| Energy spectrum measurement is an important method for studying the properties of radiation fields and the information it carries.It is widely used in fields such as nuclide identification,nuclear reaction diagnosis,and dark matter detection.Measurement targets applied in different fields have extremely high energy resolution requirements.How to improve the energy resolution of the detection system has always been the focus of attention in the field of nuclear radiation detection.The traditional methods are mainly to improve the performance of the detector and the continuous optimization of the electronic system.At this stage,as the intrinsic resolution of the detector and the improvement of electronic hardware are gradually approaching the physical bottleneck,improving the energy resolution by optimizing the data processing algorithm has gradually become an important new direction.Seeded Localized Averaging(SLA)algorithm,which uses average calculation method to process and output the signals of multiple channel sites.It gives different weights,introduces mean inequality,optimizes iteration parameters and other improvement methods.This article is based on SLA proposes a weighted average transform(WAT)algorithm based on probability density function iteration,which realizes the highlighting and extraction of signal peak information to be measured,and further improves the energy resolution of the detection system.The main research work and conclusions are as follows:(1)Perform mathematical modeling of the weighted average transformation algorithm based on the probability density function,describe and process the random input signal of the detection signal,and complete the feasibility verification of the algorithm through a complete mathematical theoretical proof.In the verification process,it is mainly proved that the points with the average value of the peak position will not drift before and after the treatment,and the symmetry after the treatment will not be changed,and the distribution FWHM after the treatment must be smaller than that before the treatment,which strengthens the randomness.The autocorrelation and time correlation between input signals.(2)A mathematical simulation program based on the weighted average transformation algorithm of the probability density function function is constructed.According to the characteristics of the current output energy spectrum,the subject considers the algorithm under typical input conditions such as Gaussian distribution,log Gaussian distribution,and multimodal distribution.Output.The simulation results of different forms not only verify the correctness and compatibility of the algorithm in this paper,but also solve the problems of trough false peaks,double peaks coincidence,trailing false peaks,etc.that occur in the seed local averaging method,and further improve the algorithm’s sensitivity to different types of peaks.Inclusiveness of distribution.(3)In the test,a set of FPGA-based energy spectrum measurement system is combined,through the use of lithium drift detector(FAST SDD)detector to measure the double-peak energy spectrum of X-ray,through digital spectrometer(MCA)and weighted average conversion Comparing the results of the algorithm,the FWHM of the two peaks is reduced from the original 16.56 and 18.0 ke V(FWHM)to 3.95 and 5.10 ke V(FWHM).The multi-peak complex energy spectrum of 137 Cs measured by lanthanum bromide,compared with the results of MCA and the weighted mean transformation algorithm,the peak half-height width is reduced from the original 6.04,7.07 and 11.09 ke V(FWHM)to3.48,4.78 and 4.52 ke V(FWHM),and has a good effect on noise reduction.The complex energy spectrum formed by the measurement of 152 Eu by the high-purity germanium detector also has the effect of improving the energy resolution.The analysis of the experimental results verifies the weighting based on the probability density function The feasibility of the average transformation algorithm under actual measurement. |
PDF Full Text Request