Sensor fusion with eddy current sensing

Abstract

Systems and methods for powder bed additive manufacturing can include sensing a powder bed with an eddy current (EC) sensor to obtain EC sensor measurements, sensing the powder bed with an auxiliary sensor to obtain topography measurements, and determining a property of the powder bed based on the EC sensor measurements and the topography measurements.

Description

[0001] Cross-reference to related applications

[0002] This disclosure claims the benefit of U.S. Provisional Patent Application No. 63 / 579,260, filed August 28, 2023, entitled “SENSOR FUSION WITH EDDY CURRENT FOR IN-SITU MONITORING”, and U.S. Provisional Patent Application No. 63 / 601,668, filed November 21, 2023, entitled “SENSOR FUSION WITH EDDY CURRENT, ON-AXIS PHOTODIODE ANDIR CAMERA”, both of which are incorporated herein by reference in their entirety. background Technical Field

[0003] This disclosure generally relates to additive manufacturing, and more specifically, to sensor fusion of eddy current sensors, coaxial photodiode sensors, and infrared cameras. Background Technology

[0004] Additive manufacturing processes can include the use of powder bed fusion (PBF) systems, which utilize a source of powder material and a set of energy sources, most commonly a high-energy laser, to selectively fuse powder material into at least one building block. The laser melts the powder layer by layer to create the building block, which ultimately becomes a solid object. In operation, a recoater distributes prescribed layers of powder onto a build platform within a powder bed. The laser is then directed to a selected location and activated to sinter the powder at predefined locations on the layer. After sintering is completed at selected locations of the current layer via laser melting, another layer of powder is applied via the recoater, and the process is repeated until the building block is complete. Summary of the Invention

[0005] The following is a simplified summary of one or more aspects of the invention to provide a basic understanding of these aspects. This summary is not a broad overview of all contemplated aspects and is neither intended to identify key or essential elements of all aspects, nor to define the scope of any or all aspects. Its sole purpose is to present some concepts of one or more aspects in a simplified form as a prelude to the more detailed description that follows.

[0006] The systems, methods, and apparatuses described herein, and various aspects thereof, can use multiple sensors to enhance the ability of additive manufacturing systems to produce high-quality building blocks. These sensors are operable to monitor and / or measure the properties of powder beds, including surface and subsurface features.

[0007] Therefore, combinations of various non-contact sensors for monitoring characteristics such as reflectivity, thermal characteristics, and voltage are described in various aspects. One example of a non-contact sensor is an eddy current sensor with a fixed "stand-off" distance above the powder bed, which can be used in several applications. Typically, these eddy current sensors can be mounted on a powder spreader or powder spreading mechanism to achieve desired proximity to the powder bed while ensuring an appropriate and regular distance for consistent sensing. The quality of powder spreading can vary based on many key variables, including the type of alloy constituting the powder, particle size distribution, moisture content, spreader travel speed, and others.

[0008] In some aspects, techniques for characterizing the uniformity of spread powder layers can include structured light scanning. This can be used to generate detailed topographic maps of the powder bed and can capture deviations from the nominal and / or average expected distribution.

[0009] One example aspect includes a method for determining the height characteristics of a powder bed captured for each layer in a build-up process. This method uses a local stand-off distance (as a function of its position above the powder bed) measured by an eddy current sensor to provide a qualitative understanding of the powder bed's morphological features. Utilizing this data by correlating it with at least one other type of sensor data can provide enhanced correction capabilities to improve the accuracy of the morphological feature knowledge, ensuring high-quality build-ups.

[0010] In another example, the sensitivity of eddy current sensors to high temperatures and temperature variations due to changes in impedance within the eddy current sensor impedance coil can be remedied by real-time monitoring of the powder bed temperature using an infrared camera during the additive manufacturing build-up operation. Thus, the powder bed temperature can be monitored using an infrared camera, and the conductivity measurements captured by the eddy current sensor impedance coil can be adjusted as a function of or relative to the temperature at each location above the powder bed.

[0011] In another example, eddy current sensors can "see through" a component down to the subsurface layer based on sensor size and sensitivity. Lack of fusion, streaks, agglomerates, and other defects that would otherwise be undetectable by photodiodes (which can only detect surface defects) can be detected by eddy current sensors. This allows for the repair of lack of fusion defects at the subsurface layer by remelting selected locations of the component at a higher layer closer to the surface.

[0012] To achieve the foregoing and related objectives, one or more aspects include the features fully described below and specifically pointed out in the claims. Certain illustrative features of one or more aspects are set forth in detail in the following description and drawings. However, these features indicate only a few of the various ways in which the principles of each aspect may be employed, and this specification is intended to include all such aspects and their equivalents.

[0013] The specific embodiments described below with reference to the accompanying drawings are intended to provide a description of various exemplary aspects, which are not intended to represent the only aspects in which this disclosure can be practiced. The term "exemplary" as used throughout this disclosure means "serving as an example, instance, or illustration" and is not necessarily to be construed as preferred or advantageous over other aspects presented in this disclosure. The detailed description includes specific details intended to provide a thorough and complete disclosure that fully communicates the scope of this disclosure to those skilled in the art. However, the techniques and methods of this disclosure can be practiced without these specific details. In some instances, well-known structures and components may be shown in block diagram form or omitted entirely to avoid obscuring the various concepts set forth throughout this disclosure. Attached Figure Description

[0014] The disclosed aspects will be described below in conjunction with the accompanying drawings, which are provided for illustrative purposes and not for limiting the disclosed aspects, wherein similar designations denote similar elements, wherein dashed lines may indicate optional elements, and wherein:

[0015] Figure 1A The PBF system according to aspects of this disclosure is shown after the slices of the component have been fused, but before the next layer of powder is deposited;

[0016] Figure 1B The PBF system according to aspects of this disclosure is shown after the slices of the component have been fused, but before the next layer of powder is deposited;

[0017] Figure 1C The present invention illustrates a PBF system in which the depositor is positioned to deposit powder on the top surface of the building block and the powder bed, in a stage bounded by the powder bed container wall.

[0018] Figure 1D A PBF system is shown in which, after a powder layer is deposited, an energy source fuses the powder with the building block in a stage according to aspects of this disclosure;

[0019] Figure 1E A functional block diagram of a 3-D printer system according to aspects of this disclosure is shown;

[0020] Figure 2This is a diagram illustrating sensor fusion opportunities according to aspects of this disclosure;

[0021] Figure 3 This is a diagram illustrating the data fusion steps according to aspects of this disclosure;

[0022] Figure 4 This is a diagram illustrating an eddy current sensor array on a powder bed with high variation, according to aspects of this disclosure;

[0023] Figure 5 This is a flowchart illustrating the transformation of raw data according to aspects of this disclosure into material data that can be used for AI / ML or other operations;

[0024] Figure 6 This is a diagram illustrating the morphology of a powder bed sensed via a structured light system according to aspects of this disclosure;

[0025] Figure 7 It is a diagram of sensor fusion according to aspects of this disclosure;

[0026] Figure 8 These are a pair of diagrams showing vortex arrays and powder beds with and without height variation according to aspects of this disclosure;

[0027] Figure 9 This is a flowchart illustrating the use of interval measurement variability according to aspects of this disclosure to adjust EC sensor readings;

[0028] Figure 10 This is a diagram illustrating the use of infrared camera-sensed temperature properties of a powder bed for fusion with data from an EC sensor array, according to aspects of this disclosure;

[0029] Figure 11 This is a flowchart illustrating the calibration of a temperature-based EC sensor array according to aspects of this disclosure;

[0030] Figure 12 This is a diagram illustrating the structural capabilities of the photodiode and EC sensor array according to aspects of this disclosure;

[0031] Figure 13 This is a flowchart illustrating the correlation analysis between verified photodiode signals and unfused (LoF) porosity based on data from the EC sensor array;

[0032] Figure 14 This is a flowchart illustrating a sensor fusion method; and

[0033] Figure 15 This is a flowchart illustrating a sensor fusion method. Detailed Implementation

[0034] The specific embodiments described below with reference to the accompanying drawings are intended to provide a description of various exemplary aspects, which are not intended to represent the only aspects in which this disclosure can be practiced. The detailed descriptions include specific details and are intended to provide a thorough and complete disclosure that fully communicates the scope of this disclosure to those skilled in the art. However, the techniques and methods of this disclosure can be practiced without these specific details. In some instances, well-known structures and components may be shown in block diagram form or omitted entirely to avoid obscuring the various concepts set forth throughout this disclosure.

