Application of a genetic algorithm to quantitative analysis of overlapped FTIR spectra

Technique for the quantitative analysis in FTIR spectrometer by Genetic Algorithm is studied without the help of more widespread methods, such as PLS and Kalman filtering. Individuals are expressed with decimal digit instead of the traditional expression-binary digit. The effect of the probability of crossover and the establishment of objective function are discussed. The method of step size mutation is used. Compared with the Kalman filtering, the Genetic Algorithm has excellent nonlinear ability of solution. The results are exciting. Mean relative error of each result is really small, generally not bigger than 1%, if the conditions are properly controlled.