Method and system for monitoring additive manufacturing processes
This invention teaches a quality assurance system for additive manufacturing. This invention teaches a multi-sensor, real-time quality system including sensors, affiliated hardware, and data processing algorithms that are Lagrangian-Eulerian with respect to the reference frames of its associated input measurements. The quality system for Additive Manufacturing is capable of measuring true in-process state variables associated with an additive manufacturing process, i.e. those in-process variables that define a feasible process space within which the process is deemed nominal. The in-process state variables can also be correlated to the part structure or microstructure and can then be useful in identifying particular locations within the part likely to include defects.
1. An additive manufacturing method, the method comprising:
depositing a layer of metal material on a build plane;
melting a build region of the layer of metal material to form a part using a heat source that scans across the build region of the layer of metal material, wherein the heat source generates a weld pool where the heat source interacts with the layer of metal;
monitoring the build region with a Lagrangian frame of reference optical sensor that follows a path of the weld pool to provide a Lagrangian sensor dataset and monitoring the build region with a Eulerian frame of reference optical sensor having a fixed field of view relative to the build plane to provide a Eulerian sensor dataset;
calibrating the Lagrangian sensor dataset using the Eulerian sensor dataset captured while the weld pool is within the fixed field of view of the Eulerian frame of reference optical sensor, wherein the fixed field of view includes the build region;
comparing the calibrated Lagrangian sensor dataset with a known-good range of a baseline dataset derived from previous additive manufacturing operations to determine whether any calibrated Lagrangian sensor dataset values are outside of the known-good range; and
in response to the comparison of the calibrated Lagrangian sensor dataset with the known-good range of the baseline dataset indicating one or more of the calibrated Lagrangian sensor dataset values are outside of the known-good range, adjusting the heat source causing subsequent calibrated Lagrangian sensor dataset values to be within the known-good range.
2. The additive manufacturing method of claim 1 , wherein a field of view of the Lagrangian frame of reference optical sensor is smaller than the fixed field of view of the Eulerian frame of reference optical sensor.
3. The additive manufacturing method of claim 2 , wherein the calibrated Lagrangian sensor dataset includes a calibrated temperature profile of the build region.
4. The additive manufacturing method of claim 1 , wherein the monitoring of the build region comprises measuring a temperature of the layer of metal material using the Lagrangian frame of reference optical sensor and calibrating the temperature measured by the Lagrangian frame of reference optical sensor using data collected by the Eulerian frame of reference optical sensor.
5. The additive manufacturing method of claim 1 , wherein calibrating the Lagrangian sensor dataset comprises converting raw sensor measurements provided by the Lagrangian frame of reference optical sensor to calibrated temperature measurements by comparing portions of the Lagrangian sensor and Eulerian sensor datasets.