[0035] Figures 1A to 1D A corresponding side view of an example 3-D printer system used for powder bed fusion build operations is shown.

[0036] In this example, the 3D printer system is a powder bed fusion (PBF) system 100. Figures 1A to 1D The PBF system 100 is shown during different phases of operation. Figures 1A to 1D The specific aspects shown are one of many suitable examples of PBF systems employing some of the fundamental principles providing this disclosure. It should also be noted that... Figures 1A to 1D Elements in other figures in this disclosure are not necessarily drawn to scale, but may be drawn larger or smaller for better illustration of the concepts described herein.

[0037] The PBF system 100 can be an electron beam PBF system 100, a laser PBF system 100, or other types of PBF system 100. Furthermore, other types of 3D printing, such as directional energy deposition, selective laser melting, binder jetting, etc., can be employed without departing from the scope of this disclosure.

[0038] The PBF system 100 may include a depositor 101 capable of depositing each layer of powder from at least one powder 117, an energy beam source 103 capable of generating an energy beam, a deflector 105 capable of applying the energy beam to fuse the powder material, and a build plate 107 capable of supporting at least one component (such as build 109). Although the terms “fusion” and / or “fusing” are used to describe the mechanical coupling of powder particles, other mechanical actions, such as sintering, melting, and / or other electrical, mechanical, electromechanical, electrochemical, and / or chemical coupling methods, are contemplated within the scope of this disclosure and / or associated with various aspects of this disclosure. In various embodiments, the energy beam source 103 may include a multimode ring laser configured to generate multiple beams, such as a first beam (which may be a point beam or a first ring beam) and a second ring beam surrounding the first beam. In various embodiments, the multimode ring laser can also be configured to generate beams of different powers (such as with beam power module 179, described in more detail below) and / or to adjust the beam with various optics, such as amplification, zooming, etc. (such as with optical module 189, described in more detail below). Although in Figures 1A to 1E While shown as separate components, the energy beam source 103, beam power module 179, and / or optical module 189 may be incorporated into a single component or two components in different ways, as will be readily understood by those skilled in the art.

[0039] The PBF system 100 may also include a build floor 111 positioned within a powder bed container. The walls 112 of the powder bed container typically define the boundaries of the container, with a portion of the build floor 111 located laterally between the walls 112 and adjacent to the underlying build floor 111. The build floor 111 may gradually descend to allow the depositor 101 to deposit the next layer. The entire assembly may reside within a chamber 113, which may enclose other components to protect the equipment, allow for atmospheric and temperature regulation, mitigate contamination risks, and permit the recycling of unused powder. The depositor 101 may include at least one hopper 115. At least one hopper 115 may contain at least one powder 117, such as metal powder, alloy, or other materials. The depositor 101 may also include at least one leveler 119, which can level the top of each layer of deposited powder. The leveler 119 may be located in different positions in different aspects.

[0040] The AM process can generate a build object, and it can also generate various support structures that maintain the structural integrity of the build object during the AM process. When the build object is completed, the support structures may be unimportant to the build object and may need to be removed to reduce weight, improve energy distribution, improve aesthetics, or for other beneficial reasons. Figures 1A to 1DThe specific aspects shown are suitable examples of PBF systems with at least one hopper employing the principles of this disclosure. Specifically, the support structure and the interface between the support structure and the building object (which has characteristics different from those of the building object itself described herein) can be used... Figures 1A to 1D At least one PBF system 100 is described herein. Methods for selectively manufacturing various aspects according to desired results are also disclosed herein. While at least one method described in this disclosure can be applied to various AM processes (e.g., using PBF systems, such as…),… Figures 1A to 1D (as shown), but it will be understood that at least one method of this disclosure can also be applied to other applications. For example, at least one method described herein can be used in other manufacturing professions or fields without departing from the scope of this disclosure. Therefore, AM processes employing at least one method of this disclosure will be considered illustrative and are not intended to limit the scope of this disclosure.

[0041] For details, please refer to the following: Figure 1A The figure shows the PBF system 100 after the slices of component 109 have been fused, but before the next layer of powder is deposited. In fact, Figure 1A This illustrates the current state of the PBF system 100 after it has deposited and fused slices in multiple layers (e.g., 150 layers) to form a building block 109 (e.g., formed from 150 slices). The multiple deposited layers create a powder bed 121, which includes deposited but not fused powder.

[0042] In various aspects, the powder in powder bed 121 can be beneficially harvested, recaptured, and / or recycled for use in the same or other projects. This can reduce waste, lower costs, and provide other benefits.

[0043] Figure 1B A PBF system 100 is shown in which the construction floor 111 can reduce the powder layer thickness 123. The reduction of the construction floor 111 causes the builders 109 and powder bed 121 to decrease in powder layer thickness 123, such that the tops of the builders and powder bed are lower than the tops of the powder bed container wall 112 by an amount equal to the powder layer thickness. In this way, for example, a space with a consistent thickness equal to the powder layer thickness 123 can be created on top of the builders 109 and powder bed 121.

[0044] Figure 1CA PBF system 100 is illustrated in which a depositor 101 is positioned to deposit powder 117 onto a space created and bounded by a powder bed container wall 112 on the top surface of a component 109 and a powder bed 121. In this example, the depositor 101 moves gradually within the defined space while releasing powder 117 from a hopper 115. A leveler 119 can level the released powder to form a powder layer 125 having a thickness substantially equal to the powder layer thickness 123 (see [link to previous section]). Figure 1B The thickness of the powder layer 125 is such that the top surface 126 of the powder layer is exposed. Therefore, the powder in the PBF system can be supported by a powder material support structure, which may include, for example, a build plate 107, a build floor 111, a build member 109, a wall 112, and the like. It should be noted that the thickness of the powder layer 125 shown (i.e., the powder layer thickness 123) is... Figure 1B ()) greater than the references in this article Figure 1A The discussion involves the actual thickness of an example of 150 previously deposited layers.

[0045] Figure 1D The powder layer 125 in which the powder is deposited is shown. Figure 1C Following this, energy beam source 103 generates at least one energy beam 127, and deflector 105 applies the energy beam to fuse the next slice of the PBF system 100 in the building block 109. In various aspects, energy beam source 103 may be a laser, in which case energy beam 127 is a laser beam. Deflector 105 may include an optical system that uses reflection and / or refraction to manipulate the laser beam to scan the selected area to be fused. Although for simplicity and clarity, Figures 1A to 1D A single energy beam 127 is shown in the disclosure, but it should be understood from this disclosure that at least one energy beam can be generated, and selectively, according to various aspects thereof. Further description of such a multi-beam configuration is shown in this disclosure.

[0046] like Figures 1A to 1EAs shown, in various aspects, the optical module 189 can be communicatively coupled to the energy beam source 103 and / or deflector 105. The optical module may include additional components configured to perform various actions and selectively generate various characteristics. For example, the energy beam 127 may be amplified through an optical device at the energy beam source 103, deflector 105, or elsewhere, thereby selectively manipulating the spot size of the energy beam 127 to produce desired effects, such as increasing or decreasing the spot size. In such an example, the optical module 189 may, depending on the circumstances, include, integrate with, couple with, and / or control at least one optical component (such as at least one lens, positioner, motor, gimbal, actuator, prism, polarizer, filter, attenuator, marker, and / or other components of the optical system) to produce the desired amplification or reduction of the spot size. The optical module 189 may include at least one communication interface, memory, processor, and / or other components, depending on the circumstances and / or as needed.

[0047] Figures 1A to 1E Beam power module 179 is also shown. In various aspects, a ring-mode laser may be employed to introduce shaping (i.e., beam shaping) of the energy beam 127, which can selectively produce the desired effect at the point of application of the energy beam spot to the powder. Beam power module 179 may include at least one component configured to selectively modify and / or tune to the energy beam source and / or power delivery within the energy beam source. In some aspects, beam power module 179 may be integrated with energy beam source 103, while in other aspects, beam power module 179 may be self-contained or distributed elsewhere and communicatively coupled to energy beam source 103. The functionality of beam power module 179 is discussed in further detail herein. Beam power module 179 may include at least one communication interface, memory, processor, and / or other components, as applicable and / or as required.

