The Development Of Multi-Channel Gamma Spectrum Measure Software Based On Ensemble Classifier

In geology and oil and gas exploration,the deep core samples collected by drilling are of great significance to the development of deep mineral resources and oil and gas resources.The field low background gamma spectrometry can realize the qualitative identification and quantitative analysis of the radionuclides in the core samples,which can greatly improve the work efficiency compared with the traditional laboratory analysis method.This paper focuses on the hardware control and measurement requirements of the field multi-channel low-background gamma spectrometry automatic measurement system,and designs and develops the corresponding control and measurement analysis software,which has important research significance and application value.The main research results are as follows:1)The field multichannel gamma-ray spectrometry measurement control software is developed,and WLAN communication technology is adopted to realize the control of the single-machine multichannel gamma-ray spectrometry measurement and the acquisition of energy spectrum measurement data;2)Multiple machine learning models including decision tree,k-means,KNN,DBSCAN,bayes,SVM,BP neural network and convolutional neural network were studied and trained,and a nuclide recognition model based on integrated classifier was established.3)Combining with the nuclide recognition model of the integrated classifier,the traditional energy spectrum analysis algorithm was studied and optimized to realize the qualitative and quantitative analysis of the radionuclides in the samples.4)A standard nuclide database shall be established,including the relevant information of all the radionuclides performing the alpha decay,such as branch ratio,energy,half-life,etc.,so as to provide operators with the information of related nuclides as a reference while satisfying the need of the integrated classifier for nuclide recognition.