Additive manufacturing method and apparatus

Abstract

A method of monitoring an additive manufacturing device. The method includes receiving one or more sensor signals from the additive manufacturing device during a build of a workpiece, comparing the one or more sensor signals to a respective acceptable process variation of a plurality of acceptable process variations, and generating a log based on the comparison. Each acceptable process variation of the plurality of acceptable process variations is associated with at least one state of a build progression of the workpiece, and the respective acceptable process variation is the acceptable process variation associated with the state of the build progression when the one or more sensor signals are generated.

Description

[0001] This application is a divisional application. The original application is PCT application No. PCT / GB2016 / 053558, filed on November 14, 2016, which entered the Chinese national phase on May 16, 2018. The application number is 201680066899.4, and the title is "Additive Manufacturing Method and Equipment". Technical Field

[0002] This invention relates to an additive manufacturing method and apparatus, and more particularly to a method and apparatus for monitoring and optionally controlling the additive manufacturing process. Background Technology

[0003] Additive manufacturing methods or rapid prototyping methods used to produce parts involve layer-by-layer curing of a flowable material. Various additive manufacturing methods exist, including powder bed systems such as selective laser melting (SLM), selective laser sintering (SLS), electron beam melting (EBM), and stereolithography, as well as non-powder bed systems such as fused deposition modeling, including arc additive manufacturing.

[0004] In selective laser melting, a powder layer is deposited on a powder bed in a build chamber, and a laser beam partially scans across the powder layer, which corresponds to a cross-section (slice) of the workpiece being built. The laser beam melts or sintersulates the powder to form a cured layer. After selective curing of the layer, the powder bed is reduced in thickness to the thickness of the newly cured layer, and another layer of powder is spread and cured on the surface as needed. In a single build, more than one workpiece can be built, with components spaced apart in the powder bed.

[0005] Various systems have been developed for monitoring additive manufacturing processes. For example, WO 2015 / 040433 discloses an apparatus and method for monitoring the molten pool passing through an optical module used to manipulate a laser beam. PCT / GB2016 / 051720, incorporated herein by reference, discloses the use of a passive acoustic sensor for sensing properties of an additive manufacturing process.

[0006] S. Clijsters, T. Craeghs, S. Buls, K. Kempen, and JP. Kruth (2014), “In situ quality control of the selective laser melting process using a high-speed, real-time melt pool monitoring system,” *International Journal of Advanced Manufacturing Technology*, pp. 1089–1101, disclose a system that enables operators to monitor the quality of SLM operations and estimate part quality online. To verify the quality, expected sensor values ​​for the melt pool are predefined. These reference values ​​are predicted based on experimental results. The aggregate of sensor values ​​for the fill and profile scan vectors can be used to calculate statistical parameters, such as the mean and standard deviation of the population distribution of a particular vector class. In addition to these standard parameters, confidence intervals can be calculated on reference data that simplify the interpretation process. Confidence intervals form the basis for quality estimation.

[0007] The goal is to provide improved in-process control for additive manufacturing processes. Summary of the Invention

[0008] According to a first aspect of the invention, a method for monitoring an additive manufacturing apparatus is provided, the method comprising: receiving one or more sensor signals from the additive manufacturing apparatus during the construction of a workpiece; comparing the one or more sensor signals with a corresponding acceptable process change among a plurality of acceptable process changes; and generating a log based on the comparison, wherein each of the plurality of acceptable process changes is associated with a state of workpiece construction progress, and the corresponding acceptable process change is an acceptable process change associated with the state of construction progress when the one or more sensor signals are generated.

[0009] In this way, taking into account factors that influence acceptable process variations as construction progresses (such as the geometry of the workpiece), multiple acceptable process variations can be customized for the construction of the workpiece. Acceptable variations can include changes in the values ​​of the sensor signals themselves or derived from the sensor signals.

[0010] Acceptable process variations can vary with build time and / or build volume and / or position within the workpiece. The state of workpiece build progress can include position on the workpiece and / or within the build volume or the time of a setup event (zero time) during build. Zero time can be the time when build begins (e.g., the execution of the first instruction for build) or the time when the equipment begins to cure a layer of material. The setup event can be chosen such that subsequent events generating one or more sensor signals during its occurrence occur a predetermined time after the setup event, making time a proxy for position with the desired resolution. This simplifies the collection of sensor data and the identification of corresponding acceptable process variations to which the sensor data will be compared, as it only requires recording the time when the sensor signal is generated, not the position of additive manufacturing equipment elements (such as steerable optics and / or the z-position of the build platform) at the time the sensor signal is generated. In one embodiment, multiple acceptable process variations are in an ordered list, and sensor signals are generated during build in a known order, wherein the order of the multiple acceptable process variations corresponds to the order in which the sensor signals are generated.

[0011] The additive manufacturing apparatus may include an energy beam and / or plasma stream for curing material, and the method may include: determining the location of material cured by the energy beam and / or plasma stream (in a workpiece or in the build volume of the additive manufacturing apparatus) when a sensor signal is generated, and determining a corresponding acceptable process variation to compare with the sensor signal from the determined location.

[0012] Additive manufacturing equipment may include a movable build support, wherein layers are cured to form a workpiece on the build support, the method comprising: determining the current number of layers being processed during the generation of one or more sensor signals, and determining a corresponding acceptable process variation to compare with the sensor signals from the determined number of layers.

[0013] The method may include: determining the time in the construction when the sensor signal is generated, and determining the corresponding acceptable process variation to compare with the sensor signal from the determined time.

[0014] The method may include: determining the order in which sensor signals are generated, and determining appropriate acceptable process variations to compare with sensor signals from the determined order.

[0015] Acceptable process changes may include one or more of the following:

[0016] a) Temperature in the build chamber of the additive manufacturing equipment

[0017] b) Temperature of the flowable material to be cured

[0018] c) Temperature of the molten pool

[0019] d) Intensity of light collected by the sensor

[0020] e) Spectral emission from the molten pool

[0021] f) Dimensions of the molten pool, such as the area of ​​the molten pool, the length and / or width of the molten pool, the length-to-width ratio, and the rate of change of the length to the width of the molten pool (which can be used to determine whether spheroidization of the molten pool has occurred).

[0022] g) Comparison of images from adjacent layers

[0023] h) Gas pressure in the construction chamber

[0024] i) Oxygen concentration in the construction chamber

[0025] j) Pump speed of the pump used to recirculate gas through the building chamber

[0026] k) Lift position, speed and / or acceleration

[0027] l) Wiper position, wiper speed and / or acceleration

[0028] m) Load on the scraper (the acceleration and / or load on the scraper can be used to detect whether the part in the cured area has been rolled up and projected onto the powder bed surface)

[0029] n) Predicted temperature of the workpiece / construction's current or future portion based on sensor signals from a thermal model.

[0030] o) Material proportions

[0031] p) Acoustic signal

[0032] q) Image of powder bed

[0033] r) Images of the cured material, and / or

[0034] The rate of change of any one of s)(a) to (r).

[0035] There can be a one-to-one correlation between each acceptable process change and the state of the artifact's construction progress among multiple acceptable process changes.

[0036] Alternatively, at least one of the acceptable process variations may be associated with multiple states of the workpiece's construction progress.

[0037] Multiple acceptable process variations may apply only to certain portions of a workpiece, as specified by the associated state of the workpiece's construction progress. For example, these portions might be parts of a workpiece with prominent protrusions or thin walls. For other portions of the workpiece, there may be no acceptable or generally acceptable process variations, such as those for the material or the scanning strategy being used. Specifically, generally acceptable process variations may apply to workpiece portions that are unlikely to fail or are non-critical, while workpiece-specific acceptable process variations may be used for portions that are more likely to fail or are critical without customized controls. In this alternative embodiment, there may be a one-to-one correlation between each workpiece-specific acceptable process variation and the state of the workpiece's construction progress, and a one-to-many correlation between each generally acceptable process variation and the state of the workpiece's construction progress.

[0038] Sensor signals may include signals from one or more of the following sensors in the additive manufacturing equipment:

[0039] a) A pyrometer, for example, used to measure the temperature of a bed of materials and / or a molten pool.

[0040] b) An acoustic sensor that can be used to obtain various aspects of the properties of a process, as described in GB 1510220.5, which is incorporated herein by reference.

[0041] c) Thermal (infrared) camera, for example, for measuring the temperature of a bed of materials.

[0042] d) Visible light cameras, for example, used to measure workpiece deformation during construction.

[0043] e) Photodiodes, for example, used to measure the temperature of a molten pool and / or a bed of materials.

[0044] f) A spectrometer for measuring the spectral emission from the plasma plume and / or molten pool generated during material melting.

[0045] g) Force feedback, for example, a load sensor on a scraper.

[0046] h) Pressure sensors, for example, for measuring gas pressure in the build chamber or differential pressure across a filter element used to filter condensate from gas recirculated through the build chamber.

[0047] i) Mass flow sensor, for example, for measuring the mass of gas recirculated through the building chamber.

[0048] j) Oxygen sensor, for example, for measuring oxygen concentration in a construction chamber.

[0049] k) Encoders, for example, for measuring the position, speed, and / or acceleration / deceleration of a scraper and / or lift used for spreading a layer of material.

[0050] l) Accelerator, for example, used to measure the acceleration / deceleration of a squeegee.

[0051] Acceptable process variation can be a change in the difference between two sensed values ​​(rather than a change in the absolute value). In this way, it may not be necessary to calibrate the sensor based on the absolute value. For example, images of the material bed can be compared after each layer is laid, and if the difference in the images of adjacent layers is outside the acceptable process variation, appropriate actions can be generated, such as stopping the build, re-dosing and laying the layers, and / or generating a warning to notify the operator of potential errors in the build.

[0052] The time allotted for completing the build, finishing layers, and / or curing areas can itself be a variable subject to acceptable process variation. For example, if the time allotted for finishing a layer falls outside of acceptable process variation, this could cause failures in the processing of subsequent layers due to build cooling, and the build could stop.