[0048] Figures 1A to 1E At least one sensor 199 is also shown. In various aspects, one or more sensors 199 can be used to sense a region of material during the AM process. In various embodiments, one or more sensors 199 may include, for example, at least a photodiode, an optical tomography (OT) camera, or an eddy current sensor.

[0049] In various aspects, deflector 105 may include at least one gimbal and actuator capable of rotating and / or translating the energy beam source to position the energy beam. In various aspects, the energy beam source 103 and / or deflector 105 may modulate the energy beam, for example, by turning the energy beam on and off during deflector scanning, such that the energy beam is applied only to appropriate regions of the powder layer. For example, in various aspects, the energy beam may be modulated by a digital signal processor (DSP).

[0050] Figure 1E A functional block diagram of a PBF system 100 according to aspects of this disclosure is shown. It should be noted that, for clarity, Figure 1E It shows Figures 1A to 1D Some components are not shown in the diagram.

[0051] In this disclosure, control devices and / or elements (including computer software) may be coupled to PBF system 100 to control at least one component within PBF system 100. Such a device may be computer 150, which may include at least one component that can assist in the control of PBF system 100. Computer 150 may be communicatively and / or communicatively coupled to PBF system 100 and / or other AM systems via at least one wired and / or wireless interface 151. Computer 150 and / or interface 151 are examples of devices that can be configured to implement the various methods described herein, which can assist in the control of PBF system 100 and / or other AM systems.

[0052] In this disclosure, computer 150 may include at least one processor 152, memory 154, signal detector 156, digital signal processor (DSP) 158, and at least one user interface 160. Computer 150 may include additional components without departing from the scope of this disclosure.

[0053] Processor 152 may assist in the control and / or operation of PBF system 100. Processor 152 may also be referred to as a central processing unit (CPU). Memory 154, which may include both read-only memory (ROM) and random access memory (RAM), may store instructions and / or data and provide them to processor 152. A portion of memory 154 may also include non-volatile random access memory (NVRAM). Processor 152 typically performs logical and arithmetic operations based on program instructions stored in memory 154. Instructions in memory 154 may be executable (e.g., by processor 152) to implement the methods described herein.

[0054] Processor 152 may include or be a component of a processing system implemented with at least one processor. The at least one processor may be implemented with any combination of a general-purpose microprocessor, microcontroller, digital signal processor (DSP), field-programmable gate array (FPGA), programmable logic device (PLD), controller, state machine, gated logic, discrete hardware component, special-purpose hardware finite state machine, or any other suitable entity capable of performing computation or other manipulation of information.

[0055] Processor 152 may also include a machine-readable medium for storing software. Software should be interpreted broadly to mean any type of instruction, whether referred to as software, firmware, middleware, microcode, hardware description language, or others. Instructions may include code (e.g., in source code format, binary code format, executable code format, RS-274 instructions (G-code), numerical control (NC) programming languages, and / or any other suitable code format). Instructions, when executed by at least one processor, cause the processing system to perform the various functions described herein.

[0056] Signal detector 156 can be used to detect and quantify signals of any level received by computer 150 for use by processor 152 and / or other components of computer 150. Signal detector 156 can detect signals such as the power of energy beam source 103, deflector 105 positioning, build-up floor 111 height, amount of powder 117 remaining in depositor 101, position of depositor 101, nozzle position of hopper 115, pixel and / or voxel position, leveler 119 positioning, and other signals. DSP 158 can be used to process signals received by computer 150. DSP 158 can be configured to generate instructions and / or instruction packets for transmission to PBF system 100.

[0057] User interface 160 may include speakers, microphones, cameras, one or more sensors, keypads or keyboards, pointing devices, and / or displays, which in some respects may be touchscreens. User interface 160 may include any element or component or combination thereof that communicates information to a user of computer 150 and / or receives input from the user.

[0058] Various components of computer 150 can be coupled together via interface 151, which may include, for example, a bus system. Interface 151 may include a data bus, for example, as well as, in addition to a data bus, a power bus, a control signal bus, and a status signal bus. Components of computer 150 may be coupled together or use some other mechanism to accept or provide input to each other.

[0059] although Figure 1E Several individual components are shown, but at least one of the components can be combined or implemented jointly. For example, processor 152 can be used not only to implement the functions described herein with respect to processor 152, but also to implement the functions described herein with respect to signal detector 156, DSP 158, and / or user interface 160. Furthermore, Figure 1E Each of the components shown can be implemented using multiple individual elements.

[0060] Figure 1E The document also shows that at least one sensor 199 may be included. In some aspects, one or more sensors 199 may include at least one sensor. In various aspects, one or more sensors 199 may be configured as part of the PBF system 100, such as... Figure 1EAs shown, or may be included in a separate component coupled to the PBF system 100. In some aspects, one or more sensors 199 may be partially or wholly located within the chamber 113 and may be physically connected (e.g., mounted) and / or communicatively coupled to at least one other component of the PBF system 100, depending on the function of the particular sensor and its use in the PBF system 100. One or more sensors 199 may include at least eddy current sensors, eddy current sensor arrays, optical sensors, temperature sensors, motion sensors, audio sensors, chemical sensors, pressure sensors, weight sensors, distance sensors, proximity sensors, orientation sensors, speed sensors, rate sensors, acceleration sensors, electromagnetic sensors, radiation sensors, humidity / moisture sensors, and / or others, as appropriate and / or as required. One or more sensors 199 may include at least one communication interface, as appropriate and / or as required.

[0061] Figure 2Figure 200 illustrates opportunities for sensor fusion. As shown, differences can be found across at least one sensor type in various aspects. Examples of sensor types include thermal energy density (TED), thermal energy Planck (TEP), layer-by-layer imaging (which may include, for example, optical imaging of a powder bed), machine logs (which may include, for example, data on various states, parameters, etc. of a 3D printer), CT scans (which may include, for example, CT scan data of a build piece), and / or others. In additive manufacturing, processes employing laser deposition technology can use or employ at least one sensor to monitor the molten pool caused by guiding and activating a laser to at least one location on the powder bed and / or the build piece. Monitoring the characteristics of the molten pool can be difficult due to the rapidly changing nature of the molten pool when a laser is applied to the powder bed and / or the build piece. The monitored characteristics may include small and localized changes in temperature, material variations, chemical processes, fluid dynamics, cavitation, and / or other aspects. Therefore, at least one sensor can be finely tuned and highly sensitive to capture the rapidly changing characteristics of the additive manufacturing process. In many respects, deep learning via artificial intelligence and / or machine learning (AI / ML) can be used to monitor and detect changing characteristics, and can be used to identify defects in building blocks that may require remediation to maintain structural integrity, comply with various parameters and / or tolerances, and other relevant details. Therefore, one or more sensors and / or AI / ML can be employed in at least one feedback loop that further trains the AI / ML model to effectively map the path for root cause analysis of ML-based puzzles and / or problems. In some respects, at least one pass / fail criterion can allow the system to efficiently perform highly detailed and / or complex processes and builds.

[0062] Figure 3Figure 300 illustrates an example of data fusion. As shown, in various aspects, the data fusion example may include identifying at least one in-situ data stream. The in-situ data stream may be used for at least one statistical process control chart. Additionally or alternatively, the in-situ data stream may be used with at least one AI / ML model. The in-situ data stream may be a data stream created using sensor data captured by at least one sensor, including temperature sensors, eddy current sensors, chemical sensors, vision sensors (such as cameras), vision+ sensors (such as long-exposure cameras), photodiodes, voltage sensors, impedance sensors, pressure sensors, proximity sensors, and others. Examples of machine sensor data streams may include oxygen level or concentration sensors, carbon level or concentration sensors, other element or chemical level or concentration sensors, pressure sensors, or others. An example of a vision sensor is an 18-megapixel camera that captures camera data layer by layer. Photodiodes may be used to monitor electromagnetic (EM) emissions from the molten pool, such as thermal emissions. As each sensor senses data (e.g., in the data stream), the data may be captured in real time and processed and / or stored in memory for later processing. Data streams may include reference information, including timing information or timestamps and / or event markers. Various AI / ML components, systems, subsystems, models, and / or algorithms may receive and / or access data streams and use them for training (e.g., model training); for collecting, developing, and / or enhancing insights (e.g., through correlations, where the occurrence of an event can be measured and correlated with at least one parameter, including temperature, light, voltage, or other parameters) to predict future events and improve future additive manufacturing operations.

