Abstract

To ensure the reliability of calculations of the parameters of the spectra obtained by various methods, a program was created for decomposing the spectra into Gaussian components in manual and automatic modes. Manual mode is used to determine the boundary conditions necessary for determining the global extremum. Next, an automatic mode is launched that provides rapid calculations of the spectrum parameters and determines their metrological characteristics, separating the value of the useful signal from the signals of the components, background and noise. The paper shows examples of separating the chromium signal from the signals of the components located on the right and left, as well as from the background signal and taking into account the noise level, which allows improving the analysis characteristics. Optimization of the program in order to determine the boundary parameters of the search for calculated parameters and its robustness is debugged by creating data similar to the experimental data using the "folding knife" or "BUTSTRAP" method.. Measurement quality is ensured by integrating experimental data with the capabilities of the SPECTRUM program. The SPECTRUM program is used as a database for metrological support of measurements and determination of reliability, expressiveness and cost-effectiveness of measurements. In examples 11 parameters: S1, S2, S3, M1, M2, M3, σ1, σ2, σ3, A, where S2 – the area of the useful signal; S1, S3 square harmful signals; M1, M2, M3 – wavelength signals; σ1, σ2, σ3 – half-width signals; A, B – parameters of background; I – noise level. Useful S2 signals depending on the chromium concentration are the basis for determining the calibration function and determining the metrological characteristics of both the calibration function and the analysis result.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call