illustrate translating a build stage of an additive manufacturing machine to an immersion position where the build stage is immersed in a volume of liquid-based slurry material. [Figure 4B] 10 is a corresponding exemplary image of a layer of liquid-based slurry material on a transport film with a build stage in a submerged position, viewed from below the transport film. [Figure 5A] 10A-10C schematically illustrate translating a build stage of an additive manufacturing machine from an immersed position to a retracted position in which the build stage is removed from a volume of liquid-based slurry material. [Figure 5B] 10 is a corresponding exemplary image of a layer of liquid-based slurry material on a transport film with the build stage in a retracted position, viewed from below the transport film and showing voids remaining in the slurry layer. [Figure 5C] 5C is a magnified and perspective-corrected image corresponding to a portion of the image of FIG. 5B. [Figure 6] 1 is an image showing multiple registration marks positioned within the field of view of an image capture device. [Figure 7A] 10A-10C illustrate schematically the feeding of a fresh layer of slurry onto the transport film of an additive manufacturing machine. [Figure 7B] 10 is a corresponding exemplary image of a layer of liquid-based slurry material on a transfer film as applied, viewed from below the transfer film. [Figure 7C] 7C is a magnified and perspective-corrected image corresponding to a portion of the image of FIG. 7B. [Figure 8A] 10A and 10B schematically illustrate translating a build stage of an additive manufacturing machine from a retracted position to an immersion position in which the build stage is immersed in a volume of liquid-based slurry material. [Figure 8B] 10 is a corresponding exemplary image of a layer of liquid-based slurry material on a transport film with the build stage in a retracted position, viewed from below the transport film and showing misalignment in the slurry layer. [Figure 8C] 8C is a magnified and perspective-corrected image corresponding to a portion of the image of FIG. 8B. [Figure 9] 1A-1C are schematic side cross-sectional views showing an additively manufactured product immersed in a slurry volume relative to an immersion position; [Figure 10A] 10 illustrates schematically subsequent steps in the production of an additively manufactured product. [Figure 10B] 10 is a corresponding exemplary image of a layer of liquid-based slurry material on a transport film, viewed from below the transport film, showing the pattern of voids resulting from modern additive manufacturing deposition processes. [Figure 11A] 1 is a flow chart illustrating steps in an embodiment of a method for manufacturing an additively manufactured product and in situ monitored manufacturing of the additively manufactured product. [Figure 11B] 1 is a flow chart illustrating steps in an embodiment of a method for manufacturing an additively manufactured product and in situ monitored manufacturing of the additively manufactured product. [Figure 11C] 1 is a flow chart illustrating steps in an embodiment of a method for manufacturing an additively manufactured product and in situ monitored manufacturing of the additively manufactured product. [Figure 11D] 1 is a flow chart illustrating steps in an embodiment of a method for manufacturing an additively manufactured product and in situ monitored manufacturing of the additively manufactured product. [Figure 11E] 1 is a flow chart illustrating steps in an embodiment of a method for manufacturing an additively manufactured product and in situ monitored manufacturing of the additively manufactured product. [Figure 11F] 1 is a flow chart illustrating steps in an embodiment of a method for manufacturing an additively manufactured product and in situ monitored manufacturing of the additively manufactured product. [Figure 12] 1A-1C are exemplary images taken at different stages in the process of forming a corrected void image. [Figure 13]1A-1D are exemplary images taken at different stages in the process of forming a corrected misregistration image; [Figure 14] 10 is an image illustrating aspects of the thresholding process. [Figure 14E] 10 is an example of a plot of the frequency of pixel values ​​used to select the constants CLower and CHigher during the dynamic thresholding process. [Figure 15] 10 is an image showing the effect of dynamic thresholding on a void image. [Figure 16] 10 is an image showing the effect of dynamic thresholding on a misaligned image. [Figure 17] 1A-1C are exemplary images used in the comparison process, including a dual void image, a dual misalignment image, a dual expected image, and a resulting comparison image, respectively. [Figure 18] 1 is an exemplary collection of comparative images and an assembled 3D rendering of the comparative images. [Figure 19] Photograph of the additively manufactured prototype sample corresponding to the assembled 3D rendering in Figure 18(b). [Figure 20A] 10 is a close-up corrected image of the void showing residual slurry within the perimeter of the void. [Figure 20B] 1 is an image showing an example of a 3D defect model including a true pore. [Figure 21] FIG. 1 is a block diagram illustrating an additive manufacturing machine (AMM) controller according to some embodiments. DETAILED DESCRIPTION OF THE INVENTION

[0016] For clarity, in some instances, only some of the named features in the figures are labeled with reference numerals.

[0017] FIG. 1 is a schematic block diagram of an additive manufacturing machine 100 according to some embodiments. The additive manufacturing machine 100 includes several subunits (also referred to as “components”) communicatively coupled to operate the additive manufacturing machine 100 to manufacture additively manufactured products. The subunits include, among others, a component 102 for supplying a material source to a build zone, a component 104 on which the additively manufactured product is built, a component 106 for depositing or curing material that forms successive layers of the additively manufactured product, a component 108 for imaging liquid-based materials during deposition of successive layers of the additively manufactured product, and a component 110 for controlling the additive manufacturing operations based on digital model data and in-situ monitoring the successive layers of the additively manufactured product for manufacturing defects. The component 110 that controls the additive manufacturing operations can be operably connected to the various subunits and components by any suitable means, such as a digital connection transmitted via a wired connection 112, a wireless connection 114, or a combination thereof. The various subunits may be separate components, may be combined, or may share components.

[0018] Details of the various subunits are shown and explained with reference to Figures 2A-2E.

[0019] 2A schematically illustrates a component 102 for supplying a supply of material to a build zone, including a reservoir 200 containing a source of liquid-based material 202, such as a slurry. The reservoir 200 is in fluid communication with a temporary reservoir 204, for example, by piping 206 having an inlet connected to the reservoir 200 and an outlet 208 (directly or indirectly) into the temporary reservoir 204. A valve 210 controls the flow of the liquid-based material 202 into the temporary reservoir 204. From the temporary reservoir 204, the liquid-based material 202 is formed by a metering device 212, such as a doctor blade, into a thin, e.g., less than 100 micron, layer 220 of the liquid-based material 202 onto a transfer film 222. The combination of an appropriate volume of liquid-based material 202 and the structural arrangement of temporary reservoir 204, metering device 212, and transport film 222 provides an appropriate head pressure to form a continuous volume of liquid-based material 202 in layer 220 on first surface 224 of transport film 222 as transport film 222 is translated (t) into and out of deposition zone 230 from deposition zone 230. Furthermore, layer 220 of liquid-based material 202 on transport film 222 has a thickness (T) across the length (L) and width (W) of build zone 240 relative to first surface 224 of transport film 222, e.g., toward build stage 250, e.g., toward deposition surface 254 of build stage 250 (see FIG. 2B ), and a bottom surface 228 oriented toward curing radiation source 280, such that a volume of liquid-based material 202 sufficient to form a to-be-deposited layer of an additively manufactured product is present in build zone 240. The thickness (T) can be controlled by the metering device 212. After passing through the build zone 240, the remaining liquid-based material 202 is collected, for example to be recycled to the reservoir 200. For dimensions consistent with those of the build zone 240, the layer 220 of liquid-based material 202 is continuous as dispensed and has a substantially consistent thickness (T) as formed by the metering device 212.

[0020] The transport film 222 may be in the form of a continuous belt disposed on a rotatable roller 216 that rotates (R) to move the transport film 222 in a first direction (M). The transport film 222 may be any suitable material that is sufficiently transparent to the curing radiation such that the curing radiation functions to solidify or otherwise form the liquid-based material into a layer of the manufactured object, sufficiently chemically inert to the liquid-based material 202 (at least over the time frame that the liquid-based material 202 is in contact with the transport film 222), does not affect the composition of the additively manufactured product, and is sufficiently transparent to allow for image capture by the image capture device 300. In some embodiments, the transport film 222 may be a film of polyethylene terephthalate, more specifically a film of biaxially oriented polyethylene terephthalate.

[0021] Figure 3A schematically illustrates an example layer 220 of liquid-based material 202 on a transport film 222 in an as-applied state, and Figure 3B is a corresponding example image of layer 220 from Figure 3A as viewed from below the transport film 222 by an image capture device 300, e.g., in the appearance shown as A-A'. As can be seen in the example image of Figure 3B, layer 220 is in an as-applied state and has a uniform appearance, indicating a uniform thickness (T) and composition.

[0022] 2B also schematically illustrates the component 104 upon which the additively manufactured product is built, including a build stage 250. The build stage 250 is multi-axis translatable and can move in any direction relative to orthogonally arranged X, Y, and Z axes 252, where the Z axis is perpendicular to a plane containing the first surface 224 of the transport film 222, the X axis is parallel to the plane containing the first surface 224 of the transport film 222 and parallel to the first direction (M) along which the transport film 222 moves, and the Y axis is parallel to the plane containing the first surface 224 of the transport film 222 and perpendicular to the first direction (M) along which the transport film 222 moves. The build stage 250 includes a surface upon which the additively manufactured product is built, such as a deposition surface 254. In some embodiments, the surface 252 is pre-disposed with a base layer of the additively manufactured product upon which subsequent layers are formed, while in other embodiments, the first layer of the additively manufactured product is formed directly on the deposition surface 254. 2B, deposition surface 254 is a surface of or attached to build stage 250. In other figures, such as FIG. 7A, build stage 250 is shown with an in-process additively manufactured product 10, and deposition surface 20 is a surface of the in-process additively manufactured product 10, or for subsequent layers of the in-process additively manufactured product 10, deposition surface 20 becomes the outer surface of the in-process additively manufactured product 10, which is regenerated during each process iteration to be the surface most distal (in the Z-axis) to deposition surface 254. For reference, the length (L) and width (W) of build zone 240 correspond to the X-axis and Y-axis directions, and the thickness (T) of layer 220 of liquid-based material 202 of transport film 222 corresponds to the Z-axis direction.