[0053] Measurements of the shape of sensor signals about their average value, such as skewness or kurtosis, can be variables exhibiting acceptable process variation. Acceptable process variation can include acceptable patterns of sensor signals across successive states of progress. For example, acceptable process variation can be correlated with patterns of sensor signals across multiple states of progress, indicating that the sensor signals are beginning to deviate from acceptable process variation. Specifically, patterns of increasing or decreasing sensor signals that are significantly different from the patterns of corresponding acceptable process variation can indicate the beginning of a deviation of the process from acceptable limits.

[0054] This method may include controlling additive manufacturing equipment during the build process based on comparisons of sensor signals with corresponding acceptable process variations.

[0055] This method may include controlling the parameters of the build-up on an additive manufacturing apparatus to maintain the sensed signal within acceptable process variations. Parameters may include one or more power of the energy beam used to cure the material, the spot size of the energy beam on the material, the scanning speed of the energy beam across the material, the spot exposure time and spot distance (for pulsed scanning of the energy beam), the hatch distance, the hatch length, and the scanning strategy (such as zigzag, checkerboard, stripe, and / or shell and core scanning strategies).

[0056] This method may include stopping the construction of a workpiece if the sensed signal falls outside an acceptable process variation. Stopping the construction of a workpiece may include halting the entire construction, or, if the workpiece is being constructed together with other workpieces, suppressing the construction of that workpiece while continuing the construction of the other workpieces.

[0057] This method may include re-baseline the build / recalibration of the additive manufacturing equipment if the sensed signal falls outside of acceptable process variations.

[0058] The method may include processing the workpiece if the sensor signal falls outside of an acceptable process variation.

[0059] The additive manufacturing apparatus may include part of a machine chain, and the method may include generating instructions for additional machines in the machine chain if the sensed signal falls outside an acceptable process variation.

[0060] This method may include storing only a subset of sensor data derived from sensor signals that fall outside the acceptable range of process variation. For example, sensor data may be stored in unforeseen circumstances, with only sensor data falling outside the acceptable range of process variation and sensor signals within a window relating to that data being stored.

[0061] The log may include location data, such as coordinate data or the construction status or progress of the workpiece, based on which the location of suspicious areas of the workpiece can be determined, generating sensor signals that fall outside the corresponding acceptable process variations during formation. The method may include using the location data to display suspicious areas of the workpiece, preferably in a 2D or 3D representation of the workpiece or construction. In this way, the user can identify suspicious areas of the workpiece from the display for further investigation and / or action to correct defects, which can be performed in later stages of the manufacturing process.

[0062] This method may include controlling additive manufacturing equipment to repair areas of the workpiece that fall outside of acceptable process variations, based on sensor signals. For example, the repair may include using an energy beam / plasma stream to re-solidify the area.

[0063] This method may include controlling an additive manufacturing apparatus to perform further inspection of a region of the workpiece that falls outside a corresponding acceptable process variation, based on sensor signals from such regions. The inspection may include, for example, using an energy beam to probe the region, such as an energy beam for curing material or other energy beams. The energy density of the energy beam used for inspection may be lower than the density required for curing the material. The inspection may include: heating the region using a heat source that can selectively heat areas of the workpiece; and monitoring the thermal properties of the region using sensors such as pyrometers or thermal cameras.

[0064] According to a second aspect of the invention, a controller for monitoring an additive manufacturing apparatus is provided, the controller comprising a processor arranged to perform the following operations: receiving one or more sensor signals from the additive manufacturing apparatus during the construction of a workpiece; comparing the one or more sensor signals with corresponding acceptable process changes of a plurality of acceptable process changes; and generating a log based on the comparison, wherein each of the plurality of acceptable process changes is associated with a state of workpiece construction progress, and the corresponding acceptable process change is an acceptable process change associated with the state of construction progress when the one or more sensor signals are generated.

[0065] A controller may be configured to receive the multiple acceptable process variations and, optionally, build instructions for constructing an artifact. The controller may include memory and a processor configured to store the multiple acceptable process variations in memory upon receiving build instructions. The multiple acceptable process variations can be erased from memory after the artifact construction is complete or when build instructions for a different artifact are uploaded to the controller. Because the multiple acceptable process variations are specific to the artifact construction, they may be contained in the same file as the instructions used to perform the construction and uploaded to the controller. Since the multiple acceptable process variations are build / artifact specific, they are redundant and can be erased once the controller is used for a different build / artifact.

[0066] The controller can be configured to store only a subset of sensor data derived from a sensor signal that falls outside the acceptable range of process variations.

[0067] According to a third aspect of the invention, an additive manufacturing apparatus including a controller according to a second aspect of the invention is provided.

[0068] According to a fourth aspect of the invention, a data carrier having instructions is provided, which, when executed by a processor, cause the processor to perform the method as described in the first aspect of the invention.

[0069] According to a fifth aspect of the invention, a method is provided for generating instructions for constructing a workpiece in an additive manufacturing apparatus, the method comprising: constructing one or more initial workpieces nominally identical to at least a portion of the workpiece using a respective additive manufacturing apparatus and / or one or more respective additive manufacturing apparatuses; receiving a set of sensor signals from the respective additive manufacturing apparatus and / or one or more respective additive manufacturing apparatuses during the construction of the initial workpiece or each initial workpiece; evaluating the one or more initial workpieces to determine whether the initial workpiece or each initial workpiece meets specified requirements; associating the set or each set of sensor signals with one or more quality identifiers, the quality identifiers identifying whether all or part of the set of sensor signals represents sensor signals for a construction that meets the specified requirements; and determining, based on the set or more sets of sensor signals and the associated quality identifiers, a plurality of acceptable process variations for generating a respective set of sensor signals during subsequent construction of the workpiece using the additive manufacturing apparatus.

[0070] Acceptable process variations can be determined by statistical analysis of natural process variations in sensor signals during the construction of one or more initial workpieces, and by whether the sensor signals are related to a portion of the workpiece being evaluated to meet the specified requirements (so-called "golden sensing data," used as a benchmark or fingerprint for judging future constructions). Acceptable process variations may have already been derived from sensor signals generated by the corresponding additive manufacturing equipment and / or one or more corresponding additive manufacturing devices during one or more initial constructions that coincide with the construction of the workpiece.

[0071] Statistical analysis may include Bayesian modeling and / or cluster analysis. Since sensor information is generated from the construction of (multiple) identical workpieces and optionally (multiple) identical constructions, the sensor data is subject to variations that may arise from different geometries and, optionally, different constructions. For example, the temperature of the region being cured (e.g., the molten pool) can vary depending on the workpiece geometry / construction setup, even when the same scanning strategy is used across the area to be cured. Furthermore, the time between curing successive layers can affect the cooling that occurs between layer formations and thus the thermal properties of the construction. The time between curing successive layers will depend on the size of the area to be cured in each layer, the so-called layer “load.” Among other things, the layer load will depend on the geometry of the workpiece being constructed, the number of workpieces to be constructed in a single construction, and the orientation of the construction workpieces. Accordingly, deriving acceptable process variations in the construction of the workpiece from sensor signals generated by the respective additive manufacturing equipment and / or one or more respective additive manufacturing equipment during one or more constructions of the initial workpiece ensures that the influence of the workpiece geometry and, optionally, the construction setup, on the sensor data is taken into account. This method is particularly suitable for manufacturing a series of identical workpieces and / or constructs using the corresponding additive manufacturing equipment and / or one or more corresponding additive manufacturing equipment.

[0072] In the manufacture of a series of nominally identical workpieces, as more workpieces are built, acceptable process variations can be improved / evolved using sensor data from each build.

[0073] An initial workpiece can be constructed at a location within the build volume that is the same as the location where subsequent workpieces are constructed. Variations in the workpiece's location within the build volume can affect the parameters required to achieve a workpiece that meets specified requirements, meaning that acceptable process variations determined for one location within the build volume may not be applicable to another location. For example, differences in airflow across different locations on the build surface may necessitate different parameters for constructing the workpiece at different locations within the build volume.

[0074] In alternative embodiments, an initial workpiece can be constructed at a location within the build volume that differs from the location where subsequent workpieces are constructed. For example, mapping can be used to transform an acceptable process variation determined for one location of the workpiece within the build volume into an acceptable process variation for another location of the workpiece within the build volume. Mapping can be determined according to suitable mapping routines, such as by constructing test blocks / elements at different locations within the build volume and comparing differences in sensor signals from test blocks / elements at different locations within the build volume.

[0075] When an artifact is one of a group of (identical or dissimilar) artifacts to be built together in a single build, the initial artifact can also be built together with a group of identical artifacts in the same layout as the one used in the build volume when the artifact is built. In this way, the same load is achieved in the build(s) of the initial artifact(s) as in the subsequently built artifacts.

[0076] In an alternative embodiment, where the workpiece is one of a group of (identical or dissimilar) workpieces in a single build to be constructed together in a first build layout, the initial workpiece(s) can be constructed in a second, different build layout within the build volume, and a mapping is used to transform acceptable process variations determined based on sensor data of the initial workpieces in the second build layout into acceptable process variations for the build of the workpieces in the first build layout. For example, the mapping can be used to transform acceptable process variations determined based on sensor data of the initial workpieces in the second build layout into acceptable process variations for the build of the workpieces in the first build layout. The mapping can be determined according to a suitable mapping routine, such as by constructing a series of builds with different loads (e.g., test blocks / elements) and comparing the differences in sensor signals from builds with different loads.

[0077] The method may include generating a mapping to transform acceptable process variations determined for a first additive manufacturing apparatus into a set of acceptable process variations for a corresponding second additive manufacturing apparatus. Specifically, the responses of sensors in the first additive manufacturing apparatus may differ from those in the second additive manufacturing apparatus. Accordingly, the mapping may transform the acceptable process variations to account for the differences in sensor responses in the two apparatuses. The mapping can be determined by constructing identical test blocks in each additive manufacturing apparatus and identifying the performance differences of the sensors.

