What question did this study set out to answer?

The aim is to present two semi-empirical mass models that achieve competitive accuracy with fewer parameters.

January 6, 20260 citationsOpen Access

Minimal Semi-Empirical Nuclear Mass Models: P1 (1 - parameter) and P5 (5 - parameter)

Key Points

The aim is to present two semi-empirical mass models that achieve competitive accuracy with fewer parameters.
Developed P1 Model with 1 parameter and P5 Model with 5 optimized parameters.
Evaluated performance on the AME2020 dataset using RMSE and MAE metrics.
Unified volume-surface coefficients were employed to reduce parametric complexity.
P1 Model achieved RMSE of 0.38158 MeV and MAE of 0.25461 MeV.
P5 Model achieved RMSE of 0.19931 MeV and MAE of 0.06685 MeV.
Performance improves with mass number, with RMSE of 0.02169 MeV for A>150 in P5.

Abstract

We present two simplified semi-empirical mass formulas achieving competitive accuracy on the AME2020 experimental dataset: P1 Model: 0. 38158 MeV RMSE 0. 25461 MeV MAE - with 1 unified parameterP5 Model: 0. 19931 MeV RMSE 0. 06685 MeV MAE - with 5 optimized parameters Both models employ unified volume-surface coefficients (aₛ = aᵥ), reducing parametric complexity while maintaining accuracy competitive with 30-40 parameter models. Performance systematically improves with mass number (RMSE = 0. 02169 MeV for A>150 - P5) This repository contains: Complete Python implementations AME2020 nuclear binding energy data (nuclearₒutput. csv) PDF descriptions Short ontological description Readme. md Results demonstrate that physically-motivated minimal parametrization can match complex model accuracy while preserving interpretability.

问 AI

Bookmark

View Full Paper