[0023] Typically, build stage 250 is positioned above build zone 240 so that build stage 250 can be translated in the X and Y axes to a desired position above the volume of liquid-based material 202 in build zone 240, and translated in the Z axis so that deposition surface 254 or deposition surface 20 (depending on the point in the iterative deposition process) on which the additively manufactured product is to be built is immersed in the volume of liquid-based material 202 in build zone 240. As shown in Figure 3A, build stage 250 (and in particular deposition surface 254) is aligned with curing radiation source 280 along axis 256.

[0024] 4A, in a first translation of the build stage 250 in the Z-axis, the deposition surface 254 (if in an initial deposition layer) or a previously deposited layer of the in-process additively manufactured product 10 (if in a subsequent or in-process deposition layer) can be immersed in a volume of liquid-based material 202 in the build zone 240. The volume of liquid-based material 202 in the build zone 240, as supplied, corresponds to the thickness (T) of the layer 220 of liquid-based material 202 as formed by the metering device 212. Immersing the deposition surface 254 (or the in-process additively manufactured product 10) causes the deposition surface to be immersed in a layer depth (D L ) (with respect to the top surface 226 of the layer 220 of as-applied liquid-based material 202). L ) is less than the thickness (T) of the layer 220 of the liquid-based material 202 in the build zone 240 as applied. L ) and the thickness (T) of layer 220 is the maximum thickness (T L ) corresponds to (T L =TD L ).

[0025] Figure 4B is a corresponding image of layer 220 of Figure 4A viewed from below transport film 222, for example, in a view indicated as B-B' by image capture device 300. As can be seen in the exemplary image of Figure 4B, in region A of layer 220 where deposition surface 254 (or in-process additively manufactured product 10) was submerged, a portion of the volume of liquid-based material 202 within build zone 240 is misaligned. The misalignment results in a reduction in the thickness of liquid-based material 202 in the region of layer 220 that corresponds to the geometry and other structural features of the deposition surface. This misalignment is observable as a change in the appearance of layer 220 (viewed from below transport film 222) relative to the appearance of layer 220 in the as-applied state (compare Figure 4B with the as-applied image of Figure 3B, as an example). In the exemplary image of FIG. 4B, layer 220 in region B (outside region A) is substantially undisturbed and remains as applied, while layer 220 in region A has a change in appearance reflecting misalignment due to immersion of deposition surface 254.

[0026] Also, in a second translational movement of the build stage 250, for example in the Z-axis, after the deposition process to form an as-deposited layer on the deposition surface (whether on the deposition surface 254 or on a subsequent or in-process deposited layer of the in-process additive manufactured product 10), the just-deposited in-process additive manufactured product 10 is withdrawn from the volume of liquid-based material 202 in the build zone 240. Typically, the second translational movement of the build stage 250 in the Z-axis withdraws the just-formed layer, e.g., the first layer (L n ) or any subsequent layer (L n+1 ) is to a retracted position in which the deposition surface 20 of the layer 220 includes the top surface 226 of the layer 220 and is spaced apart from a plane spanning the build zone 240. By being spaced apart, as the transport film 222 moves in the first direction (M) to transport a new portion of the layer 220 of liquid-based material 202 into the build zone 240, the deposition surface 20 of the just-formed layer does not contact the top surface 226 of the layer 220 of liquid-based material 202 and does not disturb the as-applied state of the layer 220.

[0027] 2B, the build stage 250 may include or be operably connected to a wireless transceiver 258. Although shown in relation to the build stage 250, any one or more of the components of the additive manufacturing machine 100 may include or be operably connected via a wireless transceiver.

[0028] FIG. 5A schematically illustrates an example of an in-process additively manufactured product 10 in a retracted position after it has been formed while immersed in the volume of liquid-based material 202 in the build zone 240 and before the transport film 222 moves in a first direction (M) to transport a new portion of the layer 220 of liquid-based material 202 to the build zone 240, and FIG. 5B is a corresponding image of the layer 220 from FIG. 5A as viewed by the image capture device 300 from beneath the transport film 222, e.g., in the appearance shown as C-C′. FIG. 5C is a perspective-corrected enlargement example of region P1 in the image of FIG. 5B. In the exemplary image, the layer 220 of liquid-based material 202 includes one or more voids 290 formed by the liquid-based material 202 formed in the as-deposited layer of the in-process additively manufactured product 10. The voids represent negative space remaining in the layer 220 of liquid-based material 202 after a portion of the liquid-based material 202 has been fabricated into the as-deposited layer of the in-process additively manufactured product 10. Furthermore, the pattern 292 of voids 290 represents a freshly formed, as-deposited layer on the deposition surface 20, where the in-process additively manufactured product 10 has been retracted from the volume of liquid-based material 202, e.g., to a retracted position. Such voids 290 are an artifact of the deposition process and correspond in geometry and other structural features to the freshly formed, as-deposited layer. In addition to the pattern 292 in region P1 in the images of FIGS. 5A and 5B, the images include a second pattern 292′ of voids. This second pattern 292′ results from an image obtained after a previous deposition process. Images obtained after the first deposition process do not have the second pattern 292′.

[0029] The process of a first translation of the build stage 250, the formation of a first as-deposited layer, a second translation of the build stage 250, and then moving the transport film 222 to transport a new portion of the layer 220 of liquid-based material 202 to the build zone 240 results in the formation of a plurality of as-deposited layers (L1, L2, ...L n-1 , L n ), where n is equal to the number of as-deposited layers that form the additively manufactured product.

[0030] Returning to the subunits of the additive manufacturing machine 100, FIG. 2C schematically illustrates the component 106 for depositing or curing material to form successive layers of an additively manufactured product, including a curing radiation source 280. As can be seen in the various figures, the curing radiation source 280 is positioned to project curing radiation 282 through the transport film 222 and into the liquid-based material 202, e.g., the layer 220 of additive manufacturing slurry, located in the build zone 240. Any suitable curing radiation source 280 capable of curing (or solidifying) the liquid-based material 202 can be used. In an exemplary embodiment, the curing radiation source 280 uses electromagnetic radiation of a specific wavelength that reacts with the photoreactive material in the liquid-based material 202. Stereolithography (SL), digital light processing (DLP), and electron beam-based technologies can be used.

[0031] 2D schematically illustrates components 108 for imaging a liquid-based material during deposition of successive layers of an additively manufactured product, including an image capture device 300 and an illumination source 310. The image capture device 300 is positioned with a field of view along an axis 302 that includes the build zone 240 as viewed through the transport film 222 so as to be able to capture images of the liquid-based material 202 during deposition of successive layers of the additively manufactured product. Such images may include, for example, (i) an image of the layer 220 of the liquid-based material 202 as it is dispensed, see e.g., Figures 3B and 7B; (ii) an image of the layer 220 of the liquid-based material 202 when the in-process additively manufactured product 10 is in a retracted position and has been immersed in the volume of liquid-based material 202 in the build zone 240, and before the transport film 222 moves in the first direction (M) to transport a new portion of the layer 220 of the liquid-based material 202 into the build zone 240, see e.g., Figure 5B; and (iii) an image of the layer 220 of the liquid-based material 202 when the in-process additively manufactured product 10 is immersed in the volume of liquid-based material 202 in the build zone 240, and before the transport film 222 moves in the first direction (M) to transport a new portion of the layer 220 of the liquid-based material 202 into the build zone 240, see e.g., Figure 5B. n-1 ) deposition surface 20 is the depth (D L ), see, for example, FIG. 8B . The axis of field of view 302 of the image capture device 300 may be at an angle (α) with respect to the axis 256 associated with the curing radiation emanating from the curing radiation source 280. The value of the angle (α) is used for image processing and analysis of the various captured images. Optionally, the image capture device 300 may be aligned with the curing radiation source 280. An exemplary image capture device 300 includes a high-resolution camera using CCD, CMOS, or hyperspectral imaging technology and having a resolution of at least 4 megapixels and up to 16 megapixels. Although described herein as an image, the image may be either a still image or a video, and may include digital formats.

[0032] The illumination source 310 is positioned to project visible light 312 towards the layer 220 of liquid-based material 202 in the build zone 240. Typically, the illumination source 310 is on the same side of the build zone 240 as the image capture device 300 so as to provide sufficient light to the build zone 240 to allow for the acquisition of an image of sufficient quality to enable subsequent image analysis.

[0033] In some embodiments, one or more, e.g., multiple, register marks are positioned within the field of view of the image capture device 300. FIG. 6 shows an example of a register mark 330 embodied as a pattern of high-contrast dots. The register mark 330 enables alignment and correspondence between two different acquired images. By way of example, the exemplary acquired image may assign each pixel a position relative to a coordinate system established by a first axis 332 and a second axis 334 defined relative to the register mark 330. The register mark 330 also enables triangulation of the image within a construction window 336, which is used for processing and analysis of the acquired image. While the register mark 330 in FIG. 6 is shown in an exemplary position relative to the construction window 3346 on the fixed target strip 338, other positions may be used. In some embodiments, the construction window 336 is a window corresponding to the construction zone 240.

[0034] Returning to the subunits of additive manufacturing machine 100, FIG. 2E schematically illustrates component 110 for controlling additive manufacturing operations based on digital model data and in-situ monitoring successive layers of the additively manufactured product for manufacturing defects. In an exemplary embodiment, component 110 includes a computer or other control device 350 having one or more processing units (processors or cores), (optionally) one or more network or other communication interfaces, memory including non-transitory computer-readable storage media, and one or more connections including a communication bus and / or wireless transceiver 352 for interconnecting the components. The connections optionally include circuitry (sometimes referred to as a chipset) that interconnects and controls communication between the system components. The controller system includes a user interface having a display device 354 and (optionally) a user interface 356, such as a keyboard / mouse or other input device. Alternatively, or in addition, display device 354 can include a touch-sensitive surface, in which case the display device is a touch-sensitive display.