[0078] The corresponding additive manufacturing equipment has the same functions as the nominally labeled additive manufacturing equipment. For example, the corresponding additive manufacturing equipment may have the same brand and model as the nominally labeled additive manufacturing equipment, or the same brand and model as an additive manufacturing equipment that can be configured to operate in a functional setting common to the nominally labeled additive manufacturing equipment, such as having the same scanning parameters, spot size, dot distance, exposure time, laser power, etc., and the same nominal gas flow rate, scraper speed, and z-plateau speed as the comparable sensor used to generate the sensor signal, so that comparative comparisons can be made between builds on different additive manufacturing equipment. The nominally labeled additive manufacturing equipment may have, or can be configured to have, the nominally labeled signal timing and / or processing delay. Of course, the additive manufacturing equipment may differ in a way that does not significantly affect the build, such as color, housing shape, and different components that may still provide the nominally labeled functional performance (such as different filter elements and / or pumps). The corresponding additive manufacturing equipment may include additive manufacturing equipment that has been configured to mimic the functions of the nominally labeled additive manufacturing equipment.

[0079] The specified requirements for the initial workpiece construction can be: completion of the workpiece construction, density of the cured material, absence or maximum number or size of cracks, voids or inclusions in the workpiece, surface finish, porosity, crystal structure of the workpiece (including grain size and morphology), and / or chemical composition. These specified requirements can also include: specified compressive strength, tensile strength, shear strength, hardness, microhardness, bulk modulus, shear modulus, elastic modulus, stiffness, elongation at break, Poisson's ratio, corrosion resistance, dissipation factor, electrical conductivity, and / or magnetic properties of the workpiece. These specified requirements can also be specified performance characteristics when performing functional testing. If only a region of the initial workpiece is considered to meet the specified requirements, the user can identify whether all sensor signals for that initial workpiece construction are identified as related to constructions that do not meet the specified requirements, or whether only sensor signals associated with regions considered to not meet the specified requirements should be identified as such, where other sensor data are related to acceptable constructions. For example, in the former case, the user could consider the fault to be a systematic error affecting most areas / all constructions. In the latter case, the user can identify a fault as a localized fault that does not significantly affect the sensor signals collected from remote areas of the workpiece.

[0080] Workpiece analysis used to determine whether a workpiece meets specified requirements may include non-destructive testing (DT) (such as dimensional testing), for example using contact sensing (such as contact-triggered probes or scanning probes) or non-contact sensing (such as optical detection) and / or CT or ultrasonic scanning, for example to determine whether a workpiece contains cracks or inclusions.

[0081] Workpiece analysis used to determine whether a workpiece meets specified requirements may include damage detection, such as physical segmentation and inspection, for example, using microscopy and / or verification testing. Workpiece analysis used to determine whether a workpiece meets specified requirements may include density measurements, such as the Archimedes test. Workpiece analysis may include analysis of values ​​obtained from process inspection / inspection. For example, additive manufacturing equipment may include layer-by-layer inspection, such as spatially resolved acoustic spectroscopy (SRAS).

[0082] The method may include: determining a set of primary acceptable process variations based on one or more sets of sensor signals and associated quality identifiers, each primary acceptable process variation being associated with a state of workpiece construction progress; generating a set of simplified secondary acceptable process variations based on the set of primary acceptable variations, at least one of which is associated with multiple states of workpiece construction progress. A simplified set of secondary acceptable process variations can be generated by grouping similar variations among the primary acceptable process variations together and generating one of the secondary acceptable process variations for each group, each secondary acceptable process variation being associated with multiple progress states corresponding to the progress states associated with the primary acceptable process variations of the corresponding group, and generating secondary acceptable process variations for the corresponding groups.

[0083] In some cases, a complete set of acceptable process changes for each state of build progress (such as each exposure point) can result in very large documentation. Accordingly, it is advantageous to simplify a set of primary acceptable process changes into a smaller, more manageable set of secondary acceptable process changes that still retain the key characteristics of the primary set. Specifically, acceptable process changes for adjacent states of build progress may be similar enough that maintaining separate primary acceptable process changes for the two states is not guaranteed.

[0084] Accordingly, each minor acceptable process change can be generated based on the characteristics of the major acceptable process change of the corresponding group, and the minor acceptable process change is generated for that corresponding group.

[0085] Values ​​derived from a similarity function can be used to group the major acceptable process variations.

[0086] This method may include receiving user input identifying different regions of a workpiece, for which different similarity criteria are used to group major acceptable process variations together. Specifically, some regions of the build can be identified as problematic, and therefore the user can deem it useful to apply acceptable process variations with higher resolution differentials to these problematic regions rather than to less problematic regions. By allowing the user to identify regions to which different similarity criteria are applied, this method allows the user to adjust the level of generalization provided to acceptable process variations based on prior knowledge of the build and workpiece requirements.

[0087] According to a sixth aspect of the invention, a system for generating instructions for controlling an additive manufacturing apparatus is provided. The system includes a processor arranged to: receive, using the respective additive manufacturing apparatus and / or one or more respective additive manufacturing apparatuses, a set of sensor signals for each build of one or more initial workpieces, the one or more initial workpieces being nominally identical to at least a portion of the workpiece; receive an indication of whether the initial workpiece or each initial workpiece meets a specified requirement; associate the set or each set of sensor signals with one or more quality identifiers, the quality identifiers identifying whether all or part of the set of sensor signals represents sensor signals for a build that meets the specified requirement; and, based on the set or more sets of sensor signals and the associated quality identifiers, determine a plurality of acceptable process variations for the respective set of sensor signals generated using the additive manufacturing apparatus during subsequent builds of the workpiece.

[0088] The processor can be configured to associate each acceptable process change with the status of the build progress. The status of progress can be location within the build volume, time during the build, and / or the order in which sensor signals are collected during the build.

[0089] The processor can be configured to generate a build file that includes the scan path and scan parameters followed by the energy beam or plasma stream when the material is cured in the formation of the workpiece in a layer-by-layer manner, as well as several acceptable process variations.

[0090] The processor can be configured to: determine a set of primary acceptable process variations based on one or more sets of sensor signals and associated quality identifiers, each primary acceptable process variation being associated with a state of the workpiece's construction progress; and generate a set of simplified secondary acceptable process variations based on the set of primary acceptable variations, at least one of which is associated with multiple states of the workpiece's construction progress.

[0091] A simplified set of minor acceptable process changes can be generated by grouping similar changes in the major acceptable process changes together and generating one of the minor acceptable process changes for each group. Each minor acceptable process change is associated with multiple progress states, which correspond to the progress states associated with the major acceptable process change in the corresponding group, and the minor acceptable process change is generated for the corresponding group.

[0092] Each minor acceptable process change can be generated based on the characteristics of the major acceptable process change for that corresponding group, and the minor acceptable process change is generated for that corresponding group.

[0093] Values ​​derived from a similarity function can be used to group the major acceptable process variations.

[0094] The processor can be configured to receive user input identifying areas of a workpiece and to group major acceptable process variations associated with the identified areas using different similarity criteria than those used to group major acceptable process variations associated with other areas of the workpiece.

[0095] According to a seventh aspect of the invention, a data carrier having instructions thereon is provided, which, when executed by a processor, causes the processor to perform according to a sixth aspect of the invention.

[0096] According to an eighth aspect of the invention, a method for controlling an additive manufacturing apparatus is provided, the method comprising: providing instructions for constructing a workpiece by curing material layer by layer using the additive manufacturing apparatus; determining a plurality of acceptable process variations for sensor signals of the additive manufacturing apparatus; and generating a build file including the instructions and the plurality of acceptable process variations.

[0097] Each of these acceptable process changes can be associated with at least one state of the workpiece's construction progress.

[0098] According to a ninth aspect of the invention, an apparatus including a processor is provided, the processor being arranged to perform the method of the eighth aspect of the invention.

[0099] According to a tenth aspect of the present invention, a data carrier having instructions stored thereon is provided, which, when executed by a processor, causes the processor to perform the method of the eighth aspect of the present invention.

[0100] According to an eleventh aspect of the invention, a data carrier having a construction file thereon is provided, the construction file being used to instruct an additive manufacturing apparatus in the construction of a workpiece in a layer-by-layer manner, the construction file including: instructions on how the additive manufacturing apparatus constructs the workpiece and acceptable process variations of sensor signals generated by sensors of the additive manufacturing apparatus during the construction of the workpiece.

[0101] Because acceptable process variations are specific to the workpiece to be built, storing both the acceptable process variations and the instructions for building the workpiece in the build file ensures that both are transmitted to the additive manufacturing equipment when the build file is uploaded.

[0102] According to a twelfth aspect of the present invention, a method for monitoring an additive manufacturing apparatus is provided, the method comprising: capturing a plurality of thermal images of a material layer when the material is cured, determining a rate of temperature change of a cured region based on a comparison of the plurality of thermal images, and storing a record of the rate of temperature change of the cured region.

[0103] The method can record the rate of temperature change in multiple regions of a material layer. A two-dimensional graph of the rate of temperature change can be stored for each layer. This method may include displaying the two-dimensional graph to a user.

[0104] Calibrate a thermal camera used to determine the rate of temperature change more easily than calibrating one used to determine the absolute temperature. However, it is argued that knowing the rate of temperature change, rather than the absolute temperature, provides useful feedback in additive manufacturing equipment. For example, thermal stresses that occur in the workpiece during build-up can be correlated with the thermal gradient across the cured material rather than the absolute temperature. The rate of temperature change can be determined based on the differences in pixel intensity across multiple images corresponding to common locations on the layer.

[0105] According to a thirteenth aspect of the present invention, an additive manufacturing apparatus is provided, comprising a support, a material source for supplying material to the support, a radiation source for curing material layers to form a workpiece in a layer-by-layer manner, a thermal camera for imaging the material when it is cured, and a processor arranged to: receive images from the thermal camera, and compare multiple images of the cured material to determine the rate of temperature change of the cured material.

[0106] According to a fourteenth aspect of the invention, a data carrier having instructions is provided, which, when executed by a processor, causes the processor to perform a method according to a ninth aspect of the invention.

