Additive manufacturing system with at least one electronic nose
AssigneesUT-Battelle, LLC, University of Tennessee Research Foundation
InventorsIlia N. Ivanov, Eric S. Muckley
Life of this patent
Estimating the term…
Abstract
An additive manufacturing system comprising at least one electronic nose (e-nose) is provided. The e-nose may comprise a housing and gas sensors. The housing may have an air channel. The active sensor portion of the sensors are positioned in the air channel. The housing may be mounted to an extruder head of an additive manufacturing device. The system may also comprise a processor. The processor may determine whether there is an abnormality in an additive manufacturing process based on one or more combinations of outputs from the gas sensors received during the additive manufacturing process input into a deployed machine learning model; and generate a report for the additive manufacturing process containing the determination.
Claims18
An additive manufacturing system comprising:
- a first electronic nose and a second electronic nose, the first electronic nose being mounted to a first side of an extruder head and the second electronic nose being mounted to a second side of the extruder head, each said electronic nose comprising:
- a housing having openings on corresponding ends thereof to enable air flow, the housing having an air channel for air to flow between the ends;
- a plurality of thin film gas sensors, each having an active sensor portion in the air channel;
- and
- a mount configured to mount the housing to the extruder head of an additive manufacturing device, and
- a processor configured to:
- supply power to the plurality of thin film gas sensors to bias the sensors;
- receive output from each of the plurality of thin film gas sensors;
- determine whether there is an abnormality in an additive manufacturing process manufacturing a product from one or more materials or composite materials based on one or more combinations of the output from each of the plurality of thin film gas sensors during the additive manufacturing process and from a deployed machine learning model;
- and
- generate a report for the additive manufacturing process containing the abnormality determination.
An additive manufacturing system comprising:
- at least one electronic nose comprising:
- a housing having openings on corresponding ends thereof to enable air flow, the housing having an air channel for air to flow between the ends;
- a plurality of thin film gas sensors, each having an active sensor portion in the air channel;
- and
- a mount configured to mount the housing to an extruder head of an additive manufacturing device, and
- a processor configured to:
- supply power to the plurality of thin film gas sensors to bias the sensors;
- receive output from each of the plurality of thin film gas sensors;
- determine whether there is an abnormality in an additive manufacturing process manufacturing a product from one or more materials or composite material based on one or more combinations of the output from each of the plurality of thin film gas sensors during the additive manufacturing process and from a deployed machine learning model;
- and
- generate a report for the additive manufacturing process containing the abnormality determination, wherein the received the output from each of the plurality of thin film gas sensors is in response to heating materials or composite materials at different temperatures, one of the materials or composite materials at a time and wherein the processor is configured to:
- generate randomly a first dataset for training and a second dataset for testing a plurality of models using the received output;
- train and test the plurality of models using one or more combinations of the output from each of the plurality of thin film gas sensors, the plurality of models is generated using a plurality of different machine learning techniques, the training based on the first dataset and the testing based on the second dataset;
- and
- evaluate a prediction accuracy of each of the plurality of models using an
- evaluation parameter and select a model from among the plurality of models to deploy as the deployed machine learning model for detecting the abnormality in the additive manufacturing process based on a comparison of the evaluation parameter for each of the plurality of models.
Dependencies are inferred from the claim text.