Build plan generation for additive manufacturing systems

Abstract

The techniques described herein relate to build plan generation for additive manufacturing (AM) systems. An example method for determining a configuration of an AM system comprises slicing a three-dimensional model of a part to be fabricated by the AM system into at least one layer including a first layer, determining a surrounding volume ratio of a first volume of fused material surrounding one or more voxels and a total volume surrounding the one or more voxels for respective features of the first layer including a first feature, identifying at least one portion of the first feature as a first subregion of the first layer based on determining that a surrounding volume ratio of the at least one portion satisfies a threshold, mapping at least the surrounding volume ratio to one or more configuration parameters of the AM system, and fabricating the first subregion using the one or more configuration parameters.

Description

BUILD PLAN GENERATION FOR ADDITIVE MANUFACTURING SYSTEMSRELATED APPLICATION

[0001] This application claims priority under 35 U.S.C. § 119(e) to U.S. Provisional Application No. 63 / 522,604, filed June 22, 2023, the content of which is incorporated by reference in its entirety for all purposes.FIELD

[0002] Disclosed embodiments are generally related to additive manufacturing systems and, more particularly, to build plan generation for additive manufacturing systems.BACKGROUND

[0003] Additive manufacturing systems employ various techniques to create three- dimensional objects from two-dimensional layers. After a layer of precursor material is deposited onto a build surface, a portion of the layer may be fused through exposure to one or more energy sources to create a desired two-dimensional geometry of solidified material within the layer according to a build plan. Next, the build surface may be indexed, and another layer of precursor material may be deposited. For example, in conventional systems, the build surface may be indexed downwardly by a distance corresponding to a thickness of a layer. This process may be repeated layer-by-layer to fuse many two-dimensional layers into a three-dimensional object.SUMMARY

[0004] In accordance with the disclosed subject matter, apparatus, systems, and methods are provided for build plan generation for additive manufacturing systems.

[0005] Some embodiments relate to an exemplary method for determining a configuration of an additive manufacturing system. The exemplary method comprises slicing a three-dimensional model of a part to be fabricated by the additive manufacturing system into at least one layer including a first layer; determining a surrounding volume ratio of a first volume of fused material surrounding one or more voxels and a total volume surrounding the one or more voxels for respective features of the first layer, the respective features including a first feature; identifying at least one portion of the first feature as a first subregion of the first layer based on determining that a surrounding volume ratio of the at least one portion satisfies a threshold; mapping at least the surrounding volume ratio to one or moreconfiguration parameters of the additive manufacturing system; and fabricating the first subregion of the first layer of the part using the one or more configuration parameters.

[0006] Some embodiments relate to an exemplary apparatus comprising instructions; at least one memory to store the instructions; and at least one processor. The exemplary apparatus comprises the at least one processor to execute the instructions to: slice a three- dimensional model of a part to be fabricated by an additive manufacturing system into at least one layer including a first layer; determine a surrounding volume ratio of a first volume of fused material surrounding one or more voxels and a total volume surrounding the one or more voxels for respective features of the first layer, the respective features including a first feature; identify at least one portion of the first feature as a first subregion of the first layer based on determining that a surrounding volume ratio of the at least one portion satisfies a threshold; identify one or more configuration parameters of the additive manufacturing system based on a mapping of at least the surrounding volume ratio to the one or more configuration parameters; and cause fabrication of the first subregion of the first layer of the part using the one or more configuration parameters.

[0007] Some embodiments relate to at least one non-transitory computer-readable medium comprising instructions that, when executed, cause at least one processor to execute the instructions to at least slice a three-dimensional model of a part to be fabricated by an additive manufacturing system into at least one layer including a first layer; determine a surrounding volume ratio of a first volume of fused material surrounding one or more voxels and a total volume surrounding the one or more voxels for respective features of the first layer, the respective features including a first feature; identify at least one portion of the first feature as a first subregion of the first layer based on determining that a surrounding volume ratio of the at least one portion satisfies a threshold; map at least the surrounding volume ratio to one or more configuration parameters of the additive manufacturing system; and cause fabrication of the first subregion of the first layer of the part using the one or more configuration parameters.

[0008] Some embodiments relate to an additive manufacturing system comprising a build surface; one or more laser energy sources; an optics assembly movable relative to the build surface and configured to direct laser energy from the one or more laser energy sources toward the build surface to melt at least a portion of a layer of material disposed on the build surface; and a controller. The exemplary system comprises the controller to slice a three-dimensional model of a part to be fabricated by the additive manufacturing system into at least one layer including a first layer; determine a surrounding volume ratio of a first volume of fused material surrounding one or more voxels and a total volume surrounding the one or more voxels for respective features of the first layer, the respective features including a first feature; identify at least one portion of the first feature as a first subregion of the first layer based on determining that a surrounding volume ratio of the at least one portion satisfies a threshold; map at least the first surrounding volume ratio to one or more configuration parameters of at least one of the one or more laser energy sources or the optics assembly; and cause the optics assembly to, based on the one or more configuration parameters, direct laser energy from the one or more laser energy sources towards the build surface to melt at least a portion of a layer of material disposed on the build surface to fabricate the first subregion of the first layer of the part.

[0009] It should be appreciated that the foregoing concepts, and additional concepts discussed below, may be arranged in any suitable combination, as the present disclosure is not limited in this respect. Further, other advantages and novel features of the present disclosure will become apparent from the following detailed description of various nonlimiting embodiments when considered in conjunction with the accompanying figures.BRIEF DESCRIPTION OF DRAWINGS

[0010] The accompanying drawings are not intended to be drawn to scale. In the drawings, each identical or nearly identical component that is illustrated in various figures may be represented by a like numeral. For purposes of clarity, not every component may be labeled in every drawing. In the drawings:

[0011] FIG. 1 is a schematic representation of an exemplary additive manufacturing system, according to some embodiments;

[0012] FIG. 2 depicts another exemplary additive manufacturing system, according to some embodiments;

[0013] FIG. 3 is a block diagram of an exemplary build plan generation system including a process controller, according to some embodiments;

[0014] FIG. 4 depicts an exemplary workflow for generating a build plan, according to some embodiments;

[0015] FIG. 5 depicts an exemplary workflow for identifying subregions of a three- dimensional part layer, according to some embodiments;

[0016] FIG. 6 is a table of exemplary surrounding volume ratio thresholds and corresponding configuration parameters, according to some embodiments;

[0017] FIG. 7 is a flowchart representative of example machine-readable instructions that may be executed by processor circuitry to implement the exemplary process controller of FIG. 3 to configure an additive manufacturing system based on subregion configurations, according to some embodiments;

[0018] FIG. 8 is a flowchart representative of example machine-readable instructions that may be executed by processor circuitry to implement the exemplary process controller of FIG. 3 to generate data associations of surrounding volume ratios and corresponding additive manufacturing system configurations, according to some embodiments;

[0019] FIG. 9 is a flowchart representative of example machine-readable instructions that may be executed by processor circuitry to implement the exemplary process controller of FIG. 3 to configure an additive manufacturing system using configurations determined by a machine-learning model, according to some embodiments; and

[0020] FIG. 10 is an example electronic platform structured to execute the machine- readable instructions of FIGS. 7, 8, and / or 9 to implement the exemplary process controller of FIG. 3 and / or, more generally, the exemplary additive manufacturing system of FIG. 3, according to some embodiments.DETAILED DESCRIPTION

[0021] Additive manufacturing (AM) techniques, such as laser bed powder fusion (LBPF), are being increasingly used in various applications, such as aerospace, energy, and medical applications. For example, LBPF can provide advantages such as producing arbitrarily complex geometries, single-step manufacture, weight saving, and / or the like.

[0022] In some embodiments, LBPF may include arranging incident laser spots on a build surface in a line with a long dimension and a short dimension, or in an array. In either case, according to some aspects, a line, or array, of incident laser energy consists of multiple individual laser energy pixels arranged adjacent to each other that can have their respective power levels individually controlled. Each laser energy pixel may be turned on or turned off independently and the power of each pixel can be independently controlled. The resultingpixel-based line or array may then be scanned across a build surface to form a desired pattern thereon by controlling the individual pixels during translation of the optics assembly.

[0023] Depending on the particular embodiment, an additive manufacturing system according to the current disclosure may include any suitable number of laser energy sources. For example, in some embodiments, the number of laser energy sources may be at least 5, at least 10, at least 50, at least 100, at least 500, at least 1,000, at least 1,500, or more. In some embodiments, the number of laser energy sources may be less than 2,000, less than 1,500, less than 1,000, less than 500, less than 100, less than 50, or less than 10. Additionally, combinations of the above-noted ranges may be suitable. Ranges both greater and less than those noted above are also contemplated as the disclosure is not so limited.

[0024] Additionally, in some embodiments, a power output of a laser energy source (e.g., a laser energy source of a plurality of laser energy sources) may be between about 50 watts (W) and about 2,000 W (2 kW). For example, the power output for each laser energy source may be between about 100 W and about 1.5 kW, and / or between about 500 W and about 1 kW. Moreover, a total power output of the plurality of laser energy sources may be between about 500 W (0.5 kW) and about 4,000 kW. For example, the total power output may be between about 1 kW and about 2,000 kW, and / or between about 100 kW and about 1,000 kW. Ranges both greater and less than those noted above are also contemplated as the disclosure is not so limited.

[0025] Depending on the embodiment, an array of laser energy pixels (e.g., a line array or a two dimensional array) may have a uniform power density along one or more axes of the array including, for example, along the length dimension (i.e. the longer dimension) of a line array. In other instances, an array can have a non-uniform power density along either of the axes of the array by setting different power output levels for each pixel’s associated laser energy source. Moreover, individual pixels on the exterior portions of the array can be selectively turned off or on to produce an array with a shorter length and / or width. In some embodiments, the power levels of the various pixels in an array of laser energy may be independently controlled throughout an additive manufacturing process. For example, the various pixels may be selectively turned off, on, or operated at an intermediate power level to provide a desired power density within different portions of the array.

[0026] Generally, laser energy produced by a laser energy source has a power area density. In some embodiments, the power area density of the laser energy transmitted throughan optical fiber is greater than or equal to 0.1 W / micrometer2, greater than or equal to 0.2 W / micrometer2, greater than or equal to 0.5 W / micrometer2, greater than or equal to 1 W / micrometer2, greater than or equal to 1.5 W / micrometer2, greater than or equal to 2 W / micrometer2, or greater. In some embodiments, the power area density of the laser energy transmitted through the optical fiber is less than or equal to 3 W / micrometer2, less than or equal to 2 W / micrometer2, less than or equal to 1.5 W / micrometer2, less than or equal to 1 W / micrometer2, less than or equal to 0.5 W / micrometer2, less than or equal to 0.2 W / micrometer2, or less. Combinations of these ranges are possible. For example, in some embodiments, the power area density of the laser energy transmitted through the optical fiber is greater than or equal to 0.1 W / micrometer2and less than or equal to 3 W / micrometer2.

[0027] Depending on the application, output of the optics assembly may be scanned across a build surface of an additive manufacturing system in any appropriate fashion. For example, in one embodiment, one or more galvo scanners may be associated with one or more laser energy sources to scan the resulting one or more laser pixels across the build surface. Alternatively, in other embodiments, an optics assembly may include an optics head that is associated with one or more appropriate actuators configured to translate the optics head in a direction parallel to a plane of the build surface to scan the one or more laser pixels across the build surface. In either case, it should be understood that the disclosed systems and methods are not limited to any particular construction for scanning the laser energy across a build surface of the additive manufacturing system.

[0028] For the sake of clarity, transmission of laser energy through an optical fiber is described generically throughout. However, with respect to various parameters such as transverse cross-sectional area, transverse dimension, transmission area, power area density, and / or any other appropriate parameters related to a portion of an optical fiber that the laser energy is transmitted through, it should be understood that these parameters refer to either a parameter related to a bare optical fiber and / or a portion of an optical fiber that the laser energy is actively transmitted through such as an optical fiber core, or a secondary optical laser energy transmitting cladding surrounding the core. In contrast, any surrounding cladding, coatings, or other materials that do not actively transmit the laser energy may not be included in the disclosed ranges.