[0107] According to a fifteenth aspect of the present invention, a method for monitoring an additive manufacturing apparatus is provided, the method comprising: capturing images of a continuous layer of material; determining differences in images between corresponding regions of the continuous layer for a plurality of corresponding regions; determining, based on the differences, a dissimilarity value quantifying the overall difference of the images of the plurality of corresponding regions; and controlling the additive manufacturing apparatus based on whether the dissimilarity value is above or below a threshold.

[0108] In this way, overall measurements are provided to identify whether build events falling outside of expected process variations have occurred, and thus to control the additive manufacturing equipment. For example, the additive manufacturing equipment may initially attempt to automatically resolve issues, such as the re-dosing of powder layers. If automatic resolution fails to bring the differences in the images of successive layers below a threshold (e.g., within expected process variations), the additive manufacturing equipment may flag a difference as having occurred to the user / operator.

[0109] Before curing using an energy beam / plasma stream, the image can be a continuous layer of material. A threshold can be set based on a desired dissimilarity value, identified for two continuous material layers that have been correctly formed (e.g., with sufficient material coverage) and do not have any cured parts protruding from them. The threshold can be determined based on a statistical analysis (natural variation) of the dissimilarity values ​​that generate continuous layers for such correct formation, with the threshold set above the statistically determined natural variation.

[0110] The corresponding regions can be individual pixels or groups of pixels in an image. Dissimilarity values ​​can be determined based on the differences between corresponding regions, where larger differences are weighted relative to their magnitude, thus contributing disproportionately to the dissimilarity value. In this way, dissimilarity values ​​will be significantly affected by large differences between a small number of corresponding regions compared to smaller differences among many corresponding regions.

[0111] According to a sixteenth aspect of the present invention, an additive manufacturing apparatus is provided, comprising: a support, a material source for supplying material to the support, and a radiation source for curing the material layer to form a workpiece in a layer-by-layer manner, a camera for imaging a continuous material layer, and a processor arranged to: determine differences in images between corresponding regions of the continuous layer for a plurality of corresponding regions, determine a dissimilarity value quantifying the overall difference of the images of the plurality of corresponding regions based on the differences, and control the additive manufacturing apparatus based on whether the dissimilarity value is above or below a threshold.

[0112] According to a seventeenth aspect of the invention, a data carrier having instructions is provided, which, when executed by a processor, causes the processor to perform the method of the fifteenth aspect of the invention.

[0113] According to an eighteenth aspect of the present invention, a method for monitoring an additive manufacturing apparatus is provided, the method comprising: capturing an image of a material layer spread by a scraper before curing the layer using an energy beam, determining a dissimilarity value that quantifies the differences between different columns of the image, and controlling the additive manufacturing apparatus to change / adjust the scraper based on whether the dissimilarity value is above or below a threshold and / or generating a warning for a user to check the scraper.

[0114] According to a nineteenth aspect of the invention, an additive manufacturing apparatus is provided, comprising: a support for supporting a material bed; a material source for providing material; a scraper for spreading the material provided by the material source across the material bed; a radiation source for generating an energy beam for curing material layers to form a workpiece in a layer-by-layer manner; a camera for capturing an image of each layer before curing the layer using the energy beam; and a processor arranged to: determine, for each image, a dissimilarity value quantifying the differences between different columns of the image; and control the additive manufacturing apparatus to change / adjust the scraper based on whether the dissimilarity value is above or below a threshold and / or generate a warning for a user to check the scraper.

[0115] According to a twentieth aspect of the invention, a data carrier having instructions is provided, which, when executed by a processor, causes the processor to perform the method of the eighteenth aspect of the invention.

[0116] According to a twenty-first aspect of the invention, a method for monitoring an additive manufacturing process is provided, wherein a workpiece is constructed by curing material in a layer-by-layer manner, the method comprising: receiving sensor signals from the additive manufacturing equipment during the construction of the workpiece, identifying areas of the workpiece from the sensor signals for further inspection, and manipulating a sensor system to capture further sensor data from the identified areas.

[0117] The sensor system may include an optical sensor system for capturing radiation generated by the area. The optical system may include steerable optics for capturing radiation emitted from a selected location on the workpiece. The method may include controlling the steerable optics for capturing radiation from the identified area. Alternatively, the optical system may be mounted on a bench system for guiding the optical system to capture radiation emitted from a selected location on the workpiece. The method may include manipulating the optical system on the bench system to capture further sensor data from the identified area.

[0118] The optical system can share optical components with an optical system used to guide the laser beam to cure the material, wherein the optical system is guided to capture further data in an area between curing of the material using the laser beam. For example, the optical system can be guided to capture radiation from an identified area during movement of a platform that forms a layer using a scraper and / or supports a bed of material. Alternatively, if the area is identified for further inspection, the predetermined movement of the optical system used to cure the material can be modified to allow further data to be captured in the area during the time period in which the material is cured. During the capture of further sensor data in the area, the laser beam used to cure the material can be turned off, its power reduced, or it can be defocused to prevent re-curing of the identified area during the capture of further sensor data. Using a laser beam with a lower energy density than required for curing the material as a laser probe can be useful to stimulate radiation emission from the identified area.

[0119] Alternatively, the optical system may include a separate optical system for manipulating the energy beam used to solidify the material.

[0120] If the sensor signal falls outside the acceptable process variation, the area can be identified for further inspection.

[0121] According to a twenty-second aspect of the invention, a controller for controlling an additive manufacturing apparatus is provided, the controller including a processor arranged to: receive sensor signals from the additive manufacturing apparatus during the construction of a workpiece, identify areas of the workpiece based on the sensor signals for further inspection, and generate instructions for manipulating a sensor system to capture further sensor data from the identified areas.

[0122] According to a twenty-third aspect of the present invention, an additive manufacturing apparatus including a controller according to a twenty-second aspect of the present invention is provided.

[0123] According to a twenty-fourth aspect of the invention, a data carrier having instructions is provided, which, when executed by a processor, causes the processor to perform the method of the twenty-second aspect of the invention.

[0124] According to a twenty-fifth aspect of the invention, a method is provided for generating instructions for constructing a workpiece in an additive manufacturing apparatus, the method comprising: receiving, in response to sensor signals generated during the construction of the workpiece using the additive manufacturing apparatus, a set of primary acceptable process variations, each primary acceptable process variation applicable to the sensor signals generated during a specific state of the workpiece's construction progress associated with the primary acceptable process variation; and generating, based on the set of primary acceptable variations, a set of simplified secondary acceptable process variations, at least one of which is associated with multiple states of the workpiece's construction progress.

[0125] A simplified set of minor acceptable process changes can be generated by grouping similar changes in the major acceptable process changes together and generating one of the minor acceptable process changes for each group. Each minor acceptable process change is associated with multiple progress states, which correspond to the progress states associated with the major acceptable process change in the corresponding group, and the minor acceptable process change is generated for the corresponding group.

[0126] Each minor acceptable process change can be generated based on the characteristics of the major acceptable process change for the corresponding group, and the minor acceptable process change is generated for that corresponding group.

[0127] Values ​​derived from a similarity function can be used to group the main acceptable process variations.

[0128] The processor can be configured to receive user input identifying different regions of a workpiece, for which different similarity criteria are used to group the main acceptable process variations together.

[0129] According to a twenty-sixth aspect of the invention, a controller including a processor is provided, the processor being arranged to perform the method of a twenty-fifth aspect of the invention.

[0130] According to a twenty-seventh aspect of the present invention, a data carrier having instructions is provided, which, when executed by a processor, causes the processor to perform the method of the twenty-fifth aspect of the present invention.

[0131] According to a twenty-eighth aspect of the invention, a method is provided for generating instructions for a machine in a manufacturing chain used to manufacture a workpiece, the manufacturing chain including additive manufacturing equipment, the method comprising: receiving sensor signals from the additive manufacturing equipment during the construction of the workpiece, and generating instructions for at least one additional machine in the manufacturing chain based on the sensor signals.

[0132] The method may include comparing a sensor signal with an acceptable process change in the sensor signal, and generating instructions for at least one additional machine in the manufacturing chain based on the comparison.

[0133] The additional machines may include subtractive manufacturing machines, polishing machines, support removal machines, measuring machines, additional additive manufacturing machines, machines for relieving thermal stress in workpieces, and / or machines for finishing the building substrate.

[0134] According to a twenty-ninth aspect of the invention, an apparatus is provided for generating instructions for a machine for a manufacturing chain used to manufacture workpieces, the apparatus including a processor arranged to perform the method of a twenty-eighth aspect of the invention.

[0135] According to a thirtieth aspect of the invention, a data carrier having machine-readable instructions stored thereon is provided, wherein, when executed by a processor, the machine-readable instructions cause the processor to perform the method of the twenty-eighth aspect of the invention.

[0136] A manufacturing chain for manufacturing workpieces, the manufacturing chain including additive manufacturing machines, at least one additional machine, and equipment according to the twenty-ninth aspect of the invention.

[0137] According to a thirty-first aspect of the invention, a method for manufacturing a workpiece is provided, the method comprising using an additive manufacturing apparatus, wherein the workpiece is formed by curing material in a layer-by-layer manner, the method comprising: receiving sensor signals from sensors of the additive manufacturing apparatus during the construction of the workpiece, and displaying a representation of each sensor signal and a corresponding acceptable process deviation of the signal from the sensor for a state of construction progress at the time the sensor signal is captured, wherein different states in the state of progress are associated with different acceptable process variations.

[0138] According to a thirty-second aspect of the invention, a visualization device is provided for use in a manufacturing process, the visualization device including a processor and a display, wherein the processor is arranged to perform the method of the thirty-first aspect of the invention, such that the display shows the representation.

[0139] According to a thirty-third aspect of the present invention, a data carrier having instructions is provided, which, when executed by a processor, causes the processor to perform the method of the thirty-first aspect of the present invention.