[0035] FIG. 7A schematically illustrates a subsequent step in the production of an additively manufactured product. By way of example, as shown in FIG. 7A, after an image of the layer 220 of liquid-based material 202 is acquired when the in-process additively manufactured product 10 is in the retracted position, as described and illustrated in connection with FIGS. 5A-5C and 6, the transport film 222 can be moved in a first direction (M) to transport a new portion of the layer 220 of liquid-based material 202 to the build zone 240. FIG. 7A schematically illustrates an example of the layer 220 of liquid-based material 202 on the transport film 222 as supplied, with the build stage 250 in the retracted position. Note that this is an in-process representation, where the in-process additively manufactured product 10 is positioned on the build stage 250, ready to deposit the next layer on the deposition surface 20. FIG. 7B is a corresponding exemplary image of the layer 220 of FIG. 7A as viewed from below the transport film 222, e.g., in the view indicated as D-D' by the image capture device 300. As seen in the exemplary image of FIG. 7B, layer 220 is in the as-applied state and has a uniform appearance, indicating a uniform thickness (T) and composition. When depositing the same slurry under the same conditions, the as-applied state of this in-process step should be substantially the same as the as-applied state at the beginning of the manufacturing process; for example, the appearance of layer 220 in the as-applied build zone 240 in the image of FIG. 7B should be substantially the same as the appearance of layer 220 in the as-applied build zone 240 in the image of FIG. 3B (assuming consistent imaging parameters). However, the image of FIG. 7B may differ from the image of FIG. 3B by having evidence of the most recent deposition process, such as pattern 292″ transported from build zone 240 for the next deposition layer. FIG. 7C is a perspective-corrected enlarged example of region P2 in the image of FIG. 7B, showing pattern 292″ of voids.

[0036] In another first translation in the Z-axis of the build stage 250, as shown in FIG. 8A , the deposition surface 20 of a previously deposited layer of the in-process additively manufactured product 10 can be immersed in a volume of the liquid-based material 202 in the build zone 240 as delivered that corresponds to the thickness (T) of the layer 220 of the liquid-based material 202 formed by the metering device 212. Immersing the deposition surface 20 causes the deposition surface 20 to be immersed in a layer depth (D L ) (with respect to the top surface 226 of the layer 220 of as-applied liquid-based material 202). L ) is less than the thickness (T) of the layer 220 of the liquid-based material 202 in the build zone 240 as applied. L ) and the thickness (T) of layer 220 is the maximum thickness (T L ) corresponds to (T L =TD L ).

[0037] FIG. 8B is a corresponding image of layer 220 of FIG. 8A viewed from below transport film 222, for example, in the view indicated as E-E′ by image capture device 300. FIG. 8C is a perspective-corrected example of region P3 in the image of FIG. 8B. As can be seen in the exemplary images of FIGS. 8B and 8C, in region C of layer 220 where deposition surface 20 was immersed, a portion of the volume of liquid-based material 202 within build zone 240 is misaligned. The misalignment results in a reduction in the thickness of liquid-based material 202 in the region of layer 220 corresponding to the geometry and other structural features of deposition surface 20. This misalignment is observable as a change in the appearance of layer 220 (viewed from below transport film 222) relative to the appearance of layer 220 in the as-applied state (compare, by way of example, FIGS. 8B and 8C with the as-applied images of FIGS. 7B and 7C). In the example images of Figures 8B and 8C, layer 220 in region D (outside of region C) is substantially undisturbed and in the as-applied state.

[0038] The misregistration in region C forms a pattern 370 representative of the deposition surface 20. Thus, the images in Figures 8B and 8C are just formed in a previous deposition process and are not intended to be a representation of any subsequent deposition process (e.g., more generally, a second layer (L n+1 ) before the deposition of the first layer (L n )) information about the surface of the deposition surface 20 prior to the first layer (L n The portion of the slurry volume between the surface of the first layer (L) and the transfer film 222 is thinner than the thickness (T) of the as-applied layer 220. Furthermore, this reduced thickness is n ) surface. Observable surface features include the geometric shape of the printed part. Surface features can be observed directly or indirectly. In direct observation, the slurry is n ) surface features on the surface of layer 220 with thickness T L The first layer (L) is sufficiently transparent to be observable through the n The surface features of the surface of the ) impart properties to the slurry that correspond to the surface features.

[0039] For illustrative purposes, referring to FIG. 9 , which is a schematic cross-sectional side view of an additively manufactured product 10 immersed in a volume of slurry within the build zone 240 to an immersion position, exemplary surface features are shown, including a mesa 400 and a channel 402, both at a distance (d1, d2, respectively) from a reference plane indicated by dashed line 404 (which may be an imaginary reference plane or, for example, the surface of the last deposited layer of the additively manufactured product 10). Due to the different distances, the surfaces of the mesa 400 and the channel 402 are at different distances relative to the transport film 222. Thus, the thickness of the layer 220 slurry between the surface of the mesa 400 and the surface of the channel 402 is different. Due to the difference in the surfaces (e.g., distance (d1, d2)), different regions A1 and A2 of the slurry have different visual appearances when viewed through the transport film 222. These different visual appearances correspond to the underlying surface and provide a secondary indication of the surface features. For example, a surface feature such as a channel 402 (or a hole in an in-process additively manufactured product 10) results in a thicker slurry between the surface feature and the transport film 222, which is observable as a visually darker or more opaque portion of the slurry. In contrast, a surface feature such as a mesa 400 results in a thinner slurry between the surface feature and the transport film 222, which is observable as a visually lighter or more transparent portion of the slurry. In some embodiments, the thinner slurry between the surface feature and the transport film is sufficiently transparent so that the actual surface feature is observable.

[0040] 10A illustrates a schematic representation of a subsequent step in the production of an additively manufactured product. As an example, as shown in FIG. 10A, after acquiring an image of a layer 220 of liquid-based material 202 while the in-process additively manufactured product 10 is in the immersion position described and illustrated in connection with FIGS. 9A-9C, and after the image of any subsequent layer (L n+1), the just-deposited in-process additively manufactured product 10 is retracted from the volume of liquid-based material 202 in the build zone 240. As discussed above in connection with Figures 5A-5C, typically this second translation of the build stage 250 is in the Z axis to a retracted position. Figure 10B is a corresponding image of the layer 220 of Figure 10A viewed from below the transport film 222, e.g., at the view designated F-F', by the image capture device 300, showing a pattern 420 of voids 422 resulting from the latest additive manufacturing deposition process.

[0041]

[0013] The present disclosure also relates to a method for in-situ monitoring of the production of an additively manufactured product during an additive manufacturing process. Figures 11A-11F show an overview of an embodiment of a method 500 for in-situ monitoring of the production of an additive manufacturing process. The various processes outlined in the steps of the flowcharts of Figures 11A-11F can be read and interpreted in conjunction with the schematic diagrams and images of Figures 3A-3B through 10A-10B.

[0042] The additive manufacturing process is carried out to form the first layer (L n The method 500 begins by forming or otherwise providing a surface, e.g., a base surface, on which the first layer (L n ) is an initial layer of the additively manufactured product 10, the base surface may be the deposition surface 254 of the build stage 250. n ) is a subsequent layer of the additively manufactured product 10, the base surface is the same as the just-deposited previous layer (L n-1 ) deposition surface 20. In the illustrated method 500, a first portion of the additively manufactured product 10 is formed in a first deposition step S505, and the additively manufactured product is attached to a build stage 250 of an additive manufacturing machine. Formation of the first portion includes depositing a first layer (L n ) on the build stage 250 or by a process of forming subsequent layers (L n+1 ) is deposited on the preceding layer (L n ) on the deposition surface 20.

[0043] Once the first portion of the additively manufactured product is formed and the additively manufactured product is attached to the build stage 250, the method 500 continues with step S510 of supplying a first volume of slurry for additive manufacturing to a build zone of the additive manufacturing machine. FIG. 3A schematically illustrates an additive manufacturing machine in the process of supplying a first volume of slurry for additive manufacturing to a build zone 240 of the additive manufacturing machine. A volume of liquid-based material 202, e.g., a slurry, present in the temporary reservoir 204 is formed into a thin, e.g., less than 100 microns, layer 220 of liquid-based material 202 on a transport film 222 by a doctor blade 212 functioning as a metering device 212. An exemplary thickness of the layer 220 is 30-100 microns, e.g., 40-80 microns. The transport film 222 advances in direction (M) to also move the layer 220 of liquid-based material 202 into the build zone 240 below the build stage 250, the surface on which the additively manufactured product 10 is formed.

[0044] After dispensing the first volume of slurry and as-dispensed layer 220 of liquid-based material 202 on transport film 222 (as described herein), method 500 continues at S515 by capturing an image of the first slurry volume within build zone 240. FIGS. 3A-3B (for the initial layer) schematically show the additive manufacturing machine in the process of capturing an image of the first slurry volume within build zone 240 and an exemplary image 510 (sometimes referred to herein as a first image of the first slurry volume within the build zone). First image 510 is captured by image capture component 108, such as image capture device 300, together with illumination source 310 (as needed to provide adequate image quality), and is of the as-dispensed first slurry volume within build zone 240. First image 510 of first slurry volume is taken through transport film 222 and shows bottom surface 228 of layer 220 of liquid-based material 202 oriented toward curing radiation source 280. In the first image 510, the slurry volume is substantially undisturbed and uniform. The first image 510 provides baseline information about the as-applied layer 220 of liquid-based material 202 for use in subsequent calibration and comparison processes. Other factors accounted for by capturing continuous, real-time baseline information include changes in environmental conditions, such as lighting. For example, the first image 510 in FIG. 3B illustrates lighting effects, including an illumination gradient (from light to dark as you progress from left to right in the image, as indicated by arrow 512) and reflections 514 from, for example, the illumination source 310 or other equipment 516. It is also possible to detect non-uniformities in the slurry, such as streaking due to insufficient layering or impurities in the liquid-based material 202 that forms the layer 220 of slurry. FIGS. 7A-7C illustrate this process for subsequent layers.

