METHOD AND SYSTEM FOR INTERLAYER FEEDBACK CONTROL AND FAILURE DETECTION IN AN ADDITIVE MANUFACTURING PROCESS

1 . An additive manufacturing system comprising:

a build window;

a projection system:

arranged beneath a lower surface of the build window; and

configured to project electromagnetic radiation toward an upper surface of the build window;

a build platform:

facing the build window; and

configured to retract and advance relative to the build window;

a sensor suite; and

a controller configured to:

during a first build cycle of an additive manufacturing process for manufacturing a first layer of a build, sample a first set of sensor data streams via the sensor suite;

calculate a first likelihood of failure of the build based on the first set of sensor data streams;

in response to calculating the first likelihood of failure within a first likelihood range, flag the build to indicate the first likelihood of failure; and

in response to calculating the first likelihood of failure within a second likelihood range greater than the first likelihood range:

pause the additive manufacturing process; and

notify an operator of the additive manufacturing system of the first likelihood of failure.

2 . The additive manufacturing system of claim 1 :

wherein the sensor suite comprises an image sensor arranged below the build window and configured to record images depicting the build window;

wherein the controller is configured to:

during the first build cycle of the additive manufacturing process:

record the first set of sensor data streams via the sensor suite, the first set of sensor data streams comprising a first image recorded by the image sensor;

access a geometry of the first layer of the build; and

calculate, via the failure detection model, the first likelihood of failure based on:

the geometry of the first layer of the build; and

the first set of sensor data streams comprising the first image.

3 . The additive manufacturing system of claim 2 , wherein the controller is configured to:

detecting a real geometry of the first layer based on the first image; and

calculate, via the failure detection model the first likelihood of failure of the build based on a comparison between the geometry of the first layer and the real geometry of the first layer and based on the first set of sensor data streams.