[0140] According to a thirty-fourth aspect of the invention, a method is provided for manufacturing a plurality of nominally identical workpieces using one or more additive manufacturing apparatuses, wherein the workpieces are formed by curing materials in a layer-by-layer manner, the method comprising: receiving sensor signals from the additive manufacturing apparatus or each additive manufacturing apparatus during the construction of the workpieces; determining a plurality of acceptable process variations based on the sensor signals for different states of the construction progress; and displaying, for the state of the construction progress at the time the sensor signals are captured, a representation of each sensor value of at least one of the workpieces and a corresponding acceptable process deviation of the plurality of acceptable process variations.

[0141] According to a thirty-fifth aspect of the invention, a visualization device for use in a manufacturing process is provided, the visualization device including a processor and a display, wherein the processor is arranged to perform the method described in the thirty-fourth aspect of the invention, such that the display shows the representation.

[0142] According to a thirty-sixth aspect of the invention, a data carrier having instructions is provided, which, when executed by a processor, causes the processor to perform the method of a thirty-fourth aspect of the invention.

[0143] The data carrier can be a suitable medium for providing instructions to the machine, such as a non-transitory data carrier, for example, a floppy disk, CD ROM, DVD ROM / RAM (including -R / -RW and +R / +RW), HD DVD, Blu-ray™ disc, memory (such as Memory Stick™, SD card, compact flash memory card, etc.), disk drive (such as hard disk drive), magnetic tape, any magneto / optical memory, or a transient data carrier, such as a signal on a wire or fiber optic cable or a wireless signal, for example, a signal transmitted over a wired or wireless network (such as Internet download, FTP transfer, etc.). Attached Figure Description

[0144] Figure 1 This is a schematic representation of a selective laser melting (SLM) apparatus according to the present invention;

[0145] Figure 2 It is a schematic representation of the gas recirculation loop in a selective laser melting device;

[0146] Figure 3 This is a schematic representation of the optical unit of a selective laser melting device;

[0147] Figure 4 This illustrates the point scanning strategy used by the SLM device;

[0148] Figure 5 It is a schematic representation of sensor signals that can be generated by multiple components of the same nominal workpiece;

[0149] Figure 6 This is a flowchart illustrating a method according to an embodiment of the present invention;

[0150] Figure 7 This is a graph showing point-by-point-based photodiode data captured as the execution layer scans multiple shadows;

[0151] Figure 8 It is a schematic representation of the correlation between major acceptable process variations and the construction progress status of the workpiece, and a hierarchical clustering algorithm for clustering major acceptable process variations based on similarity measurements;

[0152] Figure 9 It is a schematic representation of the relationship between minor acceptable process changes and the progress status of the workpiece construction in a one-to-many manner;

[0153] Figure 10 Images of layers in a selective laser melting apparatus, taken by a visible camera before and after the material in the layers has solidified, are shown; and

[0154] Figure 11 shows a visualization device for representing sensor data and acceptable process variations, wherein Figure 11(a) is the first part of Figure 11, Figure 11(b) is the second part of Figure 11, and Figure 11(c) is the third part of Figure 11. Detailed Implementation

[0155] refer to Figure 1 A selective laser melting (SLM) apparatus according to an embodiment of the present invention includes a build chamber 101 having partitions 114, 115 defining a build volume 116 and a surface on which powder can be deposited. A build platform 102 defines a working area in which a workpiece 103 is built by selectively laser melting powder 104. As a continuous layer of workpiece 103 is formed, the platform 102 can be lowered within the build volume 116 using a lifting mechanism 117. The available build volume is defined by the extent to which the build platform 102 can be lowered into the build volume 116. The build platform 102 divides the build chamber 101 into an upper chamber 120 and a lower chamber 121 according to the concept disclosed in WO 2010 / 007394.

[0156] During the fabrication of workpiece 103 using dispensing device 109 and scraper 108, a layer of powder 104 is formed. For example, the dispensing device may be as described in WO 2010 / 007396. Scraper 108 is mounted to allow upward movement of scraper 108 relative to a bias member (not shown), and load sensor 193 is provided to detect vertical deflection of scraper 108. This vertical deflection of scraper 108 can be caused by cured portions of the fabrication protruding from the bed, such as curling or deformation caused by thermal stress during fabrication. As scraper 108 crosses protrusions, load sensor 193 provides feedback on the presence of these protrusions.

[0157] Encoders 194 and 195 are used to measure the positions of the construction platform 102 and the scraper 108. The position information is fed back to the computer 160.

[0158] Laser module 105 generates a laser for melting powder 104, and under the control of computer 160, guides the laser onto powder bed 104 according to the requirements of optical module 106. The laser enters chamber 101 through window 107.

[0159] Computer 160 includes a processor unit 161, memory 162, display 163, user input device 164 (such as a keyboard, touchscreen, etc.), data connection to modules of the laser melting equipment (such as optical module 106, laser module 105), and motors (not shown) that drive the dispensing device 109, scraper 108, and build platform 102 to move. External data connection 166 is used to upload build files to computer 160. The movement of laser unit 105, optical unit 106, and build platform 102 is controlled by computer 160 based on scan instructions contained in the build files.

[0160] refer to Figure 2 The device includes a gas nozzle 140 and a gas outlet 141 for generating an airflow across the build platform 102 and through the upper chamber 120. The airflow acts as an air knife, carrying condensate generated by melting powder using a laser located away from the build area. The device includes additional gas nozzles 144 for generating an airflow across window 107. This airflow can prevent condensate from collecting on window 107, which could affect the quality of the laser beam 118 delivered through window 107.

[0161] Vent 143 provides means for venting / removing gas from chambers 120, 121. Backfill inlet 145 provides an inlet for backfilling chambers 120, 121 with inert gas. Lower chamber 121 may include additional inlets 146 for holding lower chamber 121 at an overpressure position relative to upper chamber 120.

[0162] The gas flow circuit includes filter assemblies 200 and 201 connected in parallel within the gas circuit for filtering particles in the recirculated gas. Each filter assembly 200 and 201 includes filter housings 202 and 203, filter elements E-5 and E-7 located within the filter housings 202 and 203, and manually operated valves V-2, V-3, V6, and V-7 for opening and closing the gas inlet and gas outlet, respectively. Each filter assembly 200 and 201 is removable from the gas circuit to replace the filter, as described in WO2010 / 026396 (see [link to WO2010 / 026396]). Figure 4 b).

[0163] Pump E-4 generates an airflow through the gas circuit. Gas is discharged from pump E-4 via gas nozzles 140, 144 to produce an air knife across the build surface and window 107. The pump can also deliver gas to inlet 146 in the lower chamber 121 to maintain the lower chamber 121 at an overpressure relative to the upper chamber 120. Outlet 141 is connected to filter assemblies 200, 201 via pressure sensor I-5 to complete the gas circuit.

[0164] Backfill inlet 145 is connected to a source of inert gas 211, and the flow of inert gas to the backfill inlet is controlled by solenoid valve V-19.

[0165] Vent 143 is connected to solenoid valve V-18 and vacuum pump E-1, providing a means for generating low pressure or vacuum in upper chamber 120 and lower chamber 121. Oxygen sensor I-4 detects the amount of oxygen present in the gas discharged from chambers 120 and 121 through vent 143. Vent 143 is also connected to pressure sensor I-2 and exhaust valve V-17. Pressure sensor I-2 measures the gas pressure at vent 143, and if pressure sensor I-2 measures excessive pressure, exhaust valve V-17 is opened. Normally, upper chamber 120 is maintained at a slightly overpressure relative to atmospheric pressure.

[0166] Figure 3 The optical module 106 is shown in detail. The optical module includes: a laser aperture 170 for connection to the laser module 105; a measuring aperture 171 for connection to the measuring devices 172, 173; and an output aperture 174 through which the laser beam is guided through the window 107 to the powder bed 104, and the radiation emitted from the powder bed is collected.

[0167] By scanning an optical system comprising two tiltable mirrors 175 (only one is shown) and focusing lenses 176, 177, the laser beam is manipulated and focused onto the desired location on the powder bed 104.

[0168] Each tiltable mirror 175 is mounted to rotate about an axis under the control of an actuator such as a galvanometer. These axes of rotation of the mirrors 175 are substantially perpendicular, such that one mirror can deflect the laser beam in one direction (X-direction), and the other mirror can deflect the laser beam in a perpendicular direction (Y-direction). However, it should be understood that other arrangements can be used, such as a single mirror rotatable about two axes, and / or the laser beam can be connected, for example, via optical fiber to a mirror mounted to move linearly in both the X and Y directions. Examples of the latter arrangement are disclosed in US 2004 / 0094728 and US 2013 / 0112672.

[0169] To ensure that the focal point of the laser beam remains in the same plane despite changes in the deflection angle, it is known to provide an f-θ lens after a tiltable mirror. However, in this embodiment, a pair of movable lenses 176, 177 are provided before the tiltable mirror 175 (relative to the direction of laser beam travel) to focus the laser beam as the deflection angle changes. The movement of the focusing lenses 176, 177 is controlled synchronously with the movement of the tiltable mirror 175. The focusing lenses 176, 177 may be movable in a linear direction toward and away from each other via an actuator such as a voice coil 184.

[0170] The tiltable mirror 175 and focusing lenses 176, 177 are suitably selected to transmit both a laser wavelength typically 1064 nm and the wavelength of the collected radiation emitted from the molten pool.

[0171] Beam splitter 178 is disposed between focusing lenses 176, 177 and laser 105 and measuring devices 172, 173. Beam splitter 178 is a notch filter that reflects light with the laser wavelength but allows light with other wavelengths to pass through it. The laser light is reflected towards focusing lenses 176, 177, and light without the laser wavelength, collected by scanning optics, is transmitted to measuring aperture 171.

