Neural network consistent empirical physical formula construction for neutron–gamma discrimination in gamma ray tracking

• Published in: Annals of nuclear energy Vol. 51; pp. 10 - 17
• Main Authors: Yildiz, Nihat, Akkoyun, Serkan
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.01.2013
• Subjects: Empirical physical formula; Gamma ray tracking; Neural network; Neutron–gamma discrimination; Gamma ray tracking; Neutron–gamma discrimination; Neural network; Empirical physical formula
• ISSN: 0306-4549; 1873-2100
• DOI: 10.1016/j.anucene.2012.07.042

► Detector responses in neutron–gamma discrimination were estimated by neural networks. ► Novel consistent neural network empirical physical formulas (EPFs) were constructed for detector responses. ► The EPFs are of explicit mathematical functional form. ► The EPFs can be used to derive various physical functions relevant to neutron–gamma discrimination in gamma ray tracking. Gamma ray tracking is an efficient detection technique in studying exotic nuclei which lies far from beta stability line. To achieve very powerful and extraordinary resolution ability, new detectors based on gamma ray tracking are currently being developed. To reach this achievement, the neutron–gamma discrimination in these detectors is also an important task. In this paper, by suitable layered feedforward neural networks (LFNNs), we have constructed novel and consistent empirical physical formulas (EPFs) for some highly nonlinear detector counts measured in neutron–gamma discrimination. The detector counts data used in the discrimination was actually borrowed from our previous paper. The counts used here had been originally measured versus the following parameters: energy deposited in the first interaction points, difference in the incoming direction of initial gamma rays, and finally figure of merit values of the clusters determined by tracking. The LFNN–EPFs are of explicit mathematical functional form. Therefore, by various suitable operations of mathematical analysis, these LFNN–EPFs can be used to derivate further physical functions which might be potentially relevant to neutron–gamma discrimination performance of gamma ray tracking.