Optical emission spectrometry (oes) data-driven inspection of inclusions in wrought aluminium alloys

Abstract

A promising approach to the rapid inspection of inclusions in wrought aluminium alloys is optical emission spectroscopy (OES). However, in order to separate the peaks corresponding to particular inclusions from the peaks obtained from various microstructural features in the matrix, an advanced filtering of the OES spectrum is necessary. The methodology developed in this work is based on big-data-driven predictions of whether an on-line analysed sample is good or bad. A sufficient amount of relevant data, necessary for data-driven predictions, was established by the systematic quality control of samples of AA6082 using optical and scanning electron microscopy and by analysing the same surface using OES. By following a machine-learning process, an algorithm was developed to enable the on-line division of the samples into good and bad, based on criteria received from the casting house. Although the obtained results are promising, further improvements are necessary before this method can be validated for use in regular production.