Trace element XRF determination method based on iterative discrete wavelet background deduction

Abstract

The invention discloses a trace element XRF determination method based on iterative discrete wavelet background deduction, and the method comprises the following steps of: performing L-layer one-dimensional discrete wavelet decomposition on an original detection spectral line signal of a to-be-detected sample to obtain a primary low-frequency approximation coefficient of each layer, and selecting an optimal decomposition layer and a corresponding primary low-frequency approximation coefficient av; carrying out iterative discrete wavelet decomposition on the primary low-frequency approximation coefficient av, stopping iteration when the difference value of two continuous N adjacent iteration results is smaller than preset precision, and taking the last iteration result as an approximate background signal, thus obtaining a background-deducted signal; respectively calculating the Compton peak scattering intensity and the characteristic X-ray fluorescence intensity of a target element, and carrying out approximate treatment to obtain a quantitative analysis value of the target element. The method provided by the invention can effectively avoid the influence of peak offset and original spectrogram signal peak area, improve the quantitative detection precision of trace elements, improve the detection signal-to-noise ratio by three times or more, and reduce the detection limit of trace elements.

Description

technical field [0001] The invention relates to the technical improvement field of analysis and processing of spectral data, in particular to an XRF determination method of trace elements based on iterative discrete wavelet background subtraction. Background technique [0002] Energy dispersive X-ray fluorescence spectrometry detection technology can meet various needs of trace element detection due to its almost no need for sample pretreatment, no pollution, and fast and convenient analysis capabilities. Spectrometer detection spectrum contains background information, characteristic X-rays, escape peaks , Overlay peak and other information. Affected by many factors such as the instrument system, the detection environment, and the nature of the sample itself, there are always different degrees of background interference in the spectral signal. In addition, the parts containing metal elements on each instrument will affect the spectrum during the air test, such as the metal ...

AI Technical Summary

Problems solved by technology

Using the peak stripping method to subtract the background can better maintain the position of the original peak, but its calculation process will have a certain impact on the peak area of ​​the original spectral line, and it largely depends on the shape of the original spectral line

Polynomial fitting often distorts the sample spectrum due to overfitting

The classic Fourier transform can reflect the overall connotation of the signal, but its expression is often not intuitive enough, and noise will complicate the spectrogram

The traditional background subtraction method will lead to peak shift and affect the peak area of ​​the original spectrum signal, and the detection accuracy is not high. For this, the present invention proposes a trace element XRF (X Ray Fluorescence, XRF) based on iterative discrete wavelet background subtraction X-ray fluorescence spectrometry) determination method

Images