Predicting distributions of values of layers for three-dimensional printing

Patent metadata

Field	Value
Publication number	US-11409261-B2
Application number	US-201716077669-A
Country	US
Kind code	B2
Filing date	Jan 27, 2017
Priority date	Jan 27, 2017
Publication date	Aug 9, 2022
Grant date	Aug 9, 2022

Abstract

In some examples, a distribution of values of a property of a given layer to be printed as part of three-dimensional (3D) printing is predicted, wherein the predicting is based on a distribution of values of the property in a previous layer that has been printed as part of the 3D printing. 3D printing of an object is controlled based on the predicted distribution of values of the property of the given layer.

First claim

What is claimed is:

1. A method comprising:
measuring a distribution of values of a property in a first layer of build material that has been printed as part of three-dimensional (3D) printing, wherein the distribution of values includes a value of the property for each of a plurality of locations on a print target;
receiving as input, dot count information for the first layer, wherein the dot count information for the first layer includes an amount of 3D printing agent provided at each of the plurality of locations within the first layer;
generating, by a system comprising a processor, a predicted thermal image of a second layer that has not yet been printed as part of the 3D printing, wherein the predicted thermal image includes a distribution of values of the property for the second layer, and is based on the distribution of values of the property in the first layer and the dot count information for the first layer;
controlling, by the system, the 3D printing of an object based on the predicted thermal image;
wherein the predicted thermal image is generated using a plurality of different models, each respective model of the plurality of different models receiving as input the distribution of values of the property in the first layer, and outputting a respective predicted distribution of values of the property of the second layer that has not yet been printed; and
determining which of the plurality of different models produces a least error;
wherein the plurality of different models use respective different layer models, each respective layer model of the different layer models comprising a first number of surface layers, a second number of internal layers, and a third number of bottom layers, wherein each of the first number, second number, and third number is one or greater, and a first layer model of the different layer models differs from a second layer model of the different layer models by at least one of the second number of internal layers or the third number of bottom layers.

2. The method of claim 1 , wherein the predicted thermal image is further generated based on dot count information for the second layer, wherein the dot count information for the second layer includes an amount of the 3D printing agent to be provided at each location of the plurality of locations.

3. The method of claim 2 , wherein the predicted thermal image is generated using a model that receives as inputs the dot count information and the distribution of values of the property in the first layer, and that outputs the predicted thermal image.

4. The method of claim 1 , further comprising:
adjusting weights used in each of the plurality of different models in response to the determining.

5. The method of claim 4 , wherein the adjusting is based on reducing an error between a predicted distribution of values of the property in a given layer and an actual distribution of values of the property in the given layer.

6. The method of claim 1 , wherein the property is selected from among a temperature, a density, and a thickness.

7. The method of claim 1 , wherein measuring the distribution of values of the property includes capturing the distribution of values of the property in the first layer by at least one sensor of a 3D printing system.

8. The method of claim 1 , wherein the predicting includes determining that a predicted distribution of temperatures of a layer that has not yet been printed is different from a target distribution of temperatures, based on a distribution of temperatures of the first layer of the 3D object.

9. The method of claim 1 , wherein controlling, by the system, the 3D printing includes modifying heating equipment of the system to produce a target level of heat, based on a deviation of a predicted temperature from a target temperature.

10. The method of claim 1 , including computing an error for each respective model, wherein the error includes a squared sum of errors of temperature values at different locations of the layer that has not yet been printed.

11. The method of claim 1 , wherein the predicted thermal image includes a graphical representation of the distribution of values of the property at different locations across the layer that has not yet been printed.

12. A non-transitory machine-readable storage medium storing instructions that upon execution cause a system to:
generate a predicted thermal image of a second layer of build material that has not yet been printed as part of three-dimensional (3D) printing, wherein the predicted thermal image is based on dot count information for a first layer of build material that has been printed as part of the 3D printing and a distribution of values of a property in the first layer, wherein the distribution of values includes a value of the property for each of a plurality of locations on a print target, and the dot count information includes an amount of 3D printing agent provided at each of the plurality of locations;
compare the predicted thermal image against a target distribution of values for the second layer;
update a model based on the predicted thermal image;
use the updated model to control 3D printing by a 3D printing system or to simulate an operation of 3D printing by a printing system using the updated model;
select the model from a plurality of different models that use different layer models representing a stack of build material to be printed by the printing system;
wherein each respective layer model of the different layer models comprises a first number of surface layers, a second number of internal layers, and a third number of bottom layers, wherein each of the first number, second number, and third number is one or greater, and a first layer model of the different layer models differs from a second layer model of the different layer models by at least one of the second number or the third number.

13. A system comprising:
a processor; and
a non-transitory storage medium storing instructions that are executable on the processor to:
predict, using a plurality of different models, respective distributions of values of a property of a given layer of build material that has not yet been printed as part of three-dimensional (3D) printing based on information of previously printed layers of build material, each respective model of the plurality of different models receiving as input dot count information for previously printed layers of build material and a distribution of values of the property in the previously printed layers, wherein the distribution of values includes a value of the property for each of a plurality of locations on a print target, and the dot count information includes an amount of 3D printing agent provided at each of the plurality of locations, and outputting a respective predicted distribution of values of the property of the given layer wherein the distribution of values of the property of the previous layer are captured by at least one sensor of a 3D printing system;
compute a respective error for each respective model of the plurality of different models based on a comparison of the predicted distributions of values of the property against a measured distribution of values of the property of the given layer of build material;
select the model from a plurality of different models that use different layer models representing a stack of build material to be printed by the printing system;
wherein each respective layer model of the different layer models comprises a first number of surface layers, a second number of internal layers, and a third number of bottom layers, wherein each of the first number, second number, and third number is one or greater, and

Assignees

• Hewlett Packard Development Co

Inventors

• Kothari Sunil
• Li Kristopher
• Sean Viseth
• Zeng Jun
• Zhao Lihua
• Obegi Goffril
• Dispoto Gary J
• Heiles Tod

Classifications

• B29C64/393 Primary

  for controlling or regulating additive manufacturing processes · CPC title

• B22F10/37

  of powder bed aspects, e.g. density · CPC title

• B22F10/80

  Data acquisition or data processing · CPC title

• B22F12/90

  Means for process control, e.g. cameras or sensors · CPC title

• B22F10/85

  for controlling or regulating additive manufacturing processes · CPC title