An evolutionary regulation algorithm for the twin laser measuring system

An instrument based on twin laser sensors for non-contact measuring is designed for a machine tool applied in the process monitoring for the thickness measurement of a print circuit board (PCB). Without proper adjustments of an experienced operator, the precision of twin laser measuring is worse than the resolution of a single laser measuring. In this paper, a hybrid system by evolving a case-based reasoning (CBR) system with a genetic algorithm (GA) is developed for the automatic regulation problem of a twin laser measuring system. The results of the thickness measurement of a PCB by the hybrid system were compared with the results of a back propagation neural network, a conventional CBR, and a multiple regression analysis method. The experimental results show that the GA/CBR is more accurate and efficient when applied to the thickness measurement of a PCB than other traditional methods.