[0029] In some embodiments, LBPF may include selectively melting a metal powder bed by an energy heat source, such as a laser beam, in a layer-by-layer manner based on adesigned scanning strategy to produce final parts. A designed scanning strategy, which may be referred to as a build plan, or portion(s) thereof, may include one or more parameters. Parameters herein may also be referred to as build plan parameters, LBPF parameters, LBPF process parameters, process parameters, or processing parameters and are interchangeable. Non-limiting examples of a parameter may be information including locations of a plurality of laser energy pixels on a build surface (e.g., the metal powder bed) during the build process, trajectories of the laser energy pixels across the build surface, a list of operating parameters for the laser energy pixels, expected intensity values of laser energy pixels projected onto the build surface, expected radiation emissions and / or reflections of welds and / or melt pools (e.g., an expected weld signature), and other appropriate parameters. These commanded parameters for each laser energy pixel may then be compared to the corresponding parameters in a captured image for the corresponding laser energy pixel of the plurality of laser energy pixels to identify an operating state of each laser energy pixel that is currently being operated. Build plans may alternatively or additionally include parameters such as, for one or more lasers of the additive manufacturing system, at least one of a pulse-width modulation (PWM) configuration, a laser power, a scanning speed, a spot size, a hatch spacing, or one or more sampling parameters of the one or more lasers. For example, the PWM configuration may include at least one of a frequency or a pulse duration of the one or more lasers. In some embodiments, build plans may include parameters associated with a sampling of a layer for a part in-situ. For example, parameters may include a sampling width associated with a scan axis, a sampling width associated with an indexing axis, a sampling depth, and / or a sampling resolution.

[0030] As used herein, the term “build plan” and other similar terms may refer to the planned paths, e.g., trajectories, traced by the energy sources in the additive manufacturing process, including the thickness of each build layer, to form the plurality of build layers of one or more parts on a build plate. Build plans may alternatively or additionally include other information such as a list of functional laser energy sources, intensities at which the energy sources may operate at different locations along the build surface in each layer, a velocity of the optics assembly and laser energy pixels relative to a build surfaces, locations of laser energy pixel on the build surface during a build process, and other appropriate processing parameters. The paths traced by the one or more laser energy sources and / or laser energy pixels, referenced as trajectories, may be determined prior to the start of the manufacturingprocess and / or may be generated or updated in real time in instances where real time planning is implemented for an additive manufacturing process. In some embodiments, processing parameters may be associated with individual layers of the build plan, such as energy source intensity, and may be adjusted during a build process. For example, a build plan may include per-layer build information which may be used to monitor the status of a part. Per-layer build information may include an expected build pattern during and / or after each layer is printed, expected dimensions of the part after each layer is printed, and any other applicable build information. Furthermore, a build plan may include parameters representing specific energy source power settings, weld or formation speed, recoating rates, expected laser energy pixel shapes and / or sizes, expected weld pool shapes and / or sizes, and / or other process related information. The various processing parameters, including power settings and weld or formation speed, may be actively changed during a build based on the measured information associated with the different laser energy pixels. In some embodiments, processing parameters, such as the expected locations of the plurality of laser energy pixels, may be obtained independently from a weld pattern and / or build plan. Additionally, some processing parameters, such as the commanded locations of the laser energy pixels on the build surface, may be calculated from other information present in the build plan, including laser energy source trajectories, operation timing, timing, and any other applicable information.

[0031] The build plan may be obtained from a build plan module configured to generate the build plan, or the build plan may be uploaded to the system, recalled from non- transitory computer readable memory, and / or obtained in any other appropriate manner. In some embodiments, the build plan may include processing parameters about the build process such as trajectory, timing, power, velocity, travel path, dimensions of the part and / or build surface, weld pattern, and / or expected locations and / or timing of the plurality of laser energy pixels firing on the build surface. In other embodiments, a build plan may not include an expected weld pattern and / or locations of the laser energy pixels, but may include other processing parameters which can be used to obtain expected locations of the plurality of laser energy pixels and / or an expected weld pattern. For example, from the timing of the firing of the plurality of laser energy sources, velocity of the optics assembly, and direction of travel, an expected weld pattern may be determined. Furthermore, in some embodiments, the expected weld pattern and / or expected locations of the plurality of laser energy pixels versus time may be obtained independently from a build plan. For example, the expected weldpattern and / or expected locations of the plurality of laser energy pixels may be directly imported and / or inputted into the AM system.

[0032] One challenge of determining parameters, such as LBPF parameters, in conventional AM systems is that they may be developed and assigned independently and / or otherwise separate from part feature attributes of a part be fabricated. For example, a conventional technique of determining LBPF parameters may be performed independently of part and / or layer geometries and response to process thermal input and residual stress. Such a conventional parameter determining technique creates challenges during LBPF process development and optimization and, in particular, for challenging features such as lattice structures, overhangs, thin walls, and through-holes. For example, an LBPF parameter that is advantageous for a first feature, such as a solid structure, may be disadvantageous for a second feature, such as lattice structure. In some such cases, determining the LBPF parameter independently of the feature may result in suboptimal and / or unusable parts.

[0033] Another challenge of determining parameters, such as LBPF parameters, in conventional AM systems is that they are typically determined through extensive experimentation. For example, due to the lack of principles and / or relationships to connect LBPF parameters to the physical behavior and / or response of printed features during LBPF process in-situ, substantial efforts are undertaken to design and execute experimental trials and finalize the process development for AM parts. Additionally, knowledge and learnings from such experimental trials obtained from the process development of one AM part does not easily transfer to the process development of a different AM part. For example, if the geometries of a previously processed part and a new part not yet processed are different from each other, then the knowledge of learnings from the previously processed part may not transfer to the new part.

[0034] The inventors have recognized and appreciated the aforementioned challenges of determining parameters in AM systems. The inventors have developed technology to determine parameters for optimized and / or otherwise improved process development of a part, or layer(s) thereof, based on relationships between the parameters and the physical behavior and response of printed features during LBPF in-situ. The inventors have developed such technology to at least overcome the aforementioned challenges.

[0035] In some embodiments, LBPF parameters can be assigned and / or identified based on a database of subregions and parameters developed for each type of subregion. Insome embodiments, a subregion is a feature, or portion(s) thereof, of a sliced layer of a three- dimensional (3-D) model for a part to be fabricated by an AM system. Non-limiting examples of a feature include a lattice structure, a solid structure, an overhand, a through-hole, and a thin wall. Any other feature is contemplated by the inventors.

[0036] In some embodiments, an AM system, or portion(s) thereof such as at least a process controller, can determine a characteristic of a feature. Non-limiting examples of feature characteristics include an angle with respect to a reference axis, a material composition (e.g., a type of powder), a volume of fused material that surrounds a voxel, a volume of unfused material that surrounds a voxel, a thickness, a thermal conductivity, an ultimate tensile strength, and a yield strength. Any other feature characteristic is contemplated by the inventors.

[0037] By way of example, the process controller can characterize a surrounding area, locality, neighborhood, vicinity, volume, etc., of one or more voxels that is available for conductance of heat away from a fused weld during formation in an AM process. For example, the process controller can determine whether a volume that surrounds a voxel, or portion(s) thereof, is fused material, unfused material, or a combination thereof. In some embodiments, a surrounding volume of a voxel that is fused material can have a relatively high thermal conductance. In some embodiments, a surrounding volume of a voxel that is unfused material can have a relatively low thermal conductance. In some embodiments, a “voxel” refers to a representation of a value on a regular grid in three-dimensional space. For example, a voxel can be a three-dimensional pixel (e.g., a three-dimensional computer graphics pixel). In some embodiments, a surrounding volume of a voxel can correspond to a volume of one or more layers associated with the voxel, such as a first layer below the voxel and / or a second layer above the voxel.

[0038] In some embodiments, the process controller can identify a voxel associated with a feature of a layer to analyze and / or process. The process controller can calculate and / or determine a ratio of a first volume of fused material surrounding the voxel and a total volume surrounding the voxel. For example, the process controller can express a volume of fused material that surrounds the voxel as a ratio where a ratio of 1 can represent fused material fully surrounding the voxel (e.g., the surrounding volume of the voxel is 100% fused material or 0% unfused material) and a ratio of 0 can represent no fused material surrounding the voxel (e.g., the surrounding volume of the voxel is 0% fused material or 100% unfusedmaterial). Alternatively, the surrounding volume ratio may be represented by any other range (e.g., 0 to 10, 1 to 10, 0 to 100, etc.). In some embodiments, a ratio of 1 can represent and / or correspond to a relatively high thermal conductance for the feature of the layer. For example, an increase in the surrounding volume ratio can correspond to an increase in the thermal conductance. In some embodiments, a ratio of 0 can represent and / or correspond to a relatively low thermal conductance for the feature of the layer. For example, a decrease in the surrounding volume ratio can correspond to a decrease in the thermal conductance. As used herein, the ratio can be referred to as a surrounding volume ratio, a surrounding volume parameter, a fused material surrounding volume ratio, a fused material surrounding volume parameter, or the like and are interchangeable terms. Additionally or alternatively, the process controller may calculate and / or determine a different ratio of a second volume of unfused material surrounding the voxel and the total surrounding volume of the voxel.

[0039] In some embodiments, the process controller can determine whether the ratio satisfies one or more thresholds, such as ratio thresholds (e.g., surrounding volume ratio thresholds, fused material surrounding volume ratio thresholds, etc.). For example, the process controller can determine that the ratio of a first portion of a feature is 0.15, the ratio of 0.15 is greater than a first ratio threshold of 0.1, and, thus, the ratio of 0.15 satisfies the first ratio threshold. In some embodiments, the process controller can determine that the ratio of 0.15 is less than a second ratio threshold of 0.2 and, thus, does not satisfy the second ratio threshold. The process controller can thusly determine that the ratio is within a ratio threshold band of 0.1-0.2. In some embodiments, the process controller can identify the first feature portion as a first subregion based on the ratio being within the ratio threshold band.

[0040] In some embodiments, the process controller can estimate and / or determine a thermal conductivity of a feature based on the ratio. For example, the process controller can determine that the ratio of a first portion of a feature is 0.15 and determine that a thermal conductivity of the first portion is 0.25 based on the ratio of 0.15. In some embodiments, the relationship between at least one of the ratio, the thermal conductivity, or a feature type (e.g., a thin wall structure, a lattice structure, etc.) can be empirically and / or experimentally derived. For example, the process controller can map the ratio to a plurality of relationships to determine the thermal conductivity.

[0041] In some embodiments, the process controller can determine whether the thermal conductivity satisfies one or more thresholds, such as thermal conductivitythresholds. For example, the process controller can determine that the ratio of a first portion of a feature is 0.15 and the thermal conductivity of the first portion is 0.25 based on the ratio of 0.15. The process controller can determine that the thermal conductivity of 0.25 is greater than a first thermal conductivity threshold of 0.2, and, thus, the thermal conductivity of 0.25 satisfies the first thermal conductivity threshold. In some embodiments, the process controller can determine that the thermal conductivity of 0.25 is less than a second thermal conductivity threshold of 0.3 and, thus, does not satisfy the second thermal conductivity threshold. The process controller can thusly determine that the thermal conductivity is within a thermal conductivity threshold band of 0.2-0.3. In some embodiments, the process controller can identify the first feature portion as a first subregion based on the thermal conductivity being within the thermal conductivity threshold band.

[0042] In some embodiments, the process controller can query the database of subregions for corresponding parameter(s) that correspond to the ratio threshold band. For example, the parameters in the database after process development can be tagged and / or otherwise associated with one or more subregions. In some embodiments, the parameters in the database to be assigned to a subregion can be related and / or dependent to each other to reflect the physical natures of the subregion and feature. For example, subregions can be defined at least in part based on a surrounding volume ratio and, therefore, part geometries.

[0043] In some embodiments, the parameter(s) in the database is / are previously determined from the fabrication of prior parts with feature portion(s) similar to or, in some embodiments, the same as the first feature portion. In some embodiments, the parameter(s) is / are determined by executing a machine-learning model trained using 3-D models of parts and corresponding LBPF parameters. For example, the machine-learning model can be executed using the first feature portion as input(s) (e.g., model input(s)) to generate output(s) (e.g., model output(s)), which can include the parameter(s).

[0044] Beneficially, by mapping at least a feature characteristic to parameter(s), such as LBPF parameter(s), cost (e.g., computational cost, monetary cost), complexity, material (e.g., powder bed material), and / or time (e.g., manpower time, AM system operational time) can be substantially reduced with respect to process development of a part to be fabricated. For example, iterative experimentation to identify LBPF parameter(s) for a particular part, or layer(s) thereof, can be eliminated and / or otherwise reduced by analyzing and / or evaluating the relationship(s) between the LBPF parameter(s) and feature characteristics. Beneficially,exemplary techniques disclosed herein are scalable and applicable to any geometry (e.g., feature geometry, part geometry) by identifying LB PF parameter(s) based at least in part on feature characteristics.

[0045] It will be appreciated that any embodiments of the systems, components, methods, and / or programs disclosed herein, or any portion(s) thereof, may be used to form any part suitable for production using additive manufacturing. For example, a method for additively manufacturing one or more parts may, in addition to any other method steps disclosed herein, include the steps of selectively fusing one or more portions of a plurality of layers of precursor material deposited onto the build surface to form the one or more parts. This may be performed in a sequential manner where each layer of precursor material is deposited on the build surface and selected portions of the upper most layer of precursor material is fused to form the individual layers of the one or more parts. This process may be continued until the one or more parts are fully formed.

