Artificial neural network for on-site quantitative analysis of soils using laser induced breakdown spectroscopy |
| |
| Affiliation: | 1. Univ. Bordeaux, LOMA, UMR 5798, F-33400 Talence, France;2. CNRS, LOMA, UMR 5798, F-33400 Talence, France;3. IVEA Solution, Centre Scientifique d''Orsay, Bât 503, 91400 Orsay, France;4. BRGM, Service Métrologie, Monitoring et Analyse, 3 avenue Claude Guillemin, B.P 36009, 45060 Orléans Cedex, France;1. Physics Department, Università degli Studi di Milano, Via Celoria 16, 20134 Milano, Italy;2. T&D Technologies Department, Ricerca sul Sistema Energetico – RSE, Via Rubattino 54, 20134 Milano, Italy;1. J. Heyrovský Institute of Physical Chemistry, Academy of Sciences of the Czech Republic, Dolejškova 3, 18223 Prague 8, Czech Republic;2. Department of Experimental Physics, Faculty of Mathematics, Physics and Informatics, Comenius University, Mlynská dolina F2, 842 48 Bratislava, Slovakia;3. Laboratoire interdisciplinaire Carnot de Bourgogne, UMR CNRS 6303, Université de Bourgogne, BP 47 870, F-21078 Dijon Cedex, France;1. Christian Doppler Laboratory for Laser-Assisted Diagnostics, Institute of Applied Physics, Johannes Kepler University Linz, Linz A-4040 Austria;2. voestalpine Stahl GmbH, Linz A-4031 Austria;1. BAM, Federal Institute for Materials Research and Testing, Richard Willstätter-Straße 11, D-12489 Berlin, Germany;2. Chemistry Department, Humboldt Univerisät zu Berlin, Brook-Taylor-Straße 2, D-12489 Berlin, Germany;3. Institute of Physical Engineering, Faculty of Mechanical Engineering, Brno University of Technology, Technická 2896/2, 61669 Brno, Czech Republic;4. Institute for Mining, Technical University Clausthal, Erzstraße 18, 38678 Clausthal-Zellerfeld, Germany |
| |
| Abstract: | Nowadays, due to environmental concerns, fast on-site quantitative analyses of soils are required. Laser induced breakdown spectroscopy is a serious candidate to address this challenge and is especially well suited for multi-elemental analysis of heavy metals. However, saturation and matrix effects prevent from a simple treatment of the LIBS data, namely through a regular calibration curve. This paper details the limits of this approach and consequently emphasizes the advantage of using artificial neural networks well suited for non-linear and multi-variate calibration. This advanced method of data analysis is evaluated in the case of real soil samples and on-site LIBS measurements. The selection of the LIBS data as input data of the network is particularly detailed and finally, resulting errors of prediction lower than 20% for aluminum, calcium, copper and iron demonstrate the good efficiency of the artificial neural networks for on-site quantitative LIBS of soils. |
| |
| Keywords: | |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
