Correlating Layerwise Monitoring with Volumetric Inspection in Additive Manufacturing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current in-situ process monitoring systems in additive manufacturing lack the ability to robustly correlate predicted defects with real material defects, relying solely on generated data and requiring costly post-inspection methods to ensure quality, which leads to inefficiencies and waste in the manufacturing process.
Innovation Solution
A computer-implemented method that correlates monitoring data from the additive manufacturing process with post-manufacturing inspection data by transforming and computing the spatial overlap between layerwise monitoring data and volumetric inspection data, allowing for the derivation of defect volumes and anomalies, thereby providing deeper insights into defect formation and improving quality assurance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If costly post-inspection methods are used to ensure quality, then quality assurance is improved, but manufacturing cost and time are increased
Solution Approach 1:
The patent performs defect detection and analysis during the additive manufacturing process itself, before the part is completed. By transforming monitoring data from the layerwise domain to the volumetric domain and correlating it with inspection data in real-time, the system identifies defects early, allowing for potential process adjustments without waiting for post-manufacturing inspection
Solution Approach 2:
The patent creates a digital twin or virtual representation of the physical part by transforming and correlating monitoring data with inspection data. This digital model allows for virtual inspection and defect analysis, replacing or reducing the need for physical post-inspection methods like computed tomography
2Productivity
If in-situ monitoring data alone is used for defect detection, then manufacturing time is reduced, but measurement precision is insufficient
Solution Approach 1:
The patent merges in-situ monitoring data with post-inspection data by transforming both to a common volumetric domain and computing their spatial overlap. This combination leverages the real-time advantages of monitoring data with the high precision of inspection data, achieving both productivity and measurement precision
Solution Approach 2:
The patent introduces a data transformation and correlation system as an intermediary between monitoring data and inspection data. This intermediary transforms monitoring data from the layerwise domain to the volumetric domain and computes spatial overlap with inspection data, enabling accurate defect correlation without directly comparing incompatible data formats
3Manufacturing precision
If extensive post-inspection is performed to ensure quality, then manufacturing precision is improved, but loss of time and productivity are worsened
Solution Approach 1:
The system performs defect detection and spatial correlation during or immediately after the manufacturing process, before final part completion. By transforming monitoring data to the volumetric domain and correlating with inspection data in real-time, comprehensive defect detection is achieved without requiring extensive post-inspection, maintaining high manufacturing throughput
Solution Approach 2:
The patent establishes a feedback loop where monitoring data is continuously transformed and correlated with inspection data to provide real-time defect information. This feedback enables immediate process adjustments and reduces the need for extensive final inspection, improving both precision and productivity
Data Source
AI summary
A computer-implemented method of correlating monitoring data and post-manufacturing inspection data in powder-bed-based additive manufacturing of workpieces includes (a) providing monitoring data including anomaly detection information of a workpiece's geometry in a layerwise domain, (b) providing inspection data including defect information of the workpiece's geometry in a volumetric domain, (c) extracting a region-of-interest information from the monitoring data, and, either, (d1) transforming the defect information of the inspection data from the volumetric domain into the layerwise domain, or, (d2) transforming the anomaly detection information of the monitoring data from the layerwise domain into the volumetric domain, as well as, (e) computing a spatial overlap of the monitoring data of interest and the transformed inspection data.