[0046] Turning to the figures, specific non-limiting embodiments are described in further detail. It should be understood that the various systems, components, features, and methods described relative to these embodiments may be used either individually and / or in any desired combination as the disclosure is not limited to only the specific embodiments described herein.

[0047] FIG. 1 shows, according to some embodiments, a schematic representation of an additive manufacturing system 100, including a plurality of laser energy sources 102 that deliver laser energy to an optics assembly 104 positioned within a machine enclosure 106. For example, the machine enclosure 106 may define a build volume in which an additive manufacturing process may be carried out. In particular, the optics assembly 104 may direct laser energy 108 towards a build surface 110 positioned within the machine enclosure 106 to selectively fuse powdered material on the build surface 110. As described in more detail below, the optics assembly 104 may include a plurality of optics defining an optical path within the optics assembly that may transform, shape, and / or direct laser energy within the optics assembly 104 such that the laser energy 108 is directed onto the build surface 110 as an array of laser energy pixels. In some embodiments, the optics assembly 104 may be movable within the machine enclosure 106 to scan the laser energy 108 across the build surface 110 during a manufacturing process. For example, the optics assembly 104 may be associated with appropriate actuators, rails, motors, and / or any other appropriate structurecapable of optics assembly relative to the surface. Alternatively, embodiments in which the optics assembly 104 includes galvomirrors or other appropriate components that are configured to scan the laser energy 108 across the build surface 110 while the optics assembly 104 is held stationary relative to the build surface 110 are also contemplated.

[0048] In some embodiments, the additive manufacturing system 100 further includes one or more optical fiber connectors 112 positioned between the laser energy sources 102 and the optics assembly 104. As illustrated, a first plurality of optical fibers 114 may extend between the plurality of laser energy sources 102 and the optical fiber connector 112. In particular, each laser energy source 102 may be coupled to the optical fiber connector 112 via a respective optical fiber 116 of the first plurality of optical fibers 114. Similarly, a second plurality of optical fibers 118 extends between the optical fiber connector 112 and the optics assembly 104. Each optical fiber 116 of the first plurality of optical fibers 114 is coupled to a corresponding optical fiber 120 of the second plurality of optical fibers 118 within the optical fiber connector 112. In this manner, laser energy from each of the laser energy sources 102 is delivered to the optics assembly 104 such that the laser energy 108 can be directed onto the build surface 110 during an additive manufacturing process (i.e., a build process). Of course other methods of connecting the laser energy sources 102 due to the optics assembly 104 are also contemplated.

[0049] FIG. 2 depicts one embodiment of an additive manufacturing system 200 at the beginning of a build process. The additive manufacturing system 200 includes a build plate 202 mounted on a fixed plate 204, which is in turn mounted on one or more vertical supports 206 that attach to a base 208 of the additive manufacturing system 200. In the depicted embodiment, the one or more vertical supports 206 may correspond to one, two, and / or any other appropriate number of supports configured to support the build plate 202, and the corresponding build surface, at a desired position and orientation. For example, the supports depicted in FIG. 2 may correspond to one or more vertical motion stages configured to control a vertical position and orientation of the build plate 202. A powder containment shroud 210 may at least partially, and in some embodiments completely, surround a perimeter of the build plate 202 to support a volume of precursor material 202a, such as a volume of powder, disposed on the build plate 202 and contained within the shroud 210. The shroud 210 may be supported on the base 208 or by any other appropriate portion of the additive manufacturing system 200.

[0050] The additive manufacturing system 200 may include a powder deposition system in the form of a recoater 212 that is mounted on a horizontal motion stage 214 that allows the recoater 212 to be moved back and forth across either a portion, or entire, surface of the build plate 202. As the recoater 212 traversers the build surface of the build plate 202, it deposits the precursor material 202a, such as a powder, onto the build plate 202 and smooths the surface to provide a layer of the precursor material 202a with a predetermined thickness on top of the underlying volume of fused and / or unfused precursor material deposited during prior formation steps.

[0051] In some embodiments, the supports 206 of the build plate 202 may be used to index the build surface of the build plate 202 in a vertical downwards direction relative to a local direction of gravity. In such an embodiment, the recoater 212 may be held vertically stationary for dispensing the precursor material 202a, such as a precursor powder, onto the exposed build surface of the build plate 202 as the recoater 212 is moved across the build plate 202 each time the build plate 202 is indexed downwards.

[0052] In some embodiments, the additive manufacturing system 200 may also include an optics assembly 218 that is supported vertically above and oriented towards the build plate 202. As detailed above, the optics assembly 218 may be optically coupled to one or more laser energy sources, not depicted, to direct laser energy in the form or one or more laser energy pixels onto the build surface of the build plate 202. To facilitate movement of the laser energy pixels across the build surface, the optics assembly 218 may be configured to move in one, two, or any number of directions in a plane parallel to the build surface of the build plate 202. To provide this functionality, the optics assembly 218 may be mounted on a gantry 220, or other actuated structure, which allows the optics assembly 218 to be scanned in plane parallel to the build surface of the build plate 202.

[0053] In the above embodiment, the build plate 202 is indexed vertically while the remaining active portions of the additive manufacturing system 200 are held vertically stationary. However, embodiments, in which the build plate 202 is held vertically stationary and the shroud 210, recoater 212, and optics assembly 218 are indexed vertically upwards relative to a local direction of gravity during formation of successive layers are also contemplated. In such an embodiment, the recoater horizontal motion stage 214 may be supported by vertical motion stages 216 that are configured to provide vertical movement of the recoater 212 relative to the build plate 202. Corresponding vertical motion stages mayalso be provided for the shroud 210, not depicted, to index the shroud 210 vertically upward relative to the build plate 202 in such an embodiment. In some embodiments, the additive manufacturing system 200 may also include an optics assembly 218 that is supported on a vertical motion stage 222 that is in turn mounted on the gantry 220 that allows the optics assembly 218 to be scanned in the plane of the build plate 202.

[0054] In the above embodiment, the vertical motion stages 216, horizontal motion stages 214, and gantry 220 may correspond to any appropriate type of system that is configured to provide the desired vertical and / or horizontal motion. This may include supporting structures such as: rails; linear bearings, wheels, threaded shafts, and / or any other appropriate structure capable of supporting the various components during the desired movement. Movement of the components may also be provided using any appropriate type of actuator including, but not limited to, electric motors, stepper motors, hydraulic actuators, pneumatic actuators, electric actuators, and / or any other appropriate type of actuator as the disclosure is not so limited.

[0055] In addition to the above, in some embodiments, the additive manufacturing system 200 may include one or more controllers 224 that is operatively coupled to the various actively controlled components of the additive manufacturing system 200. For example, the one or more controllers 224 may be operatively coupled to the one or more supports 206, the recoater 212, the optics assembly 218, the various motion stages 214, 216, and / or any other appropriate component of the additive manufacturing system 200. In some embodiments, the one or more controllers 224 may include one or more processors and associated non- transitory computer readable memory. The non-transitory computer readable memory may include and / or store processor executable instructions that, when executed by the one or more processors, cause the additive manufacturing system 200 to perform at least any of the methods disclosed herein.

[0056] FIG. 3 is a block diagram of an exemplary build plan generation system 300 including a process controller 302, according to some embodiments. The process controller 302, and / or, more generally, the build plan generation system 300, can configure and / or determine a configuration of an additive manufacturing system 304 based at least in part on characteristic(s) of feature(s) of layer(s) of a component, part, structure, etc., to be fabricated.

[0057] In some embodiments, the additive manufacturing system 304 of the illustrated example can be implemented by additive manufacturing system 100 of FIG. 1and / or the additive manufacturing system 200 of FIG. 2. For example, the additive manufacturing system 304 can carry out and / or perform additive manufacturing techniques, such as laser bed powder fusion (LBPF). The additive manufacturing system 304 of the illustrated example includes a vision system 306, an optics assembly 308, a part 310, and a build plate 312. Other additive manufacturing system components, such as those depicted in the examples of FIGS. 1 and / or 2, are contemplated but are not shown for clarity. In some embodiments, the optics assembly 308 can be implemented by the optics assembly 104 of FIG. 1 and / or the optics assembly 218 of FIG. 2.

[0058] In some embodiments, the vision system 306 can include one or more photosensitive detectors. Non-limiting examples of photosensitive detectors include cameras, photodiodes, and spectrometers. Non-limiting examples of cameras include infrared cameras and high frame rate (HFR) cameras. For example, the vision system 306 can include one or more infrared cameras or other suitable sensor(s) that can capture thermal responses associated with the build plate 312, such as a powder bed disposed on the build plate 312 to form a build surface. Additionally or alternatively, the vision system 306 may include one or more HFR cameras and / or high speed cameras. In the illustrated example, the vision system 306 is coupled (e.g., mounted) to the optics assembly 308. For example, the vision system 306 can be mounted to the optics assembly 308 to provide a fixed offset between a position of the optics assembly 308 and the position of the vision system 306. Similarly, a target, such as the part 310 or portion(s) thereof, is positioned at a fixed and known offset distance relative to the vision system 306, such that upon determining a position of a laser energy pixel, the additive manufacturing system 304 may also determine an offset distance between the target and the laser energy pixels of the optics assembly 308. During an additive manufacturing process, the vision system 306 may identify one or more target features on the build surface on which one or more features may be formed (e.g., via a laser bed fusion process), and the additive manufacturing system 304 may use these two known offset distances to move the optics assembly 308 such that a laser array (or a desired subset of laser energy pixels of the laser array) is projected onto the target feature identified by the vision system 306. In some embodiments, the vision system 306 can identify features on the build surface (e.g., shapes such as circular shapes, polygonal shapes, irregular shapes, overhangs, through holes, walls (e.g., thin walls), solid structures, lattice structures, etc.) andautomatically move the optics assembly 308 such that a desired pattern of laser energy is projected onto the features.

[0059] In some embodiments, the additive manufacturing system 304 can be configured to construct and / or fabricate the part 310, or layer(s) thereof, using parameters, such as LBPF parameters. For example, the process controller 302 can generate instructions 314 representative of LBPF parameters and provide the instructions 314 to the additive manufacturing system 304 to cause the additive manufacturing system 304 to fabricate the part 310 according to the LBPF parameters.

[0060] In some embodiments, the process controller 302 can determine the LBPF parameters based at least in part on feature(s) of the part 310 in-situ or feature(s) in a three- dimensional (3-D) model of the part 310. For example, the process controller 302 can obtain and / or receive a 3-D part file 316 from a server 318 via a network 320. The 3-D part file 316 can be a 3-D model of the part 310, or portion(s) thereof, such as one or more layers to be fabricated to form the part 310. In some embodiments, the 3-D part file 316 is a computer aided design (CAD) model, such as a solid model, a surface model, or a wireframe model. The 3-D part file 316 can be a file in any format. Non-limiting examples of file formats include STEP, IGES, 3D PDF, JT, STL, ACIS, PARASOLID, and QIF.

[0061] The server 318 of the illustrated example can be a computer and / or electronic server that can store data, such as the 3-D part file 316, for access via the network 320. For example, the server 318 can be a blade server or a rack-mounted server. Alternatively, the server 318 may be implemented by a cellular network device (e.g., an Internet-enabled smartphone, a telecommunications server), a desktop computer, a laptop computer, a workstation, or any other type of computing and / or electronic device.

[0062] The network 320 of the illustrated example can be implemented by any wired and / or wireless network(s) such as one or more cloud networks (e.g., one or more private cloud networks, one or more public cloud networks), one or more cellular networks (e.g., 4G LTE cellular networks, 5G cellular networks, 6G cellular networks, etc.), one or more data buses, one or more edge networks, one or more local area networks (LANs), one or more optical fiber networks, one or more private networks, one or more public networks, one or more wireless local area networks (WLANs), etc., and / or any combination(s) thereof. For example, the network 320 can be the Internet, but any other type of private and / or public network is contemplated.

[0063] The process controller 302 of the illustrated example includes and / or implements a data interface module 330, a feature extraction module 332, a feature characteristic determination module 334, a subregion identification module 336, a configuration determination module 338, a relationship determination module 340, a machine learning (ML) module 342, and a first datastore 344 (identified by DATASTORE A). The first datastore 344 of this example includes, implements, and / or stores thresholds 346 and an ML model 348. Component(s) of the process controller 302 can be in communication with one(s) of each other via a bus 350. In some embodiments, the bus 350 can be any type of computing and / or electrical bus, such as an Inter- Integrated Circuit (I2C) bus, a Peripheral Component Interconnect (PCI) bus, a Peripheral Component Interconnect Express (PCIe) bus, a Serial Peripheral Interface (SPI) bus, and / or the like.