[0172] The optical module 106 further includes a heat absorber 181 for capturing laser light transmitted through the beam splitter 178. As expected, most of the laser light is reflected by the beam splitter 178. However, a very small percentage of the laser light passes through the beam splitter 178, and this small fraction is captured by the heat absorber 181. In this embodiment, the heat absorber 181 includes a central cone 182 that reflects light onto a scattering surface 183 located on the wall of the heat absorber 181. The scattering surface 183 can be a surface with corrugated or ridged surfaces that disperse the laser light. For example, the scattering surface 183 can include ridges with a helical or spiral shape. The scattering surface can be made of anodized aluminum. A photodiode can be provided in the beam collector as a device for detecting laser powder.

[0173] Various measuring devices can be connected to the measuring aperture 171. In this embodiment, a spectrometer 172 and a photodiode 173 are provided for measuring the radiation collected by the optical scanner. A additional beam splitter 185 decomposes the radiation reflected to the aperture 171 so as to direct a portion of the radiation to the spectrometer 172 and a portion to the photodiode 173.

[0174] The selective laser curing apparatus further includes a visible camera 191 and an infrared (thermal) camera 192 for capturing images of the material bed 104, and acoustic sensors 197, 198 for recording acoustic signals generated during the curing process.

[0175] In use, the calculator 160 receives a build file that includes instructions in the form of scan paths and scan parameters, as well as multiple acceptable process variations for each sensor during workpiece construction. These scan paths and parameters will be used in the formation of each layer of the additive build. The acceptable process variations include the acceptable range of values ​​for the device's sensor signals. The acceptable range of values ​​can be the absolute value of the sensor signal, derived from the sensor signal value, the rate of change of the sensor signal, the cumulative error in the sensor signal, and / or the integral of the window function. For example, acceptable process variations could be acceptable variations in the position of platform 102 relative to its nominal position, acceptable variations in the position and / or speed of scraper 108 relative to its nominal position / speed, acceptable variations in the air pressure of build chamber 101, acceptable variations in the oxygen concentration in build chamber 101, acceptable variations in signals from spectrometer 172 and / or photodiode 173, acceptable variations in signals from acoustic sensors 197, 198, acceptable variations in the temperature of powder bed 104 (derived from thermal camera 192), acceptable variations in signals from photodiodes measuring the delivered laser power, and acceptable process variations in the differences between adjacent layer images captured by visible camera 191. For each sensor signal, the acceptable range of values ​​depends on the state of build progress; for example, the acceptable range of values ​​could vary with the position of the workpiece being formed, the number of layers being processed, and / or the build time.

[0176] This acceptable process variation is specific to the given build and thus forms part of the build file, which is sent to the additive manufacturing equipment to indicate the build of the workpiece.

[0177] Computer 160 controls various modules of the selective laser melting equipment according to scanning instructions for workpiece construction. During construction, the computer receives signals from sensors 172, 173, 191, 192, 193, 194, 195, 197, 198, I-4, and I-2. Computer 160 determines whether the signal or the value derived from the signal (such as the temperature of the molten pool) is within the acceptable process variation specified in the construction file.

[0178] If a signal or its value is outside the acceptable process variation, computer 160 controls the selective laser melting equipment to take appropriate action. For example, computer 160 can cause the additive manufacturing process to stop. This can be done if the oxygen concentration or pressure in the build chamber is outside the acceptable process variation. If the melt pool temperature is outside the acceptable process variation or moves toward an external limit of the acceptable process variation, computer 160 can change the laser beam power to bring the melt pool temperature within the acceptable process variation. If a protrusion is detected on the powder bed 104 by the load sensor on the scraper 108, the computer generates a user warning to draw the operator's attention to the protrusion / defect. If the visible image of a powder layer differs from that of a previous powder layer outside the acceptable process variation, computer 160 can instruct the powder layer to be re-dosed. If an imaging processing technique detects a line in the visible image of a powder layer outside the acceptable process variation, scraper damage may be suspected, and the computer can generate a warning to indicate the possible damage to the scraper to the operator. In cases where the thermal camera image undergoes variations beyond acceptable process variations, scanning parameters and / or scanning sequence / strategy can be varied to alter the energy density input from the laser beam to subsequent layers. (Reference) Figure 4 The scanning parameters that can be changed are laser power, spot size, dot distance, and exposure time.

[0179] Additive manufacturing equipment can also be controlled to re-examine areas of the build that have generated sensor signals beyond acceptable process variations. For example, during periods when no laser beam is used to cure the material, such as when spreading a layer using a scraper 108 on or lowering the build platform 102, the steering mirror 175 of the optical module 106 can be directed to capture additional radiation emitted from the identified area. The equipment can be controlled to expose the identified area to a lower-power laser beam than required to cure (melt to sinter) the material, in order to heat the area so that defects / faults in the build can be determined based on the thermal characteristics generated during heating of the area using the lower-power laser beam.

[0180] Computer 160 is configured to store a log of the build progress status in memory 162a for the build, where sensor signals fall outside the corresponding acceptable process variations. The storage of sensor signals may include pre-triggering and / or post-triggering, where sensor signals surrounding the stored sensor signals that triggered the sensor data are stored. Processor 161 may enable display 163 to show a 2D or 3D representation of workpiece 103, indicating areas of workpiece 103 where sensor signals outside the acceptable process variations are generated as these areas are formed. Computer 160 may also store sensor data for these areas, making this data available to a user who wishes to further investigate the occurring process variations.

[0181] refer to Figure 5 and Figure 6 The method for determining acceptable process variations in the construction of a workpiece will now be described. Acceptable process variations are determined by constructing a series of identical workpieces according to a set of general construction instructions. Initial settings for acceptable process variations are used to construct the first workpiece, such as setting limits to protect the equipment from damage, such as damage to the scraper caused by areas protruding from the powder bed, or, as determined by previous constructions, such as constructions of other workpieces with the same material or previous iterations of multiple constructions. The initial settings for acceptable process variations should be set wide enough to capture at least a majority of sensor values ​​that would likely allow the construction to meet user-specified requirements. For example, initial settings for acceptable process variations could exclude laser parameters that would not achieve sufficient energy density for melting powder.

[0182] A first workpiece in a series of workpieces is constructed 401, and sensor data 402, including sensor signals generated during construction, is retrieved. When the sensor signals are recorded / captured, they are correlated with the state of construction progress (e.g., position on the workpiece or layer being cured or construction time). The sensor signals can be plotted as a function of position p or time t, such as... Figure 5 The figure shown illustrates sensor data V plotted on the same graph for builds 1 through 5.

[0183] Then, the workpiece is inspected, for example, using non-destructive and destructive techniques, and workpiece 403 is evaluated to determine whether the workpiece meets specified requirements. Specified requirements may, for example, be workpiece density, the number of inclusions in the workpiece, workpiece strength and / or workpiece surface finish, average grain size and / or dimensions, as determined according to standard methodologies. If an area of ​​the workpiece is evaluated as not meeting specified requirements, the user identifies whether the fault is a systemic fault affecting the entire or majority of the workpiece, or a fault localized within the area. Figure 5 Two instances are shown, including unrecoverable fault 301, where the sensor value does not return to the acceptable process variation limit T. Uand T L Within, and recoverable build failures, where the sensor value returns to the acceptable process variation limit T. U and T L The former indicates a systemic failure, and all sensor data (or at least sensor data from the point where the sensor signal begins to disperse) is identified as related to an unacceptable build (i.e., non-golden data). The latter indicates a localized failure, and the user can assume that this failure will not adversely affect the rest of the build; accordingly, only sensor data from the faulty area is identified as related to the faulty build (i.e., non-golden data), and the rest of the sensor data for the build is identified as related to an acceptable build (i.e., golden data).

[0184] Based on sensor data and identifiers for gold and non-gold sensor data, acceptable process variations 404 are determined for each state of construction progress, and sensor data is generated for each. Acceptable process variations can be determined through any suitable statistical analysis. This may result in an envelope of acceptable sensor values, such as... Figure 5 The middle is composed of higher T U Threshold and lower T L The threshold is shown.

[0185] The method then includes constructing the next artifact in the series and updating acceptable process variations based on sensor data generated during the construction of this artifact. This process is performed until predetermined conditions are met, such as when the variation in acceptable process variations between constructions falls below a predetermined level.

[0186] Figure 7 The sensor signal received from photodiode 173 is shown during a raster scan of a square geometry at an angle of 45 degrees to the edge of the square. The photodetector voltage is plotted as a function of the number of pulses (i.e., the number of point exposures). For an exposure time of 60 microseconds, approximately 30 photodetector samples were collected for each point exposure. The top graph 7A shows the raw data, and the bottom graph 7B shows the average value plus and minus the average of two standard deviations for each pulse number. The separation between each shading line can be identified by the large dip in the photodetector voltage. Furthermore, it can be seen that the fingerprint of the photodetector voltage used for each shading scan changes during each shading period and depends (among other things) on the shading length and shading direction relative to the airflow through the construction chamber 120 (the photodetector voltage increases or decreases during shading periods depending on the shading direction relative to the airflow).

[0187] Based on this data, it is clear that the acceptable process variation of the photodetector voltage, defined by the type of scan being performed (in this case, raster scan), cannot alone be sufficient to capture the true range of acceptable process variations required to produce a workpiece that meets specified requirements.

[0188] Accordingly, as described above, acceptable process variation is determined for each exposure point, and the sensor signal collected for that exposure point is compared with the acceptable process variation for that exposure point. Exposure points may be defined by (as shown in the figure) number / sequence, location on the workpiece, or build time. Acceptable process variation may be the variation in average photodetector voltage at that point, maximum and minimum photodetector voltage and / or acceptable standard deviation, RMS, cluster analysis, histogram shape of photodetector values, standard deviation variation, variation in different signal ratios, rolling average, and / or filtered or smoothed average.

[0189] For additive manufacturing apparatuses with only one or a few detectors for sensing properties of the additive process, a one-to-one correlation between acceptable process variation and build progress status (such as exposure points and / or layer number) is feasible. However, as the number of detectors used to monitor the additive process and the complexity of acceptable process variation increase, the amount of data on acceptable process variation becomes particularly large. For example, in the apparatus disclosed above, the sensor signal for each exposure point can be generated by a photodetector 173, a spectrometer 172, and acoustic sensors 197, 198, and multiple acceptable process ranges can be used for the sensor signals generated by each detector 172, 173, 197, and 198. Storing such acceptable process variation (specifically, each exposure point) for every state of build progress can result in excessively large data files.

