Additive Build Health Analytics for In-Process Machine Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing additive manufacturing processes face challenges in diagnosing aborted or failed builds and identifying performance issues, requiring significant manual effort and time, with the root cause often difficult to determine during the build process.
Innovation Solution
A system is implemented that separates the monitoring and analysis processes from the additive manufacturing machine, using separate computing devices with specialized hardware to analyze data without interrupting the build, allowing for predictive trend identification and real-time adjustments to maintain build integrity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual diagnosis is used to identify build failures and performance issues, then human expertise can be applied to complex problems, but significant time and human labor are required
Solution Approach 1:
The patent replaces manual mechanical diagnosis with an automated analytics processor that uses machine learning models and algorithms to analyze build data, sensor data, and machine parameters. This substitution eliminates the need for human experts to manually diagnose build failures, significantly reducing diagnosis time while maintaining or improving diagnostic accuracy through consistent automated analysis.
Solution Approach 2:
The system enables self-diagnosis capabilities where the additive manufacturing machine automatically monitors its own performance, detects anomalies, and identifies failure causes without external human intervention. The analytics processor continuously analyzes operational data and provides automated diagnostics, allowing the system to serve itself in terms of health monitoring and troubleshooting.
2Reliability
If root cause analysis is performed during the build process, then real-time corrections can be made, but the process becomes complex and time-consuming
Solution Approach 1:
The patent implements predictive analytics that identify potential build failures and root causes before they occur during the additive manufacturing process. By analyzing historical data, machine parameters, and operational patterns in advance, the system predicts likely failure modes and prepares corrective actions beforehand, enabling real-time interventions without adding complex monitoring during the actual build process.
Solution Approach 2:
The analytics processor serves as an intermediary between the additive manufacturing machine and the operator. It simplifies complex data analysis by processing sensor readings, machine parameters, and build data through machine learning models, then presenting simplified diagnostic information and recommended actions to operators. This intermediary layer handles the complexity internally while providing clear, actionable outputs.
3Productivity
If separate computing devices are used for monitoring and analysis, then build processing is not interrupted, but system complexity increases
Solution Approach 1:
The patent segments the monitoring and analysis functions into a separate analytics processor that operates independently from the main additive manufacturing control system. This segmentation allows the build process to continue uninterrupted on the manufacturing machine while the separate analytics device processes data in parallel, analyzing build health and predicting failures without interfering with production throughput.
Solution Approach 2:
The separate analytics processor acts as an intermediary system that receives data from the additive manufacturing machine via communication interfaces without disrupting the manufacturing process. It processes analytical workloads independently and provides feedback through standardized interfaces, effectively mediating between the production system and the diagnostic functions while maintaining system modularity and reducing coupling complexity.
Data Source
AI summary
Systems, apparatus, computer-readable medium, and associated methods to monitor, analyze, and adjust at least one of 1) additive machine health and configuration or 2) build health and configuration are disclosed. An example apparatus includes an analytics processor, separate from and in a trusted relationship with an additive manufacturing machine building a part, to process, based on a trigger, data from monitoring of the additive manufacturing machine and the build of the part, the analytics processor including a hybrid model fusing additive process physics and data science to process the data to identify an abnormality in at least one of the build or the additive manufacturing machine and to adjust a configuration of the additive manufacturing machine during the build to address the abnormality.