Real time inconsistency detection during composite material manufacturing

1 . An inconsistency detection system comprising:

a computer system; and

an inconsistency analyzer located in the computer system, wherein the inconsistency analyzer is configured to:

detect fiber inconsistencies in sections of fibers input into a composite material manufacturing system in real time during operation of the composite material manufacturing system;

detect material inconsistencies in an unconsolidated composite material in real time as the unconsolidated composite material is being manufactured by the composite material manufacturing system using the fibers;

determine whether the material inconsistencies are present in a number of the sections in which the fiber inconsistencies are detected in real time as the unconsolidated composite material is being manufactured by the composite material manufacturing system; and

track the fiber inconsistencies in the number of the sections in which the fibers with fiber inconsistencies resulted in the material inconsistencies in the number of the sections in the unconsolidated composite material, wherein in tracking the fiber inconsistencies, the inconsistency analyzer is configured to: track the fiber inconsistencies in the number of the sections in which the fibers with fiber inconsistencies resulted in the material inconsistencies in the number of the sections in the unconsolidated composite material using a Kalman filter,

wherein the inconsistency analyzer is configured to: perform a number of actions in response to the material inconsistencies being present in the number of the sections in which the fiber inconsistencies were detected, and

wherein the number of actions is selected from at least one of:

generate an alert indicating the fiber inconsistencies;

generate an alert indicating the material inconsistencies;

predict material inconsistencies prior to the material inconsistencies occurring;

halt production of the unconsolidated composite material in response to the material inconsistencies being present in a number of the sections in which the fiber inconsistencies were detected; and

select new fibers for input into the composite material manufacturing system.

2 . The inconsistency detection system of claim 1 , wherein the inconsistency analyzer is configured to:

display an image of the fibers with a fiber inconsistency; and

display a graphical indicator in association with the fiber inconsistency in the image.

3 . The inconsistency detection system of claim 1 , wherein the inconsistency analyzer is configured to:

display an image of the unconsolidated composite material with a material inconsistency; and

display a graphical indicator in association with the material inconsistency in the image.

4 . The inconsistency detection system of claim 1 , wherein in detecting the fiber inconsistencies, the inconsistency analyzer is configured to:

generate fiber images of the fibers input into the composite material manufacturing system in real time during operation of the composite material manufacturing system;

detect the fiber inconsistencies in the sections of the fibers input into the composite material manufacturing system in real time during operation of the composite material manufacturing system using the fiber images and a first machine learning model trained to detect the fiber inconsistencies in the fiber images of the fibers; and

wherein in detecting the material inconsistencies, the inconsistency analyzer is configured to:

generate material images of the unconsolidated composite material in real time as the unconsolidated composite material is being manufactured by the composite material manufacturing system; and

detect the material inconsistencies in the unconsolidated composite material in real time as the unconsolidated composite material is being manufactured by the composite material manufacturing system using the material images and a second machine learning model trained to detect the material inconsistencies in the material images of the unconsolidated composite material.

5 . The inconsistency detection system of claim 4 further comprising:

a first camera system positioned to generate the fiber images of fibers in a location in which the fibers are input into the composite material manufacturing system; and

a second camera system positioned to generate the material images in a location prior to the unconsolidated composite material being wound onto a spool.

6 . The inconsistency detection system of claim 1 , wherein the Kalman filter tracks a distance d traveled by a section of fibers within a standard deviation of δ.

7 . The inconsistency detection system of claim 1 , wherein the fiber inconsistencies are selected from at least one of a gap between the fibers, a loose fiber, or a foreign object in the fibers.

8 . The inconsistency detection system of claim 1 , wherein the inconsistency analyzer is configured to:

determine an amount of fiber inconsistencies that affects the material inconsistencies.

9 . The inconsistency detection system of claim 8 , wherein the amount of fiber inconsistencies that results in the material inconsistencies in the unconsolidated composite material is as follows:

PFI

=

n

Inconsistency

Both

n

Inconsistency

Material

wherein n Inconsistency Both is a first number of the sections in images that had fiber inconsistencies and material inconsistencies and n Inconsistency Material is a second number of the sections in the images that had the material inconsistencies.

10 . A method for detecting inconsistencies in an unconsolidated composite material, the method comprising:

detecting fiber inconsistencies in sections of the fibers input into a composite material manufacturing system in real time during operation of the composite material manufacturing system;

detecting material inconsistencies in the unconsolidated composite material in real time as the unconsolidated composite material is being manufactured by the composite material manufacturing system using the fibers;

determining whether the material inconsistencies are present in a number of the sections in which the fiber inconsistencies are detected in real time as the unconsolidated composite material is being manufactured by the composite material manufacturing system;

tracking the fiber inconsistencies in the number of the sections in which the fibers with fiber inconsistencies resulted in the material inconsistencies in the number of the sections in the unconsolidated composite material, wherein tracking the fiber inconsistencies comprises:

tracking the fiber inconsistencies in the number of the sections in which the fibers with fiber inconsistencies resulted in the material inconsistencies in the number of the sections in the unconsolidated composite material using a Kalman filter; and

performing a number of actions in response to the material inconsistencies being present in the number of the sections in which the fiber inconsistencies were detected,

wherein the number of actions is selected from at least one of:

generate an alert indicating the fiber inconsistencies;

generate an alert indicating the material inconsistencies;

predict material inconsistencies prior to the material inconsistencies occurring;

halt production of the unconsolidated composite material in response to the material inconsistencies being present in a number of the sections in which the fiber inconsistencies were detected; and

select new fibers for input into the composite material manufacturing system.