[0064] In some embodiments, the process controller 302 includes the data interface module 330 to transmit data. For example, the data interface module 330 can transmit data to a system, such as by transmitting the instructions 314 to the additive manufacturing system 304. In some embodiments, the data interface module 330 can transmit data to a datastore, such as transmitting a first subregion identifier 354 (identified by SUBREGION IDENTIFIER-1 (SL1)) to a second datastore 356 (identified by DATASTORE B).

[0065] In some embodiments, the data interface module 330 can receive data. For example, the data interface module 330 can receive data via a network, such as by receiving the 3-D part file 316 from the server 318 via the network 320. In some embodiments, the data interface module 330 can receive data from a system, such as by receiving image data 358 from the vision system 306 of the additive manufacturing system 304. In some embodiments, the data interface module 330 can receive data from a datastore, such as by receiving a first configuration 360 (identified by CONFIGURATION- 1 (CONFIG- 1)) from the second datastore 356.

[0066] In some embodiments, the process controller 302 includes the feature extraction module 332 to extract and / or identify a feature associated with a part, such as the part 310. For example, the feature extraction module 332 can extract and / or identify a feature associated with the part 310 such as a lattice structure, a solid structure, an overhand, a through-hole, and / or a thin wall.

[0067] By way of example, the 3-D part file 316 can include and / or implement a 3-D model of the part 310. In some embodiments, the feature extraction module 332 can processthe 3-D part file 316 by slicing a the 3-D model of the part 310 into one or more layers. The feature extraction module 332 can select a layer of the one or more layers to process. For example, the feature extraction module 332 can identify a feature of the layer and / or portion(s) of the feature. In some embodiments, the feature extraction module 332 can identify the feature as a lattice structure and / or a portion of the lattice structure.

[0068] In some embodiments, the feature extraction module 332 can extract and / or identify a feature in the image data 358 that corresponds to the part 310. For example, the vision system 306 can capture an image of a layer of the part 310, which can be an image of a portion of the build surface of the build plate 312. The vision system 306 can process and / or package the image, or portion(s) thereof, into the image data 358 for transmission to the data interface module 330. The feature extraction module 332 can obtain the image data 358 from the data interface module 330. The feature extraction module 332 can identify a feature, such as a wall (e.g., a thin wall), in the image data 358 for subsequent processing.

[0069] In some embodiments, the process controller 302 includes the feature characteristic determination module 334 to calculate, determine, and / or identify a characteristic of a feature. For example, the feature characteristic determination module 334 can determine a characteristic of a feature such as an angle with respect to a reference axis, a material composition (e.g., a type of powder), a surrounding volume of fused material, a surrounding volume of unfused material, a thickness, a thermal conductivity, an ultimate tensile strength, and / or a yield strength.

[0070] In some embodiments, the feature characteristic determination module 334 can determine that a feature has a range of characteristics. For example, the feature characteristic determination module 334 can determine that a thin wall feature spans several portions of the layer and / or, more generally, the build surface. In some embodiments, the feature characteristic determination module 334 can determine that a first portion of the thin wall feature has a first surrounding volume ratio, a second portion of the thin wall feature has a second surrounding volume ratio different from the first surrounding volume ratio, etc. Alternatively, in some embodiments, the feature characteristic determination module 334 may determine that a feature has a uniform characteristic, such as a thin wall feature having the same or substantially the same surrounding volume ratio throughout the layer.

[0071] In some embodiments, the feature characteristic determination module 334 can determine that a feature has a first characteristic based on having a second characteristic. Forexample, the feature characteristic determination module 334 can determine that a thin wall feature has a thermal conductivity of 0.4 based on the thin wall having a surrounding volume ratio of 0.3, such as 30% of the surrounding volume of the thin wall being fused material.

[0072] In some embodiments, the feature characteristic determination module 334 can determine a feature characteristic based on a comparison of a feature to a baseline feature. For example, the feature characteristic determination module 334 can calculate a baseline surrounding volume ratio for a solid structure. In some embodiments, the feature characteristic determination module 334 can identify a feature as a lattice structure. The feature characteristic determination module 334 can compare the lattice structure to the solid structure and determine, at least based on the comparison, that a surrounding volume ratio of the lattice structure is a portion of the baseline surrounding volume ratio. For example, the feature characteristic determination module 334 can determine that the surrounding volume ratio of the lattice structure is a percentage (e.g., 20%, 50%, 80%, etc.) of the baseline surrounding volume ratio. Beneficially, by determining feature characteristics for a baseline feature, such as a solid structure (but any other baseline feature is contemplated), feature characteristics of new features can be accurately and / or quickly ascertained by comparing the new features to at least the baseline feature and, thusly, may reduce iteration and / or experimentation of conventional LB PF parameter determination techniques.

[0073] In some embodiments, the process controller 302 includes the subregion identification module 336 to identify a subregion of a layer of the part 310 based at least in part on a feature characteristic and / or, more generally, a feature of the part 310. For example, the subregion identification module 336 can compare feature characteristic(s) of feature(s) (and / or feature portion(s)) to one of the thresholds 346, such as a feature characteristic threshold. In some embodiments, the subregion identification module 336 can identify a first portion of a layer, which can include a first feature portion, as a first subregion based on a determination that a corresponding feature characteristic satisfies one or more of the thresholds 346.

[0074] By way of example, the subregion identification module 336 can determine that a first portion of a feature on a layer has a first surrounding volume ratio of 0.25, which is greater than a first surrounding volume ratio threshold of 0.2 but less than a second surrounding volume ratio threshold of 0.3. The subregion identification module 336 can determine that a second portion of the feature on the layer has a second surrounding volumeratio of 0.35, which is greater than the second surrounding volume ratio threshold of 0.3 but less than a third surrounding volume ratio threshold of 0.4. In some embodiments, the subregion identification module 336 can generate one or more contours (e.g., feature characteristic contours, visualization contours), which can be two-dimensional (2-D) contours along a surface of a sliced layer to represent range(s) of feature characteristics, such as the first and second surrounding volume ratios. For example, the subregion identification module 336 can generate a first contour along which the first feature portion has the first surrounding volume ratio, a second contour along which the second feature portion has the second surrounding volume ratio, etc. In some embodiments, the feature characteristic determination module 334 can generate the first contour to represent first subregion(s) of which the first portion has the first surrounding volume ratio, the second contour to represent second subregion(s) of which the second portion has the second surrounding volume ratio, etc.

[0075] In some embodiments, the process controller 302 includes the configuration determination module 338 to determine a configuration of the additive manufacturing system 300, or portion(s) thereof such as at least one of the vision system 306 or the optics assembly 308, for fabrication of the part 310, or layer(s) thereof. For example, the configuration determination module 338 can determine and / or generate parameters, such as LBPF parameters, to configure the additive manufacturing system 304. In some embodiments, the configuration determination module 338 can generate a file, such as a print file, based on the parameters. The configuration determination module 338 can generate the instructions 314 to be provided to the additive manufacturing system 304.

[0076] In some embodiments, the configuration determination module 338 can determine LBPF parameters for one or more lasers of the optics assembly 308, such as, but not limited to, at least one of a pulse- width modulation (PWM) configuration, a laser power, a scanning speed, a spot size, a hatch spacing, or one or more sampling parameters of the one or more lasers. For example, the PWM configuration may include at least one of a frequency or a pulse duration of the one or more lasers. A PWM configuration of a laser may refer to controlling the output of the laser by adjusting the pulse repetition frequency (PRF) and / or duty cycle (DC) of the laser. Any other type of parameter is contemplated such as information including locations of a plurality of laser energy pixels on a build surface (e.g., the metal powder bed) during the build process, trajectories of the laser energy pixels across the build surface, a list of operating parameters for the laser energy pixels, expected intensityvalues of laser energy pixels projected onto the build surface, expected radiation emissions and / or reflections of welds and / or melt pools (e.g., an expected weld signature), and / or any other appropriate parameter.

[0077] In some embodiments, the configuration determination module 338 can determine an LBPF parameter, such as a PWM configuration of a laser, on a subregion-by- subregion basis. For example, the configuration determination module 338 can configure the laser using a first PWM configuration for a first subregion of the layer in-situ, a second PWM configuration for a second subregion of the layer in-situ, and so forth. Alternatively, the configuration determination module 338 may configure the laser using the same PWM configuration for an entirety of the layer.

[0078] In some embodiments, the process controller 302 includes the relationship determination module 340 to identify and / or establish relationships between LBPF parameters and the physical behavior and / or response of printed features during LBPF process in-situ. For example, the relationship determination module 340 can determine that the LBPF parameters are a function of printed material (e.g., powder) and part geometries. By way of example, the 3-D part file 316 can be processed with a first partitioning to determine whether the 3-D part is a solid part or a lattice part. The first partitioning can be processed with a second partitioning using feature characteristic thresholds, such as surrounding volume ratio thresholds, to determine subregions. Accordingly, in some embodiments, the subregions can be identified and / or determined as a function of at least one of the printed material, the first partitioning, or the second partitioning. In some embodiments, LBPF parameters are identified for the subregions. For example, a set of the LBPF parameters can include laser power, scanning speed, laser PWM, scanning direction, trajectory designs, etc., and / or any combination(s) thereof.

[0079] In some embodiments, the relationship determination module 340 can establish a relationship between a subregion of a feature (e.g., a lattice structure, a solid structure, an overhang, etc.) having a surrounding volume ratio of 0.4 (or any other value) and an LBPF parameter, such as a PWM configuration of 50%. In some embodiments, the relationship determination module 340 can establish the relationship by generating a data association between the subregion and the PWM configuration. In some embodiments, the relationship determination module 340 can store the data association in the second datastore 356 as a first one (identified by SL1 CONFIG- 1) of a plurality of data associations 362.For example, the first one of the plurality of data associations 362 can represent a first subregion (identified by SI-1) with a first feature characteristic (e.g., a first surrounding volume ratio) having a first corresponding configuration (identified by CONFIG- 1). The first corresponding configuration of this example can include one or more LBPF parameters that, when used to configure the additive manufacturing system 304, can fabricate a subregion of a layer of the part 310 that has substantially similar features, properties, etc., to the first subregion.

[0080] In some embodiments, the process controller 302 includes the ML module 342 to execute an ML model 348 to generate and / or predict a configuration of the additive manufacturing system 304 based at least in part of identified feature(s), or portion(s) thereof, associated with a part, such as the part 310. ML generally refers to the field of deploying computer algorithms (and / or associated hardware) that improve (e.g., automatically improve, iteratively improve) by the use of data in applications (e.g., real-world applications, simulated applications) to generate outputs and through feedback of an evaluation of the outputs to the computer algorithms. Typically, performing ML involves creating a statistical model (or simply a “model”), which is configured to process data to output inferences and / or predictions. Some ML models may be built and / or generated through an iterative process (referred to as “training”) of ingesting data and evaluating outputs of the ML models.Exemplary training of an ML model may be supervised training, which may include instantiating the ML model, providing the ML model with sample data (referred to as “training data”) that may have labels (e.g., metadata, data tags) to describe the sample data, and comparing output(s) of the ML model with the labels to evaluate accuracy of the ML model based on the comparison(s).

[0081] In some embodiments, the ML module 342 can instantiate one or more ML models, such as the ML model 348, for training. Non-limiting examples of the ML model 348 include a deep learning model, a clustering model, a decision tree, a support vector machine (SVM), a Bayesian network, a hidden Markov model, and / or any combination(s) thereof. Non-limiting examples of a deep learning model include a convolutional neural network (CNN), a graph neural network (GNN), a recurrent neural network (RNN), a multi-layer perceptron, an autoencoder, a generative adversarial network (GAN), a CTC-fitted neural network model, and / or any combination(s) thereof.

[0082] In some embodiments, the ML module 342 can train (and / or retrain) one or more ML models, such as the ML model 348, using training data. For example, the ML module 342 can train the ML model 348 by applying a supervised learning training algorithm using labeled training data. As an example, the ML module 342 can train a deep learning model (e.g., a neural network) by using stochastic gradient descent. As another example, the ML module 342 can train an SVM to identify decision boundaries of the SVM by optimizing a cost function. By way of example, the ML module 342 can: (1) generate inputs to the ML model 348 using the training data; (2) label the training data using data values from additive manufacturing experts; and (3) apply a supervised training algorithm to the generated inputs and corresponding labels. Additionally or alternatively, the ML module 342 can train the ML model 348 by applying an unsupervised learning algorithm and / or a semi- supervised learning algorithm to the training data.