[0063] Figures 4 to 7 An example of sensor fusion using EC sensor measurements (which employ an eddy current sensor array) is shown.

[0064] Figure 4Figure 400 illustrates an example eddy current (EC) sensor array 199a on a powder bed 121 with high variation in various aspects. As shown, when powder is deposited in the powder bed 121, it may not result in a completely uniform or flat surface or distribution. This non-uniform distribution of powder can lead to a morphology in the powder within the powder bed 121 that includes various hills and valleys or troughs. This non-uniform distribution of powder can cause localized differences in material fusion and may result in small or even large variations in the components. This is important because in certain industries with high-performance vehicles (e.g., automotive or aerospace), tolerances are required within very narrow ranges to ensure the safety, stability, and integrity of vehicles (e.g., supercars, aircraft, or spacecraft). Manufacturing components outside the acceptable tolerances can lead to the need for scrapping or discarding the components, wasting valuable resources, including time and money. Using sensor arrays, such as the EC sensor array 199a, offers the advantage of ensuring higher measurement accuracy compared to a single sensor or a small number of sensors. In some respects, the EC sensor array 199a can detect anomalies, defects or other deviations that can be remedied by the system using components for smoothing the upper surface of the powder bed 121, depositing (or removing) powder from the powder bed 121 to create a more uniform surface for fusion with or as a component, and for remelting the component 109 when a defect is detected in the component 109.

[0065] Figure 5 This is a flowchart 500 illustrating the transformation of raw data into material data that can be used for AI / ML or other operations. In block 502, data can be acquired. Data acquisition may include using at least one sensor 199 to sense changes in height, temperature, voltage, impedance, and / or other properties of the powder bed 121 and / or component 109, and generating at least one data stream that can be used by the system. Data acquisition may cause sensor responses that result in or lead to the development of a measurement grid in block 504. The measurement grid 504 may be a highly detailed grid of the powder bed and / or component topography, which may be processed and / or stored in memory for later processing. The measurement grid 504 may be processed to generate property estimates of at least one property of the powder bed and / or component 109, including changes in height at various locations within the powder bed and / or component 109. These property estimates may be used to generate tables and / or be visually displayed via a user interface display for review by the system and / or operator, which may aid or assist in component development using the system.

[0066] Impedance is a complex property with variable aspects, including resistance (real part) and reactance (imaginary part). In practice, while eddy current sensor arrays (e.g., EC sensor array 199a) measure voltage changes, these changes indicate impedance variations caused by the interaction of the sensor with the powder and / or components in the powder bed material being analyzed. Impedance variations can be used to infer material properties such as conductivity, defects, or changes in material composition.

[0067] Once the EC sensor array 199a traverses the powder bed (e.g., via at least one path controlled by at least one processor via a motor and gantry or other mechanical or electromechanical components), EC sensor measurements are acquired for each section of the powder bed, and ultimately for selected sections or the entire powder bed. The EC sensor measurement data, in the form of a data stream, can be compared with topographic maps (e.g., point clouds) acquired using another sensor (such as structured light) (see...). Figure 6 (As described in connection with this article) in combination or in relation to other methods. Such combination and / or in relation to other methods, such as eddy current sensor arrays or structured light alone, can lead to an improved level of confidence in defect detection on powder beds and / or components.

[0068] Figure 6 Figure 600 illustrates the sensing of the morphology of a powder bed and / or component via a structured light system 199b. The structured light system 199b may include multiple cameras 199c and a projector 199d, which operate in combination to measure the morphological features of the powder bed 121 and / or component 109. In operation, the projector 199d may be pointed at selected locations on the powder bed 121 and / or component 109, and the cameras 199c may capture the reflectivity of light from the surfaces of the powder bed 121 and / or component 109, thereby determining the morphological features of the surface based on the light measurements.

[0069] In each aspect, the process of establishing relationships between anomalies identified by the EC sensor array 199a and / or the structured light system 199b involves utilizing statistical analysis or AI / ML methodologies. Given the sensitivity of sensor data from the EC sensor array 199a and / or the structured light system 199b to fluctuations under static and even rapidly changing conditions caused by the fusion operation, the effectiveness of correlations based solely on individual EC sensor arrays 199a and / or structured light systems 199b can be constrained by a relatively low level of accuracy. To address this limitation, a more robust approach can be employed using the aspects described herein, which involves incorporating multiple sensor modalities, as referenced... Figure 7 Further described. This integration is used to improve the accuracy and reliability of forecast results, thereby enhancing the overall confidence in the predicted results.

[0070] Figure 7 Figure 700 illustrates sensor fusion according to aspects of this disclosure. As shown, EC sensor data measured solely by EC sensor 199a can result in approximately 65% ​​defect prediction accuracy. Point cloud data measured solely by structured light system 199b can result in approximately 60% defect prediction accuracy. Combined EC sensor data measured by EC sensor array 199a and structured light system 199b can result in approximately 90% defect prediction accuracy. Therefore, the combined data, as a result of data fusion, leads to a significant increase in defect prediction and can thus be used to generate a much better understanding of the morphological features of powder bed 121 and / or component 109. Data fusion can be achieved by cross-referencing and / or correlating data, and by using statistical and / or AI / ML processes to identify inaccuracies and / or missing data in individual EC sensor data or point cloud data.

[0071] The EC sensor invariance transformation used for eddy current nondestructive evaluation signals will be discussed next. Figures 8 to 9 It is described that it illustrates sensor fusion using interval measurement results.

[0072] Regarding EC sensor distance management, one factor affecting the EC signal is the EC sensor measurement results. In practice, it can be difficult to track the values ​​used for EC sensor measurements, which can be used to accurately interpret EC data. Therefore, depending on various aspects, a scheme that keeps the influence of EC data on EC sensor measurement results constant may be beneficial.

[0073] Calibration via topography mapping can improve the accuracy of EC sensor array outputs, and the calibration procedure will now be described in detail from various aspects. Calibration can be achieved by utilizing topography maps of powder beds and / or components to correlate specific distance variations with voltage readings obtained from the EC sensor array 199a. By establishing a robust calibration model, uncertainties caused by interval distance fluctuations can be effectively compensated for, and their negative impacts minimized.

[0074] Figure 8 Figures 800a and 800b respectively illustrate EC sensor array 199a and powder beds 121 and / or components with and without height variations. As shown in Figure 800a, depending on various aspects, EC sensor array 199a on powder beds 121 and / or components 109 with no height variation (e.g., having a uniform powder distribution height) can result in constant interval measurement readings. As shown in Figure 800b, depending on various aspects, EC sensor array 199a on powder beds 121 and / or components 109 with height variations (e.g., without a uniform powder distribution height) can result in variable interval measurement readings.

[0075] Figure 9 This is a flowchart 900 illustrating the use of variability in interval measurement results to adjust EC sensor readings according to aspects of this disclosure. As shown, in block 902, an EC sensor array (e.g., EC sensor array 199a) can be used to measure voltage in block 904. In block 906, structured light (e.g., structured light 199b) can sense and generate a measured stand-off distance (MSOF). This MSOF can be compared with the nominal stand-off distance (NSOF). In block 908, if the MSOF is not equal to the NSOF (e.g., precisely or within an acceptable selected or preset range of variability), the EC sensor readings can be adjusted in block 910.

[0076] In some respects, sensor fusion can benefit from measuring the temperature at the location of the powder bed and / or the component.

[0077] Figure 10 Figure 1000 illustrates the use of infrared camera 199e to sense temperature properties of powder bed 121 and / or component 109 for fusion with data from EC sensor array 199a. As shown, EC sensor array 199a can be used to measure properties of powder bed and / or component, as described earlier herein. Using infrared (IR) camera 199e to measure temperature properties of powder bed 121 and / or component at selected locations may be advantageous, as temperature can be a variable that can affect the measurement results from EC sensor array 199a.

[0078] In detail, this may include various non-contact sensors for monitoring thermal or other properties. Although EC sensing via the EC sensor array 199a can be used to solve the transimpedance equation to calculate conductivity and EC sensor measurements, temperature variations in the powder bed 121 and / or building block 109 can adversely affect the accuracy of these heuristics.

[0079] In practice, the temperature sensitivity of the EC sensor array 199a is a result of the change in coil impedance due to its sensitivity to temperature variations. Therefore, the EC sensor array can be calibrated using an IR camera 199e, which monitors the temperature of the powder bed and / or building block at selected locations in real time.

