Design Of Double Peak Decomposition Method And Analytical System For Low Atomic Number Element EDXRF

Energy dispersive X-ray fluorescence analysis(EDXRF)technology has the advantages of multi-element nondestructive testing,fast,portable,safe and reliable.It has been widely concerned and applied in many fields such as mining,metallurgy,geological exploration,alloy analysis and detection.The overlapping peak decomposition of X-ray fluorescence spectrum is the most important step.This paper focuses on the spectral solution method of low separation degree overlap peak separation,MCNP software was used to simulate the spectral peak data,The obtained data were processed by spectral solution method.Finally,a set of human-computer interaction system about energy dispersive X-ray fluorescence data based on MATLAB GUI platform is made.The main work of this paper is as follows:(1)Due to the existence of overlapping peaks,The qualitative and quantitative analysis of energy dispersive X fluorescence spectrum brings some difficulties.Therefore,it has been a difficult and hot spot.In this paper,by analyzing the characteristics of the traditional derivative method and spline wavelet,Combined with its characteristics,A new method to decompose the overlapping peaks of the fourth derivative combined with the three-spline wavelet is proposed.And in the simulation experiment,the low separation degree of 0.33 overlap peak was used for verification.It can be broken down,Then the feasibility of the method was verified by simulation experiments.Using K element data,The results show that the overlapping peaks can be effectively decomposed and the peak position error is within1%.Finally,the method is used to decompose the measured data.It had the desired effect,And the peak position error is within 0.5%.(2)The high cost of the detector and the use of various restrictions limit the relevant research of some researchers.In response to this problem,Firstly,this paper uses an example to illustrate the motion trajectory of particles in MCNP software.The gas detector,scintillation detector and semiconductor detector are introduced respectively.Its characteristics are discussed respectively.And comparing the characteristics of these three detectors,The SI-PIN detector was selected for MCNP software modeling.Comparing the data without Gaussian broadening and with Gaussian broadening,Finally,the new spectral solution method mentioned above is used to process the obtained data and the ideal result is achieved.(3)Because there are many mathematical tools and formulas involved in the spectral solution process,Combined with the characteristics of MATLAB GUI platform,this paper designed a set of human-computer interaction interface for processing energy dispersive X-ray fluorescence data.Composed of data processing window and result display window,The data processing window contains five function buttons: data loading,spectral line denoising,background deduction,peak finding and overlapping peak decomposition.And write the corresponding function buttons.The result display module is to display the peak position and peak boundary.Finally,the actual data is used to detect the operation of the interface.To sum up,The fourth derivative and three spline wavelet decomposition method proposed in this paper can effectively decompose the overlapping peaks with low separation degree.Improve the precision of spectrum solution;And it is feasible to use MCNP software to simulate the energy dispersive X-ray fluorescence data to solve the spectrum.Finally,the designed human-computer interaction system can run normally using the measured data.To achieve the purpose of normal use.