[0083] In some embodiments, the training data can include data representative of relationships between at least one of powder bed material, geometry partitionings (e.g., slices) of 3-D features, or configurations of the additive manufacturing system 304. In some embodiments, the relationships were established in connection with previously fabricated parts. For example, the relationships may have been established from previous iterative experimentation for process development of a part.

[0084] In some embodiments, the ML module 342 can retrain the ML model 348 using new and / or revised portion(s) of the training data on a periodic or aperiodic basis. For example, the ML module 342 can be configured to update a previously trained neural network by updating values of one or more parameters of the neural network using new training data. In some embodiments, the ML module 342 can be configured to update the neural network by training a new neural network using a combination of previously obtained training data and new training data.

[0085] In some embodiments, the ML module 342 can be configured to update the ML model 348 in response to any one of different types of events. For example, in some embodiments, the ML module 342 can be configured to update the one or more ML models in response to a user command. As an example, the ML module 342 can generate and / or present a graphical user interface (GUI) via which the user may command performance of a training process. In some embodiments, the ML module 342 can be configured to update the ML model 348 automatically (e.g., not in response to a user command), for example, inresponse to a software command. As another example, in some embodiments, the ML module 342 can be configured to update the ML model 348 in response to detecting one or more conditions. For example, the ML module 342 can update the ML model 348 in response to detecting expiration of a period of time. As another example, the ML module 342 can update the ML model 348 in response to receiving a threshold amount of new training data.

[0086] In some embodiments, the ML module 342 can deploy the ML model 348 as an executable construct after the training process. For example, after a determination that an accuracy of the ML module 342 satisfies a threshold (e.g., an accuracy threshold), the ML module 342 can compile the trained ML model 348 into an executable file that, when instantiated and / or executed, can perform inference operations. Non-limiting examples of inference operations include outputting LBPF parameters (e.g., values of LBPF parameters) and predictions of configurations of the additive manufacturing system 304 in accordance with the LBPF parameters.

[0087] In some embodiments, the process controller 302 includes the first datastore 344 to record data, such as the thresholds 346 and the ML model 348. In some embodiments, the build plan generation system 300 includes the second datastore 356 to record data, such as the plurality of data associations 362. In some embodiments, the first datastore 344 and / or the second datastore 356 can be implemented by any technology for storing data. For example, the first datastore 344 and / or the second datastore 356 can be implemented by a volatile memory (e.g., a Synchronous Dynamic Random Access Memory (SDRAM), a Dynamic Random Access Memory (DRAM), a RAMBUS Dynamic Random Access Memory (RDRAM), etc.) and / or a non-volatile memory (e.g., flash memory). The first datastore 344 and / or the second datastore 356 may additionally or alternatively be implemented by one or more double data rate (DDR) memories, such as DDR, DDR2, DDR3, DDR4, mobile DDR (mDDR), etc. The first datastore 344 and / or the second datastore 356 may additionally or alternatively be implemented by one or more mass storage devices such as hard disk drive(s) (HDD(s)), compact disk (CD) drive(s), digital versatile disk (DVD) drive(s), solid-state disk (SSD) drive(s), etc. While in the illustrated example the first datastore 344 and the second datastore 356 are illustrated as single datastores, the first datastore 344 and / or the second datastore 356 may be implemented by any number and / or type(s) of datastore. Furthermore, the data stored in the first datastore 344 and / or the second datastore 356 may be in any dataformat. Non-limiting examples of data formats include a flat file, binary data, comma delimited data, tab delimited data, and structured query language (SQL) structures.

[0088] In some embodiments, the first datastore 344 and / or the second datastore 356 may be implemented by a database system, such as one or more databases. The term “database” as used herein means an organized body of related data, regardless of the manner in which the data or the organized body thereof is represented. For example, the organized body of related data may be in the form of one or more of a table, a log, a map, a grid, a packet, a datagram, a frame, a file, an e-mail, a message, a document, a report, a list or in any other form.

[0089] While an example implementation of the process controller 302, and / or, more generally, the build plan generation system 300, is depicted in FIG. 3, other implementations are contemplated. For example, one or more blocks, components, functions, etc., of the process controller 302, and / or, more generally, the build plan generation system 300 may be combined or divided in any other way. The process controller 302, and / or, more generally, the build plan generation system 300 of the illustrated example may be implemented by hardware alone, or by a combination of hardware, software, and / or firmware. For example, the process controller 302, and / or, more generally, the build plan generation system 300 may be implemented by one or more analog or digital circuits (e.g., comparators, operational amplifiers, etc.), one or more hardware-implemented state machines, one or more programmable processors (e.g., central processing units (CPUs), digital signal processors (DSPs), field programmable gate arrays (FPGAs), etc.), one or more network interfaces (e.g., network interface circuitry, network interface cards (NICs), smart NICs, etc.), one or more ASICs, one or more memories (e.g., non-volatile memory, volatile memory, etc.), one or more mass storage disks or devices (e.g., hard-disk drives (HDDs), solid-state disk (SSD) drives, etc.), etc., and / or any combination(s) thereof.

[0090] FIG. 4 depicts an exemplary workflow 400 for generating a build plan, according to some embodiments. In some embodiments, the workflow 400 of this example can be carried out, performed, and / or implemented by the additive manufacturing system 100 of FIG. 1, the additive manufacturing system 200 of FIG. 2, and / or the build plan generation system 300 of FIG. 3.

[0091] The workflow 400 of the illustrated example begins at block 402, at which part geometry is received, which may be in the form of a CAD file (e.g., an .STL file). Forexample, the part geometry can correspond to a geometry of the part 310 of FIG. 3 and / or the 3-D part file 316 of FIG. 3. At block 404, a slicer processes the part geometry into a plurality of slices and outputs the plurality of slices as sliced geometry at block 406. In some embodiments, the slicer can be implemented by the feature extraction module 332 of FIG. 3. In some embodiments, the sliced geometry can be implemented by a common layer interface (CLI) file (e.g., a .CLI file). For example, the CLI file can store data representing the geometrical information of the intersection of a 3-D model with a plane (which may also be referred to as a slice). In some embodiments, the volume between two parallel slices is called a layer.

[0092] At blocks 408 and 410, automatic feature recognition is performed. In some embodiments, the automatic feature recognition can be implemented by the feature extraction module 332. For example, the feature extraction module 332 can recognize and / or identify feature(s) in a layer of the sliced geometry, such as a solid part (e.g., a solid structure) or a lattice part (e.g., a lattice structure). In some embodiments, at blocks 408 and 410, the feature extraction module 332 can recognize feature(s) in the layer responsive to user-defined configuration parameters. For example, the feature(s) and / or the user-defined configuration parameters can be obtained and / or retrieved from a parameter database 412. In some embodiments, the parameter database 412 can be implemented by the first datastore 344 and / or the second datastore 356 of FIG. 3.

[0093] At block 414, partitioned geometry is generated. For example, the feature extraction module 332 can define and / or generate the partitioned geometries by extracting feature(s) from a layer and / or isolating the feature(s) from the other portion(s) of the layer. In some embodiments, the feature extraction module 332 can identify a solid part in a layer of the sliced geometry as a partitioned geometry.

[0094] At block 416, conductivity region parameter assignment is performed. For example, the feature characteristic module 334 can define and / or determine a thermal conductivity of a baseline part, such as a solid metal part. The feature characteristic module 334 can normalize the thermal conductivity to a baseline conductivity of 1 (e.g., a baseline conductivity of 1 considering the sampling dimensions of the solid metal part to be a diameter of 500 micrometers and a depth of 500 micrometers). For example, the feature characteristic module 334 can determine that the surrounding volume ratio of a solid metal part is 1, which can correspond to a baseline conductivity of 1.

[0095] At block 418, thermal conductivity calculation(s) are performed. For example, the feature characteristic module 334, and / or, more generally, the process controller 302, can calculate a feature characteristic, such as a thermal conductivity, for respective portion(s) of the partitioned geometry. For example, the feature characteristic module 334 can calculate the surrounding volume ratio of an entire XY scanning area of the layer. In some embodiments, the surrounding volume ratio(s) can be mapped to one or more corresponding thermal conductivities. In some embodiments, the feature characteristic module 334 can generate a 2- D contour mapping of thermal conductivity based on thresholds from 0 to 1 of which 1 corresponds to the thermal conductivity of the baseline part. Additionally or alternatively, the feature characteristic module 334 can generate a 2-D contour mapping of surrounding volume ratios based on thresholds from 0 to 1 of which 1 corresponds to a surrounding volume ratio of the baseline part. At block 418, the configuration determination module 338 can determine parameters, such as LBPF parameters, which correspond to the surrounding volume ratio and / or thermal conductivity calculation(s). For example, the configuration determination module 338 can determine values for sampling parameters (e.g., width, depth, resolution), conductivity thresholds that may be used to determine values of PWM configurations and / or power levels of laser energy sources, etc.

[0096] At block 422, the final partitioned geometry is generated. For example, the subregion identification module 336 can identify subregion(s) of a layer based at least in part on the surrounding volume ratio and / or thermal conductivity calculation(s). In some embodiments, the configuration determination module 338 can generate a configuration for the additive manufacturing system 304 in accordance with LBPF parameters. For example, the LBPF parameters of the configuration can cause the additive manufacturing system 304 to fabricate the subregion(s) in accordance with the LBPF parameters.

[0097] At block 424, toolpath generation is performed. For example, the configuration determination module 338 can generate and / or output a print file 426 for fabrication of the part 310. In some embodiments, the print file can include the instructions 314 of FIG. 3 that, when executed by the additive manufacturing system 304, fabricates the part 310, or layer(s) thereof, in accordance with the LBPF parameters.

[0098] FIG. 5 depicts an exemplary workflow 500 for identifying subregions of a three-dimensional part layer, according to some embodiments. The workflow 500 begins at first operation(s) 502, of which a 3-D model 504 (e.g., the 3-D part file 316 of FIG. 3) issliced and feature(s) is / are identified. For example, the 3-D model 504 can be of a part 506 with a feature 507, which is a 30 degree overhang in this example. The 3-D model 504 of the part 506 with the feature 507 can be sliced into a plurality of layers.

[0099] During second operation(s) 503, subregion generation by feature characteristic(s), such as a surrounding volume ratio, is performed. For example, the subregion identification module 336 of FIG. 3 can generate a visualization 508 of a single layer with a part boundary 510 and subregions 512, 514, 516 defined by contours. For example, the feature characteristic module 334 can determine surrounding volume ratios for portion(s) of the feature 507. The subregion identification module 336 can generate a first contour representative of a first subregion 512 (identified by Subregion 0), a second contour representative of a second subregion 514 (identified by Subregion 1), a third contour representative of a third subregion 516 (identified by Subregion 2), and a fourth contour representative of the part boundary 510. In this example, the first subregion 512 has a greater surrounding volume ratio (and thereby a higher thermal conductivity) than the second subregion 514 and the third subregion 516. In this example, the second subregion 514 has a greater surrounding volume ratio (and thereby a higher thermal conductivity) than the third subregion 516.

[0100] During third operation(s) 518, toolpath and LB PF process parameter assignment in different subregions is performed. For example, the configuration determination module 338 can determine LBPF process parameters that correspond to the subregions 512, 514, 516 and / or, more generally, that correspond to surrounding volume ratios represented by the subregions 512, 514, 516. For example, the configuration determination module 338 can determine values of an LBPF process parameter, such as a laser control duty cycle, which can optimize and / or otherwise improve fabrication of a subregion that has a particular feature characteristic.

[0101] In an exemplary visualization 520, the configuration determination module 338 can determine a first value of a laser control duty cycle parameter of 100% for the first subregion 512. The configuration determination module 338 can determine second and third values of the laser control duty cycle for the second and third subregions 514, 516, respectively. In some embodiments, the first, second, and / or third values of the laser control duty cycle parameter are linked through the surrounding volume ratio natures (and therebythe thermal conductivity natures) of each subregion, and therefore can be scaled and / or optimized with respect to thermal history and residual stress during the LBPF process.

[0102] Beneficially, the configuration determination module 338, and / or, more generally, the process controller 302 of FIG. 3, can assign different values of the same LBPF process parameter to different subregions in accordance with their feature characteristic(s). Beneficially, the additive manufacturing system 304 of FIG. 3 can optimize and / or otherwise improve fabrication of the subregions 512, 514, 516 based on their respective feature characteristic(s).

[0103] FIG. 6 is a table 600 of exemplary surrounding volume ratio thresholds 602 and corresponding configuration parameters 604, 606, according to some embodiments. In some embodiments, the table 600 can be representative of relationships between the configuration parameters and the physical behavior and response of printed features during LBPF in-situ. For example, the table 600 can be representative of the plurality of data associations 362 of FIG. 3.

