3D Quality Voxel Mapping for Real-Time Additive Manufacturing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current quality control methods for additive manufacturing are time-consuming and expensive, often relying on visual inspection or post-processing analysis, which can be inefficient and costly.
Innovation Solution
A method and system for additive manufacturing quality control that involves receiving physical property measurement data from various spatial positions, defining a three-dimensional quality measurement volume of voxels with varying sizes, and applying quality evaluation methods to ensure real-time monitoring and evaluation, using machine learning and feedback loops for optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If visual inspection or post-processing analysis is used for quality control, then quality evaluation can be performed, but the process becomes time-consuming and expensive
Solution Approach 1:
The patent implements real-time quality monitoring during the additive manufacturing process itself, rather than performing inspection after manufacturing is complete. Sensors continuously collect data about the manufacturing process, and machine learning models evaluate quality metrics in real-time, enabling quality control to occur concurrently with production and eliminate separate post-processing inspection time
Solution Approach 2:
The patent replaces manual visual inspection with automated sensor-based measurement systems and machine learning algorithms. Optical sensors, thermal sensors, and other detection devices automatically collect manufacturing data, which is then processed by machine learning models to evaluate quality, substituting human inspection with automated intelligent systems that operate continuously without additional time cost
2Measurement precision
If visual inspection or post-processing analysis is used for quality control, then quality evaluation can be performed, but the cost increases
Solution Approach 1:
The patent implements a quality control system where the additive manufacturing process itself generates the data needed for quality evaluation through integrated sensors. The system monitors its own manufacturing process in real-time, eliminating the need for separate, expensive post-processing inspection equipment and reducing overall quality control costs
Solution Approach 2:
The patent replaces expensive manual inspection processes with automated machine learning-based quality evaluation systems. The machine learning models process sensor data to identify defects and quality issues, reducing reliance on costly human expert inspection and enabling more efficient quality control at lower cost
3Ease of manufacture
If uniform voxel sizes are used in quality measurement volume, then data processing is simpler, but detail resolution in critical areas is reduced
Solution Approach 1:
The patent divides the quality measurement volume into voxels with varying sizes based on local requirements. Critical areas with higher quality requirements are represented by smaller voxels for finer resolution, while less critical areas use larger voxels. This adaptive voxel sizing allows the system to maintain high measurement precision in important regions while keeping overall data processing manageable
4Measurement precision
If high-resolution quality measurement is performed throughout the entire part, then measurement precision is improved, but data processing complexity and time increase
Solution Approach 1:
The patent applies different voxel resolutions to different regions of the part based on their quality importance. Critical areas such as load-bearing regions or areas with complex geometry use finer voxel resolution for detailed quality assessment, while non-critical areas use coarser resolution. This localized approach maintains high measurement precision where needed while reducing overall data volume and processing complexity
Solution Approach 2:
The patent segments the quality measurement volume into multiple regions with different resolution requirements. By dividing the part into zones based on quality criticality and applying appropriate voxel sizes to each zone, the system processes data in manageable segments rather than uniformly across the entire part, reducing computational complexity while maintaining necessary precision
Data Source
AI summary
Disclosed, in one general aspect, is an additive manufacturing quality control method that includes receiving physical property measurement data values from different spatial positions for a three-dimensional part manufactured with an additive manufacturing process, and defining a three-dimensional quality measurement volume of quality measurement voxels that are each associated with at least one quality control measure and at least one quality evaluation method. The three-dimensional quality measurement volume of quality measurement voxels is formatted in a three-dimensional format in which one set of the quality measurement voxels are voxels of one size, and at least another set of the quality measurement voxels are voxels of a different size. The quality measurement voxels can define voxels each associated with a vector of the quality measures and a vector of the quality evaluation methods. The measurement data values may for example be acquired from various sensors during the additive manufacturing process.