[0045] The method 500 continues, and then the additively manufactured product 10 is immersed (S520) into a first slurry volume within the build zone 240 to a first immersion position, depositing a first layer (L n) is formed. Figures 4A-4B show this process (for an initial layer) schematically. In the first immersion position, a first portion of the additively manufactured product 20 (or a first portion of the build stage 250 in the case of an initial deposition of the additively manufactured product 10) is immersed at a first layer depth (D Ln ), which also forms the first slurry volume in the build zone 240. Ln ) is less than the thickness (T) of the first slurry volume in the as-applied build zone 240 and the first layer depth (D Ln ) and the thickness (T) of layer 220 is the maximum thickness (T) of a deposited layer of the additively manufactured product. L ) corresponds to (T L =TD Ln ). First layer (L n ) is formed on the deposition surface 20 of at least a first portion of the first slurry volume located between the deposition surface 20 and the bottom surface of the first slurry volume S525. n ) is formed on the deposition surface 20 by exposing the first slurry volume to curing radiation 282 from a curing radiation source 280 based on parameters in a 3D model or other electronic data source used to operate the additive manufacturing machine to create the additively manufactured product. The exposure connects the newly formed material to the deposition surface to form a continuum. Additionally, the distance between the deposition surface 20 and the bottom surface of the first slurry volume is increased by a factor of 100 to form a first layer (L n ) layer thickness (TL n 8A-C illustrate this process for subsequent layers.

[0046] First layer (L n After forming the first layer (L), the method 500 continues at S530 by withdrawing the additively manufactured product 10 from the first slurry volume to a retracted position. FIG. 5A shows the additively manufactured product 10 in a schematic representation in the retracted position. In the retracted position, the just-formed first layer (L n ) is spaced from a plane (P) that includes the upper surface 226 of the first slurry volume and spans the build zone 240 (e.g., spaced a distance SA from the upper surface 226 of the layer 220, as shown in FIG. 5A). n) during its subsequent movement in the direction (M) of the transport film 222, the just-formed first layer (L n ) is positioned so that it is not in contact with the layer 220 of liquid-based material 202, thereby moving the just-used portion of the layer 220 of liquid-based material 202 out of the build zone 240 and moving the new portion of the layer 220 of liquid-based material 202 (which remains as supplied) into the build zone 240.

[0047] The volume of material between the deposition surface 20 and the bottom surface 228 of the first slurry volume is the first layer (L n ), so that by retracting the additively manufactured product 10, the first layer (L n ), leaving a series of openings or voids in the layer 220 of liquid-based material 202 corresponding to the receded material formed in the first layer (L ). The method 500 captures S535 an image of the slurry in this state (also referred to herein as a second image of the first slurry volume in the build zone). An exemplary image 530 may be, for example, an image of the first layer (L n 5 is an image of the first slurry volume in the build zone after a curing radiation source 280 has been repositioned to a retracted position. Image 530 is captured by image capture component 108, such as image capture device 300, together with illumination source 310 (as needed to provide adequate image quality), of the first slurry volume in build zone 240 in a post-deposition state. Image 530 of the first slurry volume was taken through transport film 222 and shows bottom surface 228 of layer 220 of liquid-based material 202 oriented toward curing radiation source 280.

[0048] The newly formed first layer (L n ) can be inferred from the information in image 530. For example, in the post-layer state, the slurry volume in the build zone contains one or more voids 290. n These voids 290 are analyzed to correspond to the pulled material formed in the just-formed first layer (L n) can be correlated with the build quality of the just-formed first layer (L ). Furthermore, if there is residual slurry around the void 290, such residual slurry may be present in the just-formed first layer (L n ) can indicate a manufacturing defect, for example, n ), which is devoid of material deposited from the slurry and therefore contains only the first layer (L n ) corresponds to the area where pores are formed in the body of the first layer (L n The presence of pores in the layer 10 can result in porosity in the as-manufactured additively manufactured product 10 over successive deposition processes.

[0049] After capturing image 530, e.g., after capturing a second image of the first slurry volume in build zone 240, method 500 continues with step S540 of delivering a slurry volume for additive manufacturing (sometimes referred to herein as a second volume of slurry) to the build zone 240 of the additive manufacturing machine. FIG. 7A schematically illustrates an additive manufacturing machine in the process of delivering a second volume of slurry for additive manufacturing to the build zone 240 of the additive manufacturing machine. A volume of liquid-based material 202, e.g., a slurry, present in temporary reservoir 204 is formed into a thin, e.g., less than 100 microns, layer 220 of liquid-based material 202 on a transport film 222 by doctor blade 212, which functions as a metering device 212. An exemplary thickness of layer 220 is between 30 and 100 microns, alternatively between 40 and 80 microns. The transport film 222 is advanced in the direction (M) to move a portion of the layer 220 of liquid-based material 202 as it was dispensed into the build zone 240 below the build stage 250. At the same time, advancing the transport film 222 in the direction (M) moves portions of the layer 220 used in the last deposition process (e.g., portions having voids 290) out of the build zone 240. Finally, after multiple cycles of advancing the transport film 222 in the direction (M), the layer 220 of liquid-based material 202 is withdrawn from the transport film 222 and collected, for example, for recirculation back into the reservoir 200.

[0050] After dispensing the second volume of additive manufacturing slurry into the build zone 240, the method 500 captures S545 an image of the second volume of slurry as dispensed (sometimes referred to herein as a first image of the second slurry volume within the build zone). Exemplary image 550 (see FIG. 7B ) is an image of the second slurry volume within the build zone as dispensed. Note that the additively manufactured product 10 is not yet in contact with the second slurry volume and preferably has not been repositioned from its retracted position. Image 550 is captured by an image capturing component 108, such as image capture device 300, along with illumination source 310 (as needed to provide adequate image quality). Image 550 of the second slurry volume was taken through the transport film 222 and shows the bottom surface 228 of layer 220 of liquid-based material 202 oriented toward the curing radiation source 280. A uniform slurry surface is observed in image 550, with minimal or no visible change in the appearance of the material on the “new” side of the film. Image 550 provides baseline information about the layer 220 of as-applied liquid-based material 202 that is used in subsequent calibration and comparison processes.

[0051] The method 500 continues by immersing the additively manufactured product 10 in a second slurry volume within the build zone 240 to a second immersion position (S550). Figure 8A shows this process diagrammatically. At the second immersion position, a second layer (L n+1 ) is deposited on the first layer (L n ) surface is located at a second layer depth (D Ln+1 ), which also forms a second slurry volume in the build zone 240. Ln+1 ) is less than the thickness (T) of the second slurry volume in the as-applied build zone 240 and the second layer depth (D Ln+1 ) and the thickness (T) of layer 220 is the maximum thickness (T) of a deposited layer of the additively manufactured product. L2 ) corresponds to (T L2 =TD Ln-1 ).

[0052] In some instances, the second immersion location is the same as the first immersion location, and the second layer depth (DLn+1 ) is the first layer depth (D Ln ) is the same as (D Ln+1 =D Ln =D L ). This is typically the case when repeatedly depositing layers of a first continuous feature of an additively manufactured product. However, in other instances, such as when transitioning from a first continuous feature to a second continuous feature of an additively manufactured product, the second layer depth (D Ln+1 ) is the first layer depth (D Ln ) can vary.

[0053] After the additively manufactured article 10 is immersed in the second slurry volume of the build zone 240 to a second immersion position, a second layer (L n+1 ), the method 500 continues by capturing S555 an image of the second volume of slurry (sometimes referred to herein as a second image of the second slurry volume of the build zone). An exemplary image 570 (see FIG. 8B) shows the second layer (L) on the deposition surface. n+1 5 is an image of a second slurry volume of a build zone 240 where the additively manufactured product 10 has been immersed in the second slurry volume of the build zone 240 to a second immersion position before forming a curable radiation source 280. Image 570 is captured by an image capturing component 108, such as image capture device 300, along with illumination source 310 (as needed to provide adequate image quality). Image 570 of the second slurry volume was taken through transport film 222 and shows bottom surface 228 of layer 220 of liquid-based material 202 oriented toward curable radiation source 280.

[0054] Second layer (L n+1 ), the method 500 continues by withdrawing the additively manufactured product 10 from the second slurry volume to a retracted position (see FIG. 10A).

[0055] Concurrently with or after capturing the desired images of the layers formed in the additive manufacturing process, method 500 continues by correcting and then analyzing the captured images. In this process, the correction removes details present in the first (baseline) image, thereby removing variations in lighting and surface texture from the second (misaligned) and third (void) images. For example, image characteristics of a second image of the first slurry volume (e.g., image 530) are corrected S560 based on the first image of the first slurry volume (e.g., image 510) to form a corrected void image. Also, image characteristics of a second image of the second slurry volume (e.g., image 570) are corrected S565 based on the first image of the second slurry volume (e.g., image 550) to form a corrected misaligned image. As part of the correction of each image (e.g., images 510, 530, 550, 570), the as-obtained image can be corrected for triangulation and offset from an axis perpendicular to the bottom surface 228 of layer 220 (e.g., using a registration mark 330 as shown and described in connection with FIG. 6 and based on an angle α), also known as "perspective correction." If desired, other corrections can be applied to individual images before correction to form corrected void images and / or corrected misalignment images, such as optical lens corrections that account for radial distortion induced by lens curvature (such optical lens corrections can be implemented using software, such as using the undistortion function in the open source software OpenCV (Open Source Computer Vision Library)).

