Improving the resilience of neural network solution of inverse problems in Raman spectroscopy of multi-component solutions of inorganic compounds to the distortions caused by frequency shift of the spectral channelsстатья Исследовательская статья

Информация о цитировании статьи получена из Scopus
Статья опубликована в журнале из списка Web of Science и/или Scopus
Дата последнего поиска статьи во внешних источниках: 25 апреля 2019 г.

Работа с статьей


[1] Improving the resilience of neural network solution of inverse problems in raman spectroscopy of multi-component solutions of inorganic compounds to the distortions caused by frequency shift of the spectral channels / I. V. Isaev, S. A. Burikov, T. A. Dolenko et al. // Journal of Physics: Conference Series. — 2018. — Vol. 1096, no. 1. — P. 012100–1–012100–8. In this study, we considered the problem of determining the concentrations of ions dissolved in water by the spectra of Raman scattering of light. At the moment, there are no adequate mathematical models describing the studied object, so in fact the only way to solve this problem is use of machine learning methods based on experimental data. As any data resulting from experimental measurements contains noise, there is a need to develop specific approaches to improving the resilience of the solution to noise in the data. Regarding the studied problem, experimental data may contain distortions of three types: variations in the concentrations of ions, error in the determination of the intensity in the channels of a spectrum, and frequency shift of the channels of a spectrum. This study is devoted to the development of approaches to improve the resilience of the neural network solution to the distortions caused by the shift of the spectral channels. [ DOI ]

Публикация в формате сохранить в файл сохранить в файл сохранить в файл сохранить в файл сохранить в файл сохранить в файл скрыть