Exergy analysis of a sustainable system with a fourth-generation nuclear power plant based on radial basis function kernel support vector regression method

Abstract

The present study proposes how to efficiently transform a landlocked city into a sustainable city by integrating a 250 MW Modular Helium Reactor with a Cu-Cl thermochemical cycle to produce low-carbon hydrogen. The system proposed for this aim consists of a three-step Cu-Cl cycle, an Organic Rankine Cycle (ORC), a gas turbine Modular Helium Reactor, and subcomponents for heating residential buildings and greenhouses. This system is analyzed thermodynamically, and some parametric studies are conducted to determine the key parameters affecting the hydrogen production rate, overall system energy and exergy efficiencies. The thermodynamic analysis is conducted using the SVR-RBF-ABC machine learning algorithm, which provides a fundamental understanding of the system behavior, while machine learning facilitates data-driven optimization and adaptation to changing conditions. The algorithm performs the simulation by randomly selecting 10 data points from the simulation dataset. As a result of the simulation, the hydrogen production capacity is 1800 kg/h, and the total exergy and energy efficiencies are found to be 55% and 57%, respectively. The statistical results regarding the exergy efficiency change with the SVR-RBF-ABC method show that the Mean Absolute Error (MAE), Mean Square Error (MSE) and coefficient of determination (R2) are 0.0347, 0.0016 and 0.9987, respectively.