[0056] Formation of a corrected void image proceeds as follows. An example of this process is illustrated by the images in FIGS. 12(a)-(c). An as-captured image 700 is corrected for perspective and offset to form a first interim corrected image 710. The first interim corrected image 710 is then corrected for environmental conditions, such as (i) reflections from surfaces in the field of view, such as any glass surfaces and the surface of the transport film 222, (ii) variations in the coloration of the slurry 202, and (iii) any lighting gradients. Such corrections can be made by normalizing based on a first image of the first slurry volume 530 (e.g., a "start of layer" image). Note that the first image of the first slurry volume may also be corrected for perspective and offset, and the corrected first image of the first slurry volume may then be used as a baseline to correct the void image for environmental conditions. The result of the correction is a corrected void image 720.

[0057] Formation of a corrected misalignment image proceeds as follows. An example of this process is illustrated by the images in FIGS. 13(a)-(c). As-captured image 730 is corrected for perspective and offset to form first interim corrected image 740. First interim corrected image 740 is then corrected for environmental conditions, such as (i) reflections from surfaces in the field of view, such as any glass surfaces and the surface of transport film 222, (ii) variations in the coloration of slurry 202, and (iii) any lighting gradients. Such corrections can be made by normalizing based on a first image of second slurry volume 570 (e.g., a "start of layer" image). Note that the first image of the second slurry volume may also be corrected for perspective and offset, and the corrected first image of the second slurry volume may be used as a baseline to correct the misalignment image for environmental conditions. The result of the correction is corrected misalignment image 750.

[0058] Each layer deposited in the iterative process has at least one corrected void image 720 and at least one corrected misalignment image 750, or alternatively, multiple corrected void images 720 and multiple corrected misalignment images 750, forming a set of corrected images associated with the production of an additively manufactured product. Typically, one corrected void image 720 is paired with one corrected misalignment image 750.

[0059] The set of corrected images is then analyzed using a thresholding process. The thresholding process classifies pixels in each image as either "dark" or "light" based on a threshold level applied region-by-region based on nearest neighbors. The threshold range is automatically set by determining the mean and standard deviation of the grayscale or color values ​​of the uncured resin in the image (outside the build region). Pixels within the build region are then classified as either "uncured resin" or "part" based on their difference from the mean. The "build region" is determined by masking a bounding box surrounding the geometry of the printed part. This thresholding is more conservative for void images (compared to misaligned images) because of their higher contrast, requiring a larger difference from the mean to be classified as a "part."

[0060] Image thresholding is used to classify pixels in an image as either "uncured resin" or "parts." The thresholding process determines whether each pixel's data value (typically 0-255) falls within a specific range. In a misalignment image, misaligned resin appears darker or lighter than uncured resin slurry, which has a fairly uniform appearance. Similarly, in a void image, voids appear darker than uncured resin slurry, which has a uniform appearance. The difference in appearance allows for classification by establishing upper and lower threshold boundary conditions for each pixel. TIFF0007748162000001.tif6150 is outside the above range, i.e., C Lower Below or C HigherAny pixel value above this is considered "misaligned" (in the case of a misaligned image) or a void (in the case of a void image) and is therefore resolved as a "part."

[0061] The above method may have some error due to overlap between color values ​​in a color image. To minimize such error, Lower and C Higher is chosen to minimize the overall error (as described below). Use of the above general formula provides a statistical method for determining the cured and uncured portions of the slurry on a binary basis, i.e., as either "uncured resin" or "part."

[0062] constant C Lower and C Higher is determined algorithmically based on the misaligned image. Lower and C HigherThe process of determining the misalignment crops the full misaligned image to a reduced area determined by the geometry file, ensuring that no image overlaps with previously constructed layers. For example, referring to Figures 14(a)-(d), cropping results in artifacts from previous processes outside the cropped area C1, such as the void artifact 760 on the left side of the image in Figure 14(a), while retaining the area to be analyzed within the cropped area C1, such as the area containing the misalignment artifact 770 in the cropped image in Figure 14(b). Next, two samples of the cropped image are acquired: a foreground image and a background image, where the foreground is the expected printed portion and the background is the uncured, unexposed resin. These two samples are then masked (usually in successive layers) using the expected geometric shape from digital model data from another electronic data source, such as a 3D model, a computer-aided design (CAD) model, an additive manufacturing file (AMF) file, or a stereolithography profile (SLC) file. These models provide image "slices" that define the shape of the layers the printer is commanded to print, which can be converted into a binary image. This image is overlaid on the captured image to mask the areas expected to be printed and those that are not. Pixel values ​​from the masked areas are converted into a value frequency graph using a binning method. Figure 14(c) is an example of a mask image, and Figure 14E is an example of a pixel value frequency graph. In this method, the number of pixels from each sample equal to a particular value between 0 and 255 is counted, and then the total count is divided by the total number of pixels to provide a frequency value. The intersection of the foreground frequency count and background frequency graph lines determines the upper and lower foreground thresholds by starting at the peak background value and iterating outward on either side of the peak background value until the foreground frequency is greater than the background frequency. In the example shown in Figure 14E, the upper threshold, i.e., C Higher is at the pixel value of 168, and the lower threshold, i.e., C Loweris at a pixel value of 156. Image pixel values ​​that fall outside the "background" threshold, i.e., below the lower threshold or above the upper threshold, are then classified as "foreground." When analyzing color images, this process is done separately for each color channel; to be classified as "foreground," all three color channels must fall outside the lower and upper "background" thresholds.

[0063] Continuing the thresholding process, pixels in the image are processed to be either white or black, and the analyzed image is converted into a binary image (based on the black or white pixels) for further classification and use. Figure 14(d) is an example of a binary image of such a conversion.

[0064] 15(a)-(b) are images showing the effect of dynamic thresholding on a void image. In particular, a corrected void image 800 is processed by dynamic thresholding to result in a binary void image 810. FIG. 16(a)-(b) are images showing the effect of dynamic thresholding on a displaced image. In particular, a corrected displaced image 820 is processed by dynamic thresholding to result in a binary displaced image 830.

[0065] Furthermore, it should be noted that the general formula and general method for thresholding can be applied using a color scale instead of a gray scale, in which case the general formula and general method would be applied to each of the three color bands (red, green, blue).

[0066] The binary void image and binary misregistration image for each layer formed in the additive manufacturing process are then compared to a binary predicted image for that layer in S570. The binary predicted layer is based on a reference binary image. The reference binary image can be generated based on a CAD model or other input and corresponds to the input used to control the layer's production by the additive manufacturing machine. Typically, as an electronic structure, the binary predicted image has a higher resolution than any of the images captured by the image capture device 300. Alternatively, the binary predicted image can be based on a sampling of the actual additively manufactured product. Figures 17(a)-(d) show exemplary images used in the comparison process, including a binary void image 850, a binary misregistration image 860, a binary predicted image 870 associated with the layer corresponding to the binary void image 850 and the binary misregistration image 860, and a comparison image 880 resulting from a pixel-to-pixel comparison 890. In the comparison 890, the binary images from both the resin misregistration and void detection are compared to the expected geometry at the pixel level. Small connection regions within manufactured parts are compared to provide a regional "confidence level."

[0067] Using one of several methods, the entire build area is divided into multiple regions that can be "present" or "absent." The "confidence level" is the ratio of pixels in the segment classified as "part" to the total number of expected pixels. If this ratio is high enough (i.e., at least 50%, although this can vary based on material and empirical experience), the entire segment is determined to be present. This method is particularly useful when the pixel classifier is tuned to minimize false "part" detections, which incorrectly classify some true "part" pixels as "uncured resin." Suitable segmentation methods include "by contour" and "tiling." The "by contour" method treats each continuous area of ​​the expected part as a single segment. This method is suitable for detecting delamination defects, as delamination defects have rarely been observed to occur across partial sections of the "contour." The tiling method divides the expected part into square tiles, allowing for the detection of partially attached areas or smaller defects.

[0068] The default assumption for the comparison is "fault" (or "defect") unless one or more strong confidence indicators exceed a threshold, in which case the connected region is considered to be successfully glued. The confidence level can be implemented based on thresholding or other suitable techniques, such as computer vision methods or neural networks.

[0069] For example, a binary predicted image typically includes one or more contiguous regions. When the corrected void image is compared to the binary predicted image from the computer-generated model, the percent coverage in the corrected void image can be quantified based on a comparison of the pixel levels within each contiguous region. The presence (or absence) of a manufacturing defect can be indicated based on percent coverage below (or above) a threshold void image value in portions of the additively manufactured product corresponding to contiguous regions in the corrected void image. In one embodiment, the threshold void image value is 97%, where percent coverage below 97% correlates with the presence of a manufacturing defect and percent coverage above 97% correlates with the absence of a manufacturing defect. Similar to comparing the corrected void image to the binary predicted image from the computer-generated model, percent coverage in the corrected misalignment image can be quantified based on a comparison of the pixel levels within each contiguous region. The presence (or absence) of a manufacturing defect can be identified S555 based on percent coverage below (or above) a threshold misalignment image value in portions of the additively manufactured product corresponding to contiguous regions in the corrected void image. In one embodiment, the threshold misregistration image value is 97%, with coverage below 97% correlating with the presence of a manufacturing defect and coverage above 97% correlating with the absence of a manufacturing defect.

[0070] In some embodiments of the method, both the threshold void image value and the threshold misregistration image value must exceed the threshold to correlate with the absence of a manufacturing defect.

