This paper presents a new procedure to optimize the geometric parameters of a n-type coaxial HPGe detector. It is based on a statistical technique called “Design of Experiments” (DoE). This technique aims to identify the most influential parameters and to determine the optimal configuration. In this work, The effects of each parameter on the detector responses have been investigated by a fractional factorial design. Only the most influential factors contributing to the detector response have been selected. Precise modeling of these factors was then performed using a full factorial design. Based on the results obtained from this design, the full energy peak efficiencies according to the geometric parameters were modeled by a multiple-linear regression. These models have been statistically validated by analysis of variance (ANOVA). The optimal combination of the geometric parameters has been identified using the desirability function approach, which is a useful tool to optimize multi-response problems. A verification test was performed to validate the results obtained. It was observed that the relative deviation found between experimental and simulated values was less than 5%.

本文提出了一种优化 n 型同轴 HPGe 探测器几何参数的新程序。它基于一种称为“实验设计”(DoE) 的统计技术。该技术旨在识别最有影响的参数并确定最佳配置。在这项工作中，通过部分因子设计研究了每个参数对探测器响应的影响。仅选择了对探测器响应有最大影响的因素。然后使用全因子设计对这些因素进行精确建模。基于从该设计获得的结果，根据几何参数的全能量峰值效率通过多元线性回归建模。这些模型已经通过方差分析 (ANOVA) 进行了统计验证。使用意愿函数方法确定了几何参数的最佳组合，这是优化多响应问题的有用工具。进行了验证测试以验证获得的结果。据观察，实验值和模拟值之间的相对偏差小于 5%。