Loading...

Simulation and Design of a One-Dimensional Position-Sensitive Scintillator detector using GEANT4 Monte Carlo Code

Amtaeh, Abolfazl | 2025

0 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 58179 (46)
  4. University: Sharif University of Technology
  5. Department: Energy Engineering
  6. Advisor(s): Hosseini, Abolfazl; Kochakpour, Javad
  7. Abstract:
  8. Precise prediction of radiation interaction position in detectors plays a crucial role in medical and industrial imaging systems. Position-sensitive detectors based on scintillators are widely used in industrial tomography and surface scanning applications such as cargo scanners. Since materials and objects in an industrial application have large dimensions and high densities for scanning these objects, we need a large detector and a high-energy gamma source or a dense detector to detect the photon transmitted from the object. Inorganic scintillator detectors with dense materials such as NaI(Tl) are widely used in industry. However, for large objects, several detectors must be used, each detector system requiring complex electronic hardware. Such detection systems are very time-consuming and expensive; therefore, for scanning a large object, a unique detector is needed as a position-sensitive detector with low cost and fast response. Organic scintillators such as plastic scintillators can be a good choice for accurate event timing and maximizing the count rate of gamma rays due to the fast response time (~2.5 ns) and short decay time. In this study, the position of gamma radiation was precisely predicted in a plastic rod scintillator using the attenuation technique and a multilayer perceptron (MLP) neural network. The plastic scintillator used in this study was BC404. Also, in order to compare with an inorganic scintillator, NaI(Tl) was selected as the second scintillator for simulation and prediction of the beam interaction location. Mylar was used as a reflector on the detector surface and quartz was used for the PMT window. The simulated detector system consists of a rod scintillator with a length of 150 cm and a diameter of 5 cm and two PMTs connected to both ends of this rod. The gamma radiation source is ⁶⁰Co. Our research shows that with the artificial neural network method, the position of radiation interaction in plastic scintillator and inorganic scintillator was estimated with a mean absolute error (MAE) of 0.005879 and 0.03723, respectively, a mean relative error percentage (MRE) of 0.017946% and 0.1269%, respectively, and a correlation coefficient of 1 and 0.9999, respectively. Also, various parameters such as source energy, number of PMTs, position calculation method, and detector length were investigated in obtaining the source position. The ANN method is a more suitable method than the NLR method. In the single PMT mode, less error is created, but the method is time-consuming. The cesium source can perform better in obtaining the source position and create less error. Finally, a BC400 plastic scintillator detector with dimensions of 50 cm in length and 5 cm in diameter, along with an electronic system, was implemented in the laboratory of the Atomic Energy Organization. Using the neural network, the results of cesium source location estimation were estimated with a mean absolute error of 0.11748, a relative error of 0.002836, and a correlation coefficient of 0.999976. Also, the investigation of changes in the energy spectrum was examined by changing the source location, which was consistent with the simulation results
  9. Keywords:
  10. Position Sensitive Detector ; Plastic Scintillator ; GEANT4 Toolkit ; Artificial Neural Network ; Multi-Layer Perceptron (MLP)

 Digital Object List

 Bookmark

No TOC