Method and system for in-process monitoring of additive manufacturing

Abstract

Methods and systems for in-process monitoring of additive manufacturing. Systems and methods for in-process monitoring of additive manufacturing are described. These systems and methods utilize a predictive model trained to identify anomalies within a component that have the likelihood of causing manufacturing defects. The predictive model is trained at least by identifying one or more manufacturing anomalies related to additive manufacturing of the specimen, identifying one or more manufacturing defects within the resulting specimen, and performing a registration between the one or more manufacturing anomalies and the one or more manufacturing defects. The predictive model is used to monitor an additive manufacturing process and optionally inform update of process parameters.

Description

Technical Field

[0001] The present disclosure relates to systems and methods for additive manufacturing. More particularly, the disclosed embodiments relate to methods and systems for monitoring additive manufacturing. Background Art

[0002] Traditional quality control methods utilized in additive manufacturing include post-manufacturing methods, such as destructive testing of test specimens and non-destructive testing of finished products. These methods can be used to identify anomalies and defects in finished products but are insufficient for in-process monitoring. Furthermore, testing for defects in finished products can lead to unnecessary waste, as defects can only be identified after the manufacturing process has concluded and the material has been used. Therefore, improved monitoring in additive manufacturing will result in fewer defective components and less wasted material. Summary of the Invention

[0003] The present disclosure provides systems, apparatus, and methods related to in-process monitoring of additive manufacturing.

[0004] In some examples, methods and systems for in-process monitoring of additive manufacturing include: receiving and / or utilizing data related to additive manufacturing of a specimen; identifying one or more manufacturing anomalies in the specimen based on the received data; identifying one or more manufacturing defects within the specimen based on non-destructive testing; and performing registration between the one or more manufacturing anomalies and the one or more manufacturing defects using a machine learning model to generate a predictive model, the predictive model being configured to identify future manufacturing anomalies that will cause future manufacturing defects.

[0005] In some examples, methods and systems for in-process monitoring of an additive manufacturing process include: monitoring the additive manufacturing process using optical tomography; and analyzing the optical tomography data using a predictive model, the predictive model being configured to identify one or more manufacturing anomalies that will cause one or more manufacturing defects in a resulting component of the additive manufacturing process; wherein the predictive model is trained using the following steps: receiving historical optical tomography data related to additive manufacturing of a specimen; identifying one or more historical manufacturing anomalies in the specimen in the historical optical tomography data; identifying one or more manufacturing defects within the specimen; and performing alignment between the one or more manufacturing anomalies and the one or more manufacturing defects to obtain the predictive model.

[0006] In some examples, a computer-implemented method includes:

[0007] receiving data related to the additive manufacturing of the test specimen;

[0008] identifying one or more manufacturing anomalies in the specimen based on the received data;

[0009] identifying one or more manufacturing defects within the specimen based on non-destructive testing; and

[0010] A registration is performed between the one or more manufacturing anomalies and the one or more manufacturing defects using a machine learning model to generate a predictive model configured to identify future manufacturing anomalies that will lead to future manufacturing defects.

[0011] The computer-implemented method according to claim 1, wherein the additive manufacturing process comprises laser bed powder fusion.

[0012] The computer-implemented method according to the foregoing description, wherein the one or more manufacturing anomalies include a spatter anomaly.

[0013] The computer-implemented method according to claim 1, wherein the non-destructive testing comprises computed tomography (CT).

[0014] The computer-implemented method according to claim 1, wherein the CT utilizes x-ray image data.

[0015] The computer-implemented method according to claim 1, wherein the received data comprises image data, and wherein identifying one or more manufacturing anomalies in the specimen comprises identifying the one or more manufacturing anomalies in the image data using a computer vision algorithm.

[0016] According to the aforementioned computer-implemented method, wherein the computer vision algorithm includes an edge detection algorithm.

[0017] In some examples, a data processing system for in-process monitoring of an additive manufacturing process includes:

[0018] one or more processors;

[0019] Memory; and

[0020] a plurality of instructions stored in the memory and executable by the one or more processors to:

[0021] receiving data related to the additive manufacturing of the test specimen;

[0022] identifying one or more manufacturing anomalies in the specimen based on the received data;

[0023] identifying one or more manufacturing defects within the specimen based on non-destructive testing; and

[0024] A registration is performed between the one or more manufacturing anomalies and the one or more manufacturing defects using a machine learning model to generate a predictive model configured to identify future manufacturing anomalies that will lead to future manufacturing defects.

[0025] According to the aforementioned data processing system, wherein the machine learning model includes a classification model.

[0026] According to the aforementioned data processing system, the received data includes optical tomography data.

[0027] The data processing system according to the foregoing embodiment, wherein the non-destructive test includes computed tomography (CT).

[0028] According to the aforementioned data processing system, wherein the plurality of instructions are further executable by the one or more processors to identify one or more fractures caused by the stress test of the specimen to identify the one or more manufacturing defects.

[0029] According to the aforementioned data processing system, wherein the plurality of instructions can also be executed by the one or more processors to:

[0030] monitoring in-process additive manufacturing of components using optical tomography to obtain optical tomography data; and

[0031] The optical tomography data is analyzed using the predictive model.

[0032] According to the aforementioned data processing system, wherein the plurality of instructions can also be executed by the one or more processors to:

[0033] Changing one or more process parameters of the additive manufacturing process; and / or

[0034] Ceasing additive manufacturing in said process; and / or

[0035] Specifies that the component in question is to be scrapped.

[0036] In some examples, a computer-implemented method for in-process monitoring of an additive manufacturing process includes:

[0037] monitoring the additive manufacturing process using optical tomography to obtain optical tomography data; and

[0038] analyzing the optical tomography data using a predictive model configured to identify one or more manufacturing anomalies that will result in one or more manufacturing defects in a resulting component of the additive manufacturing process;

[0039] The prediction model is trained using the following steps:

[0040] receiving historical optical tomography data related to additive manufacturing of the specimen;

[0041] identifying one or more historical manufacturing anomalies in the specimen in the historical optical tomography data;

[0042] identifying one or more manufacturing defects within the specimen; and

[0043] A registration between the one or more manufacturing anomalies and the one or more manufacturing defects is performed to obtain the prediction model.

[0044] According to the aforementioned computer-implemented method, the method further includes:

[0045] monitoring the power output of a laser used in the additive manufacturing process to obtain laser power data; and

[0046] One or more process parameters are modified using the predictive model and the laser power data.

[0047] The computer-implemented method according to claim 1, wherein the one or more process parameters include at least one of a molten pool size, a molten pool temperature, and / or a laser power output.

[0048] The computer-implemented method according to the foregoing description, wherein the one or more historical manufacturing anomalies in the specimen are identified in the historical optical tomography data using a computer vision algorithm.

[0049] The computer-implemented method according to the foregoing description, wherein identifying the one or more manufacturing defects within the specimen comprises identifying one or more fractures resulting from stress testing of the specimen.

[0050] In some examples, an additive manufacturing apparatus is configured to monitor the manufacture of a resulting component using the aforementioned computer-implemented method.

[0051] The features, functions, and advantages can be achieved independently in various embodiments of the present disclosure or may be combined in yet other embodiments further details of which can be seen with reference to the following description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 is a schematic diagram depicting a system for in-process monitoring of an additive manufacturing process according to aspects of the present disclosure.

[0053] Figure 2 Is described by the configuration Figure 1 Schematic diagram of the additive manufacturing device monitored by the system.

[0054] Figure 3 is a flow chart depicting the steps of a method for in-process monitoring of an additive manufacturing process according to aspects of the present disclosure.

[0055] Figure 4 It is a depiction Figure 3 Flowchart of the steps of a registration method between optical tomography data used in the method and calculated tomography data.

[0056] Figure 5 is a schematic diagram depicting a data processing system in accordance with aspects of the present disclosure.

[0057] Figure 6 is a schematic diagram depicting a network data processing system according to aspects of the present disclosure. DETAILED DESCRIPTION

[0058] Various aspects and examples of methods and systems for monitoring in the process of additive manufacturing are described below and shown in the relevant drawings. Unless otherwise indicated, the additive manufacturing system and / or its various components in accordance with the process of the present teachings may include at least one of the structures, components, functions and / or variations described, shown and / or included herein. In addition, unless specifically excluded, the process steps, structures, components, functions and / or variations described, shown and / or included herein in conjunction with the present teachings may be included in other similar devices and methods, including interchangeable between the disclosed embodiments. The following description of various examples is merely illustrative in nature and is in no way intended to limit the present disclosure, its applications or uses. In addition, the advantages provided by the examples and embodiments described below are illustrative in nature, and not all examples and embodiments provide the same advantages or the same degree of advantages.

