摘要： 提出一种基于小生境免疫遗传算法的多级序列优化方法，并解决硅钢片优化排样问题。以免疫算法为基础，通过遗传算法进化抗体群，利用小生境技术保持抗体群的多样性。遗传算子和免疫记忆策略加快了优良个体的产生，提高了算法的收敛速度。共享机制和克隆抑制策略提高了算法的全局搜索能力，有效地避免早熟收敛现象。实际生产数据排样结果表明，该算法是有效、可行的。

关键词: 小生境技术, 免疫算法, 遗传算法, 多级序列优化方法, 硅钢片

Abstract: A multilevel sequential optimization algorithm based on Niche Immune Genetic Algorithm(NIGA) is proposed to solve the problem of optimal layout of silicon steel sheet. NIGA based on immune algorithm employs genetic operators to improve antibodies, and niche technique to increase the population diversity. The combination between genetic operators and the immune memory strategy accelerates the generation of elite. Furthermore, the global searching performance is improved, and premature is avoided because of clone restraint strategy. The algorithm is applied to practical layout problem, and the results show the effectiveness and feasibility of the algorithm.

Key words: niche technique, immune algorithm, genetic algorithm, multilevel sequential optimization algorithm, silicon steel sheet

中图分类号: 

• TP18