A method for broadening the characteristic peak of CdZnTe γ-ray energy response based on double Gaussian function

Abstract

The present invention relates to the field of gamma energy spectrum measurement and analysis, and discloses a method for broadening characteristic peaks of gamma energy response of cadmium zinc telluride based on double Gaussian functions, which solves the problem that a conventional single Gaussian function cannot show the actual characteristic peak shape in the Monte Carlo simulation calculation of the energy response function of the cadmium zinc telluride detector. The method comprises the steps of: S1, measurement and preprocessing of the gamma energy spectrum; S2, characteristic peak fitting based on a double Gaussian function model; fitting a single characteristic peak in the gamma energy spectrum using a double Gaussian function model; S3, establishing a curve of the relationship between broadening function parameters and energy; S4, determining the broadening function parameters of the characteristic energy of the simulated spectrum; S5, calculating the counting ratio distribution of each channel address in the characteristic peak region; S6, broadening of the characteristic peak in the simulated spectrum. The method can quickly calculate the characteristic peak broadening function parameters required at the characteristic energy, and conveniently and effectively realize the broadening of the energy response characteristic peak in the gamma simulation spectrum of the cadmium zinc telluride detector.

Description

Technical Field

[0001] The invention relates to the field of gamma energy spectrum measurement and analysis, in particular to a method for broadening characteristic peaks of cadmium zinc telluride gamma energy response based on double Gaussian functions. Background Art

[0002] A series of important nuclear facilities, such as research reactors, isotope production reactors, nuclear fuel production lines, nuclear waste treatment production lines, and irradiation stations, are important components of my country's nuclear technology application industry. In the process of carrying out nuclear technology applications such as medical isotope preparation, fuel element production, radioactive waste treatment, and irradiation sterilization, it is necessary to conduct on-site radiation monitoring, production line pipeline monitoring, and radioactive sample monitoring to quickly and accurately determine the type and activity of γ nuclides and ensure product quality and radiation safety. The cadmium zinc telluride detector is a new type of room-temperature semiconductor detector with an energy resolution between that of a scintillator detector and a high-purity germanium detector. It has the advantages of small size, high intrinsic detection efficiency, wide detection energy range, working at room temperature, insensitive to humidity, low price, and complete localization. It is particularly suitable for measuring the type and activity of γ nuclides in the process of nuclear technology applications.

[0003] At the monitoring site of nuclear technology application, facing the measurement objects with many types of nuclides and wide characteristic gamma ray energy distribution, the gamma energy spectrum measured by CdZnTe gamma spectrometer is complex, and the existing instruments cannot accurately distinguish the near-heavy peaks, resulting in inaccurate nuclide identification, which ultimately affects the accuracy of nuclide qualitative analysis. In order to improve the application limitations of gamma energy spectrum measurement and analysis caused by the insufficient energy resolution of CdZnTe detector, a variety of energy spectrum inversion methods are used for low-resolution gamma energy spectrum at home and abroad to achieve the resolution of near-heavy peaks in the gamma energy spectrum and nuclide identification, including energy spectrum window method, function fitting method, energy response function method, etc. Among them, the gamma energy spectrum inversion method based on energy response function establishes a direct inversion equation through the gamma energy spectrum and the energy response matrix of the detector, realizes high-resolution inversion of the incident gamma original spectrum line, and accurately realizes the resolution of characteristic peaks and nuclide identification. It is a better method.

[0004] In the inversion of the gamma spectrum based on the energy response function, the energy response function is a necessary condition for establishing the high-resolution inversion equation of the energy spectrum. The gamma spectrometers based on detectors such as high-purity germanium, sodium iodide, and lanthanum bromide have symmetrical peak shapes in the gamma spectrum they measure. In the analysis of the gamma spectrum, a conventional single Gaussian function is usually used as the peak shape function for fitting. For the acquisition of the energy response function of the cadmium zinc telluride detector, a Monte Carlo program is usually used to carry out simulation calculations. The function model for characteristic peak broadening in the Monte Carlo program is also a single Gaussian function. However, for the gamma spectrum obtained by the cadmium zinc telluride gamma spectrometer, a tail appears at the front end of the characteristic peak, and the peak shape of the characteristic peak is not symmetrical. The conventional single Gaussian function is no longer applicable. At this time, it is necessary to find an accurate characteristic peak shape function as the broadening function model of the characteristic peak in the gamma energy response function.