[0059] This detailed description includes the following sections: (1) Definitions; (2) Overview; (3) Examples, Components, and Alternatives; (4) Advantages, Features, and Benefits; and (5) Conclusion. The "Examples, Components, and Alternatives" section is further divided into subsections, which are labeled accordingly.

[0060] definition

[0061] Unless otherwise indicated, the following definitions apply herein.

[0062] "Comprise," "include," and "have" (and conjugations thereof) are used interchangeably to mean inclusion, but not necessarily limitation, and are open-ended terms that are not intended to exclude additional unrecited elements or method steps.

[0063] Terms such as "first," "second," and "third" are used to distinguish or identify various members of a group and the like and are not intended to imply sequential or numerical limitations.

[0064] "AKA" means "also known as," and may be used to indicate alternative or corresponding terms for a given element.

[0065] "Processing logic" describes any suitable device or hardware configured to process data by performing one or more logical and / or arithmetic operations (e.g., executing coded instructions). For example, processing logic may include one or more processors (e.g., central processing units (CPUs) and / or graphics processing units (GPUs)), microprocessors, clusters of processing cores, FPGAs (field programmable gate arrays), artificial intelligence (AI) accelerators, digital signal processors (DSPs), and / or any other suitable combination of logic hardware.

[0066] A "controller" or "electronic controller" includes processing logic programmed with instructions to perform control functions with respect to a control element. For example, an electronic controller may be configured to receive an input signal, compare the input signal to a selected control value or setpoint value, and determine an output signal to a control element (e.g., a motor or actuator) to provide corrective action based on the comparison. In another example, an electronic controller may be configured to interface between a host device (e.g., a desktop computer, mainframe, etc.) and a peripheral device (e.g., a memory device, an input / output device, etc.) to control and / or monitor input and output signals to and from the peripheral device.

[0067] Overview

[0068] In general, methods and systems for in-process monitoring of additive manufacturing include: receiving data related to additive manufacturing of a specimen; identifying anomalies related to the additive manufacturing process (e.g., spatter, etc.) within the specimen based on the received data; identifying post-manufacturing defects (e.g., fractures) in the specimen; training a predictive model by performing alignment between the anomalies and the resulting defects; and utilizing the predictive model to monitor the production of additional components.

[0069] In some examples, the additive manufacturing process includes laser bed powder fusion. Additionally or alternatively, the additive manufacturing process can include direct energy deposition. Thus, the data received related to the additive manufacturing of the specimen can include data related to an additive manufacturing method that utilizes a focused energy source (e.g., a laser, plasma arc, or electron beam). For example, the received data can include melt pool data (e.g., melt pool size, melt pool temperature, etc.), power output and / or beam diameter of the focused energy source, scan speed, and / or other process parameters utilized in the additive manufacturing process.

[0070] In some examples, the received data includes image data, such as optical tomography (OT) image data from an optical tomography device. In some examples, the OT image data may be captured by the optical tomography device continuously or at a predetermined sampling rate. Additionally or alternatively, the OT image data may be captured by the optical tomography device at predetermined stages of the additive manufacturing process.

[0071] Identifying anomalies in the specimen includes analyzing received data (e.g., image data) to identify one or more anomalies (e.g., spatter) that occur at any stage of the additive manufacturing process. In some examples, OT image data of the additive manufacturing process can be analyzed by one or more computer vision algorithms to determine the presence of one or more anomalies in the manufactured material. For example, the one or more computer vision algorithms can include edge detection algorithms and / or image recognition algorithms configured to detect anomalies within a manufactured layer of the specimen.

[0072] In some examples, identifying anomalies in the received data may include using an image classification model configured to classify portions of the OT image data into different categories based on the presence of features corresponding to spatter and / or other manufacturing anomalies. For example, identifying anomalies may include classifying an area of the manufactured material as normal or irregular.

[0073] In some examples, identifying anomalies may include segmenting the OT image data into two or more portions, such as segmenting a melt pool from surrounding material in the OT image data. In this manner, irregularity identification may be performed on one, multiple, and / or all segmented portions in the OT image data.

[0074] In some examples, identifying anomalies in the received data may include analyzing gradients in the OT image data, such as identifying portions of the OT image data that include irregular gradients and / or areas of high contrast between adjacent pixels, which may indicate anomalies in the material.

[0075] In some examples, one or more mechanical fatigue testing processes (AKA stress testing) may be applied to the specimen after fabrication. For example, the specimen may be subjected to one or more compression tests, axial fatigue tests, torsional fatigue tests, and the like.

[0076] After manufacturing and / or stress testing of the specimen, one or more manufacturing defects (e.g., fractures) may be identified. Identifying the one or more manufacturing defects may include utilizing one or more non-destructive testing procedures, such as computed tomography (CT). In some examples, CT may utilize x-ray imaging to obtain CT image data.

[0077] In some examples, identifying manufacturing defects in the specimen includes analyzing CT image data to identify one or more defects (e.g., fractures) present within the specimen. In some examples, the CT image data can be analyzed by one or more computer vision algorithms to determine the presence of one or more fractures in the manufactured material. For example, the one or more computer vision algorithms can include an edge detection algorithm and / or an image recognition algorithm configured to detect fractures within the specimen.

[0078] In some examples, as with the irregularity identification described above, identifying manufacturing defects in the CT image data may include using an image classification model configured to classify portions of the CT image data into different classifications based on the presence of features corresponding to manufacturing defects.

[0079] In some examples, identifying manufacturing defects in CT image data may include analyzing gradients in the CT image data, such as identifying portions of the CT image data that include irregular gradients and / or areas of high contrast between adjacent pixels, which may indicate a fracture in the material.

[0080] Typically, a predictive model is trained on identified anomalies and manufacturing defects to predict whether an anomaly detected during manufacturing will result in a manufacturing defect. Thus, the predictive model is configured to generate a defect prediction for a given detected anomaly, i.e., predict whether the anomaly will result in a manufacturing defect.

[0081] In some examples, the defect prediction may include a classification between two anomaly types (e.g., a benign anomaly type and a defect-causing anomaly type). In some examples, the defect prediction may include a percentage likelihood that the anomaly will develop into a manufacturing defect. In some examples, the defect prediction may include a ranking scale for anomaly types, e.g., the prediction model may rank anomalies on a scale of potential defect severity.

[0082] Typically, the predictive model is trained by performing a registration between one or more manufacturing anomalies and one or more manufacturing defects, such that the predictive model is configured to identify manufacturing anomalies that will result in manufacturing defects.

[0083] In some examples, performing registration between anomalies and defects can include utilizing a classification model configured to classify the anomaly into two or more categories based on a comparison with the resulting defect. For example, the prediction model can classify the anomaly as: a) an anomaly that will not result in a manufacturing defect or b) an anomaly that will result in a manufacturing defect.

[0084] In some examples, performing registration between anomalies and defects may include utilizing feature extraction on relevant image data (e.g., OT image data and CT image data) related to the detected anomaly and the detected defect, such as performing edge detection, pixel density analysis, etc., to determine corresponding features of the anomaly and the defect. The extracted features may then be matched between the anomaly and the defect, for example, by identifying a correspondence between features in the OT image data and corresponding features in the CT image data.

[0085] After training, the predictive model is configured to receive in-process monitoring data related to the additive manufacturing of non-test components (e.g., production components) and identify anomalies that may cause defects in the finished product, thereby enabling proactive adjustments to the manufacturing process.

[0086] For example, during manufacturing, an optical tomography imaging device is used to monitor the manufacturing process, thereby generating in-process OT image data. This in-process OT image data can be processed using one or more computer vision algorithms, as described above for test specimens, to identify anomalies. The detected anomalies can be analyzed using a predictive model to determine a corresponding defect prediction for each detected anomaly.

[0087] In the event that a detected anomaly is predicted to result in a defect, the methods and systems described herein can be configured to modify the current manufacturing process by: (i) changing one or more process parameters of the additive manufacturing process; (ii) stopping the additive manufacturing process; and / or (iii) designating a component to be scrapped.

[0088] In some examples, changing one or more process parameters of the additive manufacturing process during the process may include changing: laser power output (e.g., adjusting the laser intensity in laser bed powder fusion and / or direct energy deposition to ensure proper melting and solidification of the material), melt pool size, melt pool temperature, scanning speed, thickness of the layer being manufactured, atmosphere control of the manufacturing chamber, gas flow rate, gas composition, powder bed temperature, etc.

[0089] Disclosed herein are technical solutions for in-process monitoring of additive manufacturing. Specifically, the disclosed systems and methods address technical problems associated with quality control techniques arising in the field of additive manufacturing, namely, ensuring that components are manufactured without defects or anomalies that would compromise the structural integrity of the manufactured component. The disclosed systems and methods provide an improved solution to this technical problem by utilizing a predictive model configured to identify anomalies in the manufacturing process that will result in defects in the finished product.

