Nuclide identification in gamma spectrometry by means of a fuzzy logic based algorithm

This work deals with the development and set-up of an auxiliary software for gamma spectrometry aimed to reduce uncertainty in quantitative analysis. Its outstanding feature is to avoid that a gamma lines set, generated from different radionuclides, might be erroneously associated to another radionuclide characterised by a gamma emission spectrum which is a random combination of those lines. This is obtained by an algorithm inspired to the so-called “Fuzzy Logic”. The input of the software is the typical output of the commercial software adopted for Gamma Spectrometry (a table which reports detected peak energy, net peak area and net area uncertainty). By means of a gamma library, the software identifies the radionuclides correspondent to the detected energies and generates a new table with the following quantities: energy of the peak detected, count rate, count rate uncertainty, peak efficiency, library energy, radionuclide name and corresponding yield. Usually in this table many lines are associated to more than one radionuclide. As first hypothesis the fuzzy logic algorithm considers as effectively existing all the radionuclides reported in the table while at the end of the process an index of compatibility is assigned to every line detected. From this value is then possible to devise the acceptability degree of every line.