The Point Reactor Kinetics Equations (PRKE) are known as a highly stiff system owing to the large time-scale difference between prompt and delayed neutrons, making the choice of numerical method critical for obtaining stable, accurate, and efficient solutions. In this study, we numerically analysed the PRKE for the Kartini reactor, TRIGA Mark II type research reactor, using its actual kinetic parameters. Four reactivity function models as a function of control rod position were examined: polynomial (from experimental data), square sine, Gaussian, and sigmoid. Jacobian matrix analysis showed that the stiffness ratio reached an order of 104, confirming that the PRKE is stiff even under undisturbed conditions. A comparison of numerical methods showed that explicit Euler and Runge-Kutta (RK4), implicit BDF, and exponential integrator (first-order Exponential Time Differencing, ETD1). RK4 method were stable only at very small-time steps ([[EQUATION]] ≤ 0.013 s and [[EQUATION]]≤ 0.018 s), whereas BDF and ETD1 remained stable up to [[EQUATION]]= 1 s. ETD1 demonstrated a computational advantage of up to 23 times faster than BDF with sufficient accuracy for Kartini transients. Validation against a step reactivity case showed agreement with exact solutions and literature, with errors below 0.13% for subcritical conditions. Among the analytical models, the Gaussian function most closely approximated the polynomial behaviour. Adding temperature feedback successfully limited uncontrolled power growth, yielding more realistic dynamics consistent with the TRIGA reactor's inherent safety features. This study confirms that ETD1 with a polynomial reactivity model provides a stable, accurate, and computationally lightweight foundation for developing the Kartini reactor's digital twin.
31 January 2026 in International Journal of Thermal Sciences
The deployment of advanced nuclear energy systems and closed fuel cycles requires improved understanding of actinide materials characteristics under electrochemical processing conditions, i.e., for spent nuclear fuel (SNF) recycling via molten-salt pyroprocessing. Meanwhile, the relationship between oxygen stoichiometry, dopant incorporation, and electronic properties in uranium and mixed-oxide fuels remains insufficiently characterized. This study addresses these challenges by developing machine learning models to predict both formation energies and band gaps of actinide materials. Using data from the Open Quantum Materials Database, this work represents the first application of ML to characterize actinide band gap evolution. The models were developed using Random Forest and Extreme Gradient Boosting (XGB) algorithms, with XGB become the best model (R2 = 0.97 for formation energy and 0.90 for band gap prediction). Feature importance analysis identified electronegativity and Mendeleev number as key descriptors governing thermodynamic stability and electronic band gap. The trained models were then applied to simulate oxygen reduction and dopant incorporation on uranium, plutonium, and mixed-oxide systems, including incorporation of fission products and noble metals, to evaluate their effects on formation energy and band gap evolution. These results demonstrate the capability of machine learning as a computational framework for the preliminary evaluation of SNF processing, providing critical insights into fuel design and electrode selection.