[0090] The disclosed systems and methods provide an integrated practical application of the principles discussed herein. Specifically, the disclosed systems and methods describe a specific way to monitor additive manufacturing during production that provides specific improvements over previous systems and results in improved manufactured components. Thus, the disclosed systems and methods apply (or utilize) the relevant principles in a meaningful and limited manner.

[0091] Various aspects of the systems and methods for in-process monitoring of additive manufacturing described herein may be embodied as computer methods, computer systems, or computer program products. Accordingly, various aspects of these systems and methods may take the form of all-hardware implementations, all-software implementations (including firmware, resident software, microcode, etc.), or implementations combining software and hardware aspects, all of which may be collectively referred to herein as "circuits," "modules," or "systems." Furthermore, various aspects of these systems and methods may take the form of a computer program product embodied in a computer-readable medium embodied with computer-readable program code / instructions.

[0092] Any combination of computer-readable media may be utilized. The computer-readable medium may be a computer-readable signal medium and / or a computer-readable storage medium. The computer-readable storage medium may include an electronic, magnetic, optical, electromagnetic, infrared, and / or semiconductor system, device, or apparatus, or any suitable combination of these. More specific examples of computer-readable storage media may include the following: an electrical connection with one or more leads, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, and / or any suitable combination of these, etc. In the context of the present disclosure, a computer-readable storage medium may include any suitable non-transitory tangible medium that can contain or store a program used by or in conjunction with an instruction execution system, apparatus, or device.

[0093] A computer-readable signal medium may include, for example, a propagated data signal embodying computer-readable program code, either in baseband or as part of a carrier wave. Such a propagated signal may take any of a variety of forms, including but not limited to electromagnetic, optical, and / or any suitable combination thereof. A computer-readable signal medium may include any computer-readable medium that is not a computer-readable storage medium and that is capable of communicating, propagating, or transporting a program for use by or in connection with an instruction execution system, apparatus, or device.

[0094] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, radio frequency (RF), etc., and / or any suitable combination of these.

[0095] The computer program code for performing the operations of various aspects of these systems and methods can be written in one or any combination of programming languages, including object-oriented programming languages (e.g., Java, C++), traditional procedural programming languages (e.g., C), and functional programming languages (e.g., Haskell). Any suitable language can be used to develop mobile applications, including those previously mentioned as well as Objective-C, Swift, C#, HTML5, etc. The program code can be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), and / or can be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0096] The various aspects of these systems and methods are described below with reference to the flow charts and / or block diagrams of the methods, devices, systems and / or computer program products. The individual blocks in the flow charts and / or block diagrams and / or the combination of blocks can be implemented by computer program instructions. The computer program instructions can be programmed into or provided to processing logic (e.g., a processor of a general-purpose computer, a special-purpose computer, a field programmable gate array (FPGA) or other programmable data processing device) to generate a machine so that (e.g., machine-readable) instructions executed via the processing logic create a means for implementing the function / behavior specified in the flow charts and / or block diagram blocks.

[0097] Additionally or alternatively, these computer program instructions may be stored in a computer-readable medium, which may direct processing logic and / or any other suitable device to function in a specific manner so that the instructions stored in the computer-readable medium generate an article of manufacture including instructions that implement the functions / behaviors specified in the flowchart and / or block diagram blocks.

[0098] Computer program instructions may also be loaded onto processing logic and / or any other suitable device to cause a series of operational steps to be performed on the device to generate a computer-implemented process, such that the executed instructions provide a process that implements the functions / behaviors specified in the flowchart and / or block diagram blocks.

[0099] Any flow chart and / or block diagram in the accompanying drawings is intended to illustrate the architecture, function and / or operation of the possible implementation of the monitoring system and method in these processes. In this regard, each square block can represent a module, a fragment or a code portion, which includes one or more executable instructions for realizing the specified logical function. In some implementations, the function annotated in the square block may not occur in the order annotated in the accompanying drawings. For example, depending on the function involved, the two square blocks shown in succession can actually be performed substantially simultaneously, or the square block can sometimes be performed in the opposite order. The combination of each square block and / or square block can be realized by a system (or a combination of special hardware and computer instructions) based on special hardware that performs the specified function or behavior.

[0100] Examples, Components, and Alternatives

[0101] The following sections describe selected aspects of exemplary systems and methods for in-process monitoring of additive manufacturing. The examples in these sections are intended to be illustrative and should not be construed as limiting the scope of the present disclosure. Each section may include one or more different embodiments or examples and / or contextual or related information, functionality, and / or structure.

[0102] A. Exemplary System for In-Process Monitoring of Additive Manufacturing

[0103] like Figure 1 and Figure 2 As shown, this section describes an illustrative system 100 for in-process monitoring of additive manufacturing. System 100 is an example of the systems described above.

[0104] System 100 is an in-process monitoring system for monitoring an additive manufacturing process 102. Additive manufacturing process 102 is an additive manufacturing process configured to produce a structured component from raw materials. In some examples, the additive manufacturing process comprises a process that utilizes a focused energy source to melt and / or fuse a portion of the raw materials in a predetermined manner, thereby producing the resulting component. For example, additive manufacturing process 102 may comprise a laser bed powder fusion process. Additionally or alternatively, additive manufacturing process 102 may comprise a direct energy deposition process.

[0105] Coupon 104 is manufactured using additive manufacturing process 102. In some examples, coupon 104 is a test component having a form and / or structure specifically selected for mechanical testing of material properties of coupon 104. For example, coupon 104 may have a substantially cylindrical shape, a substantially planar shape, or another suitable shape for testing material properties. Additionally or alternatively, coupon 104 may have a form and / or structure substantially similar to a non-test (e.g., production) component. For example, coupon 104 may have a shape configured to replicate a portion of a production component.

[0106] During fabrication, optical tomography (OT) image data related to the additive fabrication of the specimen 104 is captured using the optical tomography device 106. In some examples, the optical tomography device 106 is a single-sensor device, such that OT image data is collected by a single source. In some examples, the optical tomography device 106 is a multi-sensor device, such that OT image data is collected by an array of OT imaging sources. In some examples, the OT image data may be captured by the optical tomography device 106 continuously or at a predetermined sampling rate. In some examples, the OT image data may be captured by the optical tomography device 106 at predetermined stages of the additive manufacturing process 102.

[0107] Analyzing the OT image data to identify anomalies (e.g., spatter) in the specimen that occurred during the additive manufacturing process. In some examples, the OT image data of the additive manufacturing process can be analyzed by one or more computer vision algorithms (e.g., edge detection algorithms and / or image recognition algorithms) to determine the presence of one or more anomalies in the manufactured material.

[0108] The OT image data may be analyzed using an image classification model configured to classify portions of the OT image data into different categories based on the presence of features corresponding to spatter and / or other manufacturing anomalies. For example, identifying anomalies may include classifying an area of manufactured material as normal or irregular.

[0109] The OT image data can be segmented into two or more portions, such as separating a melt pool of the additive manufacturing process 102 from surrounding material in the OT image data. In this manner, irregularity identification can be performed on one, multiple, and / or all of the segmented portions of the OT image data. Additionally or alternatively, identifying anomalies in the specimen 104 can include analyzing gradients in the OT image data, for example, to identify portions of the OT image data that include irregular gradients and / or areas of high contrast between adjacent pixels, which can indicate an anomaly in the specimen.

[0110] After fabrication, mechanical testing 108 may be applied to the specimen 104. Mechanical testing 108 is configured to subject the specimen 104 to one or more processes configured to provide mechanical fatigue. Thus, mechanical testing 108 may include any mechanical testing process suitable for testing the mechanical and / or material properties of the specimen 104. For example, mechanical testing 108 may include one or more compression tests, axial fatigue tests, and / or torsional fatigue tests.

[0111] In some examples, such as where coupon 104 has a substantially similar form / shape to a production component, mechanical testing 108 may include subjecting coupon 104 to conditions similar to those encountered by a production component. Similarly, in some examples, mechanical testing 108 may include subjecting coupon 104 to worst-case conditions that a production component may encounter. For example, mechanical testing 108 may include higher levels of stress, compression, and / or shock than a production component would typically encounter.

[0112] After mechanical testing 108, a fractographic analysis 110 is performed on the specimen 104. The fractographic analysis 110 includes identifying one or more manufacturing defects (e.g., breaks / discontinuities) within the specimen 104. Identifying the one or more manufacturing defects may include utilizing one or more non-destructive testing processes, such as computed tomography (CT) 112. In some examples, the CT 112 may utilize x-ray imaging, thereby obtaining CT image data.

