Copper particle detection method by in-situ resistance background implemented in conductance-path recognition algorithm (iRB-CPR) among dominant aluminum particles

Abstract

Copper particle has been detected among dominant aluminum particles by in-situ resistance background implemented in conductance-path recognition algorithm (iRB-CPR). The iRB is a particle volume fraction prediction from a resistance background database measured by shifted-four-wire measurement (SRM) in various particle arrangements. The CPR is adopted from a decision tree algorithm to predict the particle conductance path by comparing the real-condition resistances classified by iRB. The selected resistance background is used as a denominator to normalize the resistance measurement by SRM under unknown particle arrangement in the real condition. Finally, Cu is detected by comparing the normalized resistances with the normalized spatial-mean resistance. The iRB-CPR successfully suppress the resistance deviation ratio in the simulation 〈r〉sim 93.8% and experiment 〈r〉exp 73.9%. Moreover, the Cu detection by iRB-CPR satisfies give a low error ES = 6.11% at S = 180 under α = 10%. This research opens a new detection modality as a part of non-ferrous metal particles recovery.