[0190] refer to Figure 8 and Figure 9 To reduce the amount of data required to control acceptable process variations for the build, the main acceptable process variations (PVs) determined for each state of build progress can be clustered (grouped) based on a measure of similarity (e.g., using a suitable clustering algorithm, such as hierarchical clustering). This is represented in... Figure 8 In the tree structure, the similarity of major acceptable process changes is indicated by the position of the branches on the vertical axis. New minor acceptable process (SV) changes are generated for each cluster from the major acceptable process (PV) changes within the cluster. The minor acceptable process (SV) changes are used for all progress states associated with the major acceptable changes falling within the cluster. The resulting one-to-many relationship is... Figure 9As shown in the diagram, the similarity values ​​S1 and S2 required for grouping major acceptable process variations (PVs) can be selected by the user and may differ for different regions of the workpiece. For example, a user can set a high similarity for problematic regions (such as protrusions) or critical regions of the workpiece where strict process control is expected, while setting a low similarity for non-problematic or non-critical regions where the failure to meet specified requirements is less important.

[0191] Reference Figure 10 To describe further examples of sensor data that can be used in the methods described above. Figure 10 Image 201 of the powder layer before processing is shown, and image 202 of the processed layer, captured as seen by a camera in an SLM device, is shown. It can be seen that the cured material 203 is significantly different from the uncured powder 204. Computer 160 analyzes image 201 of the powder layer (before curing) by comparing images of consecutive layers. First, the images are corrected to ensure orthogonality, and then a histogram of the resulting image is generated when the image 201 of the previous powder layer is subtracted from the image 201 of the current powder layer using each color channel, and the results are summed to provide a histogram of pixel counts for each bar (each bar is related to the difference interval between images 201). A single output value for the difference between the two images 201 is then generated by weighting the histogram data, such that pixel counts with larger difference intervals are more heavily weighted to contribute more to the single output value. In this way, counts corresponding to bars with larger contrast contribute more to the output value relative to the magnitude of the contrast than counts in bars with smaller contrast. Accordingly, pixels with smaller differences have a smaller impact on the final output value, effectively filtering out small differences in output that may occur due to noise inherent in image 201.

[0192] Computer 160 determines whether the output value representing the difference between the two images 201 is within acceptable process variations for that layer, such as being below a threshold level. If the value is outside (above) acceptable process variations, it is determined that the cured material is more pronounced in one image than in the other, and an abnormal build-up event has occurred. In response, computer 160 may instruct dispensing scraper mechanisms 108, 109 to re-dispense the powder layer to determine whether the significant difference in image 201 is caused by incomplete dispensing. If the image of the re-dispensed layer still produces an output value outside acceptable process variations compared to the image 201 of the previous layer, an alert may be generated for the operator to visually inspect the layer where the abnormality has occurred, such as protrusion of solid material through the powder layer.

[0193] Figure 10The graph in the image is an image of the individual output values ​​of a certain number of layers, showing the peak values ​​in the output values ​​of some layers. Computers can generate graphs like... Figure 10 The display shown allows the operator / user to quickly identify the image to be viewed by selecting a value on the graph (such as a high value (peak)), and the computer 160 displays an image 201 of the powder layer that generated that output value on the monitor 163. In this way, the operator / user can visually view the image to determine what caused the peak in the output value.

[0194] Instead of generating a single output value for a comparison of the entire image, output values ​​can be generated in a similar manner for portions of the image being compared. This allows for the identification of portions of the powder layer that cause significant differences between the powder layer image and previous layer images. This can aid in problem identification. For example, warnings generated for operators could specify the appropriate location on the powder layer where the warning was issued.

[0195] This section can be a column or row of an image extending in the direction of travel of the squeegee blade. The output value of each column / row will give an indication of the presence of linear artifacts (such as ridges of powder) in that column or row. The presence of raised ridges can indicate that the squeegee 108 has been damaged. Accordingly, when detecting the output value of a column / row above a threshold, the computer 160 can generate a warning for the operator to check for damage to the squeegee 108.

[0196] When the cured material generates a thermal image that differs from the powder thermal image, a similar process can be performed on the thermal image to identify workpiece misalignment or curling faults.

[0197] Furthermore, each image can be formed based on the photodiode output of the layer. Exposure points (for which photodiode data is collected) effectively form an image of the workpiece, and acceptable variations in photodiode data between successive layers can be known based on previous workpiece construction. The method described above for analyzing successive images can provide a rapid indication that the construction process has moved outside the acceptable process window. Specifically, the photodiode response to a laser beam impacting a cured material differs significantly from the response when impacting powder.

[0198] Images from a thermal camera can be further used to determine the cooling rate of the cured material. A series of thermal images of the layer can be captured during material curing, and the cooling rate is determined based on the intensity variation of the most recently cured area between different thermal images of the region captured at different times. The determined cooling rate is then compared to an acceptable process variation in the cooling rate. If the cooling rate is outside the acceptable process variation, the computer 160 can change the scanning parameters, thereby changing the energy input per unit area to the material. Based on the geometry of the workpiece, different acceptable process variations in the cooling rate may exist for different regions of the cured material. Specifically, the acceptable process variation in the cooling rate may differ for a cured material layer (such as a protrusion) formed on a previously cured layer of cured material, compared to a cured material layer formed on a previously cured layer of cured material. In a further embodiment, the cooling rate of the workpiece is modeled, and the acceptable process variation in the cooling rate is derived from the modeled variation in the cooling rate.

[0199] According to reference Figure 5 and Figure 6 The method described may have determined acceptable process variations in cooling rates based on the materials used in the previous build.

[0200] A graph showing the cooling rate measured across layers can be displayed to the user on monitor 163. A color-coded graph can be displayed, with different colors corresponding to different cooling rates.

[0201] This invention provides a method for controlling an additive manufacturing apparatus based on acceptable process variations determined from data generated by previous layers and / or one or more previous builds for a workpiece. This avoids the need for a comprehensive understanding of the complex processes occurring within the additive manufacturing apparatus.

[0202] Acceptable process variations are formed as part of a build file sent to the additive manufacturing equipment, which instructs the equipment to perform the build. Because acceptable process variations are embedded in the build file rather than in a separate, potentially overwritten file on the SLM machine, the build file provides a traceable source for identifying the evolution of acceptable process variations. This traceability is useful in certain manufacturing industries, such as the aerospace industry, where a product history needs to be documented.

[0203] Regression testing can be used to demonstrate that acceptable process changes remain effective for new material manufacturing machines with design modifications and / or new features / functions. Such regression testing can be performed on new machines operating according to new specifications and / or on new machines configured to simulate the functions of older machines.

[0204] Figure 11 illustrates an embodiment of a visualization device according to an embodiment of the present invention. The visualization device is arranged to allow the display of sensor data from multiple nominally identical builds to be displayed for comparison, and to allow the user to determine whether the sensor signals fall outside acceptable process variations for the state of build progress when the sensor signals are captured. At the top 501 of the display are two builds, B1406.mtt and B1405.mtt, selected by the user, both of which have been performed using the same additive manufacturing equipment V57. The selection of builds can be made using any suitable input device, such as a pointing device from a suitable menu. The next box 502 shows the sensor data selected for the display. In this representation, the sensor data are chamber pressure, chamber temperature, and build height.

[0205] Time bars 503a and 503b show the time during the build process. Time bar 503a is the time for building B1406.mtt, and time bar 503b is the time for building B1405.mtt. A shaded window 503c shows the time period of the build, during which the sensor signals are displayed in the graph below 505. The shaded window 503c can be moved to select different times during the build process, and the width of the window can be adjusted to display more or less data on the graph. Marker 503d is positioned to mark the sensor signals of interest. More than one marker may be positioned on time bars 503a and 503b. In Figure 11, time bars 503a and 503b are aligned based on the start time; however, other alignments may be possible, for example, based on the end time.

[0206] Bars 504a and 504b represent the layers (using lines) that fall within window 503c for each build. Users can manipulate bars 504 and 504b to align similar layers, allowing for comparison of data from these similar layers. Due to build differences, it's possible that nominally identical layers from the same build may not be aligned based on their start time, and therefore, further adjustments may be necessary.

[0207] Graph 505 shows the values ​​of the sensor signal for the selected series in the selected window 503d. Average values ​​are provided to indicate the full range of sensor values ​​for each series. In this embodiment, the full range is indicated by a thin line on a scale bar. The user can zoom in on each series from this full range; in this embodiment, the scale of the graph for that series is adjusted to the indicated sub-range by manipulating the dark shaded bars 505a, 505b, 505c.

[0208] Marker 505d can be moved over sensor data displayed in the graph. Acceptable process changes for each series at the progress state where marker 505d is located are indicated on the scale by lightly shaded bars 505e, 505f, and 505g. This acceptable process change can also be projected onto the graph as a reference. Figure 5 The change boundary T shown in the figure U and T L Similar semi-transparent areas. Users may be able to select which acceptable process variations are displayed as semi-transparent images by using a pointing device to select the appropriate data series. Overlay of multiple semi-transparent areas across multiple data series can provide an additional, obfuscated representation of the data.

[0209] Based on these indications of acceptable process variations, users can quickly determine whether sensor signal values ​​fall outside the acceptable range. Furthermore, it is possible to compare builds by comparing nominally identical builds and aligning the data according to the status of the build progress. This comparison is useful in determining why two nominally identical builds lead to different results. Thus, visualization devices provide useful tools for use in manufacturing processes and for designing manufacturing processes.