[0080] Figure 11This is a flowchart 1100 illustrating the calibration of an EC sensor array based on temperature (e.g., as measured by an IR camera 199e). At block 1102, temperature data (Ti) can be measured. At block 1104, the EC sensor array voltage (Vi) can be measured. This can result in a dataset (Ti, Vi) in block 1106. In block 1108, the process can be iterated (i=i+1) and return to block 1102. Also at block 1106, the dataset (Ti, Vi) can continue to block 1110 to be used to generate a function of the sensor array voltage (V=f(T)) as a function of temperature, which can be used to calibrate the voltage readings of the eddy current sensor as a function of temperature.

[0081] Figure 12 Figure 1200 illustrates the structural capabilities of a photodiode and an EC sensor array. As shown, the photodiode may be limited in its ability to sense the condition of a powder bed and / or component because it can only detect light emitted from the top or surface layer of the powder bed and / or component, not the subsurface layer. The EC sensor array is not subject to this limitation and can sense the surface layer of the powder bed and / or component, as well as at least one subsurface layer, based on layer height values.

[0082] The morphology data generated from the measurements via the photodiode can be checked for accuracy and verified against measurements taken using the EC sensor array 199a in at least one subsurface layer of the powder bed 121 and / or component 109. If inaccuracies in the morphology data measured by the photodiode are determined by checking against the measurements taken using the EC sensor array 199a, these inaccuracies can be corrected in several ways. For example, if the morphology problem is determined to be a result of lack of fusion (LoF), and if the LoF defect is repaired, the morphology data from the photodiode can be corrected.

[0083] Figure 13This is a flowchart 1300 illustrating a correlation analysis between verified photodiode signals and unfused (LoF) porosity based on an EC sensor array. As shown, for layer n, where n represents the surface layer of powder bed 121 and / or component 109, block 1302 may include photodiode data measurements. These photodiode measurements can be used to generate a topography map in block 1304. For layer n+3 (i.e., the subsurface layer), block 1306 may include EC sensor array data measurements. These EC sensor array data measurements can be used to generate a topography map of at least one layer in block 1308. The topography map from the surface layer in block 1304 and the topography map of at least one layer in block 1308 can be input into an AI / ML algorithm in block 1310, which can output a LoF prediction in block 1312.

[0084] Figure 14 This is a flowchart illustrating method 1400 of a sensor fusion method. At block 1402, method 1400 includes sensing a powder bed with an eddy current (EC) sensor to obtain EC sensor measurements. For example, in one aspect, at least one processor 152, at least one memory 154, and / or at least one sensor (e.g., EC sensor array 199a) may be configured or may include means for sensing a powder bed 121 (which may include, for example, sensing a building block 109) to obtain EC sensor measurements. As used herein, "powder bed" generally includes powder in a bed, as well as other objects in the bed, such as building blocks (i.e., fused portions of the powder, which may include support structures or other structures that may eventually be removed from the finished portion), and other aspects or features of the bed. Thus, "sensing a powder bed" includes sensing the powder in the powder bed and / or sensing the building blocks in the powder bed and / or sensing other aspects of the powder bed, etc. Similarly, "height maintained above the powder bed" includes height maintained above the powder, height maintained above the building blocks, etc.

[0085] Sensing at block 1402 may include the EC sensor array 199a detecting selected locations of the building block and / or powder bed 121 and / or building block 109. This may include measurements based on a measurement grid and property estimation, which may be used to generate tables or displays, such as those guided and / or controlled by one or more processors 152. One or more processors 152 may execute instructions stored in one or more memories 154 to move the EC sensor array 199a to the selected and / or desired location.

[0086] Sensing powder bed 121 and / or component 109 using EC sensor array 199a to obtain EC sensor measurements can be performed to generate a first set of measurements that can be used in one or more sensor fusion operations to accurately predict the effect of energy beams applied to at least one location of powder bed 121 and / or component 109. This can lead to optimized resource utilization (e.g., time and money resources consumed during AM build operations) and the manufacture of accurate components.

[0087] At block 1404, method 1400 includes sensing a powder bed with at least one auxiliary sensor 199 (e.g., one or more structured light systems 199b, one or more photodiodes, one or more cameras, and / or one or more IR cameras) to obtain topographical measurements. For example, in one aspect, the topographical measurements may be obtained by one or more auxiliary sensors 199, which may be used to generate a topographic map as directed and / or controlled by one or more processors 152. The one or more processors 152 may execute instructions stored in one or more memories 154 to move one or more auxiliary sensors 199 and / or guide one or more auxiliary sensors 199 to selected and / or desired locations.

[0088] Sensing the powder bed 121 and / or the component 109 using one or more sensors 199 to obtain morphological measurements can be performed to generate a second set of measurements that can be used in one or more sensor fusion operations to accurately predict the effect of an energy beam applied to at least one location of the powder bed 121 and / or the component 109. This can lead to optimized resource utilization (e.g., time and money resources consumed during AM build operations) and the manufacture of accurate components.

[0089] At block 1406, method 1400 includes determining the properties of the powder bed and / or component based on EC sensor measurements and topographic measurements; in other words, based on a first set of measurements and a second set of measurements. For example, in one aspect, the EC sensor measurements and topographic measurements may be input to at least one algorithm stored in one or more memories 154, which is guided and / or controlled by one or more processors 152. The one or more processors 152 may execute instructions stored in one or more memories 15, which in some aspects may be AI / ML algorithms to accurately predict the effects of energy beams applied to at least one location of the powder bed 121 and / or component 109, which can lead to optimized resource utilization (e.g., time and monetary resources consumed during AM build operations) and the manufacture of accurate components.

[0090] In an alternative or additional aspect, at block 1402, method 1400 may further include maintaining the height of the EC sensor array 199a above the powder bed. For example, in one aspect, the height of the EC sensor array 199a may be maintained via execution of at least one algorithm stored in a memory / multiple memories 154, directed and / or controlled by one or more processors 152. The one or more processors 152 may execute instructions stored in the memory / multiple memories 154, which in some aspects may be AI / ML algorithms, to accurately maintain the height of the EC sensor array 199a above the powder bed 121 and / or the building block 109, such that measurement accuracy remains high, and the effect of the energy beam applied to at least one location of the powder bed 121 and / or the building block 109 is normalized and / or optimized, which may result in optimized resource utilization (e.g., time and monetary resources consumed during AM build operations) and the manufacture of accurate building blocks.

[0091] Additionally or alternatively, method 1400 may also include determining voltage variations at multiple locations of the powder bed 121 and / or the component 109. For example, in one aspect, measurements obtained by the EC sensor array 199a may be compared via execution of at least one algorithm stored in a memory / multiple memories 154, directed and / or controlled by one or more processors 152. The one or more processors 152 may execute instructions stored in the memory / multiple memories 154 to determine morphological variations in the powder bed 121 and / or the component 109 that can be considered and / or remedied, and / or to identify LoF effects in the component.

[0092] In alternative or additional aspects, method 1400 may include determining impedance variations based on voltage variations at multiple locations of the determined powder bed and / or component. For example, in one aspect, the voltage variations may be determined according to at least one algorithm stored in a memory / multiple memories 154, directed and / or controlled by one or more processors 152. The one or more processors 152 may execute instructions stored in the memory / multiple memories 15, which in some aspects may be AI / ML algorithms to accurately predict the effects of energy beams applied to at least one location of the powder bed 121 and / or component 109, which may lead to optimized resource utilization (e.g., time and monetary resources consumed during AM build operations) and the manufacture of accurate components.

[0093] In alternative or additional aspects, sensing at block 1402 may include moving the spreader and / or depositor 101 across the powder bed 121 and / or component 109. This can be achieved by instructions stored in one or more memories 154 that, when executed by one or more processors 152, cause the processors 152 to maintain the height of the EC sensor by engaging and / or operating at least one motor and / or other mechanical or electromechanical component coupled to the EC sensor array 199a, to hold the EC sensor array 199a at a specific height above the powder bed and / or component.

[0094] In alternative or additional aspects, the determination at block 1406 may include performing at least one statistical analysis or machine learning (AI / ML) operation. For example, one or more processors 152 may execute instructions stored in one or more memories 154, which may be statistical analysis and / or AI / ML algorithms, to determine at least one property in the powder bed and / or component based on inputs including lift-off measurement results and / or topography measurement results.