[0071] In other embodiments, a threshold void image value above the threshold is sufficient to correlate with the absence of a manufacturing defect even if the threshold misregistration image value is below the threshold, since the void image provides more reliable information about the condition of the as-fabricated layer. However, in other such embodiments, a threshold void image value below the threshold is sufficient to correlate with the presence of a manufacturing defect even if the threshold misregistration image value is above the threshold.

[0072] In FIG. 17(d), the areas labeled F1 and F2 represent regions of the layer where comparison resulted in a defect indication. Such a defect indication corresponds to a lack of deposited material, poor curing, delamination, or some other defect in the additive manufacturing at that location within the layer. Multiple comparison images can be collected and assembled into a 3D model that corresponds to the as-manufactured additively manufactured product or at least a portion of the as-manufactured additively manufactured product. FIG. 18(a) is an example collection of comparison images 880 (including defect areas F1 and F2), and FIG. 18(b) is an assembled 3D rendering 900 of the comparison images 880. Defect areas F1 and F2 are visible in the 3D rendering 900. FIG. 19 is a photograph of an additively manufactured prototype sample corresponding to the assembled 3D rendering 900 of FIG. 18(b), clearly showing defect areas F1 and F2. Being able to identify and visualize defect areas such as F1 and F2 enables defect detection. Defect detection can be monitored remotely and can be done in situ during the manufacturing process after production, such as during quality control, when qualifying manufactured parts.

[0073] The disclosed method also provides a way to detect pores within an additively manufactured product. As previously described herein, the presence of residual slurry within the periphery of the void 290 indicates that the first layer (L n These manufacturing defects can indicate defects in the manufacturing of the first layer (L n ) which may be the pores in the body of the just-formed first layer (L n) and, over successive deposition processes, the porosity of the as-manufactured additively manufactured product 10. To detect residual slurry that may become pores, analysis of the corrected void image can optionally detect residues that may become porosity in the printed layer. Knowing that pores in the printed layer leave isolated pockets of slurry on the transport film 222 in areas defining the voids 290, any slurry that is separated and not continuous from the layer 220 within the perimeter of the voids 290 that is not part of the expected geometry of the additively manufactured product can be assumed to be evidence of pores. This approach works regardless of the flow of the slurry itself (which can cover various areas of the contour depending on the viscosity) because any isolated droplets of slurry were not the result of flow after the deposition event.

[0074] FIG. 20A is a zoomed-in, corrected void image 910 showing multiple residual slurry 912 within the perimeter of void 290. The residual slurry 912 is separated and not continuous with layer 220. Image processing identifies the residual slurry 912, typically in the form of droplets, at a resolution of 30 microns and determines which residual slurry 912 is separated from any potential slurry flow. In an embodiment of the "pore" classification process, the image is classified as either "resin" or "void" using the thresholding method previously described herein. Next, a "fill" algorithm seeded with "resin" is used to classify pixels outside the expected part geometry and remove all connected "resin" classified pixels, including those within the geometry. All remaining pixels are then considered defects because they are not connected to uncured resin and therefore cannot be the result of resin flow. Once identified, the residual slurry 912 is classified as a pore site, and multiple pore sites in the layer are cataloged. For each layer, the catalogs of pore sites are compared, and any pore site that persists through more than three layers, or even more than one layer, is considered to be a true pore. Furthermore, the true pores identified by this process can be incorporated into a 3D defect model. Figure 20B is an image showing an example of a 3D defect model 920 that includes a true pore 922.

[0075] The true pores identified by this process allow for the quantification of overall part density and the identification of potential weaknesses in the as-manufactured part. The high accuracy of pore identification by this process has been confirmed by forensic analysis of as-manufactured parts. In particular, a 3D defect model of the as-manufactured part was created, and areas of expected high porosity were exposed, e.g., by slicing, and examined using a microscope. The locations of the pores observed using the microscope matched the locations of the pores in the 3D defect model.

[0076] In some embodiments, component 110 for controlling additive manufacturing operations based on digital model data and in-situ monitoring successive layers of the additively manufactured product for manufacturing defects is embodied in a computer system or computer-aided machine, such as a computer-controlled additive manufacturing machine. The computer system or computer portion of the computer-aided machine can be a general-purpose computer, a special-purpose computer, or a server that includes, among other things, a non-transitory computer-readable storage medium containing instructions for operating and controlling additive manufacturing machine 100 and electronic data sources such as computer-aided design (CAD) models or additive manufacturing file (AMF) files or stereolithography contour (SLC) files (typically for successive layers) related to the additively manufactured product 10.

[0077] FIG. 21 is a block diagram illustrating an additive manufacturing machine (AMM) controller 1000 according to some embodiments. The AMM controller 1000 typically includes one or more processing units (processors or cores) 1002, (optionally) one or more network or other communication interfaces 1004, memory 1006, and one or more wired or wireless connections 1008 for interconnecting these components. For example, such connections may optionally include a communication bus that optionally includes circuitry (sometimes referred to as a chipset) that interconnects and controls communication between the system components. Alternatively, the components may communicate wirelessly using wireless transceivers. The AMM controller system 1000 includes a user interface 1010. The user interface 1010 may include a display device 1012 and optionally include input devices 1016, such as a keyboard / mouse, trackpad, and / or input buttons. Alternatively, or in addition, the display device 1012 may include a touch-sensitive surface 1014, in which case the display device is a touch-sensitive display. The connections 1008 of the AMM controller system 1000 also operably connect to and interface with various subunits communicatively coupled to operate the additive manufacturing machine 100 to produce additively manufactured products. Thus, the connections 1008 of the controller system 1000 operably connect to and interface with components 102 for providing a supply of material to the build zone (e.g., components associated with storing, supplying, and transporting slurry 202), components 104 (such as build stage 250) on which the additively manufactured product is built, components 106 (such as curing radiation source 280) for depositing or curing material that forms successive layers of the additively manufactured product, and components 108 (such as image capture device 300 and illumination source 310) for imaging the liquid-based material during deposition of successive layers of the additively manufactured product. Other components for controlling additive manufacturing operations based on digital model data and for in-situ monitoring of successive layers of the additively manufactured product for manufacturing defects may also be included.Furthermore, the various subunits may be separate components, may be combined, or may share components.

[0078] Memory 1006 may include high-speed random-access memory such as DRAM, SRAM, DDR RAM, or other random-access solid-state memory devices, and may include non-volatile memory such as one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, and / or other non-volatile solid-state storage devices. In some embodiments, memory 1006 includes one or more storage devices located remotely from processor 1002. Memory 1006, or a non-volatile memory device within memory 1006, includes a non-transitory computer-readable storage medium. In some embodiments, memory 1006, or the computer-readable storage medium of memory 1006, stores instructions for executing (e.g., by a processor) the methods described herein. For example, the memory stores the following programs, modules, and data structures, or a subset or superset thereof: an operating system 1020 that contains instructions for handling various basic system services and performing hardware-dependent tasks; a network communication module 1022 used to connect the controller system 1000 to other computers via one or more connections 1008 (wired or wireless) and one or more network communication interfaces 1004; an image / video capture module 1024 (e.g., a camera module) for processing images or video captured by the imaging component 108; One or more AMM modules 1030, including the following modules (or instruction sets), or a subset or superset thereof: o A slurry module 1032 for interfacing with and controlling the operation of slurry-related components within component 102 to provide a supply of material to the build zone, including reservoir 200, (optionally) valve 210, metering device 212, and interim reservoir 204; o A transport film module 1034 for interfacing with and controlling the operation of transport film related components within component 102 to provide a supply of material to the build zone, including the transport film 222 and the rotatable roller 216; o A build stage module 1036 for interfacing with and controlling the operation of build stage related components in the component 104 upon which the additively manufactured product is built, including the build stage 250 and wireless transceiver 258; o a curing radiation source module 1038 for interfacing with and controlling the operation of curing radiation source related components in component 106 to deposit or cure material forming successive layers of the additively manufactured product, including the curing radiation source 280; o an imaging module 1040 for interfacing with and controlling the operation of imaging-related components in component 108 for imaging the liquid-based material during deposition of successive layers of the additively manufactured product, including an image capture device 300 and an illumination source 310; and a digital model data module 1042 for interfacing with and controlling the operation of components associated with storing, sharing, and accessing electronic information related to the digital model of the part being manufactured, as well as digital image information obtained from the image module; and One or more defect detection modules 1060, including the following modules (or instruction sets), or a subset or superset thereof: o variously, an image correction module 1062 and an image comparison module 1064 for (a) interfacing with image-related components and digital information, (b) controlling the operation and application of various image processing functions, including general formulas for classification, perspective correction, and thresholding, and (c) comparing images and image-related data and information; and o A product reconstruction module 1066 for interfacing with and controlling components related to the storage, sharing, and access of electronic information related to digital models of parts to be manufactured, as well as digital image information captured from the imaging module, and related to the correction and analysis of captured images and visualization of such information, for example by computer-aided three-dimensional rendering.

[0079] Each of the above-identified modules corresponds to a set of executable instructions for performing one or more functions described above and / or in the methods described herein (e.g., additive manufacturing methods, computer-implemented methods, and other information processing methods described herein). However, these modules (e.g., sets of instructions) need not be implemented as separate software programs, procedures, or modules; thus, various subsets of these modules are optionally combined or otherwise rearranged in various embodiments. In some embodiments, memory 1006 stores a subset of the above-identified modules and data structures. In some embodiments, memory 1006 stores additional modules and data structures not described above.

[0080] Suitable additive manufacturing equipment can be utilized that can accommodate the specific requirements of the materials used to manufacture the components (such as chemical resistance), the specific requirements of the application of the equipment itself (such as specific atmospheric or vacuum requirements), and the size and geometry of the manufactured components. Examples of suitable additive manufacturing equipment include SLA and DLP equipment, electron beam-based additive manufacturing equipment, and DLP stereolithography equipment, any one of which can be modified or adapted to suit specific requirements.

