UNIVERSITY OF BUCHAREST
FACULTY OF PHYSICS

Guest
2024-11-23 17:39
 HOME     CONFERENCES     SEARCH            LOGIN     NEW USER     IMAGES   


Conference: Bucharest University Faculty of Physics 2023 Meeting


Section: Nuclear and Elementary Particles Physics


Title:
Computational Characterization of Coaxial HPGe detectors using Monte Carlo Simulation and Nonlinear Least Squares Optimization


Authors:
D. GURAU, D. STANGA, L. DONE, O. SIMA, E. IONESCU


Affiliation:
National Institute of R&D for Physics and Nuclear Engineering-Horia Hulubei (IFIN-HH)-Romania


E-mail
doru@nipne.ro


Keywords:
gamma spectrometry, Monte Carlo simulation, computer model, model calibration


Abstract:
Monte Carlo (MC) simulation models, implemented as computer codes, are widely used for computing the full-energy peak (FEP) efficiency of gamma spectrometry (GS) systems because the relative method of measurement has severe restrictions. The input variables of these computer models consist of model variables and model parameters. The geometric model of the detector is the key component of MC simulation models because it defines the parameters on which the FEP efficiency of the detector depends. Because computer models are always imperfect, there always exists a discrepancy (discrepancy function) between the true and computed values of the FEP efficiency. The goal of the model calibration is to estimate the optimal values of the model parameters and the discrepancy function. The calibration of MC simulation models for computing the FEP efficiency is known as detector characterization. MC simulation codes are computationally expensive and this can be alleviated via the use of surrogate models. A nonlinear surrogate model was developed, which well approximates the FEP efficiency provided by MC simulation codes using grid-based interpolation. A simple methodology for computational characterization of coaxial HPGe detectors using GESPECOR software and nonlinear least squares is described and applied in practice using a GS system equipped with a Canberra p-type coaxial HPGe detector model GC3018. The geometrical model of the detector contains fifteen parameters but many of them are known accurately. Consequently, five parameters (crystal radius and length, dead-layer face and side and distance holder-endcap) were optimized using the surrogate model mentioned above, lsqnonlin function from Matlab software and a set of reference values of the FEP efficiency. The optimal model for the p-type HPGe detector GC3018 was verified in the energy range of 60-2000 keV. The discrepancies between measured and computed values were smaller than 4 % for point sources and 3 % for cylindrical sources.


References:

1. Kennedy M. C. and O’Hagan, A, 2001. Journal of the Royal Statistical Society B, 63.

2. Tuo R. and Wu C.F.J., 2015. The Annals of Statistics, 45.

3. Stanga D. and Gurau D., 2021. Applied Radiation and Isotopes, 172, 109689.



Acknowledgement:
This work was supported by CORE/NUCLEU Program within the National Research Development and Innovation Plan 2022-2027, carried out with the support of Management Center of Research, Development and Innovation (CMCDI), project code PN 23-21.