Use of LIBS technique for identification of type of pollutant and ESDD level on epoxy-alumina nanocomposites using ANN. (13th July 2021)

• Record Type: Journal Article
• Authors: Babu, Myneni Sukesh; Neelmani, ; Vasa, Nilesh Jayantilal; Sarathi, Ramanujam; Imai, Takahiro
• Is Part Of: Measurement science & technology. Volume 32:Number 11(2021)
• Publication Date: 2021-07-13
• Subjects: alumina -- ANN -- conductivity -- epoxy -- ESDD -- intensity ratio -- LIBS -- nanocomposites -- pollutants; Physical measurements -- Periodicals; Scientific apparatus and instruments -- Periodicals; Equipment and Supplies -- Periodicals; Science -- instrumentation -- Periodicals; Technology -- instrumentation -- Periodicals; Mesures physiques -- Périodiques; Physical measurements; Scientific apparatus and instruments; Periodicals; 502.87
• DOI: 10.1088/1361-6501/ac0d22
• Languages: English
• ISSNs: 0957-0233

Abstract

Abstract: Epoxy-alumina nanocomposites have been coated with different types of pollutants at various concentrations in order to analyze their pollution performance using laser induced breakdown spectroscopy (LIBS) study. Elemental analysis of spectral peaks has been successful in determining the elemental composition of the pollutant present on the surface of the specimens. The conductivity and equivalent salt deposition density (ESDD) values have been determined for all the pollutants at different concentrations. A direct correlation between ESDD and the normalized intensity ratio of LIBS spectral data is noticed. Regression coefficient ( R 2 ) has been used as a performance parameter in correlating the ESDD and normalized intensity ratio. Artificial neural network (ANN) technique has been adopted to LIBS spectral data for the classification of the pure epoxy and 3 wt% epoxy-alumina nanocomposite specimen based on conductivity and type of pollutant. ANN developed with conjugate gradient backpropagation with Polak-Ribiere updates training algorithm has reflected higher classification accuracy and required lesser epochs to converge, compared to other training algorithms in classifying the contaminated specimens.