[0005] In summary, the existing single Gaussian function model is only suitable for the characteristic peak fitting of high-purity germanium, sodium iodide, lanthanum bromide and other detectors. The characteristic peak broadening method based on the single Gaussian function model is only suitable for their characteristic peak broadening in the calculation of the detector energy response function through Monte Carlo simulation. For the γ energy spectrum obtained by the CdZnTe detector, a tail appears at the front end of the characteristic peak, and the peak shape of the characteristic peak is not symmetrical. The conventional single Gaussian function cannot show the actual characteristic peak shape. When using the existing characteristic peak broadening method to broaden the characteristic peak of the γ energy response function of the CdZnTe detector, its practicality is limited. Summary of the invention

[0006] The technical problem to be solved by the present invention is to provide a method for broadening the characteristic peak of cadmium zinc telluride gamma energy response based on a double Gaussian function, so as to solve the problem that a conventional single Gaussian function cannot show the actual characteristic peak shape in the Monte Carlo simulation calculation of the energy response function of the cadmium zinc telluride detector.

[0007] The technical solution adopted by the present invention to solve the technical problem is: a method for broadening the characteristic peak of cadmium zinc telluride gamma energy response based on a double Gaussian function, comprising the following steps:

[0008] S1. Measurement and preprocessing of γ energy spectrum;

[0009] The environmental background spectrum is measured using a CdZnTe detector; the experimental γ spectrum containing multi-energy γ characteristic peaks is measured using a CdZnTe detector; the environmental background is deducted from the γ spectrum and the γ spectrum data is smoothed; the characteristic peak background is deducted and the data of each channel address of the characteristic peak after the background is deducted is normalized;

[0010] S2, characteristic peak fitting based on double Gaussian function model;

[0011] Using the preprocessed normalized characteristic peak data, a double Gaussian function model is used to fit a single characteristic peak in the γ energy spectrum. The double Gaussian function model is as follows:

[0012] ;

[0013] In the formula, a i (i=1,2,…,5) is the function parameter, E is the energy corresponding to each channel address in the characteristic peak area, E0 is the characteristic γ energy corresponding to the characteristic peak, and a set of function parameters is obtained for each characteristic peak;

[0014] f(a i ,E,E0) means the proportion of the counts corresponding to each channel energy E in the characteristic peak region corresponding to the characteristic energy E0 to the total counts of the characteristic peak, where the first item is the characteristic peak fidelity item and the second item is the characteristic peak front tail item.

[0015] S3, establishing a curve of the relationship between broadening function parameters and energy;

[0016] The double Gaussian function model is used to broaden the energy response characteristic peak in the γ simulation spectrum of the CdZnTe detector. The function parameter a in the double Gaussian function model is i (i=1,2,…,5), which is the broadening function parameter;

[0017] For a single broadening function parameter, according to the broadening function parameter a i (i=1,2,…,5) With the change trend of energy E, the Function is fitted to establish the broadening function parameter a i The relationship curve between (i=1,2,…,5) and energy E is as follows:

[0018] ;

[0019] ;

[0020] ;

[0021] ;

[0022] ;

[0023] Where b ij (i=1,2,…,5; j=1,2,…) i is the function parameter, E is the characteristic peak energy;

[0024] b ij Method for determining:

[0025] Each characteristic peak in S2 obtains a set of function parameters a i (E, i=1,2,…,5), multiple characteristic peaks can obtain multiple sets of function parameters: a i (E1), a i (E2), a i (E3),….