[0095] Figure 15 This is a flowchart illustrating a sensor fusion method 1900. At block 1902, method 1900 includes sensing a powder bed and / or component using an eddy current (EC) sensor array to obtain EC sensor measurements. For example, in one aspect, at least one processor 152, at least one memory 154, and / or at least one sensor (e.g., EC sensor array 199a) may be configured or may include means for sensing the powder bed 121 and / or component 109 to obtain EC sensor measurements.

[0096] Sensing at block 1902 may include the EC sensor array 199a detecting selected locations of powder bed 121 and / or component 109. This may include measurements from one or more EC sensors based on a measurement grid and property estimation, which may be used to generate tables or displays, such as those guided and / or controlled by one or more processors 152. One or more processors 152 may execute instructions stored in one or more memories 154 to move the EC sensor array 199a to the selected and / or desired location.

[0097] Sensing powder bed 121 and / or component 109 using EC sensor array 199a to obtain at least one EC sensor measurement can be performed to generate a first set of EC sensor measurements that can be used in one or more sensor fusion operations to accurately predict the effect of energy beams applied to at least one location of powder bed 121 and / or component 109. This can lead to optimized resource utilization (e.g., time and money resources consumed during AM build operations) and the manufacture of accurate components.

[0098] At block 1904, method 1400 includes sensing the powder bed and / or component with at least one auxiliary sensor 199 (e.g., one or more structured light systems, one or more photodiodes, one or more cameras, and / or one or more IR cameras) to obtain auxiliary measurement results. For example, in one aspect, the topography measurement results may be obtained by one or more auxiliary sensors 199, which may be used to generate auxiliary measurement results as directed and / or controlled by one or more processors 152. The one or more processors 152 may execute instructions stored in one or more memories 154 to move one or more auxiliary sensors 199 and / or guide one or more auxiliary sensors 199 to selected and / or desired locations.

[0099] Sensing powder bed 121 and / or component 109 using one or more sensors 199 to obtain morphological measurements can be performed to generate one or more auxiliary measurements that can be used in one or more sensor fusion operations to accurately predict the effect of energy beams applied to at least one location of powder bed 121 and / or component 109. This can lead to optimized resource utilization (e.g., time and money resources consumed during AM build operations) and the manufacture of accurate components.

[0100] At block 1906, method 1400 includes modifying one or more EC sensor measurements based on one or more auxiliary measurements, in other words, based on one or more auxiliary measurements and / or a set of measurements. For example, in one aspect, the one or more auxiliary measurements may be input to at least one algorithm stored in a memory / multiple memories 154, which is directed and / or controlled by one or more processors 152. The one or more processors 152 may execute instructions stored in the memory / multiple memories 15, which in some aspects may be AI / ML algorithms, to accurately modify the one or more EC sensor measurements. This can be performed to accurately predict the effects of energy beams applied to at least one location of the powder bed 121 and / or the building block 109, which can lead to optimized resource utilization (e.g., time and monetary resources consumed during AM build operations) and the manufacture of accurate building blocks.

[0101] In an alternative or additional aspect, at block 1906, method 1900 may further include modifying one or more EC sensor measurements obtained by the EC sensor array 199a above the powder bed and / or the build piece. For example, in one aspect, calibrating one or more EC sensor measurements based on one or more auxiliary measurements may include executing at least one algorithm stored in a memory / multiple memories 154, directed and / or controlled by one or more processors 152. The one or more processors 152 may execute instructions stored in the memory / multiple memories 154, which in some aspects may be AI / ML algorithms, to calibrate the EC sensor measurements such that the EC sensor array 199a can have improved accuracy for subsequent measurements, and the effect of the energy beam applied to at least one location of the powder bed 121 and / or the build piece 109 is normalized and / or optimized, which can lead to optimized resource utilization (e.g., time and money resources consumed during AM build operations) and the manufacture of accurate build pieces.

[0102] In an alternative or additional aspect, at block 1906, method 1900 may further include modifying one or more EC sensor measurements obtained by the EC sensor array 199a above the powder bed and / or the build piece. For example, in one aspect, calibrating one or more EC sensor measurements based on one or more auxiliary measurements may include executing at least one algorithm stored in a memory / multiple memories 154, directed and / or controlled by one or more processors 152. The one or more processors 152 may execute instructions stored in the memory / multiple memories 154, which may be algorithms configured to compare the measured interval distance of the EC sensor array 199a with the nominal interval distance of the EC sensor array to obtain a comparison result that normalizes and / or optimizes the effect of the energy beam applied to at least one location of the powder bed 121 and / or the build piece 109, which may result in optimized resource utilization (e.g., time and monetary resources consumed during AM build operations) and the manufacture of accurate build pieces.

[0103] Alternatively or additionally, the measurements from one or more EC sensors may be modified based on comparisons, such that the EC sensor array 199a can have improved accuracy for subsequent measurements, and the effect of the energy beam applied to at least one location of the powder bed 121 and / or the build 109 is normalized and / or optimized, which can lead to optimized resource utilization (e.g., time and money resources consumed during AM build operations) and the manufacture of accurate builds.

[0104] In alternative or additional aspects, method 1900 may include determining at least one attribute in the sensor bed based on modified EC sensor measurements. For example, in one aspect, an attribute / multiple attributes may be determined according to at least one algorithm stored in one or more memories 154, directed and / or controlled by one or more processors 152. The one or more processors 152 may execute instructions stored in one or more memories 154, which in some aspects may be AI / ML algorithms to accurately predict the effects of energy beams applied to at least one location of the powder bed 121 and / or the building block 109, which may lead to optimized resource utilization (e.g., time and monetary resources consumed during AM build operations) and the manufacture of accurate building blocks.

[0105] In an alternative or additional aspect, at block 1902, method 1900 may further include maintaining the height of the EC sensor array 199a above the powder bed and / or the building block. For example, in one aspect, the height of the EC sensor array 199a may be maintained via executing at least one algorithm stored in a memory / multiple memories 154, directed and / or controlled by one or more processors 152. The one or more processors 152 may execute instructions stored in the memory / multiple memories 154, which in some aspects may be AI / ML algorithms, to accurately maintain the height of the EC sensor array 199a above the powder bed 121 and / or the building block 109, such that measurement accuracy remains high, and the effect of the energy beam applied to at least one location of the powder bed 121 and / or the building block 109 is normalized and / or optimized, which may result in optimized resource utilization (e.g., time and monetary resources consumed during AM build operations) and the manufacture of accurate building blocks.

[0106] Additionally or alternatively, method 1900 may also include determining voltage variations at multiple locations of the powder bed 121 and / or the component 109. For example, in one aspect, measurements obtained by the EC sensor array 199a may be compared via execution of at least one algorithm stored in a memory / multiple memories 154, directed and / or controlled by one or more processors 152. The one or more processors 152 may execute instructions stored in the memory / multiple memories 154 to determine morphological variations in the powder bed 121 and / or the component 109 that can be considered and / or remedied, and / or to identify LoF effects in the component.

[0107] In alternative or additional aspects, method 1900 may include performing at least one statistical analysis or machine learning (AI / ML) operation. For example, one or more processors 152 may execute instructions stored in one or more memories 154, which may be statistical analysis and / or AI / ML algorithms, to determine at least one property in the powder bed and / or component based on inputs including lift measurement results and / or topography measurement results.

[0108] In an alternative or additional aspect, at block 1904, method 1900 may further include storage. For example, in one aspect, the auxiliary measurement result may be a temperature or thermal measurement result captured or acquired by an IR camera 199e. The IR camera 199e may be controlled according to one or more algorithms stored in at least one memory 154 and executed by at least one processor 152. Temperature data may be stored in one or more memories 154.

[0109] One or more processors 152 can execute instructions stored in one or more memories 154, which in some respects may be AI / ML algorithms to correlate measurements from one or more EC sensor arrays with temperature data.

[0110] At least one sensor calibration model can be generated based on relevant instructions stored in one or more memories 154 and executed by one or more processors 152, such that measurement accuracy remains high and the effect of energy beams applied to at least one location of powder bed 121 and / or component 109 is normalized and / or optimized, which can lead to optimized resource utilization (e.g., time and money resources consumed during AM build operations) and the manufacture of accurate components.