[0210] Visualization devices can provide means for uploading methods for determining acceptable process variations based on sensor signals from multiple builds. In this way, users can customize statistical analysis to their needs. Users can indicate which builds meet acceptance requirements and should be considered "golden builds," and which builds fail to meet these requirements. Statistical methods can use these indications to determine acceptable process variations. Acceptable process variations can be uploaded as new build data is added to the device. For example, when manufacturing multiple nominally identical parts in a manufacturing facility, each additive manufacturing device can report data during and / or after a build, and the device can modify the acceptable process variations based on the reported sensor data.

[0211] It will be understood that modifications and changes can be made to the above embodiments without departing from the invention as defined herein. For example, the invention is not limited to the powder bed additive manufacturing process described above, but can be used in other additive manufacturing processes, such as arc additive manufacturing in which a plasma arc is used to melt metal wires to form a workpiece layer by layer.

Claims

1. A method for monitoring additive manufacturing equipment, comprising: One or more sensor signals are received from the additive manufacturing equipment during the construction of the workpiece. The one or more sensor signals are compared with corresponding acceptable process changes of a plurality of acceptable process changes, and Logs are generated based on the comparisons. Each acceptable process variation is derived from sensor signals generated by the additive manufacturing equipment and / or one or more corresponding additive manufacturing devices during one or more initial builds of nominally identical workpieces. Each of the plurality of acceptable process variations is associated with the order in which the sensor signals are generated, and the corresponding acceptable process variation is determined by the order in which the one or more sensor signals are generated during the construction period. The acceptable process variation includes the acceptable range of values ​​for the sensor signal.

2. The method according to claim 1, wherein, The acceptable process variation includes acceptable variations in the signal from the photodiode.

3. The method according to claim 2, wherein, The photodiode measures the radiation collected by the optical scanner.

4. The method according to claim 3, wherein, The radiation is light with a wavelength different from that of the laser guided to the powder bed by the optical scanner.

5. The method according to any one of claims 1 to 4, wherein, There is a one-to-one correlation between each of the plurality of acceptable process changes and each state of the progress of the construction of the workpiece for which sensor signals are generated.

6. The method according to any one of claims 1 to 4, wherein, At least one of the acceptable process variations is associated with multiple states of the progress of the construction of the workpiece.

7. The method according to claim 1, wherein, The acceptable process change is the change in the difference between two sensed values.

8. The method of claim 1, further comprising controlling the additive manufacturing equipment during the construction process based on a comparison of the sensor signals with the corresponding acceptable process variations.

9. The method of claim 1, further comprising controlling parameters of the buildup on the additive manufacturing equipment to return the sensor signal to the corresponding acceptable process variation and / or maintain the sensor signal within the corresponding acceptable process variation.

10. The method of claim 1, further comprising displaying a 2D or 3D representation of the workpiece, the representation indicating a region of the workpiece, and generating sensor signals beyond the corresponding acceptable process variations as the region is formed.

11. The method of claim 1, wherein each acceptable process variation includes the average value and standard deviation of the one or more sensor signals.

12. A controller for monitoring additive manufacturing equipment, the controller comprising a processor arranged to perform the following operations: One or more sensor signals are received from the additive manufacturing equipment during the construction of the workpiece. The one or more sensor signals are compared with corresponding acceptable process changes among a plurality of acceptable process changes, and Logs are generated based on the comparisons. Each acceptable process variation is derived from sensor signals generated by the additive manufacturing equipment and / or one or more corresponding additive manufacturing devices during one or more initial builds of nominally identical workpieces. in, Each of the plurality of acceptable process changes is associated with at least one state of the progress of the construction of the workpiece, and the corresponding acceptable process change is the acceptable process change associated with the state of the progress of the construction when the one or more sensor signals are generated. The acceptable process variation includes the acceptable range of values ​​for the sensor signal.

13. The controller according to claim 12, wherein, The processor is configured to receive the plurality of acceptable process variations and instructions for constructing the workpiece.

14. The controller of claim 12 or claim 13, comprising a memory, and the processor being configured to: store the plurality of acceptable process variations in the memory upon receiving instructions for the construction, and erase the plurality of acceptable process variations from the memory upon completion of the construction of the workpiece or when instructions for the construction of a different workpiece are loaded to the controller.

15. An additive manufacturing apparatus comprising a controller according to claim 13 or 14.

16. A method for generating instructions for constructing a workpiece in an additive manufacturing apparatus, the method comprising: A set of sensor signals is received from the additive manufacturing equipment or one or more corresponding additive manufacturing equipment for each build of one or more initial workpieces, the set of sensor signals or each set of sensor signals being generated during the build of the corresponding initial workpiece among the one or more initial workpieces, the one or more initial workpieces being nominally identical to at least a portion of the workpiece. Evaluate the one or more initial workpieces to determine whether the initial workpiece or each initial workpiece meets the specified requirements, and The set of sensor signals, or each set of sensor signals, is associated with one or more quality identifiers, which indicate whether all or part of the sensor signals in the set represent constructed sensor signals that meet the specified requirements. Based on the set of sensor signals, or each set of sensor signals and an associated quality identifier, determine multiple acceptable process variations for the corresponding set of sensor signals generated during the subsequent construction of the workpiece using the additive manufacturing equipment. The acceptable process variation includes the acceptable range of values ​​for the sensor signal.

17. The method according to claim 16, wherein, The acceptable process variation is determined by statistical analysis of the natural process variations in the sensor signals during the construction of the one or more initial workpieces and by whether the sensor signals are related to a portion of the workpiece that is evaluated to meet the specified requirements.

18. The method according to claim 16 or claim 17, wherein, The method is used in the manufacture of a series of nominally identical workpieces, and as more workpieces are built, the acceptable process variation is improved / evolved using the sensor signals from each build.

19. The method of claim 16, wherein, In the case where only a region of the initial workpiece is considered not to meet the specified requirements, the method includes: receiving an identifier from a user regarding whether all sensor signals for the construction of the initial workpiece are identified as being associated with a construction that does not meet the specified requirements, or regarding whether only the sensor signals associated with the region considered not to meet the specified requirements should be identified as such, wherein other sensor signals are associated with an acceptable construction.

20. The method of claim 16, further comprising associating each acceptable process change with at least one state of progress in the construction of the workpiece.

21. The method of claim 16, comprising: A set of primary acceptable process variations is determined based on the set or more sets of sensor signals and associated quality identifiers, each primary acceptable process variation being associated with a state of progress in the construction of the workpiece; a set of simplified secondary acceptable process variations is generated based on the set of primary acceptable variations, at least one of the secondary acceptable process variations being associated with multiple states of progress in the construction of the workpiece.

22. The method according to claim 21, wherein, The simplified set of minor acceptable process changes is generated by grouping similar changes in the major acceptable process changes together and generating one of the minor acceptable process changes for each group. Each minor acceptable process change is associated with multiple progress states, which correspond to the progress states associated with the major acceptable process change of the corresponding group. The minor acceptable process change is generated for the corresponding group.

23. The method according to claim 22, wherein, Each minor acceptable process change is generated based on the characteristics of the major acceptable process change of the corresponding group, and the minor acceptable process change is generated for the corresponding group.

24. The method according to claim 22 or claim 23, wherein, The main acceptable process variations are grouped using values ​​derived from a similarity function.

25. The method of claim 16, comprising constructing the or each initial workpiece using the one or more corresponding additive manufacturing apparatuses.

26. A system for generating instructions for controlling additive manufacturing equipment, the system comprising a processor arranged to perform the method as described in any one of claims 16 to 24.

27. A method for controlling additive manufacturing equipment, comprising: The method provides instructions for constructing a workpiece by curing material layer by layer using the additive manufacturing apparatus, for determining a plurality of acceptable process variations of sensor signals for the additive manufacturing apparatus according to any one of claims 16-25, and for generating a build file including the instructions and the plurality of acceptable process variations.

28. The method according to claim 27, wherein, Each of the acceptable process variations is associated with at least one state of the progress of the construction of the workpiece.

29. An apparatus including a processor, the processor being configured to perform the method as claimed in claim 27 or claim 28.

30. A data carrier having a build file thereon, the build file being used to indicate additive manufacturing equipment in the construction of a workpiece in a layer-by-layer manner, the build file comprising: Instructions regarding how the additive manufacturing equipment constructs the workpiece, and acceptable process variations for sensor signals generated by sensors of the additive manufacturing equipment during the construction of the workpiece, wherein a plurality of acceptable process variations are determined by the method according to any one of claims 16-25.

31. A method for manufacturing a workpiece, comprising using additive manufacturing equipment, wherein, The workpiece is formed by curing material in a layer-by-layer manner, the method comprising: receiving sensor signals from sensors of the additive manufacturing equipment during workpiece construction; and displaying a representation of each sensor signal and a corresponding acceptable process deviation from the signal at the point where the sensor signal is captured, indicating the state of progress of the construction, wherein different states of progress are associated with different acceptable process variations, each acceptable process variation being derived from sensor signals generated by the additive manufacturing equipment and / or one or more corresponding additive manufacturing devices during one or more initial constructions of nominally identical workpieces. The acceptable process variation includes the acceptable range of values ​​for the sensor signal.

32. A visualization device to be used in a manufacturing process, the visualization device comprising a processor and a display, wherein, The processor is configured to perform the method of claim 31, such that the display shows the representation.

33. A method for manufacturing a plurality of nominally identical workpieces using one or more additive manufacturing apparatuses, wherein, Each workpiece is formed by curing material in a layer-by-layer manner, the method comprising: receiving sensor signals from sensors of the one or more additive manufacturing devices during the construction of the workpiece; determining a plurality of acceptable process variations based on the sensor signals for different states of the progress of the construction according to any one of claims 16-25; and displaying, for the state of the progress of the construction at the time the sensor signals are captured, a representation of each sensor value of at least one of the workpieces and a corresponding acceptable process deviation of the plurality of acceptable process variations.

34. A visualization device to be used in a manufacturing process, the visualization device comprising a processor and a display, wherein, The processor is configured to perform the method of claim 33, such that the display shows the representation.

35. A data carrier having instructions that, when executed by a processor, cause the processor to perform the method as described in any one of claims 1 to 11, 16 to 24, 27, 28, 31 and 33.

Images