# US 12,222,838 B2 · System and methods for generating fault indications for an additive manufacturing process based on a probabilistic comparison of the outputs of multiple process models to measured sensor data > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,222,838 B2 - Document ID: US12222838B2 - Status: Granted - Kind: B2 - Application number: 17/724,704 - Filed: 2022-04-20 - Published: 2022-08-04 - Granted: 2025-02-11 - Earliest claimed: 2019-01-08 - Assignees: General Electric Company - Applicants: General Electric Company - Inventors: Harry Kirk Mathews, Jr.; Sarah Felix; Subhrajit Roychowdhury; Saikat Ray Majumder; Thomas Spears - Web page: https://patentgrep.com/patent/US12222838B2 ## CPC - G05B 23/0254: {based on a quantitative model, e.g. mathematical relationships between inputs and outputs; functions: observer, Kalman filter, residual calculation, Neural Networks} - G06F 11/3447: {Performance evaluation by modeling} - B29C 64/386: Data acquisition or data processing for additive manufacturing - B29C 64/393: for controlling or regulating additive manufacturing processes - B33Y 50/00: Data acquisition or data processing for additive manufacturing - G06F 11/3409: {for performance assessment} - G06F 17/18: for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)} - G06F 30/00: Computer-aided design [CAD] - B33Y 50/02: for controlling or regulating additive manufacturing processes - G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - B29C: SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING (making preforms B29B11/00; making laminated products by combining previously unconnected layers which become one product whose layers will remain together B32B37/00 - B32B41/00) - B33Y: ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING ## Abstract Generating fault indications for an additive manufacturing machine based on a comparison of the outputs of multiple process models to measured sensor data. The method receiving sensor data from the additive manufacturing machine during manufacture of at least one part. Models are selected from a model database, each model generating expected sensor values for a defined condition. Difference values are computed between the received sensor data and an output of each of the models. A probability density function is computed, which defines, for each of the models, a likelihood that a given difference value corresponds to each respective model. A probabilistic rule is applied to determine, for each of the models, a probability that the corresponding model output matches the received sensor data. An indicator is output of a defined condition corresponding to a model having the highest match probability. ## Claims ### Claim 1 A non-transitory computer-readable medium comprising computer-executable instructions, which, when executed by a processor associated with an additive manufacturing machine, cause the processor to perform a method comprising: receiving, via a communication interface, sensor data comprising one or more sensor values from one or more sensors of an additive manufacturing machine during an additive manufacturing process to manufacture at least one part; computing, for respective ones of a plurality of models, a probability density function defining a probability that a selected set of the sensor data matches an output of a corresponding one of the plurality of models; and outputting an indication of a defined condition being present during the additive manufacturing process, the defined condition corresponding to a selected one of the plurality of models, wherein the selected one of the plurality of models has at least a specified probability that the selected set of the sensor data matches the output of the selected one of the plurality of models. ### Claim 2 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein the probability density function is determined based at least in part on a difference between the selected set of the sensor data and the output of the corresponding one of the plurality of models. ### Claim 3 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein respective ones of the plurality of models are configured to determine expected sensor values indicative of the defined condition being present during the additive manufacturing process, and wherein the output comprises the expected sensor values. ### Claim 4 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein the selected one of the plurality of models has the highest probability from among the plurality of models that the selected set of the sensor data matches the output. ### Claim 5 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein the defined condition comprises a process condition. ### Claim 6 (depends on 5) The non-transitory computer-readable medium of claim 5, wherein the process condition comprises a characteristic of a melt pool. ### Claim 7 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein at least one of the plurality of models comprises a time-dependent model. ### Claim 8 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein the selected set of the sensor data comprises a plurality of sensor values that span at least one of a temporal scale or a spatial scale. ### Claim 9 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein the sensor data comprises at least one of: an on-axis photodiode sensor data and an off-axis photodiode sensor data. ### Claim 10 (depends on 1) The non-transitory computer-readable medium of claim 1, comprising: comparing the selected set of the sensor data to the output corresponding to respective ones of the plurality of models, wherein respective ones of the plurality of models are configured to determine expected sensor values indicative of the defined condition being present during the additive manufacturing process, and wherein the output comprises the expected sensor values. ### Claim 11 (depends on 10) The non-transitory computer-readable medium of claim 10, wherein the selected set of the sensor data comprises a first sensor data stream and a second sensor data stream, the first sensor data stream comprising on-axis photodiode sensor data, and the second sensor data stream comprising off-axis photodiode sensor data. ### Claim 12 (depends on 1) The non-transitory computer-readable medium of claim 1, comprising: determining, for respective ones of the plurality of models, one or more residual outputs based at least in part on a comparison of the selected set of the sensor data to the output corresponding to respective ones of the plurality of models, and computing the probability density function corresponding to respective ones of the plurality of models based at least in part on the one or more residual outputs. ### Claim 13 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein the indication comprises at least one of: a fault indication and a quality metric. ### Claim 14 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein at least one of the plurality of models utilizes data from at least one of: an actuator sensor, a galvanometer position sensor; an environmental sensor, and a melt pool monitoring sensor. ### Claim 15 (depends on 1) The non-transitory computer-readable medium of claim 1, determining, for the respective ones of the plurality of models, whether a probabilistic match exists between the selected set of the sensor data and the output, wherein the probabilistic match is based at least in part on the specified probability, wherein the selected one of the plurality of models exhibits the probabilistic match. ### Claim 16 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein the defined condition is selected from among: normal operations, a known fault, and an unknown fault. ### Claim 17 (depends on 1) The non-transitory computer-readable medium of claim 1, wherein the specified probability is determined based at least in part on a signal noise parameter. ### Claim 18 (depends on 17) The non-transitory computer-readable medium of claim 17, wherein the signal noise parameter comprises at least one of: a distribution parameter and a covariance parameter. ### Claim 19 (depends on 17) The non-transitory computer-readable medium of claim 17, wherein the signal noise parameter depends at least on a plurality of laser-on transients. ### Claim 20 (depends on 19) The non-transitory computer-readable medium of claim 19, wherein the plurality of laser-on transients are treated as random variables.