[0111] In an alternative or additional aspect, at block 1904, method 1900 may further include correlating voltage data with temperature data. For example, in one aspect, one or more EC sensor measurements acquired or captured by EC sensor array 199a may include voltage data, and one or more auxiliary measurements may be one or more temperature or thermal measurement values ​​captured or acquired by IR camera 199e. IR camera 199e may be controlled according to one or more algorithms stored in at least one memory 154 and executed by at least one processor 152. The correlation between voltage data and temperature data may occur via the execution of at least one algorithm stored in one or more memories 154, directed and / or controlled by one or more processors 152, wherein the at least one algorithm may be an AI / ML algorithm in some aspects.

[0112] In an alternative or additional aspect, at block 1902, method 1900 may further include sensing along the build axis of the additive manufacturing process. For example, in one aspect, sensor 199 may be a coaxial sensor of an energy beam. Sensing may occur via the execution of at least one algorithm stored in a memory / multiple memories 154, directed and / or controlled by one or more processors 152, wherein the at least one algorithm may be an AI / ML algorithm in some aspects.

[0113] In an alternative or additional aspect, at block 1902, method 1900 may further include sensing deviations from the build axis of the additive manufacturing process. For example, in one aspect, sensor 199 may be an off-axis sensor deviating from the beam axis of the energy beam. Sensing may occur via the execution of at least one algorithm stored in one or more memories 154, directed and / or controlled by one or more processors 152, wherein the at least one algorithm may be an AI / ML algorithm in some aspects.

[0114] In an alternative or additional aspect, at block 1902, method 1900 may further include sensing at least one layer beneath the surface layer of powder bed 121 and / or component 109. For example, in one aspect, sensor 199 may be a sensor with subsurface sensing capability, such as EC sensor array 199a. Sensing may occur via the execution of at least one algorithm stored in one or more memories 154, directed and / or controlled by one or more processors 152, wherein the at least one algorithm may be an AI / ML algorithm in some aspects.

[0115] In alternative or additional aspects, method 1900 may also include performing a machine learning process via at least one processor 152 of the relevant subsystem.

[0116] In alternative or additional aspects, method 1900 may also include predicting at least one non-fusion defect via a prediction subsystem and using at least one processor 152 based on a machine learning process.

[0117] For clarity, not all conventional features of these aspects are disclosed herein. It will be understood that in the development of any actual implementation of this disclosure, numerous implementation-specific decisions must be made to achieve the developer's specific objectives, and these specific objectives will vary for different implementations and different developers. It should be understood that such development efforts may be complex and time-consuming, but will in any case be routine engineering tasks for those skilled in the art who benefit from this disclosure. Elements shown by dashed lines in the figures should be considered optional in various respects.

[0118] Furthermore, it should be understood that the wording or terminology used herein is for descriptive rather than limiting purposes, and that the terminology or terminology in this specification will be interpreted by those skilled in the art based on the teachings and guidance presented herein, in conjunction with the knowledge of those skilled in the art. Moreover, no term in the specification or claims is intended to be attributed to uncommon or special meanings unless expressly stated otherwise.

[0119] The various aspects disclosed herein cover current and future known equivalents of the known modules mentioned herein by way of illustration. Furthermore, while aspects and applications have been shown and described, it will be apparent to those skilled in the art, who will benefit from this disclosure, that many more modifications than those mentioned above are possible without departing from the inventive concept disclosed herein.

[0120] In yet another variation, aspects of this disclosure can be implemented using a combination of hardware and software.

[0121] While the aspects described herein have been described in conjunction with the exemplary aspects outlined above, various alternatives, modifications, variations, improvements, and / or substantial equivalents, whether known or currently unforeseen or possibly currently unforeseen, may become apparent to those skilled in the art. Therefore, the exemplary aspects set forth above are intended to be illustrative and not limiting. Various changes may be made without departing from the spirit and scope of this disclosure. Therefore, this disclosure is intended to encompass all known or subsequently developed alternatives, modifications, variations, improvements, and / or substantial equivalents.

[0122] Therefore, the claims are not intended to be limited to the aspects shown herein, but should be given the full scope consistent with the language of the claims, wherein reference to an element in the singular is not intended to mean "one and only one," but rather "one or more," unless specifically stated otherwise. All structural and functional equivalents of elements known or to be known later by one of ordinary skill in the art throughout the various aspects described in this disclosure are expressly incorporated herein by reference and are intended to be covered by the claims. Furthermore, nothing disclosed herein is intended to be intended for the public, whether or not such disclosure is expressly stated in the claims. No claim element will be construed as means plus function unless the element is expressly stated using the phrase "means for..."

[0123] Furthermore, the word "example" is used herein to mean "serving as an example, instance, or illustration." Any aspect described herein as an "example" is not necessarily to be construed as preferred or superior to other aspects. Unless otherwise specified, the term "some" means one or more. Combinations such as "at least one of A, B, or C," "at least one of A, B, and C," and "A, B, C, or any combination thereof" include any combination of A, B, and / or C, and may include multiples of A, multiples of B, or multiples of C. Specifically, combinations such as "at least one of A, B, or C," "at least one of A, B, and C," and "A, B, C, or any combination thereof" may be only A, only B, only C, A and B, A and C, B and C, or A and B and C, wherein any such combination may contain one or more members of A, B, or C. Nothing disclosed herein is intended for public use, whether or not such disclosure is expressly stated in the claims.

[0124] Aspects of this disclosure may be systems, methods, and / or computer program products. A computer program product may include a computer-readable storage medium (or media) having computer-readable program instructions thereon for causing a processor to perform aspects of this disclosure.

[0125] Computer-readable storage media can be tangible devices that can hold and store program code in the form of instructions or data structures, which can be accessed by a processor of a computing device, such as computer system 150. Computer-readable storage media can be electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. For example, such computer-readable storage media can include random access memory (RAM), read-only memory (ROM), EEPROM, compact disc read-only memory (CD-ROM), digital versatile disk (DVD), flash memory, hard disk, portable computer disk, memory stick, floppy disk, or even mechanical encoding devices, such as punch cards or raised structures in grooves on which instructions are recorded. As used herein, computer-readable storage media should not be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or transmission media, or electrical signals transmitted through wires.

[0126] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to a suitable computing device, or downloaded via a network (e.g., the Internet, a local area network, a wide area network, and / or a wireless network) to an external computer or external storage device. This network may include copper cables, optical fibers, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network interface in each computing device receives the computer-readable program instructions from the network and forwards them to a computer-readable storage medium within the respective computing device.

[0127] Computer-readable program instructions used to perform the operations of this disclosure may be assembly instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of at least one programming language, including object-oriented programming languages ​​and conventional procedural programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone 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 scenario, the remote computer may be connected to the user's computer via any type of network (including LAN or WAN) or may be connected to an external computer (e.g., via the Internet). In some aspects, electronic circuits including, for example, programmable logic circuits, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs) may execute computer-readable program instructions to personalize the electronic circuits for performing aspects of this disclosure by utilizing the status information of the computer-readable program instructions.

[0128] In various aspects, the systems and methods described in this disclosure can be solved according to modules. As used herein, the term "module" refers to a real-world device, component, or arrangement of components implemented using hardware (such as by an application-specific integrated circuit (ASIC) or FPGA), for example, or as a combination of hardware and software (such as by a microprocessor system and a set of instructions implementing the functionality of the module), which (when executed) transform the microprocessor system into a special-purpose device. A module can also be implemented as a combination of both, where some functions are facilitated solely by hardware, while others are facilitated by a combination of hardware and software. In some embodiments, at least a portion of the module, and in some cases, all of it, can be implemented in a computer system (such as those described above). Figure 1E The module (which is described in more detail herein) executes on a processor. Therefore, each module can be implemented in a variety of suitable configurations and should not be limited to any particular implementation illustrated herein.

Claims

1. A method for powder bed additive manufacturing, comprising: The powder bed is sensed using an eddy current (EC) sensor to obtain EC sensor measurement results; The powder bed is sensed using an auxiliary sensor to obtain morphology measurement results; and The properties of the powder bed are determined based on the EC sensor measurement results and the morphology measurement results.

2. The method of claim 1, wherein, The attribute includes defects in the powder bed.

3. The method according to claim 2, wherein, The defects include unfused (LoF) defects in the components of the powder bed.

4. The method according to claim 1, wherein, Sensing the powder bed using the EC sensor also includes: Maintain the EC sensor at the same height above the powder bed; and Determine the voltage changes at multiple locations in the powder bed.

