Research On Nuclear Spectrum Simulation Method Based On C# And DSP

The nuclear science and technology researches greatly boosted the development of the human and provided convenience for human in all aspects.However,nuclear pollution caused by nuclear weapon tests and nuclear power plant leakage has also brought various degrees of environmental and physical health impacts and injuries.Strengthening nuclear technology research and nuclear radiation monitoring and reducing nuclear pollution have become the focus and difficulty for nuclear workers and the entire human research.Among them,nuclear signals are used as a research basis and are used extensively in nuclear signal processing methods and techniques.However,the original nuclear signal is generated by radioactive sources,and there are limitations in the time and place and the life of the detector.The flexibility of the nuclear signal and the flexibility of the parameters are not strong,and it is hard to meet the requirements of flexible,diverse and reproducible signals.Moreover,the highly radioactive sources with strong radiation have unavoidable physical harm to the staff engaged in nuclear signal acquisition and the implementation of nuclear measurement in actual scientific research has been challenged.Through the monitoring of gamma rays released by radioactive substances in the radiation environment,we can know the composition of radioactive substances,we can judge the content and activity and this can reduce environmental pollution and physical injuries.However,the measured spectral signals are often seriously overlapped due to the environmental complexity,interference ray and detector energy resolution in the actual measurement.NaI?Tl?detector is widely used due to its advantages of high detection efficiency,convenient maintenance and moderate cost as a commonly used gamma ray detector,but the ability of distinguish low-energy overlapping spectra is not strong,making the decomposition of overlapping spectra become a problem.At the same time,in order to achieve more accurate and timely energy spectrum measurement analysis and abnormal condition processing in the radiation monitoring process,and make it suitable for real-time measurement and analysis in the field,it is often necessary to measure nuclear energy spectrum in real-time,accurate inversion.Based on this background,this paper is supported by the project of the Sichuan Science and Technology Support Program“Study of Random Characteristics of Nuclear Radiation Signals”?2014GZ0020?.Combining the previous studies on imitation nuclear signal,nuclear spectrum simulation,and decomposition of overlapping peaks,relying on C#and DSP,a set of nuclear spectrum simulation system was built.At the same time,with the aid of MATLAB analysis tools to signal simulation as the main line,the imitation nuclear signal generation,with random properties index S-K Gaussian pulse forming and trapezoidal shaping filter,the field of arbitrary nuclear energy spectrum in the process of radiation monitoring real-time realistic simulation,the decomposition of the overlapping spectrum and energy spectrum drift are discussed in all aspects of research.In this paper,the main work and achievements have the following several aspects:1.By means of MATLAB simulation platform,different attenuation constants and different amplitudes are simulated for the exponential nuclear pulse signal before shaping.The S-K Gaussian and trapezoidal shaping algorithms were discussed,and the exponential pulse Gaussian shaped waveforms with different adjustment parameters K were simulated.And complete the trapezoidal shaping or filtering simulation of no-noise and noisy exponential pulses.2.The statistical properties of energy spectrum and its mathematical model are discussed and the basic principles of particle swarm optimization are described.The original overlapping peaks are simulated on MATLAB.The relation between the particle swarm algorithm and the overlapping peaks decomposition is analyzed.The fitness function of the particle swarm decomposing overlapping peaks is determined and the two-peak overlapping spectrum,three-peak overlapping spectrum and four-peak overlapping spectrum are decomposed.The ability of the algorithm to decompose overlapping peaks is completed.The new overlapped peaks are formed by a certain proportion of Gaussian peaks which are decomposed and the dynamic simulation of the formation process is made by adding sampling method.3.The mathematical model of full spectrum is proposed and the original composite spectra of radium(²²⁶Ra),thorium(²³²Th)and cesium(¹³⁷Cs)are completed in MATLAB.The evaluation function of particle swarm optimization is given for the particle of the whole spectrum decomposition.The full-spectrum decomposition involving three single-spectral weight unknowns is achieved.The three single spectra are combined to form a new hybrid spectrum and the formation process is dynamically simulated.4.The drift of energy spectrum is discussed,the drift multiple is set and the simulation of three kinds of nuclear energy spectrum drift of ²²⁶Ra,²³²Th and ¹³⁷Cs is completed in MATLAB.The original drift mixture spectra is obtained by using the drift energy spectrum in a certain proportion.The particle swarm algorithm is used to decompose the original drift mixed spectrum,involving six unknowns.The six unknowns are related to the weights and drift multiples of the three single spectra.5.A set of nuclear energy spectrum simulation system based on C#and DSP is constructed.The hardware setup,the design of the UI interface on C#,and the implementation and validation of the function algorithm were completed.The exponential signal simulation of different attenuation constants and different amplitudes,Gaussian shaping of different forming parameters are realized in C#.Linear congruential method and discrete direct sampling method were used to realize the dynamic inversion of ²²⁶Ra,²³²Th,and ¹³⁷Cs three kinds of nuclide energy spectrums and their mixed spectra.The energy spectrum simulation and error calculation and analysis were completed on the software environment CCS of DSP.