Additive Manufacturing Control With Adaptive Anomaly Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In additive manufacturing processes like wire arc additive manufacturing (WAAM) and laser metal deposition, the lack of repeatability and inability to effectively utilize in-situ monitoring data for process adjustments leads to high rejection rates, as defects are identified but not adequately addressed during the manufacturing process.
Innovation Solution
Implementing a method that includes creating machine code, generalized and adaptive anomaly detection models, and a digital twin to monitor and adjust process parameters in real-time, allowing for early anomaly detection and elimination, reducing rejection rates by using sensor data to predict and correct anomalies during the manufacturing process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If in-situ monitoring systems are implemented to detect defects during the manufacturing process, then the ability to identify defects is improved, but the rejection rate remains high because the detected defects cannot be effectively addressed
Solution Approach 1:
The system implements a feedback mechanism where sensor data from in-situ monitoring is continuously fed back to the control system, which automatically adjusts process parameters to eliminate detected defects. This closed-loop feedback enables real-time correction of anomalies without aborting the process, thereby maintaining high defect detection capability while improving component acceptance rates through automatic remediation.
Solution Approach 2:
The manufacturing system performs self-correction by automatically adjusting its own process parameters based on detected defects. The control system autonomously modifies welding parameters, robot motion, or other process variables to eliminate anomalies, enabling the system to service itself without external intervention and convert detected defects into acceptable components.
2Reliability
If the manufacturing process is aborted when anomalies are detected to ensure quality, then component quality is maintained, but productivity decreases due to frequent process interruptions and rejections
Solution Approach 1:
The system takes preliminary action by detecting anomalies early in the manufacturing process and immediately initiating corrective measures before defects propagate or become irreparable. This early intervention allows the system to prevent quality issues rather than merely detecting and aborting them, maintaining component quality while avoiding process interruptions and preserving manufacturing throughput.
Solution Approach 2:
The manufacturing process continues without interruption by implementing real-time defect elimination. Instead of aborting when anomalies are detected, the control system continuously adjusts process parameters to correct defects on-the-fly, maintaining the continuity of useful manufacturing action and preventing productivity loss from process abortions while ensuring component quality.
3Manufacturing precision
If process parameters are adjusted based on detected defects, then defect elimination is improved, but process complexity increases due to the need for real-time monitoring and control adjustments
Solution Approach 1:
The control system integrates multiple functions into a unified platform that performs sensor data acquisition, anomaly detection, process parameter analysis, and automatic adjustment. This multi-functional system consolidates what would otherwise be separate monitoring and control devices, reducing overall system complexity while maintaining advanced defect elimination capabilities through integrated real-time processing.
Data Source
AI summary
Various embodiments of the teachings herein include a method for the additive manufacture of a component. The method may include: training a representation with datasets from a previously executed manufacturing process with a known process result; calculating output data from input data; and creating an adaptive anomaly detection model trained on a parameter-set-specific basis with available training data. The method may include transferring the machine code and the detection models to a control system; starting the manufacturing process; monitoring the process with sensors; evaluating sensor signals of the manufacturing process using the generalized anomaly detection model; training a specialized anomaly detection model in parallel from an adaptive anomaly detection model using process data of the running manufacturing process; and detecting anomalies in the manufacture of the component using the specialized anomaly detection model during the manufacturing process.