Proton--proton and proton--antiproton differential elastic cross sections modeling at high and ultra-high energies using a hybrid computing paradigm


Abstract: This work presents a hybrid computing technique for modeling the differential elastic cross section of both proton--proton ``pp'' and proton--antiproton ``pp(bar)'' collisions from high to ultra-high energy regions (from 13.9 GeV to 14 TeV) as a function of the center-of-mass energy ``s'' squared and four momentum transfer squared ``t''. We proposed a genetic algorithm (GA) and support vector regression (SVR) hybrid techniques to calculate and predict the ``differential elastic cross section'' of both ``pp'' and ``pp(bar)''. Our proposed GA-SVR hybrid model shows a good match to the available experimental data, as well as predicting the latest and future ``TOTEM'' experiments for differential elastic cross sections that are not utilized in the training phase. An extrapolation of the differential cross section to $\vert t\vert \to $0 gives the total cross section through the optical theorem. The model calculations and predictions for ``pp'' and ``pp(bar)'' total cross-sections over a wide collision energy ranging from ISR to LHC are given. All the results for the total cross-sections matched well with the last and the more recent measurements ($\sqrt{ s}$= 7 TeV and 8 TeV), as well as the results of other models for the future ``TOTEM'' measurements at ($\sqrt{ s}$ = 10 TeV and 14 TeV).

Keywords: Deferential elastic cross sections, proton--proton (antiproton) collision, empirical modeling, support vector regression, genetic algorithm,

Full Text: PDF