[0026] Using a1(E1), a1(E2), a1(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 1j ;

[0027] Using a2(E1), a2(E2), a2(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 2j ;

[0028] Using a3(E1), a3(E2), a3(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 3j ;

[0029] Using a4(E1), a4(E2), a4(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 4j ;

[0030] Using a5(E1), a5(E2), a5(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 5j ;

[0031] S4, determining broadening function parameters of the simulated spectrum characteristic energy;

[0032] According to the characteristic energy E0 in the simulated spectrum, the broad function parameter a obtained in step S3 is substituted i The formula of the relationship curve between (i=1,2,…,5) and energy E, from the relationship curve between function parameters and energy On the graph, find the corresponding broadening function parameter a i (i=1,2,…,5), establish the corresponding characteristic peak function:

[0033] ;

[0034] S5, calculating the count ratio distribution of each channel address in the characteristic peak area;

[0035] According to the established characteristic peak function f(a i ,E,E0), calculate the count proportion distribution corresponding to each channel energy E in the characteristic peak area, that is, the count proportion distribution F in each channel energy in the characteristic peak area;

[0036] S6, broadening of characteristic peaks in the simulated spectrum;

[0037] The Monte Carlo method is used to simulate and calculate the simulated spectrum of the CdZnTe detector to the radiation source. The counts in the unbroadened characteristic peak region in the simulated spectrum are redistributed on each channel in the characteristic peak region according to the proportional distribution F to achieve the broadening of the characteristic peak in the simulation.

[0038] Furthermore, the measurement and preprocessing of the gamma energy spectrum in step S1 includes the following steps:

[0039] S11. Use a cadmium zinc telluride detector to measure the environmental background spectrum;

[0040] S12. Based on a radioactive γ source, a cadmium zinc telluride detector is used to carry out experimental measurements to obtain an experimental γ energy spectrum containing multi-energy γ characteristic peaks;

[0041] S13, subtracting the environmental background from the experimental γ spectrum and smoothing the γ spectrum data;

[0042] S14, determining the channel energy of the left and right boundaries of the characteristic peak by observing the peak shape of the characteristic peak;

[0043] S15. Deduct the characteristic peak background by using the linear background deduction method. Calculate the γ spectrum background using the linear background method. The background calculation formula is as follows:

[0044]

[0045] Where BD is the background area, Z is the left channel address, and Spe z is the energy spectrum count corresponding to the energy of the left channel address, Y is the right channel address, Spe Y is the energy spectrum count corresponding to the energy of the right channel address;

[0046] The net peak area of ​​the characteristic peak is obtained by subtracting the background from the total peak area of ​​the characteristic peak;

[0047] S16, normalizing the data of each channel address of the characteristic peak after background deduction, that is, dividing the count at each channel address under the characteristic peak by the total count of the corresponding characteristic peak.

[0048] Furthermore, in step S13, the gamma spectrum data is smoothed by a five-point smoothing method, and the calculation formula of the five-point smoothing method is as follows:

[0049]

[0050] Where, SPE i is the energy spectrum count at the energy of the ith channel after smoothing, Spe i-2、 Spe i-1、 Spe i、 Spe i+1、 Spe i+2 They are the energy spectrum counts at the i-2, i-1, i, i+1, and i+2 channel energies in the original energy spectrum respectively.

[0051] Beneficial effects of the present invention

[0052] Compared with the prior art, the method for broadening the characteristic peak of the cadmium zinc telluride gamma energy response based on double Gaussian function of the present invention has the following advantages and beneficial effects:

[0053] 1) The double Gaussian function model composed of two Gaussian functions is used to fit a single characteristic peak in the γ spectrum, which can improve the fitting effect and better express the peak shape of the characteristic peak of the γ spectrum of CdZnTe;

[0054] 2) The relationship curve between the established function parameters and energy can quickly obtain the characteristic peak broadening function parameters required at the characteristic energy;

[0055] 3) Based on the double Gaussian function model, the characteristic peaks of the simulated spectrum are broadened, which can well show the actual situation of the front end tailing of the characteristic peaks;

