| About this Abstract |
| Meeting |
6th World Congress on Integrated Computational Materials Engineering (ICME 2022)
|
| Symposium
|
6th World Congress on Integrated Computational Materials Engineering (ICME 2022)
|
| Presentation Title |
Image Processing Based Failure Site Prediction of PC Wires During Wiredrawing Using Supervised Machine Learning Approach |
| Author(s) |
Mohamed Imad Eddine Heddar, Mehdi Brahim, Nedjoua Matougui |
| On-Site Speaker (Planned) |
Mohamed Imad Eddine Heddar |
| Abstract Scope |
The aim of this work is to develop an image texture and defect morphology based method for predicting failure site whether it is during decoiling or wiredrawing probabilistically. Traditional image segmentation of micrographs for acquiring stereological particles data is used coupled with image processing for extracting texture based information. The data are modeled through supervised classification using Random Forest Classifier (RFC) with hyper parameter optimization for model calibration. Based on comparison between experimental results and model testing points, it was found that this model can achieve decent predictive results with classification accuracy rate around ~ 71.3%. The obtained results can be improved using of multisource datasets with the addition of local mechanical properties. |
| Proceedings Inclusion? |
Definite: Other |