Data-driven quality monitoring of bending processes in hairpin stator production using machine learning techniques

Abstract

In the course of series production of high-performance automotive traction motors, the so-called hairpin technology is increasingly coming into focus. An essential process step is the bending of the hairpins, which is significantly influenced by variations in the wire material. Therefore, this paper will investigate how machine learning techniques can be used to monitor the dimensional accuracy of hairpins solely based on bending process data such as torque curves. In this way, deviations can be detected at an early stage and measures can be taken to prevent further rejects.