Fabricated shape estimation for droplet based additive manufacturing

Patent metadata

Field	Value
Publication number	US-11741273-B2
Application number	US-202016898994-A
Country	US
Kind code	B2
Filing date	Jun 11, 2020
Priority date	Jun 11, 2020
Publication date	Aug 29, 2023
Grant date	Aug 29, 2023

Abstract

A geometry of a substrate surface is received at a neural network. The neural network is trained using one or more training sets. Each training set comprises a different type of substrate geometry and a collection of manufacturing process parameters. The substrate is configured to receive at least one liquid droplet. A shape of the at least one droplet after it has been deposited on the substrate is determined based on the received geometry. An output representing the determined shape of the at least one droplet is produced.

First claim

What is claimed is:

1. A method comprising:
inputting a geometry of a substrate surface and manufacturing process parameters to a neural network, the neural network comprising a combination of a linear layer, non-linear activation functions, and a convolution layer, the neural network trained using one or more training sets, each training set comprising a different type of substrate geometry, and a collection of the manufacturing process parameters, the substrate surface configured to receive at least one droplet;
determining a shape of the at least one droplet after it has been deposited on the substrate surface based on an output of the neural network; and
producing a simulation output representing the determined shape of the at least one droplet.

2. The method of claim 1 , wherein the shape comprises a 3D representation of the at least one droplet after it has deposited on the substrate surface.

3. The method of claim 1 , wherein the one or more training sets comprise a surfaces of varying curvature, comprising one or more of a smooth surface, a rough surface, and a step-like surface.

4. The method of claim 1 , wherein the geometry of the substrate surface comprises a 3D representation.

5. The method of claim 1 , wherein the simulation output is a 3D representation of the shape obtained by determining the shape based on the geometry.

6. The method of claim 1 , wherein the training set comprises droplet shapes determined using one or more of a high-fidelity model and a steady-state model.

7. The method of claim 1 , further comprising estimating a shape of a product part based on the simulation output, the product part comprising a plurality of droplets.

8. A method comprising:
training a neural network using one or more training sets, each training set comprising a different type of substrate surface geometry, and a collection of manufacturing process parameters, the neural network comprising a combination of a linear layer, non-linear activation functions, and a convolution layer;
inputting a geometry of a substrate and the manufacturing process parameters to the neural network, the substrate configured to receive at least one droplet;
determining a shape of the at least one droplet after it has been deposited on the substrate based on an output of the neural network; and
producing a simulation output representing the determined shape of the at least one droplet.

9. The method of claim 8 , wherein the shape comprises a 3D representation of the at least one droplet after it has deposited on the substrate.

10. The method of claim 8 , wherein the one or more training sets comprise surfaces of varying curvature, comprising one or more of a curved surface, a highly curved surface, a rough surface, and a step-like surface.

11. The method of claim 8 , wherein the geometry of the substrate comprises a 3D representation.

12. The method of claim 8 , wherein the simulation output is a 3D representation of the shape obtained by determining the shape based on the geometry.

13. The method of claim 8 , wherein the training set comprises droplet shapes determined using one or more of a high-fidelity model and a steady-state model.

14. The method of claim 8 , further comprising estimating a shape of a product part based on the simulation output, the product part comprising a plurality of droplets.

15. A system, comprising:
a processor; and
a memory storing computer program instructions which when executed by the processor cause the processor to perform operations comprising:
inputting a geometry of a substrate surface and manufacturing process parameters to a neural network, the neural network comprising a combination of a linear layer, non-linear activation functions, and a convolution layer, the neural network trained using one or more training sets, each training set comprising a different type of substrate geometry, and a collection of manufacturing process parameters, the substrate surface configured to receive at least one droplet;
determining a shape of the at least one droplet after it has been deposited on the substrate surface based on an output of the neural network; and
producing a simulation output representing the determined shape of the at least one droplet.

16. The system of claim 15 , wherein the shape comprises a 3D representation of the at least one droplet after it has deposited on the substrate surface.

17. The system of claim 15 , wherein the one or more training sets comprise surfaces of varying curvature, comprising one or more of a curved surface, a highly curved surface, a rough surface, and a step-like surface.

18. The system of claim 15 , wherein the geometry of the substrate surface comprises a 3D representation.

19. The system of claim 15 , wherein the simulation output is a 3D representation of the shape obtained by determining the shape based on the geometry.

20. The system of claim 15 , further comprising estimating a shape of a product part based on the output, the product part comprising a plurality of droplets.

Assignees

• Palo Alto Res Ct Inc

Inventors

• Korneev Svyatoslav
• Thiagarajan Vaidyanathan
• Nelaturi Saigopal

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06F30/17 Primary

  Mechanical parametric or variational design · CPC title

• G06N3/04

  Architecture, e.g. interconnection topology · CPC title

• G06N3/08

  Learning methods · CPC title