Abstract: Based on an HT voltage value of each data point applied to a light detector and a plurality of reference spectra measured, in measurement of a reference substance performed for a plurality of times, a weighting function deriver of a spectrum measuring device derives a relation between the HT voltage value (HTi) and a degree of dispersion ?i of a plurality of spectral values as a weighting function (?=f(HT)). Moreover, a numerical evaluator of the spectrum measuring device is configured to calculate a degree of individual coincidence between a sample spectrum and the reference spectrum, acquire the degree of dispersion ?i by applying the HT voltage value at measuring the sample spectrum to the weighting function as a weighting value, and evaluate the difference between the reference spectrum and the sample spectrum based on the degree of individual coincidence to which the weighting value is applied.

Abstract: A spectral analysis method for thermodynamically analyzing performs multivariate analysis to a plurality of measurement spectra measured under a plurality of measurement parameters, using MCR method that separates a pure spectrum from the plurality of measurement spectra, and calculates a concentration value of each pure spectrum. In this method, an equilibrium model corresponding to an equilibrium state of a sample in which three or more types of chemical species coexist is set, and at least one chemical-equilibrium equation corresponding to the equilibrium model is set. In the MCR method, a concentration curve based on the chemical-equilibrium equation is fitted to the calculated concentration values, an optimal value of a thermodynamic parameter is acquired, a new concentration value is acquired from the chemical-equilibrium equation based on the optimal value, and thus the concentration value is constrained.

Abstract: Based on an HT voltage value of each data point applied to a light detector and a plurality of reference spectra measured, in measurement of a reference substance performed for a plurality of times, a weighting function deriver of a spectrum measuring device derives a relation between the HT voltage value (HTi) and a degree of dispersion ?i of a plurality of spectral values as a weighting function (?=f(HT)). Moreover, a numerical evaluator of the spectrum measuring device is configured to calculate a degree of individual coincidence between a sample spectrum and the reference spectrum, acquire the degree of dispersion ?i by applying the HT voltage value at measuring the sample spectrum to the weighting function as a weighting value, and evaluate the difference between the reference spectrum and the sample spectrum based on the degree of individual coincidence to which the weighting value is applied.