| Nuclear energy spectrum measurement technology as an important means of information acquisition,with high precision,high sensitivity.Digital nuclear spectrometer as an important tool for nuclear energy spectrum measurement,performance has been much concerned by researchers in the field.Due to the characteristics of the digital nuclear spectrometer itself and the nuclear pulse signal itself,the energy resolution of the digital nuclear spectrometer has always been the focus of the research on digital nuclear spectrometer.In this paper,the main factors that affect the energy resolution of digital nuclear spectrometer are listed.Based on these factors,the idea of improving the error of digital pulse signal is proposed.After comparing the classical Kalman filter with the robust H∞ filtering,the robust H∞ filtering algorithm is used to realize the filtering and baseline recovery of the digital pulse signal.Under the premise of significantly improving the accuracy of pulse baseline recovery,the classical Kalman filter algorithm is neglected to rely too much on the previous time estimate,and the result of processing is not ideal when the noise is unstable.In the process of nuclear pulse signal formation,this paper uses the Z transform method,and on the basis of studying the existing algorithms,the improved algorithm is proposed to improve the influence of the existing trapezoidal forming algorithm on the forming result in the error accumulation process.The energy resolution of the spectrometer is improved.In this paper,a set of digital nuclear spectrometer system is designed and implemented.The system consists of basic detector system,digital signal acquisition system and digital core pulse signal processing system.It can realize the data acquisition and forming of pulse signal,Function.In the event of a nuclear accident,it is possible to use a digital nuclear spectrometer to measure the type of nuclear radiation and the energy,time and other information of the nuclear accident site. |