5. The method according to claim 4, further comprising: The impedance change is determined based on the voltage changes at multiple locations of the determined powder bed.

6. The method according to claim 4, wherein, The EC sensor is mounted on the powder spreader, and sensing the powder bed using the EC sensor also includes: The powder spreader is moved across the powder bed.

7. The method according to claim 1, wherein, The auxiliary sensor includes a light sensor.

8. The method according to claim 7, wherein, The optical sensor includes a structured light sensor, and the topography measurement results include point clouds at least partially obtained from the structured light.

9. The method according to claim 8, wherein, The structured light sensor includes at least one camera and a projector.

10. The method according to claim 1, wherein, Determining the properties of the powder bed includes performing at least statistical analysis or machine learning.

11. A system for powder bed additive manufacturing, comprising: Eddy current (EC) sensor array, which is configured to acquire EC sensor measurement results; An auxiliary sensing subsystem is configured to sense a powder bed to obtain morphology measurement results; as well as At least one processor; as well as At least one memory, The at least one memory stores instructions that, when executed by the at least one processor, cause the at least one processor to determine the properties of the powder bed based on the EC sensor measurement results and the morphology measurement results.

12. The system according to claim 11, wherein, The attribute includes defects in the powder bed.

13. The system according to claim 12, wherein, The defects include unfused (LoF) defects in the components of the powder bed.

14. The system according to claim 11, wherein, The eddy current sensor array is also configured to: Maintain a height above the powder bed; and Voltage changes at multiple locations in the powder bed are sensed.

15. The system according to claim 14, wherein, The instructions also cause the at least one processor to determine impedance changes based on sensed voltage changes at multiple locations of the powder bed.

16. The system according to claim 14, wherein, An EC sensor is mounted on a powder spreader configured to move across the powder bed.

17. The system according to claim 11, wherein, The auxiliary sensing subsystem includes a light sensor.

18. The system according to claim 17, wherein, The optical sensor includes a structured light sensor, and the topography measurement results also include point clouds at least partially obtained from the structured light.

19. The system according to claim 18, wherein, The structured light sensor includes at least one camera and a projector.

20. The system according to claim 11, wherein, Determining the properties of the powder bed includes performing at least statistical analysis or machine learning.

21. A method for powder bed additive manufacturing, comprising: The measurement results of the EC sensor are obtained by sensing in the powder bed using an eddy current (EC) sensor. An auxiliary sensor is used to sense the powder bed to obtain auxiliary measurement results; and The EC sensor measurement results are modified based on the auxiliary measurement results.

22. The method of claim 21, further comprising: The properties of the powder bed are determined based on measurements from a modified EC sensor.

23. The method according to claim 22, wherein, The attribute includes defects in the powder bed.

24. The method according to claim 23, wherein, The defects include unfused (LoF) defects in the components of the powder bed.

25. The method according to claim 21, wherein, Sensing in the powder bed using the EC sensor also includes: Maintain the EC sensor at the same height above the powder bed; and Determine the voltage changes at multiple locations in the powder bed.

26. The method of claim 25, further comprising: The impedance change is determined based on the voltage changes at multiple locations of the determined powder bed.

27. The method according to claim 25, wherein, The EC sensor is mounted on the powder spreader, and sensing in the powder bed using the EC sensor also includes: The powder spreader is moved across the powder bed.

28. The method according to claim 21, wherein, The auxiliary sensor includes a light sensor.

29. The method according to claim 28, wherein, The optical sensor includes a structured light sensor, and the auxiliary measurement results include point clouds at least partially obtained from the structured light.

30. The method according to claim 29, wherein, The structured light sensor includes at least one camera and a projector.

31. The method according to claim 21, wherein, The EC sensor measurement results include EC sensor measurement results, the auxiliary measurement results include topography measurement results, and modifying the EC sensor measurement results includes calibrating the EC sensor measurement results based on the topography measurement results.

32. The method according to claim 31, wherein, The topography measurement results include the measured interval distance, and modifying the EC sensor measurement results includes: The measured interval distance is compared with the nominal interval distance of the EC sensor to obtain a comparison result; and The EC sensor measurement results are modified based on the comparison results.

33. The method according to claim 21, wherein, The EC sensor measurement result includes one EC sensor measurement result, and the modified EC sensor measurement result includes one modified EC sensor measurement result.

34. The method according to claim 21, wherein, The EC sensor measurement results include voltage measurement results, and the modified EC sensor measurement results include modified voltage measurement results.

35. The method according to claim 21, wherein, The auxiliary sensor includes a thermal sensor, and the auxiliary measurement results include thermal measurement results.

36. The method according to claim 35, wherein, The thermal measurement results include temperature data, and sensing in the powder bed using the thermal sensor includes: The temperature data is stored in a memory; The EC sensor measurement results are correlated with the temperature data; and Based on the correlation, a sensor calibration model is generated.

37. The method of claim 36, wherein, The EC sensor measurement results include voltage data, and correlating the EC sensor measurement results includes correlating the voltage data with the temperature data.

38. The method according to claim 35, wherein, Sensing using the thermal sensor in the powder bed includes sensing on the build axis during the additive manufacturing process.

39. The method according to claim 38, wherein, The thermal sensor includes a photodiode.

40. The method of claim 35, wherein, Sensing in the powder bed using the thermal sensor includes sensing deviations from the build axis of the additive manufacturing process.

41. The method according to claim 40, wherein, The thermal sensor includes an infrared camera.

42. The method according to claim 36, wherein, The correlation of the EC sensor measurement results also includes: Perform the machine learning process.

43. The method of claim 42, further comprising: Predict at least one unfused (LoF) defect based on the machine learning process.

44. The method according to claim 21, wherein, Sensing in the powder bed using the EC sensor includes sensing at least one layer beneath the surface layer of the powder bed.

45. A system for calibrating an array of eddy current (EC) sensors used to monitor a powder bed during additive manufacturing, the system comprising: The first sensing subsystem includes: At least one eddy current (EC) sensor array; The second sensing subsystem; and Related subsystems, including: At least one processor, The EC sensor measurement results captured by the first sensing subsystem and the topography map generated by the second sensing subsystem are correlated by at least one processor of the correlation subsystem and are used to calibrate at least one sensor in the EC sensor array.

46. ​​The system according to claim 45, wherein, At least one processor of the correlation subsystem estimates at least one property of the powder bed based on the correlation between the EC sensor measurement results and the topography map.

47. The system according to claim 46, wherein, The at least one property includes variations in electrical conductivity, defects, and material composition.

48. The system according to claim 45, wherein, The EC sensor measurements include the voltage difference at the location of the powder bed.

49. The system according to claim 48, wherein, The impedance change is inferred by the relevant subsystem based on the voltage difference at multiple locations in the powder bed.

50. The system according to claim 45, wherein, The EC sensor array is mounted on the powder spreader.

51. The system according to claim 45, wherein, The second sensing subsystem includes structured light, and the topography map includes a point cloud at least partially obtained from the structured light.

52. The system according to claim 51, wherein, The structured light includes: At least one camera; and Projector.

53. The system according to claim 45, wherein, At least one processor of the correlation subsystem is configured to perform at least structural analysis or machine learning during the correlation of the EC sensor measurements and the topography map.

54. The system according to claim 53, wherein, At least one processor of the related subsystem is configured to predict defects at a higher rate than predictions made solely by analyzing EC sensor measurements or topography maps.

55. A sensor fusion system for additive manufacturing (AM), comprising: Eddy current (EC) sensor, which is configured to sense in a powder bed to obtain EC sensor measurement results; An auxiliary sensor is configured to sense in the powder to obtain auxiliary measurement results; At least one processor; as well as At least one memory storing instructions that, when executed by the at least one processor, cause the processor to modify the EC sensor measurement results based on the auxiliary measurement results.

56. The sensor fusion system according to claim 55, wherein, The instructions also cause the processor to determine the properties of the powder bed based on modified EC sensor measurements.

57. The sensor fusion system according to claim 56, wherein, The attribute includes defects in the powder bed.

58. The sensor fusion system according to claim 57, wherein, The defects include unfused (LoF) defects in the components of the powder bed.

59. The sensor fusion system according to claim 55, wherein, Sensing in the powder bed using the EC sensor also includes: Maintain the EC sensor at the same height above the powder bed; and Determine the voltage changes at multiple locations in the powder bed.

Images