NEURO-FUZZY MODELING OF SPECTROSCOPIC DATA .A. MODELING OF DYE SOLUTIONS

The applicability of an adaptive network-based fuzzy inference system for modelling of spectroscopic absorbance data in the visible region h as been studied. Some basic principles of neuro-fuzzy modeling are dis cussed. The samples were of a two-component dye solution over a concen tration range of 0-900 mg/l for the one component (red) while the conc entration of the other component (yellow) was kept constant. The model led system had two inputs (wavelength and concentration) with the abso rbance values as output. Input/output data sets were computed. General ised bell-shaped membership functions were used for the inputs.