COMPOSITE MANUFACTURING WITH A MULTI-DIMENSIONAL ARRAY OF INDEPENDENTLY-CONTROLLABLE THERMAL ZONES

摘要

A system for manufacturing a composite structure includes a tool with a surface for supporting component elements of the composite structure. The surface is divided into a multi-dimensional array of thermal zones for in-process control of the temperature of a component element (e.g. resin) of the composite structure. The sensors sense a characteristic of the component element and provide sensor data, which is applied to a machine-learning algorithm configured to generate control data to achieve a defined quality goal. A controller then independently controls the thermal zones to locally heat, cool or maintain the temperature of the component element according to the control data to advance the component element or composite structure to the defined quality goal. This may be performed over a plurality of instances during which the machine-learning algorithm learns to increase advancement of the component element or composite structure to the defined quality goal.

主权项

1. A system for manufacturing a composite structure, the system comprising:
a tool having a surface configured to support component elements of a composite structure, and a plurality of heating/cooling devices configured to provide a plurality of thermal zones into which the surface is divided, the plurality of heating/cooling devices and thereby the plurality of thermal zones being arranged in a multi-dimensional array and independently controllable to locally heat, cool or maintain a temperature of a component element of the component elements during manufacture of the composite structure therefrom; a plurality of sensors configured to sense a characteristic of the component element at a respective plurality of spatially-distributed points, and provide sensor data corresponding thereto; a computing device coupled to the plurality of sensors and configured to apply the sensor data to a machine-learning algorithm configured to generate control data to achieve a defined quality goal for the component element or composite structure; and a controller coupled to the computing device and tool, and configured to independently control the plurality of heating/cooling devices and thereby the plurality of thermal zones to locally heat, cool or maintain the temperature of the component element according to the control data to advance the component element or composite structure to the defined quality goal, wherein the plurality of sensors is configured to sense the characteristic and provide sensor data corresponding thereto, the computing device is configured to apply the sensor data to the machine-learning algorithm, and the controller is configured to independently control the plurality of heating/cooling devices and thereby the plurality of thermal zones, at a plurality of instances during manufacture of the composite structure from the component elements, the machine-learning algorithm learning over the plurality of instances to increase advancement of the component element or composite structure to the defined quality goal.