| Summary: | BackgroundThe measurement and analysis of γ ray spectrum are important means of qualitative identification and quantitative analysis of radionuclides. The analysis of overlapping peaks in complex spectra has always been a difficult problem. The traditional fitting method has greater uncertainty when processing data with high complexity and high statistical fluctuations.PurposeThis study aims to propose an improved method to achieve more accurate overlapping peak analysis.MethodsFirst of all, an overlapping peak analysis method based on constrained nonlinear optimization was proposed by using system calibration data, hence the peak width constraint was added to the conventional nonlinear least squares fitting of overlapping peaks, making the results more physically consistent. Then a simple iterative method was designed to solve the constrained optimization problem by transforming the constraints into real-time updates of peak width and weight coefficients, and its performance was investigated by simulated data under different statistical fluctuations, peak distance and peak area ratio. Finally, the method was verified by the spectra of 152Eu and 133Ba point sources measured by LaBr3(Ce) and NaI(Tl) spectrometers.ResultsThe results show that the errors of peak parameters such as peak position, full width at half maximum (FWHM) and peak area obtained by the proposed method are significantly lower than those of the unconstrained method. For spectra with high statistical fluctuations, the relative error of the analytic peak area can be reduced by more than 50%.ConclusionsMaking full use of known information such as FWHM calibration can constrain the results of nonlinear fitting in a more accurate range, effectively improving the accuracy of overlapping peak analysis. This method is suitable for complex spectrum analysis.
|
|---|