Research On The Optimization Strategy Of Neutron Shielding Based On Genetic Algorithm

Radiation shielding design is an important part of nuclear engineering design.In addition to taking into account the most critical radiation shielding,the design takes into account the material characteristics,engineering mechanics,geometric structure,device layout and engineering cost required for the construction of nuclear facilities,especially those with special uses,For example,nuclear power reactors for naval vessels,space reactors for aerospace,floating nuclear power plants,and so on,need to consider the volume and weight of radiation shielding systems.Therefore,radiation shielding design is a typical multi-objective optimization problem.The traditional radiation shielding design method is usually based on manual iteration and experience judgment to carry out the reactor radiation shielding optimization design,resulting in the final radiation shielding scheme non-optimal,uncertainty factors,long design cycle and so on.For the multi-objective optimization problem,in order to avoid the human error factor and eliminate the influence of subjective experience caused by the traditional normalized weight factor,a multiobjective optimization design strategy based on non-dominated sorting genetic algorithm is also proposed.Based on the genetic algorithm and radiation transport method,the radiation shielding optimization program is developed,which has the advantages of strong objectivity and small computational complexity.Starting with the neutron shielding problem,the multi-objective optimization strategy of this paper is verified with the American commercial icebreaker Savannah reactor shielding device as the benchmark problem.The optimization strategies based on the non-normalized and the traditional normalized genetic algorithms are compared and analyzed.In this paper,the strategies are used to optimize the structures and the materials of shielding layers,respectively,and the results of convergence data obtained from optimization are compared and verified.The effectiveness and feasibility of the multi-objective optimization strategy based on genetic algorithms in radiation shielding design optimization are demonstrated,which lays a foundation for the development of radiation shielding optimization platform.