[0113] Thus, the fractographic analysis 110 may include analyzing CT image data from the CT 112 to identify and / or analyze defects present within the specimen. In some examples, the fractographic analysis 110 includes utilizing one or more computer vision algorithms (e.g., edge detection algorithms and / or image recognition algorithms) to determine the presence of one or more defects in the specimen 104.

[0114] In some examples, the fracture analysis 110 may include using an image classification model configured to classify a portion of the CT image data into different categories based on the presence of features corresponding to a fracture. For example, the image classification may be used to classify a portion of the CT image data into: a) a portion of material without a defect and b) a portion of material with a defect.

[0115] In some examples, fracture analysis 110 may include analyzing gradients in the CT image data to, for example, identify portions of the CT image data that include irregular gradients and / or areas of high contrast between adjacent pixels, which may indicate fracture defects in the material.

[0116] A registration process 114 is performed on the anomalies identified in the OT image data and the defects identified in the CT image data, resulting in a predictive model 116 configured to identify manufacturing anomalies that will result in manufacturing defects.

[0117] The registration 114 may include utilizing a classification model configured to classify anomalies into two or more categories based on comparison with the resulting defects. For example, the prediction model may classify anomalies as: a) anomalies that will not result in a manufacturing defect or b) anomalies that will result in a manufacturing defect.

[0118] The registration 114 may include utilizing feature extraction of relevant image data (e.g., OT image data and CT image data) related to the detected anomaly and the detected defect, such as performing edge detection, pixel density analysis, etc., to determine corresponding features of the anomaly and the defect. The extracted features may then be matched between the anomaly and the defect, such as by identifying a correspondence between features in the OT image data and corresponding features in the CT image data.

[0119] After registration 114, the predictive model 116 is configured to predict whether an anomaly detected during manufacturing will result in a manufacturing defect. In other words, the predictive model 116 is configured to generate a defect prediction for a given detected anomaly, i.e., a prediction of whether the anomaly will result in a manufacturing defect. Thus, the predictive model is capable of receiving in-process monitoring data related to additive manufacturing of non-test components (e.g., production components) and identifying anomalies that may result in defects in the finished product, thereby enabling proactive adjustments to the manufacturing process.

[0120] In some examples, the defect prediction may include a classification between two anomaly types (e.g., a benign anomaly type and a defect-causing anomaly type). In some examples, the defect prediction may include a percentage likelihood that the anomaly will develop into a manufacturing defect. In some examples, the defect prediction may include a ranking scale for anomaly types, e.g., the prediction model may rank anomalies on a scale of potential defect severity.

[0121] In the event that the detected anomaly is predicted to result in a defect, the system 100 may optionally generate updated manufacturing instructions 118 and provide them to the additive manufacturing process 102 to modify the manufacturing process. The updated manufacturing instructions 118 may include at least one of: (i) modifying one or more process parameters of the additive manufacturing process; (ii) stopping the additive manufacturing process; and / or (iii) specifying a component to be scrapped.

[0122] In some examples, changing one or more process parameters of the additive manufacturing process during the process may include changing: laser power output (e.g., adjusting the laser intensity in laser bed powder fusion and / or direct energy deposition to ensure proper melting and solidification of the material), melt pool size, melt pool temperature, scanning speed, thickness of the layer being manufactured, atmosphere control of the manufacturing chamber, gas flow rate, gas composition, powder bed temperature, etc.

[0123] Steering Figure 2, an example additive manufacturing apparatus 200 is shown for use with the predictive model 116 and configured to perform additive manufacturing of a non-test component (e.g., manufactured component 202). In some examples, the additive manufacturing apparatus 200 is the same additive manufacturing apparatus that performs the additive manufacturing process 102. Alternatively, the additive manufacturing apparatus is a different additive manufacturing apparatus configured to utilize the same or substantially similar additive manufacturing process 102.

[0124] Additive manufacturing apparatus 200 includes an optical tomography (OT) imaging apparatus 204. In some examples, OT imaging apparatus 204 is OT imaging apparatus 106. Alternatively, OT imaging apparatus 204 may be a different OT imaging apparatus than apparatus 106, but configured to utilize the same or substantially similar OT process.

[0125] Optical tomography imaging device 204 is used to monitor the manufacture of manufactured component 202, thereby obtaining in-process OT image data. The in-process OT image data associated with manufactured component 202 may be processed using one or more computer vision algorithms (substantially similar to those described above with respect to test specimen 104) to identify in-process anomalies. The detected anomalies are analyzed by prediction model 116 to determine a corresponding defect prediction for each detected anomaly.

[0126] Therefore, in the event that a detected anomaly is predicted to result in a defect, the system 100 may optionally generate updated manufacturing instructions 118 and provide them to the additive manufacturing apparatus 200 to modify the manufacturing process. The updated manufacturing instructions 118 may include at least one of the following: (i) modifying one or more process parameters of the additive manufacturing process; (ii) stopping the additive manufacturing process; and / or (iii) specifying a component to be scrapped.

[0127] Changing one or more process parameters of the additive manufacturing process during the process may include changing: the laser power output of the additive manufacturing device 200 (for example, adjusting the laser intensity in laser bed powder melting and / or direct energy deposition to ensure proper melting and solidification of the material), the melt pool size, the melt pool temperature, the scanning speed, the thickness of the layer being manufactured, the atmosphere control of the manufacturing chamber, the gas flow rate, the gas composition, the powder bed temperature, etc.

[0128] B. Exemplary Methods for In-Process Monitoring of Additive Manufacturing

[0129] This section describes the steps of an exemplary method 300 for in-process monitoring of additive manufacturing; see Figure 3 and Figure 4 Aspects of the above-described systems and methods may be used in the following method steps. Where appropriate, references may be made to components and systems that can be used to perform the various steps. These references are for illustrative purposes only and are not intended to limit the possible ways of performing any particular step of the method.

[0130] Figure 3 is a flow chart showing steps performed in an exemplary method and may not list the complete process or all steps of the method. Figure 3 Various steps of method 300 are depicted in FIG. 3 , but these steps do not necessarily need to be performed in their entirety and, in some cases, may be performed simultaneously or in an order different from that shown.

[0131] Step 302 of method 300 includes receiving data related to additive manufacturing of a specimen, the received data including optical tomography (OT) image data from an optical tomography imaging device. The OT image data may include continuously captured image data and / or image data captured at a predetermined sampling rate.

[0132] In some examples, the received data includes data related to process parameters of the additive manufacturing method, such as melt pool data (e.g., the size of the melt pool, the temperature of the melt pool, etc.), the power output and / or beam diameter of the focused energy source, the scanning speed and / or other process parameters utilized in the additive manufacturing process.

[0133] Step 304 of method 300 includes identifying one or more manufacturing anomalies in the specimen. Identifying one or more manufacturing anomalies in the specimen includes analyzing OT image data of the additive manufacturing process using one or more computer vision algorithms (e.g., edge detection algorithms and / or image recognition algorithms) to determine the presence of one or more anomalies in the manufactured material.

[0134] In some examples, the OT image data is analyzed using an image classification model that is configured to classify portions of the OT image data into different categories based on the presence of features corresponding to spatter and / or other manufacturing anomalies. For example, identifying anomalies may include classifying an area of manufactured material as normal or irregular.

[0135] In some examples, the OT image data can be segmented into two or more portions, such as to separate a melt pool of an additive manufacturing process from surrounding material in the OT image data. In this manner, irregularity identification can be performed on one, multiple, and / or all of the segmented portions in the OT image data. Additionally or alternatively, identifying one or more anomalies can include analyzing pixel gradients in the OT image data to, for example, identify portions of the OT image data that include irregular gradients and / or areas of high contrast between adjacent pixels, which can indicate an anomaly in the specimen.

[0136] Step 306 of method 300 includes identifying one or more manufacturing defects within the specimen. Identifying the one or more manufacturing defects may include utilizing one or more non-destructive testing processes, such as computed tomography (CT). Accordingly, step 306 includes analyzing CT image data using one or more computer vision algorithms (e.g., edge detection algorithms and / or image recognition algorithms) to identify defects present within the specimen.

[0137] In some examples, step 306 includes utilizing an image classification model configured to classify the portion of the CT image data into different classifications based on the presence of features corresponding to the fracture. For example, image classification can be used to classify the portion of the CT image data into: a) a portion of material without a defect and b) a portion of material with a defect.

[0138] In some examples, step 306 may include analyzing gradients in the CT image data to, for example, identify portions of the CT image data that include irregular gradients and / or areas of high contrast between adjacent pixels, which may indicate fracture defects in the material.

[0139] Step 308 of method 300 includes performing a registration between one or more manufacturing anomalies and one or more manufacturing defects using a machine learning model to generate a prediction model configured to identify future manufacturing anomalies that will lead to future manufacturing defects. Figure 4 A further description of step 308 is provided.