[0081] An exemplary method of additive manufacturing can include providing a controller of an additive manufacturing device with a design of a component to be manufactured, which can be incorporated into an additive manufacturing protocol.

[0082] Additive manufacturing protocols can be developed and / or adapted for use with any suitable additive manufacturing process. Examples of suitable additive manufacturing processes are disclosed in ISO / ASTM 52900-15, which defines categories of additive manufacturing processes including binder jetting, directed energy deposition, material extrusion, material jetting, powder bed fusion, sheet lamination, and photopolymerization. The contents of ISO / ASTM 52900-15 are incorporated herein by reference. Stereolithography is a form of additive manufacturing that uses a photopolymerization process. In exemplary embodiments, stereolithography additive manufacturing techniques involve photoinitiation by exposure to ultraviolet or beta radiation. In some exemplary embodiments, ultraviolet radiation is generated in a digital light processor (DLP) or stereolithography (SLA) device. In other exemplary embodiments, beta radiation is generated in an electron beam (EBeam) or electron beam irradiation (EBI) device. While the methods and compositions disclosed herein are described within the context of stereolithography, it is expressly contemplated that such methods and compositions may be extended and / or adapted to other additive manufacturing processes.

[0083] In another aspect of the exemplary method, a supply volume of a slurry composition, such as a tank or reservoir, is established. A base portion of the green body of the part is then formed by curing a portion of the slurry composition in contact with a movable base of the additive manufacturing apparatus. Alternatively, the base portion can be pre-fabricated before the start of the additive manufacturing process. Additional portions of the green body of the part are formed layer-by-layer by first curing a portion of the slurry composition in contact with the base portion to form a first layer of the green body, and then curing a portion of the slurry composition in contact with a previously deposited layer of the green body to form the additional portion while translating the movable base relative to the interface between the surface of the supply volume and the last-formed additional portion of the green body. The translation of the movable base is typically dependent on the component design and dictated by the additive manufacturing protocol. In an exemplary embodiment, the translation of the movable base relative to the interface between the surface of the supply volume and the last-formed additional portion of the green body has an X-axis resolution and a Y-axis resolution of 50 microns or better, and a Z-axis resolution of 20 microns or better. In an exemplary embodiment, each additional portion of the layer-by-layer formed green body of the component has a thickness of at least 25 microns, alternatively between 25 microns and 50 microns. Upon completion of the layer-by-layer fabrication of the green body of the component, the green body of the component may be removed from the additive manufacturing apparatus and sintered (or processed by other debinding / densification techniques) to form a densified ceramic. The image capture, image correction, and image comparison processes described herein may be suitably incorporated into this method.

[0084] Based on one or more of the identified defects and / or other deviations, the additive manufacturing process or slurry composition can be adjusted to correct, mitigate, or compensate for the defects and / or deviations. For example, the components of the composition (either the material chemistry or the amount of the components) can be adjusted and changed. As another example, one or more parameters of the additive manufacturing technique can be adjusted and changed, such as the movement of the build stage (spatially and temporally), the temperature associated with deposition, and the sequence of steps. Other process parameters that can be adjusted include increasing the reservoir temperature to a higher temperature, decreasing viscosity, creating a more uniform layer thickness, adjusting a pre-exposure delay to allow for proper planarization of the additively manufactured printed layer, adjusting the movement speed to allow for mitigation of hydraulic bearing forces and printing window peeling, applying multiple exposures per layer to limit scattering and increase the depth of cure, and continuously varying intensity exposures (moving images) to optimize desired properties. Other design parameters that can be adjusted include parameters such as: Altering the design of thin geometries that are too thin to print, increasing / removing pores that are too small and may be blocked by scattering during exposure, adding drainage or cleaning holes to parts to aid in the removal of trapped slurry, or combining gyro and lattice morphologies to support delicate geometries with structurally and neutrally useful materials. Composition and / or parameter adjustments can be made independently or in combination. Also, composition and / or parameter adjustments can be implemented in subsequent iterations of an ongoing iterative deposition process or in subsequent iterative deposition processes. Alternatively, adjustments to the slurry composition or additive manufacturing parameters can be made to determine the effects of changing such composition / parameters. Information about such causes and effects can be developed and used in subsequent iterations of an ongoing iterative deposition process or in subsequent iterative deposition processes.

[0085] In some manufacturing methods or manufacturing method steps, additively manufactured product features and structures (or portions thereof) are fabricated as an integral, monolithic structure, for example, using an additive manufacturing process. As used herein, an additive manufacturing process includes any technique for building a 3D object by adding material layer by layer. In one example, the disclosed method can be applied to the manufacture of nuclear fission reactor structures and auxiliary equipment. One example of a suitable additive manufacturing process utilizes 3D printing of metal alloys, such as molybdenum-containing metal alloys, Zircaloy-4, or Hastelloy X, to form the aforementioned structural features, such as cladding. In other embodiments, fissionable nuclear fuel compositions and / or heat transfer agents and / or moderator materials and / or poisons used as part of a nuclear fission reactor structure can be included in the integral, monolithic structure when a suitable multi-material additive manufacturing process having multiple metals in the feedstock is used. If molten metal is not included in the additive manufacturing process, the additive manufacturing process can be paused, a volume of molten metal (either in liquid or solid form) is placed into the fuel cavity, and the additive manufacturing process is continued to complete the construction of the closed chamber. Other alloys that can be used in nuclear fission reactor structures and auxiliary equipment when using appropriate multi-material additive manufacturing processes having multiple metals in the feedstock include steel alloys, zirconium alloys, and molybdenum-tungsten alloys (for the cladding and / or containment structure), beryllium alloys (for the reflector), and stainless steel (for the containment structure). Even when not produced by additive manufacturing processes, the above materials can be used in manufacturing various features and structures of such nuclear fission reactors and auxiliary equipment.

[0086] The additive manufacturing techniques disclosed herein can include additional steps of (a) predictive and causal analysis, (b) in-situ monitoring combined with machine vision and accelerated processing during layer-by-layer manufacturing of the structure, (c) automated analysis combined with machine learning components, and (d) virtual inspection of the digital representation of the as-built structure. Furthermore, additive manufacturing techniques can create complex geometries and, when coupled with in-situ sensors, machine vision images, and artificial intelligence, allow for the tuning of manufacturing quality as components are built on a layer-by-layer basis (often these layers are on the scale of 50 microns), resulting in predictive quality assurance of the manufacturing of such reactors and structures.

[0087] A variety of materials can be used to additively manufacture the components of the fission reactor and auxiliary equipment. For cladding, typically a corrosion-resistant material with a low thermal neutron absorption cross section is used. Exemplary materials include Zircaloy or steel, but other materials may be used if appropriate for the reactor conditions, such as compositions including metals and ceramic systems (Be, C, Mg, Zr, O, Si), and their alloys, including molybdenum, tungsten, rhenium, tantalum, hafnium, and carbides. For fissionable nuclear fuel, the composition should be greater than 5% and less than 20% U. 235 Assays containing high-assay low-enriched uranium (HALEU) or 20% or more U 235The fuel element structure may be highly enriched uranium (HEU) containing uranium. Suitable fissile nuclear fuel compositions applicable to the disclosed fuel element structure include uranium oxide (UO2) enriched less than 20% uranium, 10% by weight molybdenum (U-10Mo), uranium nitride (UN), and other stable fissile fuel compounds. Burn-in poisons may also be included. Typically, the fissile nuclear fuel composition is in the form of a ceramic material (cermet), such as UO2 with W or Mo and UN with W or Mo. If used, a heat transfer agent, such as a salt or metal that melts at operating temperatures, may be included in the fuel element structure to improve thermal bonding between the fuel composition body and the cladding body. Additionally, the heat transfer agent may occupy cracks or other defects in the fuel element structure (whether initially present or developed during reactor operation) to promote thermal bonding. Molten metals suitable for inclusion in the disclosed nuclear propulsion fission reactor structures and in the fuel element structures to provide heat transfer contact include sodium (Na), sodium-potassium (NaK), potassium (K), and iron (Fe).

[0088] It should also be understood that although the terms "first" and "second" are used herein to describe various elements, these elements should not be limited by these terms. These terms are used only to distinguish one element from another. For example, a first image could be referred to as a second image, and similarly, a second image could be referred to as a first image, without departing from the scope of various described embodiments. The first image and the second image are both images, but are not the same image.

[0089] While some of the various figures depict some logical stages in a particular order, stages that are not order-dependent may be rearranged, and other stages may be combined or separated. While some rearrangements or other groupings are specifically mentioned, others will be apparent to those skilled in the art, and thus the rearrangements and groupings presented herein are not an exhaustive list of alternatives. Furthermore, it should be recognized that the stages may be implemented in hardware, firmware, software, or any combination thereof.

[0090] Furthermore, the terminology used in the description of various embodiments set forth herein is for the purpose of describing particular embodiments only and is not intended to be limiting. As used in the description of the various described embodiments and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly dictates otherwise. It will also be understood that the term "and / or," as used herein, refers to and encompasses any and all possible combinations of one or more of the associated listed terms. It will be further understood that the terms "includes," "including," "comprises," and / or "comprising," when used herein, specify the presence of stated features, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, elements, components, and / or groups thereof.

[0091] Although particular embodiments have been referenced, it will be apparent that other embodiments and variations may be devised by others skilled in the art without departing from the spirit and scope thereof, and it is intended that the appended claims be construed to include all such embodiments and equivalent variations.