[0056] In summary, the method for broadening the characteristic peak of the cadmium zinc telluride gamma energy response based on the double Gaussian function described in the present invention can solve the problem that a single Gaussian function cannot show the front-end tailing shape of the actual characteristic peak in the gamma energy spectrum of the cadmium zinc telluride detector. By establishing a relationship curve between the broadening function parameters and energy, the characteristic peak broadening function parameters required at the characteristic energy can be quickly calculated, and the broadening of the energy response characteristic peak in the gamma simulation spectrum of the cadmium zinc telluride detector can be conveniently and effectively achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 It is a flow chart of a method for broadening a characteristic peak of a cadmium zinc telluride gamma energy response based on a double Gaussian function in an embodiment of the present invention;

[0058] Figure 2 To measure the CdZnTe detector using the double Gaussian function model 235 The fitting effect of the characteristic peak of U source (185keV);

[0059] Figure 3 To measure the CdZnTe detector using the double Gaussian function model 137 The fitting effect of the characteristic peak of Cs source (662keV);

[0060] Figure 4 To measure the CdZnTe detector using the double Gaussian function model 60 The fitting effect of the characteristic peak of Co source (1173keV);

[0061] Figure 5 To measure the CdZnTe detector using the double Gaussian function model 60 The fitting effect of the characteristic peak of Co source (1332keV);

[0062] Figure 6 It is the direct simulation spectrum of the characteristic energy of 200keV~1200keV for CdZnTe detector;

[0063] Figure 7 Simulate the spectrum of characteristic peak broadening of 200keV~1200keV characteristic energy for CdZnTe detector;

[0064] Figure 8 Measured for CdZnTe detectors 152 Comparison of the broadened characteristic peaks at the characteristic energies of 779keV, 964keV, 1085keV, and 1112keV in the Eu source with the experimental characteristic peaks. DETAILED DESCRIPTION

[0065] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention.

[0066] like Figure 1 As shown, the present invention provides a method for broadening the characteristic peak of cadmium zinc telluride gamma energy response based on a double Gaussian function, comprising the following steps:

[0067] 1. Measurement and preprocessing of experimental γ-ray spectrum

[0068] 1.1 Use CdZnTe detector to measure the environmental background spectrum;

[0069] 1.2 Based on the radioactive γ source, the experimental measurement is carried out using a cadmium zinc telluride detector to obtain an experimental γ energy spectrum containing multi-energy γ characteristic peaks;

[0070] 1.3 Subtract the environmental background from the experimental γ spectrum and smooth the γ spectrum data using the five-point smoothing method. The calculation formula of the five-point smoothing method is as follows:

[0071]

[0072] Where, SPE i is the energy spectrum count at the energy of the ith channel after smoothing, Spe i-2、 Spe i-1、 Spe i、 Spe i+1、 Spe i+2 They are the energy spectrum counts at the i-2, i-1, i, i+1, and i+2 channel energies in the original energy spectrum respectively.

[0073] 1.4 Determine the channel energy of the left and right boundaries of the characteristic peak by observing the peak shape;

[0074] 1.5 The characteristic peak background is deducted by the linear background method, and the γ spectrum background is calculated by the linear background method. The background calculation formula is as follows:

[0075]

[0076] Where BD is the background area, Z is the left channel address, and Spe zis the energy spectrum count corresponding to the energy of the left channel address, Y is the right channel address, Spe Y is the energy spectrum count corresponding to the energy of the right channel address;

[0077] 1.6 After deducting the background, the data of each channel of the characteristic peak is normalized, that is, the counts at each channel under the characteristic peak are divided by the total counts of the corresponding characteristic peak.

[0078] 2. Characteristic peak fitting based on double Gaussian function model

[0079] Using the preprocessed normalized characteristic peak data, a double Gaussian function model is used to fit a single characteristic peak in the γ energy spectrum. The double Gaussian function model is as follows:

[0080] ;

[0081] In the formula, a i (i=1,2,…,5) is the function parameter, E is the energy corresponding to each channel address in the characteristic peak area, E0 is the characteristic γ energy corresponding to the characteristic peak, and a set of function parameters is obtained for each characteristic peak;

[0082] f(a i ,E,E0) means the proportion of the counts corresponding to each channel energy E in the characteristic peak region corresponding to the characteristic energy E0 to the total counts of the characteristic peak, where the first item is the characteristic peak fidelity item and the second item is the characteristic peak front tail item.