[0140] Step 310 of method 300 includes monitoring the in-process additive manufacturing of a non-test component using optical tomography to acquire optical tomography data; and analyzing the optical tomography data using a predictive model. As described with respect to the test specimen in step 304, the optical tomography data is processed using one or more computer vision algorithms to identify in-process manufacturing anomalies. The detected anomalies are then analyzed using the predictive model to determine a corresponding defect prediction for each detected anomaly, where the defect prediction corresponds to a probability that the anomaly will result in a defect.

[0141] Optional step 312 of method 300 includes updating the manufacturing process based on the defect prediction. In some examples, this includes selecting one of the following updating processes: (i) changing one or more process parameters of the additive manufacturing process during the process; (ii) stopping the additive manufacturing process during the process; and / or (iii) designating a component to be scrapped.

[0142] Reference Figure 4 , the registration step 308 of method 300 includes one or more sub-steps. Although described below and in Figure 4Various sub-steps of step 308 are depicted in FIG, but these sub-steps do not necessarily need to be performed in full, and in some cases may be performed simultaneously or in an order different from that shown.

[0143] Sub-step 308A of step 308 includes extracting features of the anomaly identified in step 304. In some examples, this includes analyzing the optical tomography data using one or more computer vision algorithms, machine learning models, and / or statistical pattern recognition algorithms. For example, edge detection, pixel density analysis, and the like may be used in conjunction with models such as Bayesian classification, artificial neural network models, k-nearest neighbor models, and / or other classification / feature extraction models to extract features of the anomaly.

[0144] Sub-step 308B of step 308 includes extracting features of the defects identified in step 306. In some examples, this includes analyzing the computed tomography data using one or more computer vision algorithms, machine learning models, and / or statistical pattern recognition algorithms. For example, edge detection, pixel density analysis, and the like may be used in conjunction with models such as Bayesian classification, artificial neural network models, k-nearest neighbor models, and / or other classification / feature extraction models to extract features of the defects.

[0145] Sub-step 308C of step 308 includes performing a registration between the anomaly and the defect based at least in part on the corresponding extracted features. In some examples, the extracted features may be compared and / or matched between the anomaly and the defect, for example, by identifying correspondences between features in the OT image data and corresponding features in the CT image data. For example, the matched features may include shape, size, location, orientation, depth, and / or other characteristics related to the geometry and / or composition of the anomaly and / or defect.

[0146] In some examples, sub-step 308C includes utilizing a classification model configured to classify the anomaly into two or more categories based on comparison with the resulting defect. For example, the anomaly may be classified as: a) an anomaly that results in a manufacturing defect or b) an anomaly that does not result in a manufacturing defect.

[0147] C. Exemplary Data Processing System

[0148] like Figure 5As shown, this example describes a data processing system 500 (also referred to as a computer, computing system, and / or computer system) according to aspects of the present disclosure. In this example, data processing system 500 is an illustrative data processing system suitable for implementing aspects of systems and methods for in-process monitoring of additive manufacturing. More specifically, in some examples, a device (e.g., a smartphone, a tablet, a personal computer) as an embodiment of the data processing system can be used to train a predictive model, operate a predictive model, analyze optical tomography data, analyze computed tomography data, and / or analyze or utilize data related to an additive manufacturing process.

[0149] In this illustrative example, data processing system 500 includes system bus 502 (also referred to as a communications framework). System bus 502 may provide communications between processor unit 504 (also referred to as a processor), memory 506, persistent storage 508, communications unit 510, input / output (I / O) unit 512, codec 530, and / or display 514. Memory 506, persistent storage 508, communications unit 510, input / output (I / O) unit 512, display 514, and codec 530 are examples of resources that processor unit 504 may access via system bus 502.

[0150] Processor unit 504 is configured to execute instructions that may be loaded into memory 506. Depending on the particular implementation, processor unit 504 may include a number of processors, multiple processor cores, and / or a specific type of processor or processors (e.g., a central processing unit (CPU), a graphics processing unit (GPU), etc.). Processor unit 504 may also be implemented using a number of heterogeneous processor systems in which a main processor is present with secondary processors on a single chip. As another illustrative example, processor unit 504 may be a symmetric multi-processor system containing multiple processors of the same type.

[0151] Memory 506 and persistent storage 508 are examples of storage devices 516. A storage device may include any suitable hardware capable of storing information (eg, digital information), such as data, program code in functional form, and / or other suitable information, either temporarily or permanently.

[0152] Storage device 516 may also be referred to as a computer-readable storage device or a computer-readable medium. Memory 506 may include volatile memory 540 and non-volatile memory 542. In some examples, a basic input / output system (BIOS), containing basic routines to transfer information between elements within data processing system 500, such as during startup, may be stored in non-volatile memory 542. Persistent storage 508 may take various forms, depending on the particular implementation.

[0153] Persistent storage 508 may include one or more components or devices. For example, persistent storage 508 may include one or more devices, such as a magnetic disk drive (also known as a hard disk drive or HDD), a solid state drive (SSD), a floppy disk drive, a tape drive, a Jaz drive, a Zip drive, a flash memory card, a memory stick, or any combination thereof. One or more of these devices may be removable and / or portable, such as a removable hard disk drive. Persistent storage 508 may include one or more storage media, alone or in combination with other storage media, including optical disk drives such as compact disk ROM devices (CD-ROMs), CD recordable drives (CD-R drives), CD rewritable drives (CD-RW drives), and / or digital versatile disk ROM drives (DVD-ROMs). To facilitate connection of persistent storage 508 to system bus 502, a removable or non-removable interface, such as interface 528, is typically used.

[0154] Input / output (I / O) unit 512 allows for input and output of data with other devices (i.e., input devices and output devices) that may be connected to data processing system 500. For example, input devices may include one or more pointing and / or information input devices such as a keyboard, a mouse, a trackball, a stylus, a touch pad or touch screen, a microphone, a joystick, a game pad, a satellite dish, a scanner, a TV tuner card, a digital camera, a digital video camera, a web camera, and the like. These and other input devices may be connected to processor unit 504 through system bus 502 via interface ports. Suitable interface ports may include, for example, a serial port, a parallel port, a game port, and / or a universal serial bus (USB).

[0155] One or more output devices may use some of the same type of ports as the input devices (in some cases, the same physical ports). For example, a USB port may be used to provide input to the data processing system 500 and to output information from the data processing system 500 to the output devices. One or more output adapters may be provided for certain output devices (e.g., monitors, speakers, printers, etc.) that require special adapters. Suitable output adapters may include circuit cards (e.g., video cards and sound cards) that provide a means of connection between the output device and the system bus 502. Other devices and / or device systems may provide both input and output capabilities, such as a remote computer 560. The display 514 may include any suitable human-machine interface or other mechanism configured to display information to a user, such as a cathode ray tube (CRT), light emitting diode (LED), or liquid crystal display (LCD) monitor or screen.

[0156] Communications unit 510 refers to any suitable hardware and / or software that provides for communicating with other data processing systems or devices. Although communications unit 510 is shown as being internal to data processing system 500, in some examples it may be at least partially external to data processing system 500. Communications unit 510 may include internal and external technologies, such as modems (including conventional telephone-grade modems, cable modems, and DSL modems), ISDN adapters, and / or wired and wireless Ethernet cards, hubs, routers, and the like. Data processing system 500 may operate in a networked environment using logical connections to one or more remote computers 560. Remote computers 560 may include personal computers (PCs), servers, routers, network PCs, workstations, microprocessor-based devices, peer devices, smartphones, tablets, other network notebooks, and the like. Remote computers 560 generally include many of the elements described with respect to data processing system 500. Remote computers 560 may be logically connected to data processing system 500 via a network interface 562 connected to data processing system 500 via communications unit 510. The network interface 562 encompasses wired and / or wireless communication networks, such as local area networks (LANs), wide area networks (WANs), and cellular networks. LAN technologies may include fiber distributed data interface (FDDI), copper distributed data interface (CDDI), Ethernet, token ring, and the like. WAN technologies include point-to-point links, circuit switching networks (e.g., Integrated Services Digital Network (ISDN) and its variants), packet switching networks, and digital subscriber lines (DSL).

[0157] The codec 530 may include an encoder, a decoder, or both, including hardware, software, or a combination of hardware and software. The codec 530 may include any suitable device and / or software configured to encode, compress, and / or encrypt a data stream or signal for transmission and storage, and to decode, decompress, and / or decrypt the data stream or signal (e.g., for playback or video editing). Although the codec 530 is depicted as a separate component, the codec 530 may be included or implemented in a memory such as the non-volatile memory 542.

