Additive Manufacturing Parameter Transfer via Digital Twin
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The transfer of scan parameter sets between additive manufacturing machines is time-consuming, labor-intensive, and wasteful, requiring reworking and recalibration to maintain material properties and quality, especially when switching between different machines or machine lines.
Innovation Solution
The implementation of system identification and transfer learning techniques to create a digital twin of an additive machine, allowing for the automatic transfer of parameter sets between machines based on monitored sensor data, reducing the need for extensive recalibration and design of experiments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If scan parameter sets are reworked and redeveloped for each machine, then material properties and quality are maintained within tolerance limits, but time and labor are consumed excessively
Solution Approach 1:
The patent creates a digital twin (virtual copy) of the additive manufacturing machine that includes a trained machine learning model. This digital twin can be replicated and transferred to other machines, allowing parameter sets to be automatically adapted without manual rework. The machine learning model captures the relationship between scan parameters and material properties, enabling automatic generation of suitable parameter sets for target machines.
Solution Approach 2:
The patent uses machine learning to automatically adjust and transform scan parameter sets when transferring between machines. The system learns the specific characteristics of each machine and automatically modifies parameters to account for variations, eliminating manual recalibration while maintaining material properties within tolerance limits.
2Reliability
If multiple builds are performed for prove-out of scan parameter sets, then acceptable parameter sets are achieved, but machine production time is reduced and raw material is consumed
Solution Approach 1:
The patent performs preliminary training of machine learning models using historical sensor data and scan parameters before actual production. The digital twin is prepared in advance with pre-trained models that can predict suitable parameter sets, eliminating the need for multiple iterative builds during production setup.
Solution Approach 2:
The system uses historical sensor data from previous builds to train machine learning models that provide feedback on optimal parameter settings. This feedback mechanism allows the system to learn from past builds and automatically generate reliable parameter sets without requiring extensive prove-out builds.
3Manufacturing precision
If tight calibration is maintained on the machine, then material properties are within tolerance limits, but the process becomes time consuming and labor intensive
Solution Approach 1:
The patent implements self-calibration through machine learning models that automatically adapt to each machine's characteristics. The digital twin learns from sensor data and automatically adjusts parameter sets without requiring manual calibration by operators, making the process both precise and easy to manufacture.
Solution Approach 2:
The patent replaces manual calibration procedures with an automated machine learning-based system. The digital twin uses computational algorithms to determine optimal parameter sets, substituting the mechanical/manual calibration process with an intelligent automated system that maintains precision while reducing labor intensity.
4Adaptability or versatility
If scan parameter sets are transferred between different machines, then production flexibility is improved, but parameter accuracy deteriorates without rework
Solution Approach 1:
The patent creates a universal digital twin framework that can be applied across different additive manufacturing machines. The machine learning model in the digital twin is designed to adapt to various machine types and configurations, enabling parameter set transfer while maintaining accuracy through automatic adjustment to target machine characteristics.
Data Source
AI summary
A method of transferring operational parameter sets between different domains of additive manufacturing machines includes creating a first machine domain parameter set in a first machine domain, accessing a model of a second additive manufacturing in a second machine domain, creating a second machine domain parameter set by applying transfer learning techniques including learning differences between the first machine domain and the second machine domain, adjusting the first machine domain parameter set using the differences before incorporation into the second machine domain to obtain the second machine domain parameter set, the second machine domain parameter set representing operational settings for the second additive manufacturing machine, the second additive manufacturing machine producing a product sample, determining if the product sample is within quality assurance metrics, and if the product sample is not within the quality assurance metrics, adjusting the second machine domain parameter set.