[0083] The double Gaussian function model improves the original Gaussian function. The model consists of two Gaussian functions, and a parameter is added in front of the two Gaussian functions. The double Gaussian function model combined with two Gaussian functions is used to fit a single characteristic peak in the γ spectrum, which can improve the fitting effect and better express the peak shape of the characteristic peak of the γ spectrum of CdZnTe.

[0084] 3. Establishment of the relationship curve between broadening function parameters and energy

[0085] The present invention adopts a double Gaussian function model to broaden the energy response characteristic peak in the γ simulation spectrum of the CdZnTe detector. The function parameter a in the double Gaussian function model is i (i=1,2,…,5), which is the parameter of the broadening function. For a single broadening function parameter, according to the broadening function parameter a i (i=1,2,…,5) With the change trend of energy E, the Function is fitted to establish the broadening function parameter a i The relationship curve between (i=1,2,…,5) and energy E is as follows:

[0086] ;

[0087] ;

[0088] ;

[0089] ;

[0090] ;

[0091] Where b ij (i=1,2,…,5; j=1,2,…) are function parameters, and E is the characteristic peak energy.

[0092] b ij Method for determining:

[0093] Each characteristic peak in S2 obtains a set of function parameters a i (E, i=1,2,…,5), multiple characteristic peaks can obtain multiple sets of function parameters: a i (E1), a i (E2), a i (E3), …;

[0094] Using a1(E1), a1(E2), a1(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 1j ;

[0095] Using a2(E1), a2(E2), a2(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 2j ;

[0096] Using a3(E1), a3(E2), a3(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 3j ;

[0097] Using a4(E1), a4(E2), a4(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 4j ;

[0098] Using a5(E1), a5(E2), a5(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 5j ;

[0099] The relationship curve between the function parameters and energy established can quickly obtain the characteristic peak broadening function parameters required at the characteristic energy.

[0100] 4. Determination of parameters of broadening function for simulated spectrum characteristic energy

[0101] According to the characteristic energy E0 in the simulation spectrum, the above five formulas are introduced, and the relationship curve between the function parameters and the energy is On the graph, find the corresponding broadening function parameter a i (i=1,2,…,5), establish the corresponding characteristic peak function:

[0102] ;

[0103] 5. Calculate the count ratio distribution of each channel in the characteristic peak area

[0104] According to the established characteristic peak function f(a i ,E,E0), calculate the count proportion distribution of each channel address in the characteristic peak area, that is, the count proportion distribution F of each channel address in the characteristic peak area;

[0105] 6. Broadening of characteristic peaks in simulated spectra

[0106] The Monte Carlo program is used to simulate and calculate the simulated spectrum of the radioactive source of the CdZnTe detector. The simulated spectrum is not broadened. The counts in the unbroadened characteristic peak region of the simulated spectrum are redistributed on each channel in the characteristic peak region according to the proportional distribution F to achieve the broadening of the characteristic peak in the simulation.

[0107] Example

[0108] 1. Measurement and preprocessing of experimental γ-ray spectrum

[0109] The crystal volume is 1500mm 3 CdZnTe detector for experimental gamma point source 235 U. 137 Cs, 60Co was used to carry out spectrum measurement, with 1024 channels selected, the detection distance was 20cm, and the single spectrum measurement time was 600s. The environmental background spectrum was measured using the CdZnTe detector, and the measurement time was 600s. The environmental background was deducted from the experimental γ spectrum, and the γ spectrum data was smoothed twice by the five-point smoothing method. By observing the peak shape of the characteristic peaks, the left and right boundary channel energies of the characteristic peaks of 185keV, 662keV, 1173keV, and 1332keV were determined to be (174.1keV, 196.1keV), (628.9keV, 679.3keV), (1149.5keV, 1195.5keV), and (1312.6keV, 1355.8keV), respectively. The background was subtracted by the linear background method, and the data of each channel of the characteristic peak was normalized, that is, the counts at each channel under the characteristic peak were divided by the total counts of the corresponding characteristic peak to complete the preprocessing of the data of the four different energy characteristic peaks in the experimental γ energy spectrum.