[0158] The non-volatile memory 542 may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, etc., or any combination thereof. The volatile memory 540 may include random access memory (RAM), which may act as external cache memory. The RAM may include static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), etc., or any combination thereof.

[0159] Instructions for the operating system, applications, and / or programs may be located in storage devices 516, which are in communication with processor unit 504 via system bus 502. In these illustrative examples, the instructions are in functional form in persistent storage 508. These instructions may be loaded into memory 506 for execution by processor unit 504. The processes of one or more embodiments of the present disclosure may be performed by processor unit 504 using computer-implemented instructions that may be located in memory (e.g., memory 506).

[0160] These instructions are referred to as program instructions, program code, computer usable program code, or computer readable program code that are executed by the processor in processor unit 504. The program code in different embodiments may be embodied on different physical or computer-readable storage media (e.g., memory 506 or persistent storage 508). Program code 518 may be located in functional form on a selectively removable computer-readable medium 520 and may be loaded or transferred to data processing system 500 for execution by processor unit 504. In these examples, program code 518 and computer-readable medium 520 form a computer program product 522. In one example, computer-readable medium 520 may include computer-readable storage medium 524 or computer-readable signal medium 526.

[0161] Computer-readable storage media 524 may include, for example, an optical or magnetic disk that is inserted or placed into a drive or other device that is part of persistent storage 508 for transfer to a storage device (e.g., a hard drive) that is part of persistent storage 508. Computer-readable storage media 524 may also take the form of persistent storage, such as a hard drive, thumb drive, or flash memory, that is connected to data processing system 500. In some cases, computer-readable storage media 524 may not be removable from data processing system 500.

[0162] In these examples, computer-readable storage media 524 is a non-transitory physical or tangible storage device used to store program code 518, rather than a medium that propagates or transmits program code 518. Computer-readable storage media 524 is also referred to as a computer-readable tangible storage device or a computer-readable physical storage device. In other words, computer-readable storage media 524 is a medium that can be touched by a person.

[0163] Alternatively, program code 518 can be transmitted to data processing system 500 (e.g., remotely via a network) using computer-readable signal media 526. Computer-readable signal media 526 can be, for example, a propagated data signal containing program code 518. For example, computer-readable signal media 526 can be an electromagnetic signal, an optical signal, and / or any other suitable type of signal. These signals can be transmitted via a communication link, such as a wireless communication link, a fiber optic cable, a coaxial cable, a lead wire, and / or any other suitable type of communication link. In other words, in the illustrative examples, the communication link and / or connection can be physical or wireless.

[0164] In some exemplary embodiments, program code 518 may be downloaded from another device or data processing system via a network to persistent storage 508 via computer-readable signal media 526 for use within data processing system 500. For example, program code stored in a computer-readable storage medium in a server data processing system may be downloaded from the server via a network to data processing system 500. The computer providing program code 518 may be a server computer, a client computer, or some other device capable of storing and transmitting program code 518.

[0165] In some examples, program code 518 may include an operating system (OS) 550. Operating system 550, which may be stored on persistent storage 508, controls and allocates resources of data processing system 500. One or more applications 552 utilize the operating system's management of resources via program modules 554 and program data 556 stored on storage device 516. OS 550 may include any suitable software system configured to manage and expose the hardware resources of computer 500 for sharing and use by applications 552. In some examples, OS 550 provides application programming interfaces (APIs) that facilitate connecting to different types of hardware and / or provides applications 552 with access to hardware and OS services. In some examples, certain applications 552 may provide additional services for use by other applications 552, such as in the case of so-called "middleware." Aspects of the present disclosure may be implemented with respect to various operating systems or combinations of operating systems.

[0166] The different components shown for data processing system 500 are not meant to provide architectural limitations to the manner in which different embodiments may be implemented. One or more embodiments of the present disclosure may be implemented in a data processing system that includes fewer components or components in addition to and / or in place of those shown for computer 500. Figure 5Other components shown in the example may vary from the depicted example. Various embodiments may be implemented using any hardware device or system capable of running program code. As an example, data processing system 500 may include organic components integrated with inorganic components and / or may be composed entirely of organic components (excluding humans). For example, a memory device may be composed of an organic semiconductor.

[0167] In some examples, processor unit 504 may take the form of a hardware unit having hardware circuitry specifically manufactured or configured for a specific purpose or to produce a specific result or progress. This type of hardware can perform operations without requiring program code 518 to be loaded from a storage device into memory to be configured to perform those operations. For example, processor unit 504 may be a circuit system, an application-specific integrated circuit (ASIC), a programmable logic device, or some other suitable type of hardware that is configured (e.g., pre-configured or reconfigured) to perform a number of operations. For a programmable logic device, for example, the device is configured to perform a number of operations and can be reconfigured at a later time. Examples of programmable logic devices include programmable logic arrays, field programmable logic arrays, field programmable gate arrays (FPGAs), and other suitable hardware devices. For this type of implementation, executable instructions (e.g., program code 518) may be implemented as hardware, for example, by specifying the FPGA configuration using a hardware description language (HDL) and then using the resulting binary file to (re)configure the FPGA.

[0168] In another example, data processing system 500 can be implemented as a collection of specialized state machines (e.g., finite state machines (FSMs)) based on an FPGA (or in some cases, an ASIC), which can allow critical tasks to be isolated and run on custom hardware. While a processor such as a CPU can be described as a shared, general-purpose state machine that executes instructions given to it, an FPGA-based state machine is constructed for a special purpose and can execute hard-coded logic without sharing resources. These systems are often used for safety-related and critical tasks.

[0169] In another illustrative example, processor unit 504 may be implemented using a combination of processors found in computers and hardware units. Processor unit 504 may have several hardware units and several processors configured to run program code 518. For this depicted example, some processing may be implemented in several hardware units, while other processing may be implemented in several processors.

[0170] In another example, the system bus 502 may include one or more buses such as a system bus or an input / output bus. Of course, the bus system may be implemented using any suitable type of architecture that provides for transmitting data between different components or devices attached to the bus system. The system bus 502 may include various types of bus structures, including a memory bus or storage controller, a peripheral bus or external bus, and / or a local bus using any available bus architecture (e.g., Industry Standard Architecture (ISA), Micro Channel Architecture (MSA), Extended ISA (EISA), Intelligent Drive Electronics (IDE), VESA Local Bus (VLB), Peripheral Component Interconnect (PCI), Card Bus, Universal Serial Bus (USB), Advanced Graphics Port (AGP), Personal Computer Memory Card International Association bus (PCMCIA), Firewire (IEEE 1394), and Small Computer System Interface (SCSI)).

[0171] In addition, communication unit 510 may include several devices for sending data, receiving data, or both. Communication unit 510 may be, for example, a modem or a network adapter, two network adapters, or some combination thereof. In addition, memory may be, for example, memory 506 or a cache such as may be found in an interface and memory controller hub that may be present in system bus 502.

[0172] D. Exemplary Distributed Data Processing System

[0173] like Figure 6 As shown, this example depicts a general network data processing system 600 (interchangeably referred to as a computer network, network system, distributed data processing system, or distributed network), aspects of which may be included in one or more exemplary embodiments of the systems and methods for process monitoring of additive manufacturing described herein. For example, communication between modules utilized in the aforementioned systems and / or devices configured to perform the steps of a computer-implemented method may be facilitated by using the network data processing system 600.

[0174] It should be understood that Figure 6 This is provided as an implementation example and is not intended to imply any limitations with regard to the environments in which different embodiments may be implemented. Many modifications to the depicted environments are possible.

[0175] Network system 600 is a network of devices (e.g., computers) and other components, each of which may be an example of data processing system 500. Network data processing system 600 may include network 602, which is a medium configured to provide communications links between the various devices and computers connected within network data processing system 600. Network 602 may include connections such as wired or wireless communication links, fiber optic cables, and / or any other suitable medium for transmitting and / or communicating data between network devices, or any combination thereof.

[0176] In the depicted example, a first network device 604 and a second network device 606 and one or more computer-readable memory or storage devices 608 are connected to the network 602. Network devices 604 and 606 are each an example of the data processing system 500 described above. In the depicted example, devices 604 and 606 are shown as server computers that communicate with one or more server data repositories 622 that can be used to store information local to the server computers 604 and 606. However, network devices may include, but are not limited to, one or more personal computers, mobile computing devices (e.g., personal digital assistants (PDAs), tablets, and smartphones), handheld gaming devices, wearable devices, tablet computers, routers, switches, voice gates, servers, electronic storage devices, imaging devices, media players, and / or other networked tools that can perform mechanical or other functions. These network devices can be interconnected via wired, wireless, optical, and other appropriate communication links.

