ML Deformation Prediction for Additive Manufacturing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing additive manufacturing techniques face challenges in controlling the shape of end objects due to porosity issues in precursor parts and limited control over sintering and fusion processes.
Innovation Solution
The use of a machine learning model, specifically a deep neural network, to predict object deformations during the sintering process by analyzing temperature profiles and voxel-level interactions, allowing for more accurate shape control and reduced simulation time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional physics-based simulation methods are used to predict object deformations during sintering, then prediction accuracy is improved, but computational time and complexity increase significantly
Solution Approach 1:
The system performs preliminary action by training a machine learning model offline using physics-based simulation data. The trained model can then rapidly predict deformations without requiring real-time physics simulations, thus achieving both high accuracy and fast prediction speed during actual sintering processes.
Solution Approach 2:
A machine learning model serves as an intermediary between complex physics-based simulations and practical deformation prediction needs. The model learns the underlying patterns from simulation data and translates them into fast, accurate predictions, bridging the gap between theoretical accuracy and practical speed requirements.
2Adaptability or versatility
If binder jet additive manufacturing is used to produce precursor parts, then manufacturing flexibility is improved, but porosity control deteriorates leading to shape deformation
Solution Approach 1:
The system performs preliminary compensation by predicting deformations before the sintering process occurs. The digital model is adjusted in advance to counteract expected deformations, ensuring that the final sintered part achieves the desired shape despite inherent porosity issues in binder jet manufactured precursor parts.
Solution Approach 2:
The system applies preliminary anti-action by pre-distorting the digital model in the opposite direction of expected deformation. This compensatory approach counteracts the harmful effects of porosity-induced shrinkage and deformation during sintering, maintaining manufacturing precision while preserving the flexibility of binder jet technology.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables faster and more accurate prediction of object deformations, improving shape control and reducing the computational complexity of traditional physics-based simulation methods.
Implementation Method 1
predict object deformations during the sintering process by analyzing temperature profiles
Data Source
AI summary
Examples of methods are described herein. In some examples, a method includes determining a graph representation of a three-dimensional (3D) object. In some examples, the graph representation includes nodes and edges associated with the nodes. In some examples, each node includes a temperature profile attribute. In some examples, the method includes predicting, using a machine learning model, a deformation of the 3D object based on the graph representation.