[0110] 2. Characteristic peak fitting based on double Gaussian function model

[0111] Using the preprocessed normalized characteristic peak data, a double Gaussian function model is used to fit a single characteristic peak in the γ energy spectrum. The double Gaussian function model is as follows:

[0112] ;

[0113] In the formula, a i (i=1,2,…,5) is the function parameter, E is the energy corresponding to each channel address in the characteristic peak area, E0 is the characteristic γ energy corresponding to the characteristic peak, and a set of function parameters is obtained for each characteristic peak;

[0114] f(a i ,E,E0) means the proportion of the counts corresponding to each channel energy E in the characteristic peak region corresponding to the characteristic energy E0 to the total counts of the characteristic peak, where the first item is the characteristic peak fidelity item and the second item is the characteristic peak front tail item.

[0115] The double Gaussian function model improves the original Gaussian function. The model consists of two Gaussian functions and adds a parameter in front of the two Gaussian functions. The fitting results are:

[0116] like Figure 2 As shown, the function parameters for the 185keV energy characteristic peak are: a1=0.127095581, a2=2.511040337, a3=1.17689901, a4=4.177910866, a5=186.2258016;

[0117] like Figure 3As shown, the function parameters for the 662keV energy characteristic peak are: a1=1.046228588, a2=5.107053158, a3=0.235429912, a4=5.250747509, a5=652.208758;

[0118] like Figure 4 As shown, the function parameters for the 1173keV energy characteristic peak are: a1=1.378381847, a2=8.857965659, a3=0.021227545, a4=2.431675825, a5=1166.626045;

[0119] like Figure 5 As shown, the function parameters for the 1332keV energy characteristic peak are: a1=1.421634439, a2=9.11728603, a3=0.00527545, a4=1.031675825, a5=1332.626045;

[0120] The double Gaussian function model is used to fit the characteristic peak data. From the characteristic peak fitting results at the above four characteristic peak energies, it can be seen that the fitting effect is good.

[0121] 3. Establishment of the relationship curve between broadening function parameters and energy

[0122] The double Gaussian function model is used to broaden the energy response characteristic peak in the γ simulation spectrum of the CdZnTe detector. The function parameter a in the double Gaussian function model is i (i=1,2,…,5), which is the broadening function parameter. For a single set of broadening function parameters, according to the broadening function parameter a i (i=1,2,…,5) With the change trend of energy E, the Function is fitted to establish the broadening function parameter a i The relationship curve between (i=1,2,…,5) and energy E is as follows:

[0123] ;

[0124] ;

[0125] ;

[0126] ;

[0127] ;

[0128] 4. Determination of parameters of broadening function for simulated spectrum characteristic energy

[0129] A Monte Carlo model was established based on the CdZnTe detector to simulate and calculate the unbroadened simulated spectra of the detector measuring various characteristic energies. The characteristic energies E0 in the simulated spectra were 200keV, 400keV, 600keV, 800keV, 1000keV, and 1200keV, respectively. The unbroadened simulated spectra are as follows: Figure 6 As shown. Substitute these energies into the five formulas in step S3 above, and the relationship curve between the function parameters and the energy On the graph, find the corresponding broadening function parameters and establish the corresponding characteristic peak function.

[0130] For E0=200keV, the characteristic peak broadening function parameters are: a1=0.195, a2=2.513, a3=1.109, a4=4.326, a5=195.360;

[0131] For E0=200keV, the characteristic peak broadening function is:

[0132] ;

[0133] For E0=400keV, the characteristic peak broadening function parameters are: a1=0.736, a2=3.390, a3=0.556, a4=5.334, a5=393.360;

[0134] For E0=400keV, the characteristic peak broadening function is:

[0135]

[0136] For E0=600keV, the characteristic peak broadening function parameters are: a1=0.998, a2=4.714, a3=0.290, a4=5.363, a5=591.360;

