Additive manufacturing with in situ defect detection
A method for monitoring and analyzing an additive manufacturing process includes heating a melt zone to fuse an additive media with an active layer to build a part being manufactured based on a part design model, capturing raw melt data of the melt zone, and generating an active layer dataset that is spatially defined. The method may also include analyzing the active layer dataset with respect to a plurality of defect signatures within a defect signature library. The defect signature library may be predefined based on a machine learning processing of historical sensor datasets with corresponding ground truth datasets. The method may also include detecting a defect in the part based on the analysis of the active layer dataset with respect to a plurality of defect signatures, simultaneously with the energy source acting upon the active layer of the part for manufacturing of the part.
1. A method for monitoring and analyzing an additive manufacturing process, the method comprising:
heating, via an energy source, a melt zone to form a melt pool to fuse an additive media on an active layer to build a part being manufactured based on a part design model;
receiving, by a sensor, reflected radiation from the melt pool due to the energy source acting upon the melt pool and generating raw melt data of the melt pool based on the reflected radiation;
generating, based on the raw melt data, an active layer dataset that is spatially defined;
analyzing the active layer dataset with respect to a plurality of defect signatures within a defect signature library to identify matches between the active layer dataset and the plurality of defect signatures, the defect signatures within the defect signature library indicating characteristics of a melt pool that are indicative of the formation of a defect due to conditions including temperatures below a target temperature for the melt pool and temperatures above the target temperature for the melt pool, the defect signature library being predefined based on a machine learning processing of historical sensor datasets with corresponding ground truth datasets; and
detecting a defect in the part based on the analyzing the active layer dataset with respect to a plurality of defect signatures.
2. The method of claim 1 , further comprising developing a defect model of the part based on the detecting the defect in combination with other defect detections identified in the part.
3. The method of claim 2 , further comprising providing the defect model for part certification upon completion of manufacturing of the part.
4. The method of claim 1 , further comprising developing a defect model of the part based on the active layer dataset and additional layer datasets corresponding to additional layers involved in manufacturing the part.
5. The method of claim 1 , further comprising, in response to detecting the defect, discontinuing construction of the part.
6. The method of claim 5 , further comprising:
controlling a laser to perform a healing operation to remove the defect based on the analyzing the active layer dataset; and
resuming construction of the part in accordance with the part design model.
7. The method of claim 1 , wherein
the raw melt data comprises light intensity data indicative of an intensity of the reflected radiation received by the sensor, and
the generating the active layer dataset includes generating the active layer dataset to include temperature data that is spatially defined based on the light intensity data.
8. The method of claim 1 , wherein
the analyzing the active layer dataset comprises analyzing the active layer dataset and previously constructed layer datasets for the part with respect to the plurality of defect signatures, and
the detecting the defect comprises detecting the defect based on the analyzing the active layer dataset and previously constructed layer datasets with respect to the plurality of defect signatures.
9. The method of claim 1 wherein
the part design model defines a plurality of model layers including an active model layer that corresponds to the active layer and future model layers corresponding to future layers to be built in construction of the part,
the analyzing the active layer dataset comprises analyzing the active layer dataset and the future model layers with respect to the plurality of defect signatures, and
the detecting the defect comprises detecting the defect based on the analyzing the active layer dataset and the future model layers with respect to the plurality of defect signatures.
10. The method of claim 1 , wherein the part is made of a material that is not penetrated by x-rays and does not permit x-ray imaging of an internal area of the part.
11. The method of claim 1 , wherein
the generating the raw melt data of the melt pool comprises receiving reflected radiation from the melt pool due to the energy source acting upon the melt pool by a plurality of sensors comprising the sensor and generating the raw melt data based on the reflected radiation received at each sensor in the plurality of sensors, and
the plurality of sensors comprise an optical imaging sensor or an infrared imaging sensor.
12. A system for monitoring and analyzing an additive manufacturing process, the system comprising:
an energy source configured to heat a melt zone to form a melt pool to fuse an additive media on an active layer to build a part being manufactured based on a part design model;
a sensor configured to receive reflected radiation from the melt pool due to the energy source acting upon the melt pool and generating raw melt data of the melt pool based on the reflected radiation; and
processing circuitry configured to:
generate, based on the raw melt data, an active layer dataset that is spatially defined;
analyze the active layer dataset with respect to a plurality of defect signatures within a defect signature library to identify matches between the active layer dataset and the plurality of defect signatures, the defect signatures within the defect signature library indicating characteristics of a melt pool that are indicative of the formation of a defect due to conditions including temperatures below a target temperature for the melt pool and temperatures above the target temperature for the melt pool, the defect signature library being predefined based on a machine learning processing of historical sensor datasets with corresponding ground truth datasets; and
detect a defect in the part based on the analyzing the active layer dataset with respect to a plurality of defect signatures.
13. The system of claim 12 , wherein the processing circuitry is further configured to develop a defect model of the part based on a detected defect in combination with other defected detections within the part.
14. The system of claim 13 , wherein the processing circuitry is further configured to provide the defect model for part certification upon completion of manufacturing of the part.
15. The system of claim 12 , wherein the processing circuitry is further configured to discontinue construction of the part in response to detecting the defect.
16. The system of claim 15 , wherein the processing circuitry is further configured to:
control the energy source to perform a healing operation to remove the defect based on the analyzing the active layer dataset; and
resume construction of the part in accordance with the part design model.
17. The system of claim 12 , wherein
the raw melt data comprises spectral data or light intensity data indicative of a wavelength or intensity, respectively, of the reflected radiation received by the sensor, and
the processing circuitry is further configured to generate the active layer dataset to include temperature data that is spatially defined based on the spectral data or light intensity data.
18. An apparatus for monitoring and analyzing an additive manufacturing process, the apparatus comprising processing circuitry configured to:
generate, based on raw melt data provided by a sensor, an active layer dataset that is spatially defined, wherein the raw melt data is based on reflected radiation from the melt pool, received by the sensor, due to an energy source acting upon the melt pool to manufacture a part;
analyze the active layer dataset with respect to a plurality of defect signatures within a defect signature library to identify matches between the active layer dataset and the plurality of defect signatures, the defect signatures within the defect signature library indicating characteristics of a melt pool that are indicative of the formation of a defect due to conditions including temperatures below a target temperature for the melt pool and temperatures above the target temperature for the melt pool, the defect signature library being predefined based on a machine learning processing of historical sensor datasets with corresponding ground truth datasets; and
detect a defect in a part being manufactured based on the analyzing the active layer dataset with respect to a plurality of defect signatures.
19. The apparatus of claim 18 , wherein the processing circuitry is further configured to discontinue construction of the part in response to detecting the defect.
20. The apparatus of claim 19 , wherein the processing circuitry is further configured to:
control the energy source to perform a healing operation to remove the defect based on the analyzing the active layer dataset; and
resume construction of the part in accordance with a part design model.