COMPUTER-IMPLEMENTED SPECTROMETRIC SYSTEM AND METHOD FOR BIOMASS PROCESS MONITORING AND DIAGNOSIS

Abstract

The present invention relates to a computer-implemented numerical-experimental system for monitoring and diagnosing biomass combustion using optical spectrometry. The system allows for the non-invasive estimation of flame temperature and the detection of spectral emissions associated with alkali metals, particularly sodium and potassium, during combustion processes. The invention comprises a combustion chamber configured to generate a flame from biomass, an optical acquisition system that captures the radiation emitted by the flame using optical fiber and a cosine corrector, a plurality of spectrometers capable of recording spectral intensities in different regions of the electromagnetic spectrum, and a computational module responsible for processing the spectral data.The computational module applies a two-color spectrometric method to calculate flame temperature from the ratio of spectral intensities measured at two selected wavelengths. The system also identifies spectral peaks characteristic of alkali metals, such as sodium and potassium. In a preferred embodiment, the system further incorporates a machine learning model trained on combustion spectral data, which improves temperature estimation and classifies the state of the combustion process. The invention provides a non-invasive, real-time, high-spectral-resolution diagnostic tool applicable to monitoring biomass combustion processes in experimental research and potentially in industrial settings.