[0137] For E0=600keV, the characteristic peak broadening function is:

[0138]

[0139] For E0=800keV, the characteristic peak broadening function parameters are: a1=1.164, a2=6.278, a3=0.143, a4=4.788, a5=789.360;

[0140] For E0=800keV, the characteristic peak broadening function is:

[0141] ;

[0142] For E0=1000keV, the characteristic peak broadening function parameters are: a1=1.284, a2=7.793, a3=0.0596, a4=3.726, a5=987.360;

[0143] For E0=1000keV, the characteristic peak broadening function is:

[0144] ;

[0145] For E0=1200keV, the characteristic peak broadening function parameters are: a1=1.376, a2=8.887, a3=0.0173, a4=2.228, a5=1185.360;

[0146] For E0=1200keV, the characteristic peak broadening function is:

[0147] ;

[0148] 5. Calculate the count ratio distribution of each channel in the characteristic peak area

[0149] According to the characteristic peak function f(E, E0) established in step S4, the count proportion distribution of each channel address in the energy characteristic peak region of 200keV, 400keV, 600keV, 800keV, 1000keV, and 1200keV is calculated, that is, the count proportion distribution of each channel address in the characteristic peak region is F(200keV), F(400keV), F(600keV), F(800keV), F(1000keV), and F(1200keV);

[0150] 6. Broadening of characteristic peaks in simulated spectra

[0151] The counts in the unbroadened characteristic peak region at the characteristic energies of 200keV, 400keV, 600keV, 800keV, 1000keV, and 1200keV in the simulated spectrum are redistributed on each channel in the characteristic peak region according to the proportional distribution F(200keV), F(400keV), F(600keV), F(800keV), F(1000keV), and F(1200keV), respectively, to achieve the broadening of the characteristic peaks in the simulation, as shown in FIG. Figure 7 shown.

[0152] Finally, the CdZnTe detector was used to 152 Eu source was used to conduct γ energy spectrum measurement experiment. There were four characteristic peaks with energies of 779keV, 964keV, 1085keV and 1112keV in the energy spectrum. The characteristic peaks at the characteristic energies were broadened by using the invention. The energy spectrum after the characteristic peaks were broadened was compared with the original experimental spectrum. Figure 8 As shown, the results show that at the characteristic energy positions of 779keV, 964keV, 1085keV, and 1112keV, it was found through observation that the peak shapes of the two are relatively consistent.

[0153] According to the broadening of the 200keV-1200keV characteristic peak in the simulated spectrum and 152 The broadening results of the characteristic peaks of 779keV, 964keV, 1085keV and 1112keV in the Eu source experimental spectrum illustrate the accuracy and feasibility of the present invention.

