Sensor-Assisted 3D Printing with Voxel-Level Quality Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current additive manufacturing techniques require time-consuming and costly processes to determine optimal printing parameters for achieving high-quality 3D printed parts, especially for complex geometries and materials used in industries like aviation and medicine, due to the need for numerous tests and evaluations.
Innovation Solution
The use of in-situ information, such as voxel-level references for meltpool characteristics, and a closed-loop control architecture that adjusts process parameters in real-time using sensors to achieve desired part quality, allowing for faster development and reduced costs by controlling part quality at the voxel level.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional additive manufacturing processes are used to determine optimal printing parameters, then part quality can be achieved, but part development time and costs increase significantly
Solution Approach 1:
The system performs preliminary action by generating a voxelized reference map before the actual printing process. This reference map contains pre-calculated optimal printing parameters for each voxel based on the desired part quality, eliminating the need for time-consuming trial-and-error testing during part development
Solution Approach 2:
The invention creates a digital copy (voxelized reference map) of the desired part geometry with associated printing parameters. This digital twin allows virtual optimization of printing parameters without physical testing, reducing development time while maintaining part quality
2Manufacturing precision
If traditional additive manufacturing processes are used to determine optimal printing parameters, then part quality can be achieved, but development costs increase due to numerous tests and evaluations
Solution Approach 1:
The system uses digital copying by creating a voxelized reference map that virtualizes the printing parameter optimization process. This eliminates costly physical trials and material waste associated with traditional trial-and-error methods while ensuring high part quality
Solution Approach 2:
The invention replaces the mechanical trial-and-error testing system with a computational system that calculates optimal printing parameters through voxelized reference maps. This substitution eliminates the need for repeated physical testing and material consumption, reducing development costs
3Manufacturing precision
If voxel-level control of part quality is implemented, then part quality precision improves, but system complexity increases
Solution Approach 1:
The system applies segmentation by dividing the part into individual voxels and assigning specific printing parameters to each voxel based on the voxelized reference map. This enables precise local control of part quality while managing complexity through automated parameter assignment
Solution Approach 2:
The voxelized reference map serves as a digital copy that encapsulates all the complex voxel-level control information. This digital representation simplifies the interface between the control system and the printer, as the complex optimizations are pre-calculated and stored in the reference map
Data Source
Figure 1A~1B
Figure 2
Figure 3
AI summary
Methods and apparatus are disclosed for sensor-assisted part development in additive manufacturing. An example apparatus includes at least one memory, instructions in the apparatus, and processor circuitry to execute the instructions to identify a reference process observable (116) of a computer-generated part (110), receive input (164) from at least one sensor during three-dimensional printing to identify an estimated process observable (168) using feature extraction (166), and adjust at least one three-dimensional printing process parameter (158) to reduce an error (152) identified from a mismatch between the estimated process observable (168) and the reference process observable (116).