Sensor based smart segmentation

Patent metadata

Field	Value
Publication number	US-10884396-B2
Application number	US-201916287192-A
Country	US
Kind code	B2
Filing date	Feb 27, 2019
Priority date	Feb 27, 2019
Publication date	Jan 5, 2021
Grant date	Jan 5, 2021

Abstract

According to some embodiments, system and methods are provided comprising receiving, via a communication interface of a platform comprising a segmentation module and a processor, a defined geometry for one or more geometric structures forming one or more parts, wherein the parts are manufactured with an additive manufacturing machine; generating a build file including an initial parameter set to fabricate each part; fabricating the part based on the build file; receiving sensor data for the fabricated part; generating a parameter set for each layer that forms the part, via execution of an iterative learning control process for each layer; generating raw power data for each layer that forms the part, using the processor, based on the generated parameter set; applying a noise reduction process to the raw power data; and generating a segmented build file, using the segmentation module, via application of the noise reduction process on the raw power data. Numerous other aspects are provided.

First claim

The invention claimed is:

1. A method comprising:
receiving, via a communication interface of a platform comprising a segmentation module and a processor, a defined geometry for one or more geometric structures forming one or more parts, wherein the parts are manufactured with an additive manufacturing machine;
generating a build file including an initial parameter set to fabricate each part;
fabricating the part based on the build file;
receiving sensor data for the fabricated part;
generating a parameter set for each layer that forms the part, via execution of an iterative learning control process for each layer, wherein the part is formed from a plurality of successive layers;
generating raw power data for each layer that forms the part, using the processor, based on the generated parameter set, wherein the raw power data is generated from a comparison of a reference thermal intensity to a thermal intensity received as part of the received sensor data for the fabricated part;
applying a noise reduction process to the raw power data; and
generating a segmented build file, using the segmentation module, via application of the noise reduction process on the raw power data.

2. The method of claim 1 , wherein a segment is a region of the part bounded with a respective parameter set.

3. The method of claim 1 , wherein the noise reduction process is a clustering algorithm.

4. The method of claim 3 , wherein the clustering algorithm is k-means.

5. The method of claim 1 , wherein the parameter set is optimized.

6. The method of claim 1 , wherein the parameter set includes values for at least one of laser power, laser scan speed, focus or spot size, layer thickness and hatch spacing.

7. The method of claim 1 , wherein the iterative learning control process is executed one or more times, and generates an optimized parameter set for each execution.

8. The method of claim 7 , wherein the noise reduction process is applied to the raw power data generated via each execution of the iterative learning control process.

9. The method of claim 1 , wherein the generated parameter set is a final iteration and wherein the noise reduction process is applied to the raw power data of the final iteration of the parameter set.

10. The method of claim 1 , further comprising:
fabricating the part based on the segmented build file.

11. A system comprising:
a segmentation module;
a segmentation processor; and
a memory storing program instructions, the segmentation processor and the segmentation module operative with the program instructions to perform the functions as follows:
receive a defined geometry for one or more geometric structures forming one or more parts, wherein the parts are manufactured with an additive manufacturing machine;
generate a build file including an initial parameter set to fabricate each part;
fabricate the part based on the build file;
receive sensor data for the fabricated part;
generate a parameter set for each layer that forms the part, via execution of an iterative learning control process for each layer, wherein the part is formed from a plurality of successive layers;
generate raw power data for each layer that forms the part based on the generated parameter set, wherein the raw power data is generated from a comparison of a reference thermal intensity to a thermal intensity received as part of the received sensor data for the fabricated part;
apply a noise reduction process to the raw power data; and
generate a segmented build file via application of the noise reduction process on the raw power data.

12. The system of claim 11 , wherein a segment is a region of the part bounded with a respective parameter set.

13. The system of claim 11 , wherein the noise reduction process is a clustering algorithm.

14. The system of claim 11 , wherein the parameter set is optimized.

15. The system of claim 11 , wherein the iterative learning control process is executed one or more times, and generates an optimized parameter set for each execution.

16. The system of claim 15 , wherein the noise reduction process is applied to the raw power data generated via each execution of the iterative learning control process.

17. The system of claim 11 , wherein the generated parameter set is a final iteration and wherein the noise reduction process is applied to the raw power data of the final iteration of the parameter set.

18. The system of claim 11 , wherein the received sensor data is from one or more sensors incorporating at least one of physical baffles to block off-axis radiation and optical filters to restrict a bandwidth of a detected radiation.

19. A non-transitory computer-readable medium storing instructions that, when executed by a computer processor, cause the computer processor to perform a method comprising:
receiving, via a communication interface of a platform comprising a segmentation module and a processor, a defined geometry for one or more geometric structures forming one or more parts, wherein the parts are manufactured with an additive manufacturing machine;
generating a build file including an initial parameter set to fabricate each part;
fabricating the part based on the build file;
receiving sensor data for the fabricated part;
generating a parameter set for each layer that forms the part, via execution of an iterative learning control process for each layer, wherein the part is formed from a plurality of successive layers;
generating raw power data for each layer that forms the part, using the processor, based on the generated parameter set, wherein the raw power data is generated from a comparison of a reference thermal intensity to a thermal intensity received as part of the received sensor data for the fabricated part;
applying a noise reduction process to the raw power data; and
generating a segmented build file, using the segmentation module, via application of the noise reduction process on the raw power data.

20. The medium of claim 19 , wherein the noise reduction process is a clustering algorithm.

Assignees

• Gen Electric

Inventors

• Roychowdhury Subhrajit
• Gupta Vipul Kumar
• Rausch Randal T
• Gambone Justin John
• Ping Xiaohu
• Chen Alexander
• Hershey John Erik

Classifications

• Y02P80/40

  Minimising material used in manufacturing processes · CPC title

• B33Y50/02 Primary

  for controlling or regulating additive manufacturing processes · CPC title

• G05B2219/35134

  3-D cad-cam · CPC title

• G05B19/4099 Primary

  Surface or curve machining, making three-dimensional [3D] objects, e.g. desktop manufacturing · CPC title

• G05B13/0265

  the criterion being a learning criterion · CPC title