Study on multi-objective optimization method for shielding structure of nuclear reactors with complex constraints

The design of the nuclear reactor shielding structure is a typical multi-objective global optimization problem, and there are complex constraint relationships among various sub-objectives and shielding structure parameters.[Purpose] This study aims to solve the multi-objective optimization problem with complex constraints of the nuclear reactor shielding structure.[Methods] This paper respectively studied the Adaptive Weight Genetic Algorithm, the Multi-Objective Optimization Algorithm Based on Objective Decomposition (MOEA/D algorithm), as well as the Non-dominated Sorting Genetic Algorithm II (NSGA-II algorithm) and the Non-dominated Sorting Genetic Algorithm III (NSGA-III algorithm).

By combining the multi-objective optimization algorithms with the three-dimensional SN program HYDRA and the Monte Carlo program MCNP, this paper innovatively developed a constraint processing method for optimization parameters and optimization sub-objectives, and applied it to the optimized design of the shielding structures of marine nuclear reactors and space nuclear reactors.[Results] The optimization study shows that the method treating weight and volume as optimization objectives while treating other radiation related sub-objectives as constraints is more suitable for the shielding structure design problem of marine nuclear reactors, and a structure different from the traditional design is obtained for space nuclear reactors with the proposed method.[Conclusion] Regarding the shielding problem of marine reactors, taking the shielding weight and volume as the optimization objectives, and using other radiation-related sub-objectives as constraint conditions is more suitable for the optimization design of marine reactor shielding problems, and can achieve good optimization results.