Powder Bed Fusion Defect Verification Using Stochastic Physics Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Additive manufacturing processes face inefficiencies due to the iterative and time-consuming process of refining parameters to achieve acceptable quality, particularly in high-tolerance components like aircraft parts, where stochastic defects can occur randomly and affect fatigue life, hindering widespread adoption in certain industries.
Innovation Solution
A method involving a multidimensional space physics model to predict stochastic defects by analyzing additive manufacturing parameters and generating random values for uncontrolled and uncontrollable variables, determining probability distributions, and categorizing part designs as defect-free when below a predefined threshold, using a controller with a processor and memory to iterate simulations and correlate variables with flaw occurrences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If iterative trial-and-error parameter refinement is used to achieve acceptable quality, then manufacturing precision is improved, but loss of time increases significantly
Solution Approach 1:
The patent performs preliminary computational analysis and simulation of stochastic defects before actual manufacturing. By using physics-based models to predict potential flaws and optimize parameters in advance, the system eliminates the need for extensive iterative trial-and-error testing, thereby reducing time loss while maintaining manufacturing precision.
Solution Approach 2:
The patent creates virtual copies of the manufacturing process through computational models and simulations. By analyzing defect probabilities in these virtual representations before physical manufacturing, the system can identify and correct parameter issues without time-consuming physical iterations, resolving the contradiction between precision and time.
2Reliability
If extensive parameter iterations are performed to ensure defect-free high-tolerance components, then reliability is improved, but productivity decreases
Solution Approach 1:
The patent replaces physical trial-and-error manufacturing iterations with computational modeling and simulation. By using physics-based stochastic defect models to predict reliability outcomes virtually, the system achieves high reliability certification without the productivity-loss-inducing extensive physical iterations, thus resolving the contradiction.
Solution Approach 2:
The system performs preliminary reliability assessment through computational analysis before manufacturing. By predicting stochastic defect probabilities using physics models in advance, the system can qualify parts for high-tolerance applications rapidly without extensive iterative testing, improving both reliability assurance and productivity.
3Manufacturing precision
If traditional iterative methods are used to qualify additively manufactured parts, then manufacturing precision is ensured, but device complexity increases due to multiple iterations
Solution Approach 1:
The patent uses virtual copies and computational models to represent the manufacturing process and predict outcomes. By analyzing defect probabilities in simulated environments before physical manufacturing, the system ensures manufacturing precision without requiring complex iterative processes, thereby reducing overall process complexity.
Solution Approach 2:
The system replaces complex physical iterative testing with computational modeling. By using physics-based simulations to predict stochastic defects and optimize parameters virtually, the system maintains manufacturing precision while significantly simplifying the qualification process and reducing device complexity.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
A method of evaluating an additive manufacturing process includes receiving a set of additive manufacturing parameters and an additive manufacturing part design at an analysis module, receiving a set of random values at the analysis module, determining a probability distribution of stochastic flaws within a resultant additively manufactured article using at least one multidimensional space physics model, and categorizing the additive manufacturing part design as defect free when the probability distribution is below a predefined threshold. Each value in the set of random values corresponds to a distinct variable in a set of variables. Each variable in the set of variables at least partially defines at least one of an uncontrolled additive manufacturing parameter and an uncontrollable additive manufacturing parameter.