A hybrid non-dominated sorting genetic algorithm and its application on multi-objective optimal design of nuclear power plant. (February 2017)

• Record Type: Journal Article
• Authors: Chen, Lei; Yan, Changqi; Liao, Yi; Song, Feifei; Jia, Zhen
• Is Part Of: Annals of nuclear energy. Volume 100:Part 2(2017:Feb.)
• Page Start: 150
• Page End: 159
• Publication Date: 2017-02
• Subjects: Multi-objective optimization -- Steam generator -- Primary loop; Nuclear energy -- Periodicals; Nuclear engineering -- Periodicals; 621.4805
• DOI: 10.1016/j.anucene.2016.09.030
• Languages: English
• ISSNs: 0306-4549

Abstract

Highlights: The optimization ability of NSGA-II is improved. The design targets can be obvious optimized through optimization methodology. Multi-objective optimization is implanted into the design of nuclear power plant. Abstract: The design of nuclear component can be optimized by seeking out the best combination of article operational and structural parameters. Through multi-objective optimization, the optimized scheme can not only meets the design requirements, but also satisfies the safety regulations. In this work, a hybrid non-dominated sorting genetic algorithm is proposed, and its performance is verified by comparing it with its prototype and immune memory clone constraint multi-objective algorithm through four test-functions; the designs of the steam generator and the primary loop of Qinshan I nuclear power plant are optimized by the proposed algorithm. The results show that the algorithm outperforms the other two through overall evaluation; the reactor inlet temperature is an important parameter which influences the distribution of the Pareto optimal front; through optimization, the weight of the steam generator can be reduced by 16.5%, and the primary flow-rate can be reduced by 17.0%, the weight of the primary loop can be reduced by 11.4%, and the volume can be reduced by 9.8%.