[0104] In some embodiments, the surrounding volume ratio thresholds 602 can implement the thresholds 346 of FIG. 3. For example, the feature characteristic determination module 334 of FIG. 3 can calculate and / or determine the surrounding volume ratio thresholds 602. In some embodiments, the configuration determination module 338 can determine the configuration parameters 604, 606. The surrounding volume ratio thresholds 602 of this example are incremented by 0.1. Alternatively, the surrounding volume ratio thresholds 602 may be incremented in any other unit (e.g., 0.025, 0.050, 0.075, etc.). In some embodiments, the surrounding volume ratio thresholds 602 are determined by users. Alternatively, the surrounding volume ratio thresholds 602 may be determined by a model, such as an ML model.

[0105] The configuration parameters 604, 606 of this example are first PWM configuration parameters 604 for solid parts and second PWM configuration parameters 606 for lattices (e.g., lattice parts). In the table 600, the configuration parameters 604, 606 can vary based at least in part on the surrounding volume ratio thresholds 602.

[0106] By way of example, the feature extraction module 332 of FIG. 3 can identify a portion of a layer of the feature 507 of FIG. 5 as a solid feature (e.g., the feature 507 is a solid structure). The feature characteristic determination module 334 can determine a surrounding volume ratio of 0.55 for the portion of the layer of the feature 507. The subregionidentification module 336 can determine that the first portion is a first subregion after a determination that the solid feature has a surrounding volume ratio of 0.55 and the surrounding volume ratio is in a surrounding volume ratio band bound by the surrounding volume ratio thresholds of 0.5 and 0.6. The configuration determination module 338 can map the first subregion identifier 354, which can correspond to and / or identify a subregion defined by a solid structure with a surrounding volume ratio in the surrounding volume ratio band, to the first configuration 360 via the first one of the plurality of data associations 362. In such embodiments, the configuration determination module 338 can determine that the first configuration 360 includes at least a value of 100% for a PWM configuration of at least one laser energy source.

[0107] By way of another example, the feature extraction module 332 can identify a portion of a layer of the feature 507 of FIG. 5 as a lattice feature (e.g., the feature 507 is a lattice structure). The feature characteristic determination module 334 can determine a surrounding volume ratio of 0.55 for the portion of the layer of the feature 507. The subregion identification module 336 can determine that the first portion is a second subregion after a determination that the lattice feature has a surrounding volume ratio of 0.55 and the surrounding volume ratio is in a surrounding volume ratio band bound by the surrounding volume ratio thresholds of 0.5 and 0.6. The configuration determination module 338 can map a second subregion identifier, which can correspond to and / or identify a subregion defined by a lattice structure with a surrounding volume ratio in the surrounding volume ratio band, to a second configuration 360 (e.g., CONFIG-2) via a second one of the plurality of data associations 362 (e.g., SI-2CONFIG-2). In such embodiments, the configuration determination module 338 can determine that the second configuration includes at least a value of 60% for a PWM configuration of at least one laser energy source.

[0108] FIGS. 7-9 are flowcharts can be representative of machine-readable instructions that may be executed by processor circuitry to implement one or more controllers, such as the process controller 302 of FIG. 3. Additionally or alternatively, block(s) of one(s) of the flowcharts of FIGS. 7, 8, and / or 9 may be representative of state(s) of one or more hardware-implemented state machines, algorithm(s) that may be implemented by hardware alone such as an ASIC, etc., and / or any combination(s) thereof.

[0109] FIG. 7 is a flowchart 700 representative of example machine-readable instructions that may be executed by processor circuitry to implement the process controller302 of FIG. 3 to configure the additive manufacturing system 100, 200, 304 of FIGS. 1, 2, and / or 3 based on subregion configurations, according to some embodiments. The flowchart 700 of FIG. 7 begins at block 702, at which the process controller 302 obtains a three- dimensional (3-D) model of a part to fabricate with additive manufacturing (AM). For example, the data interface module 330 of FIG. 3 can obtain the 3-D part file 316 from the server 318 via the network 320. In some embodiments, the 3-D part file 316 can be a digital representation of the part 310 to be fabricated by the additive manufacturing system 304 of FIG. 3. Alternatively, the data interface module 330 may obtain the image data 358, which can include image(s) of the build surface of the build plate 312.

[0110] At block 704, the process controller 302 slices the 3-D model into layers. For example, the feature extraction module 332 of FIG. 3 can slice the 3-D part file 316 into a plurality of layers.

[0111] At block 706, the process controller 302 determines whether to invoke machine learning to predict an AM configuration based on the layers. For example, the ML module 342 of FIG. 3 can determine, based on user input, whether to execute the ML model 348 using one(s) of the layers as input(s) to generate output(s), which can include one or more LB PF parameters that optimize and / or otherwise improve fabrication of the one(s) of the layers.

[0112] If, at block 706, the process controller 302 determines to invoke machine learning to predict an AM configuration based on the layers, control proceeds to block 708. At block 708, the process controller 302 executes a machine learning model to predict an AM configuration for respective subregions of the layers. For example, the ML module 342 can execute and / or instantiate the ML model 348 using one(s) of the layers as input(s) to generate output(s), which can include one or more LBPF parameters that optimize and / or otherwise improve fabrication of the one(s) of the layers. After executing a machine learning model to predict an AM configuration for respective subregions of the layers at block 708, control proceeds to block 716.

[0113] If, at block 706, the process controller 302 determines not to invoke machine learning to predict an AM configuration based on the layers, control proceeds to block 710. At block 710, the process controller 302 determines a surrounding volume ratio for portion(s) of respective features of the layers. For example, the feature characteristic determination module 334 of FIG. 3 can calculate a surrounding volume ratio for portion(s) of the feature507 on a layer of the part 506 of FIG. 5, such as a ratio of fused material of a voxel of the feature 507 and a total surrounding volume of the voxel.

[0114] At block 712, the process controller 302 identifies the portion(s) as subregions of the layers based on determinations that one(s) of the surrounding volume ratios satisfy a threshold. For example, the subregion identification module 336 of FIG. 3 can compare the surrounding volume ratio for the portion(s) of the feature 507 to the surrounding volume ratio thresholds 602 of FIG. 6. In some embodiments, the subregion identification module 336 can determine that the surrounding volume ratios satisfy one or more thresholds. After the determination(s), the subregion identification module 336 can identify the portion(s) of the feature 507 as subregion(s). For example, the subregion identification module 336 can identify a first portion of the feature 507 as a first subregion, which can be representative of the first portion being a wall overhang with a surrounding volume ratio of 0.45. In some embodiments, the subregion identification module 336 can assign the first subregion identifier 354 to the first portion of the feature 507.

[0115] At block 714, the process controller 302 maps the surrounding volume ratios to AM configuration(s). For example, the configuration determination module 338 can query the second datastore 356 with the first subregion identifier 354. In some embodiments, the second datastore 356 can return the first configuration 360 based on a mapping of the first subregion identifier 354 to the first one of the plurality of data associations 362. In some embodiments, the first configuration 360 can include one or more LBPF parameters, such as a power level for one or more laser energy sources, a PWM configuration for one or more laser energy pixels, etc., and / or any combination(s) thereof.

[0116] At block 716, the process controller 302 configures an AM system based on the AM configuration(s). For example, the configuration determination module 338 can compile and / or generate a print file for fabrication of one or more layers of the part 310 and / or one or more layers of the part 506. In some embodiments, the configuration determination module 338 can generate the instructions 314, which can be print instructions, based on the print file. The configuration determination module 338 can output and / or cause transmission of the instructions 314 to the additive manufacturing system 304 to cause the additive manufacturing system 304 to be configured in accordance with the one or more LBPF parameters.

[0117] At block 718, the process controller 302 fabricates the subregions of the part using the AM system configured using the AM configuration(s). For example, the additive manufacturing system 304 can fabricate one(s) of the subregions 512, 514, 516 of FIG. 5 using the one or more LB PF parameters.

[0118] At block 720, the process controller 302 determines whether to configure another AM system. For example, the data interface module 330 can determine whether another 3-D part file and / or additional image data has been received for reconfiguration of the additive manufacturing system 304 and / or a different additive manufacturing system. If, at block 720, the process controller 302 determines to configure another AM system, control returns to block 702. Otherwise, the flowchart 700 of FIG. 7 concludes.

[0119] FIG. 8 is a flowchart 800 representative of example machine-readable instructions that may be executed by processor circuitry to implement the process controller 302 of FIG. 3 to generate data associations of surrounding volume ratios and corresponding additive manufacturing system configurations, according to some embodiments. The flowchart 800 begins at block 802, at which the process controller 302 slices a digital representation of a surface geometry of a three-dimensional object into layers. For example, the feature extraction module 332 of FIG. 3 can slice the 3-D part file 316 into a plurality of layers. In some embodiments, the 3-D part file 316 can be a digital representation, such as a 3-D model, of a 3-D part, such as the part 506 of FIG. 5.

[0120] At block 804, the process controller 302 selects a layer to process. For example, the feature characteristic determination module 334 of FIG. 3 can select a first layer of the plurality of layers to process.

[0121] At block 806, the process controller 302 determines a surrounding volume ratio for respective feature(s) of the layer. For example, the feature characteristic determination module 334 can determine that the first layer has a portion (e.g., a slice) of the feature 507 of FIG. 5. In some embodiments, the feature characteristic determination module 334 can calculate a surrounding volume ratio of the portion of the feature 507.

[0122] At block 808, the process controller 302 identifies subregion(s) of the layer based on comparison(s) of the surrounding volume ratios and surrounding volume ratio thresholds. For example, the subregion identification module 336 of FIG. 3 can generate contours representative of different surrounding volume ratios of the portion of the feature 507. In some embodiments, the subregion identification module 336 can determine thatone(s) of the surrounding volume ratios satisfy one or more thresholds. After the determination(s), the subregion identification module 336 can identify the contours as defining the subregions 512, 514, 516. For example, the subregion identification module 336 can determine that a first one of the contours is associated with a first surrounding volume ratio that satisfies a first surrounding volume ratio threshold. The subregion identification module 336 can identify the first one of the contours as the first subregion 512 based on the first surrounding volume ratio that satisfying the first surrounding volume ratio threshold. In some embodiments, the subregion identification module 336 can generate a subregion identifier, such as the first subregion identifier 354 of FIG. 3, to be representative and / or symbolize that the portion of the feature 507 is a wall overhang having a corresponding surrounding volume ratio.

[0123] At block 810, the process controller 302 determines an additive manufacturing (AM) system configuration for respective subregion(s) of the layer. For example, the configuration determination module 340 of FIG. 3 can determine a first configuration to include one or more LBPF parameters, such as a power level of one or more laser energy sources, a PWM configuration associated with one or more laser pixels, a scanning speed, a sampling width of a scan axis of a build plate, a sampling width of an indexing axis of a build plate, a sampling depth, a sampling resolution, etc., and / or any combination(s) thereof.

[0124] At block 812, the process controller 302 determines whether to select another layer to process. For example, the feature characteristic determination module 334 can determine that there is another layer of the plurality of layers to process.

[0125] If, at block 812, the process controller 302 determines to select another layer to process, control returns to block 804 to select another layer to process. If, at block 812, the process controller 302 determines not to select another layer to process, control proceeds to block 814.

[0126] At block 814, the process controller 302 stores data associations of the surrounding volume ratios and corresponding AM configuration(s). For example, the relationship determination module 340 of FIG. 3 can store a data association of the first subregion identifier 354 and the first configuration 360 as one of the plurality of data associations 362 in the second datastore 356 of FIG. 3.

[0127] At block 816, the process controller 302 determines whether to process another digital representation. For example, the feature extraction module 332 can determinewhether there is another 3-D part file to process. If, at block 816, the process controller 302 determines to process another digital representation, control returns to block 802 to slice another digital representation of a surface geometry of a three-dimensional object into layers. Otherwise, the flowchart 800 of FIG. 8 concludes.

[0128] FIG. 9 is a flowchart 900 representative of example machine-readable instructions that may be executed by processor circuitry to implement the process controller 302 of FIG. 3 to configure the additive manufacturing system 100, 200, 304 of FIGS. 1, 2, and / or 3 using configurations determined by a machine-learning model, according to some embodiments. The flowchart 900 of FIG. 9 begins at block 902, at which the process controller 302 configures a machine learning (ML) model. For example, the data interface module 330 can obtain an ML model, such as a neural network, from a network, a server, etc. In some embodiments, the ML module 342 can configure the neural network by creating a number of input, hidden, and / or output layers, determining initial values of neuron weights, setting a plurality of hyperparameters, etc., and / or any combination(s) thereof.

