Support vector machine-based X-ray fluorescence spectrum analysis method and support vector machine-based X-ray fluorescence spectrum analysis device

Abstract

The invention provides a support vector machine-based X-ray fluorescence spectrum analysis method and a support vector machine-based X-ray fluorescence spectrum analysis device. The method comprises the following steps: collecting an X-ray fluorescence spectrum of a training set sample and extracting a plurality of data point intensity values of the fluorescence spectrum; measuring the contents of target elements in the training set sample; building a prediction model of the support vector machine by taking the intensity values as input layer data and taking the contents of the target elements as output layer data; training the prediction model by using a genetic algorithm to determine parameters of a supporting vector machine model of X-ray fluorescence spectrum analysis target elements; collecting a plurality of data point intensity values of the X-ray fluorescence spectrum of a prediction set sample and taking the intensity values as input layer data of the supporting vector machine model of the X-ray fluorescence spectrum analysis target elements so as obtain the contents of the corresponding target elements of the prediction set sample. By virtue of the method, the model parameters can be effectively obtained, so that a good prediction effect can be achieved.

Description

technical field [0001] The invention relates to the technical field of X-ray spectrum detection, in particular to an X-ray fluorescence spectrum analysis method and device based on a support vector machine. Background technique [0002] X-ray Fluorescence Spectroscopy (XPF for short) uses X-rays to excite the outer electrons of elements, uses a spectrometer to obtain the fluorescence spectrum during electronic transitions, and conducts qualitative and quantitative analysis based on the energy and intensity of the spectral lines. [0003] X-ray fluorescence spectroscopy is a kind of atomic emission spectroscopy, which is widely used in element determination. It can measure a wide range of elements, including all elements with Z≥3 (Li) in the periodic table, and can measure a wide range of concentrations. It has the characteristics of fast and accurate, simple operation, and can simultaneously measure multiple elements. It is suitable for multiple Determination of various typ...

AI Technical Summary

Problems solved by technology

[0004] Due to the variety of soil types, complex properties, and the constraints of soil matrix effects and self-absorption effects, there is no universal standard curve or model established in the prior art, and there is no reliable XPF method for detecting heavy metals in farmland soil. The detection accuracy is not high

Images