[0177] Additionally, client electronic devices 610 and 612 and / or client smart device 614 may be connected to network 602. Each of these devices is described above with respect to Figure 5 An example of a data processing system 500 is described. Client electronic devices 610, 612, and 614 may include, for example, one or more personal computers, network computers, and / or mobile computing devices, such as personal digital assistants (PDAs), smartphones, handheld gaming devices, wearable devices, and / or tablet computers. In the depicted example, server 604 provides information (e.g., boot files, operating system images, and applications) to one or more of client electronic devices 610, 612, and 614. In the context of their relationship to a server (e.g., server computer 604), client electronic devices 610, 612, and 614 may be referred to as "clients." The client devices may communicate with one or more client data repositories 620, which may be used to store information local to the client (e.g., cookies and / or relevant contextual information). The network data processing system 600 may include more or fewer servers and / or clients (or no servers or clients) as well as other devices not shown.

[0178] In some examples, the first client electronic device 610 can transmit the encoded file to the server 604. The server 604 can store the file, encode the file, and / or send the file to the second client electronic device 612. In some examples, the first client electronic device 610 can send an uncompressed file to the server 604, which can compress the file. In some examples, the server 604 can encode the text, audio, and / or video information and send the information to one or more clients via the network 602.

[0179] The client smart device 614 may include any suitable portable electronic device capable of wireless communication and executing software, such as a smartphone or tablet. In general, the term "smartphone" may describe any suitable portable electronic device configured to perform the functions of a computer, typically with a touch screen interface, Internet access, and an operating system capable of running downloaded applications. In addition to making phone calls (e.g., via a cellular network), a smartphone may be capable of sending and receiving email, text and multimedia messages, accessing the Internet, and / or acting as a web browser. A smart device (e.g., a smartphone) may include features of other known electronic devices, such as a media player, a personal digital assistant, a digital camera, a video camera, and / or a global positioning system. A smart device (e.g., a smartphone) may be capable of communicating with a user via, for example, near field communication (NFC), Wi-Fi or mobile broadband network wirelessly connects with other smart devices, computers or electronic devices. A wireless connection can be established between smart devices, smart phones, computers and / or other devices to form a mobile network that can exchange information.

[0180] The data and program code located in the system 600 can be stored in or on a computer-readable storage medium (e.g., a networked storage device 608 and / or a persistent storage 508 of one of the network computers) as described above and can be downloaded to a data processing system or other device for use. For example, program code can be stored on a computer-readable storage medium on a server computer 604 and downloaded to a client 610 via the network 602 for use on the client 610. In some examples, the client data repository 620 and the server data repository 622 reside on one or more storage devices 608 and / or 508.

[0181] The network data processing system 600 can be implemented as one or more different types of networks. For example, the system 600 may include an intranet, a local area network (LAN), a wide area network (WAN), or a personal area network (PAN). In some examples, the network data processing system 600 includes the Internet, wherein the network 602 represents a global collection of networks and gateways that use the Transmission Control Protocol / Internet Protocol (TCP / IP) protocol suite to communicate with each other. The core of the Internet is the backbone of high-speed data communication lines between major nodes or host computers. Thousands of commercial, government, educational, and other computer systems can be used to route data and messages. In some examples, the network 602 can be referred to as a "cloud." In those examples, each server 604 can be referred to as a cloud computing node, and the client electronic device can be referred to as a cloud consumer, etc. Figure 6 It is intended as an example, not as an architectural limitation to any exemplary embodiments.

[0182] E. Illustrative Combinations and Additional Examples

[0183] This section describes additional aspects and features of systems and methods for in-process monitoring of additive manufacturing, which are presented as a series of paragraphs (without limitation), some or all of which may be designated alphanumerically for clarity and efficiency. Each of these paragraphs may be combined in any suitable manner with one or more other paragraphs and / or with the disclosure elsewhere in this application. Some of the following paragraphs explicitly reference and further qualify other paragraphs, thereby providing (without limitation) examples of some suitable combinations.

[0184] A0. A computer-implemented method comprising:

[0185] receiving data related to the additive manufacturing of the test specimen;

[0186] identifying one or more manufacturing anomalies in the test specimen based on the received data;

[0187] identifying one or more manufacturing defects within the test specimen based on non-destructive testing; and

[0188] A registration is performed between the one or more manufacturing anomalies and the one or more manufacturing defects using a machine learning model to generate a predictive model configured to identify future manufacturing anomalies that will lead to future manufacturing defects.

[0189] A1. The computer-implemented method of paragraph A0, wherein the additive manufacturing process comprises laser bed powder fusion.

[0190] A1.1 The computer-implemented method of paragraph A0, wherein the additive manufacturing process comprises direct energy deposition.

[0191] A2. The computer-implemented method of paragraph A1, wherein the data related to the additive manufacturing of the specimen includes melt pool data.

[0192] A3. The computer-implemented method of paragraph A2, wherein the melt pool data includes a size of the melt pool.

[0193] A4. The computer-implemented method of paragraphs A2 and / or A3, wherein the data related to the additive manufacturing of the specimen further includes a power output of a laser utilized in laser bed powder fusion.

[0194] A5. The computer-implemented method of any of paragraphs A2-A4, wherein the one or more manufacturing anomalies include a spatter anomaly.

[0195] A6. The computer-implemented method of any of paragraphs A0-A5, wherein the non-destructive testing comprises computed tomography (CT).

[0196] A6.1 The computer-implemented method of paragraph A6, wherein the CT utilizes x-ray image data.

[0197] A7. A computer-implemented method according to any of paragraphs A0-A6, wherein the machine learning model includes a classification model.

[0198] A8. The computer-implemented method of any of paragraphs A0-A7, wherein the received data includes image data; and

[0199] Wherein identifying one or more manufacturing defects in the test specimen includes identifying one or more defects in the image data using a computer vision algorithm.

[0200] A9. The computer-implemented method of paragraph A8, wherein the computer vision algorithm comprises an edge detection algorithm.

[0201] A10. A computer-implemented method as described in any of paragraphs A0-A9, wherein the received data includes optical tomography data.

[0202] A11. The computer-implemented method of any of paragraphs A0-A10, wherein identifying one or more manufacturing defects comprises identifying one or more fractures resulting from stress testing of the specimen.

[0203] A12. The computer-implemented method of any of paragraphs A0-A11, further comprising:

[0204] Monitoring in-process additive manufacturing of components using optical tomography to obtain optical tomography data; and

[0205] Analyzing optical tomography data using predictive models.

[0206] A12.1. The computer-implemented method of paragraph A12, further comprising:

[0207] Based on the analysis, (i) one or more process parameters of the in-process additive manufacturing are changed, (ii) the in-process additive manufacturing is stopped, and / or (iii) the component is designated for scrapping.

[0208] B0. A data processing system for in-process monitoring of an additive manufacturing process, the system comprising:

[0209] one or more processors;

[0210] Memory; and

[0211] A plurality of instructions stored in memory and executable by one or more processors to:

[0212] receiving data related to the additive manufacturing of the test specimen;

[0213] identifying one or more manufacturing anomalies in the test specimen based on the received data;

[0214] identifying one or more manufacturing defects within the test specimen based on non-destructive testing; and

[0215] A registration is performed between the one or more manufacturing anomalies and the one or more manufacturing defects using a machine learning model to generate a predictive model configured to identify future manufacturing anomalies that will lead to future manufacturing defects.

[0216] B1. The data processing system of paragraph B0, wherein the additive manufacturing process comprises laser bed powder fusion.

[0217] B1.1 A data processing system according to paragraph B0, wherein the additive manufacturing process includes direct energy deposition.

[0218] B2. The data processing system of paragraph B1, wherein the data related to the additive manufacturing of the specimen includes melt pool data.

[0219] B3. A data processing system according to paragraph B2, wherein the melt pool data includes the size of the melt pool.

[0220] B4. The data processing system of paragraphs B2 and / or B3, wherein the data related to the additive manufacturing of the specimen further includes the power output of the laser utilized in the laser bed powder fusion.

[0221] B5. The data processing system of any of paragraphs B2-B4, wherein the one or more manufacturing anomalies include a spatter anomaly.

[0222] B6. A data processing system according to any of paragraphs B0-B5, wherein the non-destructive testing includes computed tomography (CT).

[0223] B6.1 A data processing system according to B6, wherein the CT utilizes x-ray image data.

[0224] B7. A data processing system according to any of paragraphs B0-B6, wherein the machine learning model includes a classification model.

[0225] B8. A data processing system according to any of paragraphs B0-B7, wherein the received data includes image data; and

[0226] Among other things, the plurality of instructions may be further executed by the one or more processors to utilize a computer vision algorithm to identify one or more manufacturing defects.