Claims

1. A method for broadening the characteristic peak of cadmium zinc telluride gamma energy response based on double Gaussian function, characterized in that: The following steps are involved: S1. Measurement and preprocessing of γ energy spectrum; The environmental background spectrum is measured using a CdZnTe detector; the experimental γ spectrum containing multi-energy γ characteristic peaks is measured using a CdZnTe detector; the environmental background is deducted from the γ spectrum and the γ spectrum data is smoothed; the characteristic peak background is deducted and the data of each channel address of the characteristic peak after the background is deducted is normalized; S2, characteristic peak fitting based on double Gaussian function model; Using the preprocessed normalized characteristic peak data, a double Gaussian function model is used to fit a single characteristic peak in the γ energy spectrum. The double Gaussian function model is as follows: ； In the formula, a i (i=1,2,…,5) is the function parameter, E is the energy corresponding to each channel address in the characteristic peak area, E0 is the characteristic γ energy corresponding to the characteristic peak, and a set of function parameters is obtained for each characteristic peak; f(a i ,E,E0) means the proportion of the counts corresponding to each channel energy E in the characteristic peak region corresponding to the characteristic energy E0 to the total counts of the characteristic peak, where the first item is the characteristic peak fidelity item and the second item is the characteristic peak front tail item; S3, establishing a curve of the relationship between broadening function parameters and energy; The double Gaussian function model is used to broaden the energy response characteristic peak in the γ simulation spectrum of the CdZnTe detector. The function parameter a in the double Gaussian function model is i (i=1,2,…,5), which is the broadening function parameter; For a single broadening function parameter, according to the broadening function parameter a i (i=1,2,…,5) With the change trend of energy E, the Function is fitted to establish the broadening function parameter a i The relationship curve between (i=1,2,…,5) and energy E is as follows: ； ； ； ； ； Where b ij (i=1,2,…,5; j=1,2,…) are function parameters, E is the characteristic peak energy; The b ij The method to determine is: In step S2, each characteristic peak obtains a set of function parameters a i (E, i=1,2,…,5), multiple characteristic peaks can obtain multiple sets of function parameters: a i (E1), a i (E2), a i (E3), …; Using a1(E1), a1(E2), a1(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 1j ; Using a2(E1), a2(E2), a2(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 2j ; Using a3(E1), a3(E2), a3(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 3j ; Using a4(E1), a4(E2), a4(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 4j ; Using a5(E1), a5(E2), a5(E3), ..., To fit the function model, we use Matlab to perform regression fitting to obtain b 5j ; S4, determining broadening function parameters of the simulated spectrum characteristic energy; According to the characteristic energy E0 in the simulated spectrum, the broadening function parameter a obtained in step S3 is substituted i The formula of the relationship curve between (i=1,2,…,5) and energy E, from the relationship curve between function parameters and energy On the graph, find the corresponding broadening function parameter a i (i=1,2,…,5), establish the corresponding characteristic peak function: ； S5, calculating the count ratio distribution of each channel address in the characteristic peak area; According to the established characteristic peak function f(a i ,E,E0), calculate the count proportion distribution of each channel address in the characteristic peak area, that is, the count proportion distribution F of each channel address in the characteristic peak area; S6, broadening of characteristic peaks in the simulated spectrum; The Monte Carlo method is used to simulate and calculate the simulated spectrum of the CdZnTe detector to the radiation source. The counts in the unbroadened characteristic peak region in the simulated spectrum are redistributed on each channel in the characteristic peak region according to the proportional distribution F to achieve the broadening of the characteristic peak in the simulation.

2. The method for broadening the characteristic peak of CdZnTe gamma energy response based on double Gaussian function according to claim 1, characterized in that: The measurement and preprocessing of the gamma energy spectrum in step S1 includes the following steps: S11. Use a cadmium zinc telluride detector to measure the environmental background spectrum; S12. Based on a radioactive γ source, a cadmium zinc telluride detector is used to carry out experimental measurements to obtain an experimental γ energy spectrum containing multi-energy γ characteristic peaks; S13, subtracting the environmental background from the experimental γ spectrum and smoothing the γ spectrum data; S14, determining the channel energy of the left and right boundaries of the characteristic peak by observing the peak shape of the characteristic peak; S15. Deduct the characteristic peak background by using the linear background deduction method. Calculate the γ spectrum background using the linear background method. The background calculation formula is as follows: ； Where BD is the background area, Z is the left channel address, and Spe z is the energy spectrum count corresponding to the energy of the left channel address, Y is the right channel address, Spe Y is the energy spectrum count corresponding to the energy of the right channel address; The net peak area of ​​the characteristic peak is obtained by subtracting the background from the total peak area of ​​the characteristic peak; S16, normalizing the data of each channel address of the characteristic peak after background deduction, that is, dividing the count at each channel address under the characteristic peak by the total count of the corresponding characteristic peak.

3. The method for broadening the characteristic peak of the cadmium zinc telluride gamma energy response based on double Gaussian function according to claim 2, characterized in that: In step S13, the gamma spectrum data is smoothed by a five-point smoothing method, and the calculation formula of the five-point smoothing method is as follows: ； Where, SPE i is the energy spectrum count at the energy of the ith channel after smoothing, Spe i-2、 Spe i-1、 Spe i、 Spe i+1、 Spe i+2 They are the energy spectrum counts at the i-2, i-1, i, i+1, and i+2 channel energies in the original energy spectrum respectively.

Images