Data-driven statistical analysis for discharge position prediction on Wire EDM

• Published in: Procedia CIRP 2022, Vol.113, p.143-148
• Main Authors: Kronauer, Samuele, Mavkov, Bojan, Mejari, Manas, Piga, Dario, Jaques, Fabrice, D'Amario, Rino, Campli, Riccardo Di, Nasciuti, Adriano
• Format: Article
• Language: English
• Subjects: Data-driven statistical analysis; Predictive Model; Wire EDM
• ISSN: 2212-8271; 2212-8271
• DOI: 10.1016/j.procir.2022.09.122

Wire-cut Electrical Discharge Machining (Wire EDM) is a machining technique widely used to cut high-precision punch tools and highly value added precision components. With increasing resource efficiency requirements and zero-defect manufacturing trend, pushing the limits of machining reliability, even at cutting speed next to the technical limits, is becoming imperative. Predicting the position of the sparks along the wire is thus needed to develop more efficient EDM processes, thanks to the suppression of discharges which are expected to happen in undesired positions. This will lead to a reduction of the number of machine stops caused by wire breaks. Motivated by this need, the paper presents a data-driven statistical analysis to get insight into the correlation between the discharge positions of two consecutive sparks, along with the relation between spark positions and discharge frequency. The underlying basis for this analysis was the possibility to obtain reliable real-time information about the position of sparks along the wire, a feature made available by a Discharge Location Tracker [4]. Results on spark position correlation are presented for cutting experiments on a steel workpiece of 50 mm in plane parallel with different machining parameters using brass wire of diameter 0.25 mm.