11 . The method of claim 10 further comprising:

displaying an image of the fibers with a fiber inconsistency; and

displaying a graphical indicator in association with the fiber inconsistency in the image.

12 . The method of claim 10 further comprising:

displaying an image of the unconsolidated composite material with a material inconsistency; and

displaying a graphical indicator in association with the material inconsistency in the image.

13 . The method of claim 10 , wherein detecting the fiber inconsistencies comprises:

generating fiber images of the fibers input into the composite material manufacturing system in real time during operation of the composite material manufacturing system;

detecting the fiber inconsistencies in the sections of the fibers input into the composite material manufacturing system in real time during operation of the composite material manufacturing system using the fiber images and a first machine learning model trained to detect the fiber inconsistencies in the fiber images of the fiber; and

wherein detecting the material inconsistencies comprises:

generating material images of the unconsolidated composite material in real time as the unconsolidated composite material is being manufactured by the composite material manufacturing system; and

detecting the material inconsistencies in the unconsolidated composite material in real time as the unconsolidated composite material is being manufactured by the composite material manufacturing system using the material images and a second machine learning model trained to detect the material inconsistencies in the material images of the unconsolidated composite material.

14 . The method of claim 13 , wherein a first camera system positioned to generate the fiber images of the fibers in a location in which the fibers are input into the composite material manufacturing system; and a second camera system positioned to generate the material images in a location prior to the unconsolidated composite material being wound onto a spool.

15 . The method of claim 10 , wherein the Kalman filter tracks a distance d traveled by a section of fibers within a standard deviation of δ.

16 . The method of claim 10 , wherein the fiber inconsistencies are selected from at least one of a gap between the fibers, a loose fiber, or a foreign object in the fibers.

17 . The method of claim 10 further comprising:

determining an amount of fiber inconsistencies that affects the material inconsistencies.

18 . The method of claim 17 , wherein the amount of fiber inconsistencies that results in the material inconsistencies in the unconsolidated composite material is as follows:

PFI

=

n

Inconsistency

Both

n

Inconsistency

Material

wherein n Inconsistency Both is a first number of the sections in images that had fiber inconsistencies and material inconsistencies and n Inconsistency Material is a second number of the sections in the images that had the material inconsistencies.

19 . A method for detecting inconsistencies in an unconsolidated composite material, the method comprising:

detecting fiber inconsistencies in sections of the fibers input into a composite material manufacturing system in real time during operation of the composite material manufacturing system;

detecting material inconsistencies in the unconsolidated composite material in real time as the unconsolidated composite material is being manufactured by the composite material manufacturing system using the fibers;

determining whether the material inconsistencies are present in a number of the sections in which the fiber inconsistencies are detected in real time as the unconsolidated composite material is being manufactured by the composite material manufacturing system;

tracking the fiber inconsistencies in the number of the sections in which the fibers with fiber inconsistencies resulted in the material inconsistencies in the number of the sections in the unconsolidated composite material;

determining an amount of fiber inconsistencies that affects the material inconsistencies, wherein the amount of fiber inconsistencies that results in the material inconsistencies in the unconsolidated composite material is as follows:

PFI

=

n

Inconsistency

Both

n

Inconsistency

Material

wherein n Inconsistency Both is a first number of the sections in images that had fiber inconsistencies and material inconsistencies and n Inconsistency Material is a second number of the sections in the images that had the material inconsistencies; and

performing a number of actions in response to the material inconsistencies being present in the number of the sections in which the fiber inconsistencies were detected,

wherein the number of actions is selected from at least one of:

generate an alert indicating the fiber inconsistencies;

generate an alert indicating the material inconsistencies;

predict material inconsistencies prior to the material inconsistencies occurring;

halt production of the unconsolidated composite material in response to the material inconsistencies being present in a number of the sections in which the fiber inconsistencies were detected; and

select new fibers for input into the composite material manufacturing system.

20 . The method of claim 19 , wherein tracking the fiber inconsistencies comprises:

tracking the fiber inconsistencies in the number of the sections in which the fibers with fiber inconsistencies resulted in the material inconsistencies in the number of the sections in the unconsolidated composite material using a Kalman filter.

21 . The method of claim 20 , wherein the Kalman filter tracks a distance d traveled by a section of fibers within a standard deviation of δ.