Predicting and optimizing the thermal-hydraulic, natural circulation, and neutronics parameters in the NuScale nuclear reactor using nanofluid as a coolant via machine learning methods through GA, PSO and HPSOGA algorithms

Abstract

This investigation studies the optimization of water-based alumina nanofluid to advance heat transfer and safety performance of the NuScale natural circulation reactor. First, comprehensive CFD and neutronic simulation is employed to design a reactor core using nanofluid coolant (0.001–10% volume fractions and 10–90 nm particle sizes). Consequently, the outlined results prove a sufficient enhancement of safety and heat transfer parameters by applying nanofluid coolant. Next, a developed Artificial Neural Network (ANN), utilizing the obtained data, predicts the thermal–hydraulic and neutronic parameters of the NuScale reactor core with Al2O3/Water nanofluid. Achieving the optimal vol% and size of nanoparticles by implementing Genetic Algorithms (GA), Particle Swarm Optimization (PSO), and Hybridized PSO-GA, based on the developed ANN results, are the main goals of this work. These optimization algorithms, which have a significant ability to attain the best solutions, also determine the optimal values of natural circulation parameters (Vmax/Vavg, Vout-Vin, and pressure drop), heat transfer coefficient, MDNBR, RPPF, and excess reactivity, for obtained vol% and size. The validation results demonstrate the efficiency of the developed ANN and these three evolutionary computation algorithms for optimization. The differences between the outcomes of implemented algorithms, focusing on how each works and affects optimal solutions in problem space, are also described. Finally, this paper compares the results of optimal design with conventional NuScale, which uses water coolant. This comparison indicates the potential of the proposed nanofluid coolant to increase thermal and safety performance.