一种基于SLM工艺参数编码的分层孔隙率预测方法及系统
By using a layered porosity prediction method that encodes SLM process parameters, and combining CT image processing and multi-dimensional sensitivity analysis with a machine learning model, the problem of porosity prediction in the SLM process is solved, achieving high-precision porosity prediction that is suitable for the production of complex parts.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUAZHONG UNIV OF SCI & TECH
- Filing Date
- 2025-11-21
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies struggle to accurately predict porosity caused by the layer-by-layer forming characteristics of selective laser melting (SLM) processes, especially in the production of parts with complex geometries. Traditional methods cannot effectively handle nonlinear effects and multi-layer interactions.
By preprocessing CT images of SLM samples, the porosity of each layer is obtained. The sensitivity of process parameters is analyzed using Pearson correlation coefficient, random forest variable importance analysis, variance inflation factor and spatial lag regression model. One-dimensional and two-dimensional feature matrices are constructed, and high-precision prediction is performed by combining SVR, BP neural network and CNN models.
It achieves high-precision prediction of porosity of SLM formed parts, improves the accuracy and reliability of the model, and is applicable to different data scales and task complexities. In particular, the prediction accuracy of the CNN model reaches 90.45%, overcoming the shortcomings of traditional methods.
Smart Images
Figure CN121544942B_ABST