[0129] At block 904, the process controller 302 obtains training data representative of relationships between at least one of powder bed material, geometry partitionings of 3-D features, or additive manufacturing configurations. For example, the ML module 342 can obtain training data, such as information associated with a plurality of powder bed materials, exemplary contours (e.g., the contours that represent the subregions 512, 514, 516 of FIG. 5), one(s) of the plurality of data associations 362, etc., and / or any combination(s) thereof.

[0130] At block 906, the process controller 302 trains the ML model using the training data. For example, the ML module 342 can train the neural network using the training data.

[0131] At block 908, the process controller 302 determines whether an accuracy of the ML model satisfies a threshold. For example, the ML module 342 can determine whether an accuracy associated with output(s) of the ML model 348 is greater than a threshold, such as an accuracy threshold, and thereby satisfies the threshold.

[0132] If, at block 908, the process controller 302 determines that an accuracy of the ML model does not satisfy a threshold, control returns to block 906 to continue training the ML model using the training data. If, at block 908, the process controller 302 determines that an accuracy of the ML model satisfies a threshold, control proceeds to block 910.

[0133] At block 910, the process controller 302 deploys the ML model as an executable construct. For example, the ML module 342 can compile the trained neural network into an executable file that, when executed, can perform inference operations. In some embodiments, the ML module 342 can compile a configuration image representative of the trained neural network that may be loaded onto an accelerator for ML operations. In some embodiments, the ML module 342 can store the executable file, the configuration image, etc., and / or any combination(s) thereof, in the first datastore 344 as the ML model 348.

[0134] At block 912, the process controller 302 receives a 3-D model of part to be fabricated by an additive manufacturing (AM) system. For example, the data interface module 330 of FIG. 3 can receive the 3-D part file 316 from the server 318 via the network 320.

[0135] At block 914, the process controller 302 executes the ML model using the 3-D model as data input(s) to generate printer configuration(s) as data output(s). For example, the ML module 342 can execute the ML model 348 using the 3-D part file 316, or slices and / or layers thereof, as data input(s) to generate data output(s), which can include a configuration of the additive manufacturing system 304. In some embodiments, the configuration can include one or more LBPF parameters, such as a laser control duty cycle, a laser power of one or more laser energy sources, etc., and / or any combination(s) thereof.

[0136] At block 916, the process controller 302 obtains feedback based on a review of the data output(s). For example, a user associated with an additive manufacturing system (e.g., an engineer, a technician, quality control personnel, etc.) can evaluate the configuration and make adjustment(s). In some embodiments, the user can provide the adjustment(s) to the ML module 342 via a graphical user interface (GUI). For example, the adjustment(s) and associated data can be used to retrain the ML model 348.

[0137] At block 918, the process controller 302 determines whether to retrain the ML model. For example, the ML module 342 can determine to retrain the ML model 348 in response to a trigger, such as a time threshold being satisfied, a quantity of training data being received exceeding a threshold, etc.

[0138] If, at block 918, the process controller 302 determines to retrain the ML model, control returns to block 904. If, at block 918, the process controller 302 determines not to retrain the ML model, control proceeds to block 920.

[0139] At block 920, the process controller 302 determines whether to receive another 3-D model of a part to be fabricated. For example, the data interface module 330 can determine whether another 3-D part file has been received to facilitate fabrication of another part.

[0140] If, at block 920, the process controller 302 determines to receive another 3-D model of a part to be fabricated, control returns to block 912 to receive another 3-D model of a part to be fabricated by an AM system. Otherwise, the flowchart 900 of FIG. 9 concludes.

[0141] The above methods may be implemented by one or more controllers including at least one processor operatively coupled to the various controllable portions of an additive manufacturing system as disclosed herein. The methods may be embodied as computer readable instructions stored on non-transitory computer readable memory associated with the at least one processor such that when executed by the at least one processor the additive manufacturing system may perform any of the actions related to the methods disclosed herein. Additionally, it should be understood that the disclosed order of the steps is exemplary and that the disclosed steps may be performed in a different order, simultaneously, and / or may include one or more additional intermediate steps not shown as the disclosure is not so limited.

[0142] The above-described embodiments of the technology described herein can be implemented in any of numerous ways. For example, the embodiments may be implemented using hardware, software, or a combination thereof. When implemented in software, the software code can be executed on any suitable processor or collection of processors, whether provided in a single computing device or distributed among multiple computing devices. Such processors may be implemented as integrated circuits, with one or more processors in an integrated circuit component, including commercially available integrated circuit components known in the art by names such as CPU chips, GPU chips, microprocessor, microcontroller, or co-processor. Alternatively, a processor may be implemented in custom circuitry, such as an ASIC, or semicustom circuitry resulting from configuring a programmable logic device. As yet a further alternative, a processor may be a portion of a larger circuit or semiconductor device, whether commercially available, semi-custom or custom. As a specific example, some commercially available microprocessors have multiple cores such that one or a subset of those cores may constitute a processor. Though, a processor may be implemented using circuitry in any suitable format.

[0143] Further, it should be appreciated that a computing device including one or more processors may be embodied in any of a number of forms, such as a rack-mounted computer, a desktop computer, a laptop computer, or a tablet computer. Additionally, a computing device may be embedded in a device not generally regarded as a computing device but with suitable processing capabilities, including a Personal Digital Assistant (PDA), a smart phone, tablet, or any other suitable portable or fixed electronic device.

[0144] Also, a computing device may have one or more input and output devices. These devices can be used, among other things, to present a user interface. Examples of output devices that can be used to provide a user interface include display screens for visual presentation of output and speakers or other sound generating devices for audible presentation of output. Examples of input devices that can be used for a user interface include keyboards, individual buttons, and pointing devices, such as mice, touch pads, and digitizing tablets. As another example, a computing device may receive input information through speech recognition or in other audible format.

[0145] Such computing devices may be interconnected by one or more networks in any suitable form, including as a local area network or a wide area network, such as an enterprise network or the Internet. Such networks may be based on any suitable technology and may operate according to any suitable protocol and may include wireless networks, wired networks or fiber optic networks.

[0146] Also, the various methods or processes outlined herein may be coded as software that is executable on one or more processors that employ any one of a variety of operating systems or platforms. Additionally, such software may be written using any of a number of suitable programming languages and / or programming or scripting tools, and also may be compiled as executable machine language code or intermediate code that is executed on a framework or virtual machine.

[0147] In this respect, the embodiments described herein may be embodied as a computer readable storage medium (or multiple computer readable media) (e.g., a computer memory, one or more floppy discs, compact discs (CD), optical discs, digital video disks (DVD), magnetic tapes, flash memories, RAM, ROM, EEPROM, circuit configurations in Field Programmable Gate Arrays or other semiconductor devices, or other tangible computer storage medium) encoded with one or more programs that, when executed on one or more computers or other processors, perform methods that implement the various embodimentsdiscussed above. As is apparent from the foregoing examples, a computer readable storage medium may retain information for a sufficient time to provide computer-executable instructions in a non-transitory form. Such a computer readable storage medium or media can be transportable, such that the program or programs stored thereon can be loaded onto one or more different computing devices or other processors to implement various aspects of the present disclosure as discussed above. As used herein, the term “computer-readable storage medium” encompasses only a non-transitory computer-readable medium that can be considered to be a manufacture (i.e., article of manufacture) or a machine. Alternatively or additionally, the disclosure may be embodied as a computer readable medium other than a computer-readable storage medium, such as a propagating signal.

[0148] The terms “program” or “software” are used herein in a generic sense to refer to any type of computer code or set of computer-executable instructions that can be employed to program a computing device or other processor to implement various aspects of the present disclosure as discussed above. Additionally, it should be appreciated that according to one aspect of this embodiment, one or more computer programs that when executed perform methods of the present disclosure need not reside on a single computing device or processor, but may be distributed in a modular fashion amongst a number of different computers or processors to implement various aspects of the present disclosure.

[0149] Computer-executable instructions may be in many forms, such as program modules, executed by one or more computers or other devices. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. Typically the functionality of the program modules may be combined or distributed as desired in various embodiments.

[0150] The embodiments described herein may be embodied as a method, of which an example has been provided. The acts performed as part of the method may be ordered in any suitable way. Accordingly, embodiments may be constructed in which acts are performed in an order different than illustrated, which may include performing some acts simultaneously, even though shown as sequential acts in illustrative embodiments.

[0151] Further, some actions are described as taken by a “user.” It should be appreciated that a “user” need not be a single individual, and that in some embodiments, actions attributable to a “user” may be performed by a team of individuals and / or an individual in combination with computer-assisted tools or other mechanisms.

[0152] FIG. 10 is an example implementation of an electronic platform 1000 structured to execute the machine-readable instructions of FIGS. 7-9 to implement a process controller, such as the process controller 302 of FIG. 3. It should be appreciated that FIG. 10 is intended neither to be a description of necessary components for an electronic and / or computing device to operate as a process controller or process control circuitry, in accordance with the techniques described herein, nor a comprehensive depiction. The electronic platform 1000 of this example may be an additive manufacturing computer, an electronic device, such as a cellular network device, a desktop computer, a laptop computer, a server (e.g., a computer server, a blade server, a rack-mounted server, etc.), a workstation, or any other type of computing and / or electronic device.

[0153] The electronic platform 1000 of the illustrated example includes processor circuitry 1002, which may be implemented by one or more programmable processors, one or more hardware-implemented state machines, one or more ASICs, etc., and / or any combination(s) thereof. For example, the one or more programmable processors may include one or more CPUs, one or more DSPs, one or more FPGAs, etc., and / or any combination(s) thereof. The processor circuitry 1002 includes processor memory 1004, which may be volatile memory, such as random-access memory (RAM) of any type. The processor circuitry 1002 of this example implements the feature extraction module 332, the feature characteristic determination module 334 (identified by FEATURE CHAR DETERM MODULE), the subregion identification module 336, the configuration determination module 338 (identified by CONFIG DETERMINATION MODULE), the relationship determination module 340 (identified by RELATIONSHIP DETERM MODULE), and the machine learning module 342 of FIG. 3.

[0154] The processor circuitry 1002 may execute machine-readable instructions 1006 (identified by INSTRUCTIONS), which are stored in the processor memory 1004, to implement at least one of the process controller 302 of FIG. 3. The machine -readable instructions 1006 may include data representative of computer-executable and / or machineexecutable instructions implementing techniques that operate according to the techniques described herein. For example, the machine-readable instructions 1006 may include data (e.g., code, embedded software (e.g., firmware), software, etc.) representative of the flowcharts of FIGS. 7, 8, and / or 9, or portion(s) thereof.

[0155] The electronic platform 1000 includes memory 1008, which may include the instructions 1006. The memory 1008 of this example may be controlled by a memory controller 1010. For example, the memory controller 1010 may control reads, writes, and / or, more generally, access(es) to the memory 1008 by other component(s) of the electronic platform 1000. The memory 1008 of this example may be implemented by volatile memory, non-volatile memory, etc., and / or any combination(s) thereof. For example, the volatile memory may include static random-access memory (SRAM), dynamic random-access memory (DRAM), cache memory (e.g., Level 1 (LI) cache memory, Level 2 (L2) cache memory, Level 3 (L3) cache memory, etc.), etc., and / or any combination(s) thereof. In some examples, the non-volatile memory may include Flash memory, electrically erasable programmable read-only memory (EEPROM), magnetoresistive random-access memory (MRAM), ferroelectric random-access memory (FeRAM, F-RAM, or FRAM), etc., and / or any combination(s) thereof.

[0156] The electronic platform 1000 includes input device(s) 1012 to enable data and / or commands to be entered into the processor circuitry 1002. For example, the input device(s) 1012 may include an audio sensor, a camera (e.g., an HFR camera, an infrared camera, still camera, a video camera, etc.), a keyboard, a microphone, a mouse, a touchscreen, a voice recognition system, etc., and / or any combination(s) thereof.

[0157] The electronic platform 1000 includes output device(s) 1014 to convey, display, and / or present information to a user (e.g., a human user, a machine user, etc.). For example, the output device(s) 1014 may include one or more display devices, speakers, etc. The one or more display devices may include an augmented reality (AR) and / or virtual reality (VR) display, a liquid crystal display (LCD), a light-emitting diode (LED) display, an organic light-emitting diode (OLED) display, a quantum dot (QLED) display, a thin-film transistor (TFT) LCD, a touchscreen, etc., and / or any combination(s) thereof. The output device(s) 1014 can be used, among other things, to generate, launch, and / or present a user interface. For example, the user interface may be generated and / or implemented by the output device(s) 1014 for visual presentation of output and speakers or other sound generating devices for audible presentation of output.

