Prediction of Wall Thickness Parameters in TPMS Models Based on CNN-SVM and MLR
View Full Paper
Ask AI
Bookmark
Share
Key Points
The aim is to accurately predict wall thickness parameters in TPMS models during the design stage to enhance forming quality.
Developed a CNN-SVM framework with MLR for prediction.
Generated 152 TPMS models to create 912 data samples.
Implemented voxel-based sampling and rasterization preprocessing for data preparation.
Used TPMS type, lattice filling type, volume fraction, and cell length as input features.
CNN-SVM model achieved a mean squared error of 0.0011 and an R2 of 0.92.
Approximately 86.9% of test samples had prediction errors within 20%.
Performance improvements of 15.8%, 10.6%, and 18.5% over traditional MLR, CNN, and SVM models were observed.
Sheet filling type has the highest positive impact on wall thickness, while Diamond TPMS negatively affects it.
Discussion
No takes yet. Share an insight, caveat, or question.
Implication
Demonstrates an effective method for predicting wall thickness parameters in TPMS models, suggesting improved design outcomes for additive manufacturing.