Quantitative Analysis and Identification Improvement in Laser-Induced Breakdown Spectroscopy by Self-Absorption Correction and Artificial Neural Network, By Seyed Sajad Moosavifard

Research

Title	Quantitative Analysis and Identification Improvement in Laser-Induced Breakdown Spectroscopy by Self-Absorption Correction and Artificial Neural Network
Type	Article
Keywords	Training , Adaptive optics , Neurons , Metals , Logic gates , Plasmas , Optical variables measurement
Researchers	Amir Hossein Farhadian، Seyed Sajad Moosavifard

Abstract

Abstract: In this research, the laser-induced breakdown spectroscopy (LIBS) technique is used for concentration prediction and identification in aluminum alloys. For this purpose, calibration-free LIBS (CF-LIBS) and artificial neural network (ANN) analyses were implemented. Self-absorption correction (SAC) and gate time improvement in CF-LIBS lead to more accurate quantitative results and concentration calculation close to real values. In addition, in identification of different Al alloys by ANNs, results show that using corrected lines intensity of fundamental species has better results in network construction and fewer errors.