Claims

1. 1. A method for in-situ monitoring production of an additively manufactured product, comprising: forming a first portion of the additively manufactured product in a first deposition step, the additively manufactured product being attached to a build stage of an additive manufacturing machine; providing a first slurry volume of a slurry for additive manufacturing to a build zone of the additive manufacturing machine; capturing a first image of the first slurry volume within the build zone, wherein in the first image the first slurry volume within the build zone is as-dispensed and has a thickness between a top surface oriented toward the build stage of the additive manufacturing machine and a bottom surface oriented toward a curing radiation source of the additive manufacturing machine; dipping the additively manufactured product into the first slurry volume within the build zone, wherein at a first immersion position, the first portion of the additively manufactured product is at a first layer depth (D Ｌｎ ) and the layer depth (D Ｌｎ ) is less than the thickness of the first slurry volume in the as-dispensed build zone; a first layer (L) formed on the deposition surface of the portion of the additive manufactured product from at least a first portion of the first slurry volume located between a deposition surface and the bottom surface of the first slurry volume; ｎ ), wherein the distance between the deposition surface and the bottom surface of the first slurry volume is less than the first layer (L ｎ ) layer thickness (TL ｎ ) defining or forming a Retracting the additive manufactured product from the first slurry volume, wherein in a first retracted position, the first layer (L ｎ ) is retracted away from a plane that includes the upper surface of the first slurry volume and spans the build zone; capturing a second image of the first slurry volume within the build zone, wherein in the second image the additively manufactured product is in the first retracted position and the first slurry volume within the build zone is in a post-layer state including one or more voids within the first slurry volume; providing a second slurry volume of the slurry for additive manufacturing to the build zone of the additive manufacturing machine; capturing a first image of the second slurry volume within the build zone, wherein in the first image of the second slurry volume, the second slurry volume within the build zone is as-dispensed and has a thickness between a top surface oriented toward the build stage of the additive manufacturing machine and a bottom surface oriented toward a curing radiation source of the additive manufacturing machine; dipping the additively manufactured product into the second slurry volume within the build zone, at a second immersion location, ｎ＋１ ) will be deposited on the first layer (L ｎ ) surface of the second slurry volume at a second layer depth (D Ｌｎ＋１ ) and the second layer depth (D Ｌｎ＋１ ) is less than the thickness of the second slurry volume in the as-dispensed build zone; capturing a second image of the second slurry volume within the build zone, wherein in the second image of the second slurry volume, the additively manufactured product is in the second immersion position and the second slurry volume within the build zone is in a pre-layered state including a reduced volume of second slurry relative to the second slurry volume as delivered; correcting image characteristics of the second image of the first slurry volume based on the first image of the first slurry volume to form a corrected void image; correcting image characteristics of the second image of the second slurry volume based on the first image of the second slurry volume to form a corrected misregistration image; and (a) comparing the corrected void image with a binary predicted image from a computer-generated model, and (b) comparing the corrected misalignment image with the binary predicted image from the computer-generated model, wherein the binary predicted image from the computer-generated model is ｎ ) is of a layer within the additive manufactured product corresponding to identifying the presence or absence of defects in the additively manufactured product based on the comparing.

2. generating a comparison image based on comparing (a) the corrected void image and the binary predicted image from a computer-generated model, and (b) the corrected misalignment image and the binary predicted image from the computer-generated model; 10. The method of claim 1, further comprising assembling a plurality of comparison images to form a 3D model corresponding to at least a portion of the additively manufactured product.

3. 3. The method of claim 1 or 2, wherein identifying the presence or absence of a defect comprises identifying a location of the defect within the additively manufactured product.

4. 4. The method of claim 3, wherein the location of the defect is identified by a built layer within the additively manufactured product and a location within the built layer.

5. 5. The method of claim 1, wherein the comparison of (a) the corrected void image with the binary predicted image from the computer-generated model, and (b) the comparison of the corrected misalignment image with the binary predicted image from the computer-generated model is a pixel-level comparison.

6. 5. The method of claim 1, wherein the binary predicted image includes one or more contiguous regions, and wherein comparing the corrected void image with the binary predicted image from the computer-generated model quantifies coverage in the corrected void image based on a pixel-level comparison within each contiguous region.

7. 7. The method of claim 6, wherein the absence of defects is indicated by a coverage that exceeds a threshold void image value in the portion of the additively manufactured product that corresponds to the continuous region of the corrected void image.

8. 8. The method of claim 1, wherein the binary predicted image includes one or more contiguous regions, and wherein comparing the corrected misaligned image with the binary predicted image from the computer-generated model quantifies coverage in the corrected misaligned image based on a pixel-level comparison within each contiguous region.

9. 9. The method of claim 8, wherein the absence of defects is indicated by a coverage that exceeds a threshold misregistration image value in a portion of the additively manufactured product that corresponds to the contiguous region of the corrected misregistration image.

10. the binary prediction image includes one or more contiguous regions; comparing the corrected void image with the binary expected image from the computer-generated model quantifies a first coverage in the corrected void image based on a pixel-level comparison within each continuous region; 3. The method of claim 1 or 2, wherein comparing the corrected misaligned image with the binary expected image from the computer-generated model quantifies a second coverage in the corrected misaligned image based on a pixel-level comparison within each continuous region.

11. 11. The method of claim 10, wherein for the contiguous regions of the corrected void image and the contiguous regions of the corrected misalignment image that correspond to the same portion of the additively manufactured product, the absence of defects in the portion of the additively manufactured product is indicated by the first coverage rate being above a threshold void image value and the second coverage rate being below a threshold misalignment image value.

12. 11. The method of claim 10, wherein for the contiguous regions of the corrected void image and the contiguous regions of the corrected misalignment image that correspond to the same portion of the additively manufactured product, the presence of a defect in the portion of the additively manufactured product is indicated by the first coverage percentage being below a threshold void image value and the second coverage percentage being above a threshold misalignment image value.

13. 13. The method of claim 1, wherein the defects comprise adhesion defects where material from the first slurry volume does not adhere to the additively manufactured product.

14. 14. The method of claim 1, wherein the first image of the first slurry volume and the second image of the first slurry volume are corrected for perspective and offset before correcting image characteristics of the second image of the first slurry volume.

15. 15. The method of claim 1, wherein the first image of the second slurry volume and the second image of the second slurry volume are corrected for perspective and offset before correcting image characteristics of the second image of the second slurry volume.

16. The additive manufacturing machine comprises: a transport film, a first surface of the transport film supporting the slurry for additive manufacturing and transporting the slurry into and out of the build zone; a curing radiation source positioned to project curing radiation through the transport film onto a layer of the additive manufacturing slurry disposed in the build zone; an illumination source positioned to project visible light onto the layer of additive manufacturing slurry disposed in the build zone; an image capture device positioned to have a field of view that includes the build zone as viewed through the transport film; The method according to claim 1 , comprising:

17. A non-transitory computer-readable storage medium storing instructions for execution by a process, the instructions comprising: forming a first portion of an additively manufactured product in a first deposition step, the additively manufactured product being attached to a build stage of an additive manufacturing machine; providing a first slurry volume of a slurry for additive manufacturing to a build zone of the additive manufacturing machine; capturing a first image of the first slurry volume within the build zone, wherein in the first image the first slurry volume within the build zone is as-dispensed and has a thickness between a top surface oriented toward the build stage of the additive manufacturing machine and a bottom surface oriented toward a curing radiation source of the additive manufacturing machine; dipping the additively manufactured product into the first slurry volume within the build zone, wherein at a first immersion position, the first portion of the additively manufactured product is at a first layer depth (D Ｌｎ ) and the layer depth (D Ｌｎ ) is less than the thickness of the first slurry volume in the as-dispensed build zone; a first layer (L) formed on the deposition surface of the portion of the additive manufactured product from at least a first portion of the first slurry volume located between a deposition surface and the bottom surface of the first slurry volume; ｎ ), wherein the distance between the deposition surface and the bottom surface of the first slurry volume is less than the first layer (L ｎ ) layer thickness (TL ｎ ) defining or forming a Retracting the additive manufactured product from the first slurry volume, wherein in a first retracted position, the first layer (L ｎ ) is retracted away from a plane that includes the upper surface of the first slurry volume and spans the build zone; capturing a second image of the first slurry volume within the build zone, wherein in the second image the additively manufactured product is in the first retracted position and the first slurry volume within the build zone is in a post-layer state including one or more voids within the first slurry volume; providing a second slurry volume of the slurry for additive manufacturing to the build zone of the additive manufacturing machine; capturing a first image of the second slurry volume within the build zone, wherein in the first image of the second slurry volume, the second slurry volume within the build zone is as-dispensed and has a thickness between a top surface oriented toward the build stage of the additive manufacturing machine and a bottom surface oriented toward a curing radiation source of the additive manufacturing machine; dipping the additively manufactured product into the second slurry volume within the build zone, at a second immersion location, ｎ＋１ ) will be deposited on the first layer (L ｎ ) surface of the second slurry volume at a second layer depth (D Ｌｎ＋１ ) and the second layer depth (D Ｌｎ＋１ ) is less than the thickness of the second slurry volume in the as-dispensed build zone; capturing a second image of the second slurry volume within the build zone, wherein in the second image of the second slurry volume, the additively manufactured product is in the second immersion position and the second slurry volume within the build zone is in a pre-layered state including a reduced volume of second slurry relative to the second slurry volume as delivered; correcting image characteristics of the second image of the first slurry volume based on the first image of the first slurry volume to form a corrected void image; correcting image characteristics of the second image of the second slurry volume based on the first image of the second slurry volume to form a corrected misregistration image; (a) comparing the corrected void image with a binary predicted image from a computer-generated model, and (b) comparing the corrected misalignment image with the binary predicted image from the computer-generated model, wherein the binary predicted image from the computer-generated model is ｎ ) is of a layer within the additive manufactured product corresponding to and identifying the presence or absence of defects in the additively manufactured product based on the comparing step.

Images