Real-time Quality Monitoring of Wire EDM Process Using Machine Learning
Abstract
Traditional quality assessment in EDM involves labor-intensive manual evaluation of geometric deviations. Anticipating curvatures post-main cut allows compensation through trim cut adaptation, streamlining the process. This paper presents a novel approach for automating and realtime quality monitoring of Wire EDM processes through the integration of machine learning models trained on historical process and quality data. The effectiveness of this method is validated on industrial Wire EDM machine along with feature importance analysis to provide insights into the underlying patterns captured. Testing on industrial unseen data demonstrates a mean absolute error of approx. 1 µm for geometric curvature deviation in the Wire EDM process.
Fields of Study & Subjects
Advanced Machining and Optimization Techniques
Advanced machining processes and optimization
Manufacturing Process and Optimization
Publication details
- Journal / source
- Procedia CIRP
- Publisher
- Elsevier BV
- Volume
- 138
- Pages
- 1015–1020
- Publication date
- 1 January 2026
- Language
- EN