[0227] B8.1. A data processing system according to paragraph B8, wherein the computer vision algorithm includes an edge detection algorithm.

[0228] B9. A data processing system according to any of paragraphs B0-B8.1, wherein the received data includes optical tomography data.

[0229] B10. The data processing system of any of paragraphs B0-B9, wherein the plurality of instructions are further executable by the one or more processors to identify one or more fractures resulting from stress testing of the specimen to identify one or more manufacturing defects.

[0230] B11. The data processing system of any of paragraphs B0-B10, wherein the plurality of instructions are further executable by the one or more processors to:

[0231] monitoring in-process additive manufacturing of components using optical tomography to obtain optical tomography data; and

[0232] Analyzing optical tomography data using predictive models.

[0233] B11.1. A data processing system according to paragraph B11, wherein the plurality of instructions are further executable by the one or more processors to:

[0234] Changing one or more process parameters of the additive manufacturing process; and / or

[0235] Stopping additive manufacturing in progress; and / or

[0236] Specify the components to be scrapped.

[0237] C0. A computer-implemented method for in-process monitoring of an additive manufacturing process, the method comprising:

[0238] Monitoring additive manufacturing processes using optical tomography; and

[0239] analyzing the optical tomography data using a predictive model configured to identify one or more manufacturing anomalies that will result in one or more manufacturing defects in a resulting component of the additive manufacturing process;

[0240] The prediction model was trained using the following steps:

[0241] receiving historical optical tomography data related to additive manufacturing of the specimen;

[0242] identifying one or more historical manufacturing anomalies in the specimen in the historical optical tomography data;

[0243] identifying one or more manufacturing defects within the test specimen; and

[0244] A registration between the one or more manufacturing anomalies and the one or more manufacturing defects is performed to obtain a predictive model.

[0245] C1. The computer-implemented method of paragraph C0, wherein one or more historical manufacturing defects within the specimen are identified using computed tomography (CT).

[0246] C2. The computer-implemented method of paragraph C0, wherein the CT utilizes x-ray image data.

[0247] C3. The computer-implemented method of any of paragraphs C0-C2, wherein the additive manufacturing process comprises laser bed powder fusion.

[0248] C3.1 A computer-implemented method as described in any of paragraphs C0-C2, wherein the additive manufacturing process includes direct energy deposition.

[0249] C3.2. The computer-implemented method of paragraph C3 or C3.1, further comprising:

[0250] Monitoring the melt pool of an additive manufacturing process to obtain melt pool data; and

[0251] The predictive model and the melt pool data are utilized to change one or more process parameters of the additive manufacturing process.

[0252] C3.2.1. The computer-implemented method of paragraph C3.2, wherein the melt pool data includes a size of the melt pool.

[0253] C3.3. The computer-implemented method of paragraph C3 or C3.1, further comprising:

[0254] Monitoring the power output of lasers used in additive manufacturing processes to obtain laser power data; and

[0255] The predictive model and the laser power data are used to change one or more process parameters.

[0256] C3.4 A computer-implemented method according to any of paragraphs C3.2-C3.3, wherein the one or more process parameters include at least one of melt pool size, melt pool temperature, and / or laser power output.

[0257] C3.5. The computer-implemented method of any of paragraphs C3-C3.4, wherein the one or more manufacturing anomalies and / or the one or more historical manufacturing anomalies include a spatter anomaly.

[0258] C4. A computer-implemented method according to any of paragraphs C0-C3.3, wherein the prediction model includes a classification model.

[0259] C5. The computer-implemented method of any of paragraphs C0-C4, wherein a computer vision algorithm is utilized to identify one or more historical manufacturing anomalies in the specimen in the historical optical tomography data.

[0260] C5.1. The computer-implemented method of paragraph C5, wherein the computer vision algorithm comprises an edge detection algorithm.

[0261] C6. The computer-implemented method of any of paragraphs C0-C5.1, wherein identifying one or more manufacturing defects within the specimen comprises identifying one or more fractures resulting from stress testing of the specimen.

[0262] C7. An additive manufacturing apparatus configured to monitor the manufacture of a resulting component using the computer-implemented method of any of paragraphs C0-C6.

[0263] Advantages, Features, and Benefits

[0264] The various embodiments and examples of methods and systems for in-process monitoring of additive manufacturing described herein offer several advantages over known solutions for monitoring additive manufacturing processes. For example, the illustrative embodiments and examples described herein allow for real-time detection and correction of manufacturing anomalies, which can reduce the incidence of defects in the final product.

[0265] Additionally, among other benefits, the illustrative embodiments and examples described herein allow for reduced material waste and associated costs by enabling proactive adjustments to the manufacturing process rather than relying on post-process testing.

[0266] Additionally, among other benefits, the illustrative embodiments and examples described herein allow for fine-tuning of process parameters such as laser power output, melt pool size, and scan speed, thereby increasing the quality of the manufactured components.

[0267] Additionally, the illustrative embodiments and examples described herein allow for improved quality control, reduced waste, and increased overall efficiency in the manufacturing process, among other benefits.

[0268] There is no known system or device that can perform these functions.However, not all embodiments and examples described herein provide the same advantages or the same degree of advantages.

[0269] in conclusion

[0270] The disclosure set forth above may encompass a plurality of different examples with independent practicality. Although each of these has been disclosed in its preferred form, the specific embodiments disclosed and illustrated herein should not be considered in a limiting sense, because numerous changes may be made. With respect to the use of section headings within this disclosure, these headings are only used for organizational purposes. The subject matter of the present disclosure includes all novel and non-obvious combinations and sub-combinations of the various elements, features, functions and / or properties disclosed herein. The accompanying claims specifically point out certain combinations and sub-combinations that are considered novel and non-obvious. Other combinations and sub-combinations of features, functions, elements and / or properties may be claimed in applications claiming priority to this application or a related application. Regardless of whether these claims are wider, narrower, equal or different in scope than the original claims, these claims are also considered to be included in the subject matter of the present disclosure.

Claims

1. A computer-implemented method (300), the method (300) comprising: receiving (302) data related to additive manufacturing of a test specimen (104); identifying (304) one or more manufacturing anomalies in the test specimen (104) based on the received data; identifying (306) one or more manufacturing defects within the test specimen (104) based on non-destructive testing; and A registration (308) is performed between the one or more manufacturing anomalies and the one or more manufacturing defects using a machine learning model to generate a predictive model (116) configured to identify future manufacturing anomalies that will lead to future manufacturing defects.

2. The computer-implemented method (300) of claim 1, wherein: The additive manufacturing process (102) includes laser bed powder fusion.

3. The computer-implemented method (300) of claim 1, wherein: The one or more manufacturing anomalies include a spatter anomaly.

4. The computer-implemented method (300) of claim 1, wherein: The non-destructive testing includes computed tomography (CT) (112).

5. The computer-implemented method (300) of claim 4, wherein: The CT (112) utilizes x-ray image data.

6. The computer-implemented method (300) of claim 1, wherein: The received data includes image data; and Wherein, identifying (304) one or more manufacturing anomalies in the specimen (104) includes identifying the one or more manufacturing anomalies in the image data using a computer vision algorithm.

7. The computer-implemented method (300) of claim 6, wherein: The computer vision algorithm includes an edge detection algorithm.

8. A data processing system (500) for in-process monitoring of an additive manufacturing process (102), the data processing system comprising: one or more processors (504); Memory (506); as well as a plurality of instructions stored in the memory (506) and executable by the one or more processors (504) to: receiving data related to additive manufacturing of a test specimen (104); identifying one or more manufacturing anomalies in the test specimen (104) based on the received data; identifying one or more manufacturing defects within the test specimen (104) based on non-destructive testing; and A registration is performed between the one or more manufacturing anomalies and the one or more manufacturing defects using a machine learning model to generate a predictive model (116) configured to identify future manufacturing anomalies that will lead to future manufacturing defects.

9. A computer-implemented method (300) for in-process monitoring of an additive manufacturing process (102), the method (300) comprising: monitoring (310) the additive manufacturing process (102) using optical tomography to obtain optical tomography data; as well as analyzing the optical tomography data using a predictive model (116) configured to identify one or more manufacturing anomalies that will result in one or more manufacturing defects in a resulting component of the additive manufacturing process (102); The prediction model (116) is trained using the following steps: receiving historical optical tomography data related to additive manufacturing of a specimen (104); identifying one or more historical manufacturing anomalies in the specimen (104) in the historical optical tomography data; identifying one or more manufacturing defects within the test specimen (104); as well as A registration between the one or more manufacturing anomalies and the one or more manufacturing defects is performed to obtain the prediction model (116).

10. An additive manufacturing apparatus (200) configured to monitor the manufacture of a resulting component using the computer-implemented method (300) according to claim 9.

Images