[0158] The electronic platform 1000 includes accelerators 1016, which are hardware devices to which the processor circuitry 1002 may offload compute tasks to accelerate their processing. For example, the accelerators 1016 may include artificial intelligence / machine-learning ( AI / ML) processors, ASICs, FPGAs, graphics processing units (GPUs), neural network (NN) processors, systems-on-chip (SoCs), vision processing units (VPUs), etc., and / or any combination(s) thereof. In some examples, one or more of the feature extraction module 332, the feature characteristic determination module 334, the subregion identification module 336, the configuration determination module 338, the relationship determination module 340, or the machine learning module 342 may be implemented by one(s) of the accelerators 1016 instead of the processor circuitry 1002. In some examples, the feature extraction module 332, the feature characteristic determination module 334, the subregion identification module 336, the configuration determination module 338, the relationship determination module 340, and / or the machine learning module 342 may be executed concurrently (e.g., in parallel, substantially in parallel, etc.) by the processor circuitry 1002 and the accelerators 1016. For example, the processor circuitry 1002 and one(s) of the accelerators 1016 may execute in parallel function(s) corresponding to the machine learning module 342.

[0159] The electronic platform 1000 includes storage 1018 to record and / or control access to data, such as the machine-readable instructions 1006. In this example, the storage 1018 implements the first datastore 344 of FIG. 3, which includes and / or stores the thresholds 346 and the ML model 348 of FIG. 3. Additionally or alternatively, the storage 1018 may implement the second datastore 356 of FIG. 3. The storage 1018 may be implemented by one or more mass storage disks or devices, such as HDDs, SSDs, etc., and / or any combination(s) thereof.

[0160] The electronic platform 1000 includes interface(s) 1020 to effectuate exchange of data with external devices (e.g., computing and / or electronic devices of any kind) via a network 1022. In this example, the interface(s) 1020 implements the data interface module 340 of FIG. 3. The interface(s) 1020 of the illustrated example may be implemented by an interface device, such as network interface circuitry (e.g., a NIC, a smart NIC, etc.), a gateway, a router, a switch, etc., and / or any combination(s) thereof. The interface(s) 1020 may implement any type of communication interface, such as BLUETOOTH®, a cellular telephone system (e.g., a 4G LTE interface, a 5G interface, a 6G interface, etc.), an Ethernet interface, a near-field communication (NFC) interface, an optical disc interface (e.g., a Blu- ray disc drive, a Compact Disk (CD) drive, a Digital Versatile Disk (DVD) drive, etc.), an optical fiber interface, a satellite interface (e.g., a beyond-line-of-site (BLOS) satelliteinterface, a line-of-site (LOS) satellite interface, etc.), a Universal Serial Bus (USB) interface (e.g., USB Type-A, USB Type-B, USB TYPE-C™ or USB-C™, etc.), etc., and / or any combination(s) thereof.

[0161] The electronic platform 1000 includes a power supply 1024 to store energy and provide power to components of the electronic platform 1000. The power supply 1024 may be implemented by a power converter, such as an alternating current-to-direct-current (AC / DC) power converter, a direct current-to-direct current (DC / DC) power converter, etc., and / or any combination(s) thereof. For example, the power supply 1024 may be powered by an external power source, such as an alternating current (AC) power source (e.g., an electrical grid), a direct current (DC) power source (e.g., a battery, a battery backup system, etc.), etc., and the power supply 1024 may convert the AC input or the DC input into a suitable voltage for use by the electronic platform 1000. In some examples, the power supply 1024 may be a limited duration power source, such as a battery (e.g., a rechargeable battery such as a lithium-ion battery).

[0162] Component(s) of the electronic platform 1000 may be in communication with one(s) of each other via a bus 1026. For example, the bus 1026 may be any type of computing and / or electrical bus, such as an I2C bus, a PCI bus, a PCIe bus, a SPI bus, and / or the like. In some embodiments, the bus 1026 of FIG. 10 can implement the bus 350 of FIG. 3.

[0163] The network 1022 may be implemented by any wired and / or wireless network(s) such as one or more cellular networks (e.g., 4G ETE cellular networks, 5G cellular networks, 6G cellular networks, etc.), one or more data buses, one or more local area networks (EANs), one or more optical fiber networks, one or more private networks, one or more public networks, one or more wireless local area networks (WEANs), etc., and / or any combination(s) thereof. For example, the network 1022 may be the Internet, but any other type of private and / or public network is contemplated. In some embodiments, the network 1022 can implement the network 320 of FIG. 3.

[0164] The network 1022 of the illustrated example facilitates communication between the interface(s) 1020 and a central facility 1028. The central facility 1028 in this example may be an entity associated with one or more servers, such as one or more physical hardware servers and / or virtualizations of the one or more physical hardware servers. For example, the central facility 1028 may be implemented by a public cloud provider, a private cloud provider, etc., and / or any combination(s) thereof. In this example, the central facility1028 may compile, generate, update, etc., the machine-readable instructions 1006 and store the machine-readable instructions 1006 for access (e.g., download) via the network 1022. For example, the electronic platform 1000 may transmit a request, via the interface(s) 1020, to the central facility 1028 for the machine-readable instructions 1006 and receive the machine- readable instructions 1006 from the central facility 1028 via the network 1022 in response to the request. In some embodiments, the central facility 1028 can implement the server 318 of FIG. 3.

[0165] Additionally or alternatively, the interface(s) 1020 may receive the machine- readable instructions 1006 via non-transitory machine-readable storage media, such as an optical disc 1030 (e.g., a Blu-ray disc, a CD, a DVD, etc.) or any other type of removable non-transitory machine -readable storage media such as a USB drive 1032. For example, the optical disc 1030 and / or the USB drive 1032 may store the machine -readable instructions 1006 thereon and provide the machine-readable instructions 1006 to the electronic platform 1000 via the interface(s) 1020.

[0166] While the present teachings have been described in conjunction with various embodiments and examples, it is not intended that the present teachings be limited to such embodiments or examples. On the contrary, the present teachings encompass various alternatives, modifications, and equivalents, as will be appreciated by those of skill in the art. Accordingly, the foregoing description and drawings are by way of example only.

Claims

CLAIMS1. A method for determining a configuration of an additive manufacturing system, the method comprising: slicing a three-dimensional model of a part to be fabricated by the additive manufacturing system into at least one layer including a first layer; determining a surrounding volume ratio of a first volume of fused material surrounding one or more voxels and a total volume surrounding the one or more voxels for respective features of the first layer, the respective features including a first feature; identifying at least one portion of the first feature as a first subregion of the first layer based on determining that a surrounding volume ratio of the at least one portion satisfies a threshold; mapping at least the surrounding volume ratio to one or more configuration parameters of the additive manufacturing system; and fabricating the first subregion of the first layer of the part using the one or more configuration parameters.

2. The method of claim 1, wherein the surrounding volume ratio is a first surrounding volume ratio, the threshold is a first threshold, the one or more configuration parameters are one or more first configuration parameters, and the method further comprising: identifying a second subregion of the first layer; determining a second surrounding volume ratio of the second subregion; determining that the second surrounding volume ratio satisfies a second threshold; mapping the second surrounding volume ratio to one or more second configuration parameters; and fabricating the second subregion of the first layer of the part using the one or more second configuration parameters.

3. The method of claim 2, wherein the first threshold and the second threshold are different.

4. The method of any one of claims 1-3, further comprising: identifying a second subregion of the first layer; andfabricating the second subregion after fabricating the first subregion.

5. The method of any one of claims 1-3, further comprising: identifying a second subregion of the first layer; fabricating a first portion of the first subregion; fabricating a second portion of the second subregion after fabricating the first portion of the first subregion; and fabricating a third portion of the first subregion after the fabricating of the second portion of the second subregion.

6. The method of any one of claims 1-5, wherein the one or more configuration parameters include, for one or more lasers of the additive manufacturing system, at least one of a pulse- width modulation configuration, a laser power, a scanning speed, a spot size, a hatch spacing, or one or more sampling parameters of the one or more lasers.

7. The method of claim 6, wherein the pulse-width modulation configuration includes at least one of a frequency or a pulse duration of the one or more lasers.

8. The method of claim 6, wherein the one or more sampling parameters include, for a photosensitive detector of the additive manufacturing system, at least one of a first sampling width associated with a scan axis of a build plate of the additive manufacturing system, a second sampling width associated with an indexing axis of the build plate, a sampling depth, or a sampling resolution.

9. The method of any one of claims 1-8, further comprising: generating control instructions based on the one or more configuration parameters; configuring one or more lasers of the additive manufacturing system using the one or more configuration parameters; and fabricating the first subregion using the one or more lasers.

10. The method of any one of claims 1-9, wherein the first feature is one of an overhang, a solid structure, a thin wall, a through hole, or a lattice structure.

11. The method of any one of claims 1-10, further comprising: providing at least the first layer to a machine-learning model as one or more inputs; and executing the machine-learning model to generate the one or more configuration parameters as one or more outputs of the machine-learning model.

12. The method of claim 11, further comprising: obtaining a plurality of 3-D models; generating training data based on the plurality of 3-D models; and training the machine-learning model using the training data.

13. The method of claim 11, wherein the machine-learning model is a neural network.

14. The method of any one of claims 1-13, wherein the additive manufacturing system is a laser bed powder fusion additive manufacturing system.

15. The method of any one of claims 1-14, further comprising: generating a visualization of a plurality of contours of the first layer, the plurality of contours representing a portion of the first feature and including at least a first contour; calculating the surrounding volume ratio based on the first contour; and identifying the first contour as the first subregion based on determining that the surrounding volume ratio satisfies the threshold.

16. The method of any one of claims 1-15, further comprising: determining that the first subregion is associated with a subregion identifier; and mapping the subregion identifier to the one or more configuration parameters in a datastore.

17. The method of claim 1, wherein the threshold is a surrounding volume ratio threshold representing a band of surrounding volume ratios.

18. The method of any one of claims 1-17, further comprising fusing precursor material with laser energy from one or more laser energy sources of the additive manufacturing system to form the part on a build surface of the additive manufacturing system.

19. A part manufactured using the method of claim 18.

20. An apparatus comprising: instructions; at least one memory to store the instructions; and at least one processor to execute the instructions to: slice a three-dimensional model of a part to be fabricated by a additive manufacturing system into at least one layer including a first layer; determine a surrounding volume ratio of a first volume of fused material surrounding one or more voxels and a total volume surrounding the one or more voxels for respective features of the first layer, the respective features including a first feature; identify at least one portion of the first feature as a first subregion of the first layer based on determining that a surrounding volume ratio of the at least one portion satisfies a threshold; identify one or more configuration parameters of the additive manufacturing system based on a mapping of at least the surrounding volume ratio to the one or more configuration parameters; and cause fabrication of the first subregion of the first layer of the part using the one or more configuration parameters.

21. At least one non-transitory computer-readable medium comprising instructions that, when executed, cause at least one processor to execute the instructions to at least: slice a three-dimensional model of a part to be fabricated by an additive manufacturing system into at least one layer including a first layer; determine a surrounding volume ratio of a first volume of fused material surrounding one or more voxels and a total volume surrounding the one or more voxels for respective features of the first layer, the respective features including a first feature;identify at least one portion of the first feature as a first subregion of the first layer based on determining that a surrounding volume ratio of the at least one portion satisfies a threshold; map at least the surrounding volume ratio to one or more configuration parameters of the additive manufacturing system; and cause fabrication of the first subregion of the first layer of the part using the one or more configuration parameters.

22. An additive manufacturing system comprising: a build surface; one or more laser energy sources; an optics assembly movable relative to the build surface and configured to direct laser energy from the one or more laser energy sources toward the build surface to melt at least a portion of a layer of material disposed on the build surface; and a controller to: slice a three-dimensional model of a part to be fabricated by the additive manufacturing system into at least one layer including a first layer; determine a surrounding volume ratio of a first volume of fused material surrounding one or more voxels and a total volume surrounding the one or more voxels for respective features of the first layer, the respective features including a first feature; identify at least one portion of the first feature as a first subregion of the first layer based on determining that a surrounding volume ratio of the at least one portion satisfies a threshold; map at least the first surrounding volume ratio to one or more configuration parameters of at least one of the one or more laser energy sources or the optics assembly; and cause the optics assembly to, based on the one or more configuration parameters, direct laser energy from the one or more laser energy sources towards the build surface to melt at least a portion of a layer of material disposed on the build surface to fabricate the first subregion of the first layer of the part.

23. An apparatus comprising instructions, at least one memory to store the instructions; and at least one processor to execute the instructions to perform the method of any one of claims 1-18.

24. At least one non-transitory computer-readable medium comprising instructions that, when executed, cause at least one processor to execute the instructions to perform the method of any one of claims 1-18.

25. An additive manufacturing system comprising: a build surface; one or more laser energy sources; an optics assembly movable relative to the build surface and configured to direct laser energy from the one or more laser energy sources toward the build surface to melt at least a portion of a layer of material disposed on the build surface; and a controller to perform the method of any one of claims 1-18.

Images